Dongwei Su Chinese Stock Markets A Research Handbook 2003

H I N E S E
J) T 0 C K
A R K E T S
A Research Handbook
T>ongwei Su
C) H % j p % S E
S) T 0 C K
M) A R K E T S
This page is intentionally left blank
C) H t JC% S E
Is) T 0 C K
:M) A R K E T S
A Research Handbook
v >
(Dongwei Su
Department of Finance
JiNan University, China
* World Scientific
New Jersey London Singapore Hong Kong
Published by
World Scientific Publishing Co. Pte. Ltd.
5 Toh Tuck Link, Singapore 596224
USA office: Suite 202, 1060 Main Street, River Edge, NJ 07661
UK office: 57 Shelton Street, Covent Garden, London WC2H 9HE
Library of Congress Cataloging-in-Publication Data
Su, Dongwei.
Chinese stock markets : a research handbook / Dongwei Su.
p. cm.
Includes bibliographical references and index.
ISBN 9810245122 (alk. paper)
1. Stock exchanges-China. 2. Stocks-China. 3. Securities-China. I. Title.
HG5782 .S8 2003
332.64'251-dc21 2002033184
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library.
Copyright 2003 by World Scientific Publishing Co. Pte. Ltd.
All rights reserved. This book, or parts thereof, may not be reproduced in any form or by any means,
electronic or mechanical, including photocopying, recording or any information storage and retrieval
system now known or to be invented, without written permission from the Publisher.
For photocopying of material in this volume, please pay a copying fee through the Copyright
Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA. In this case permission to
photocopy is not required from the publisher.
Printed in Singapore by World Scientific Printers (S) Pte Ltd
to Jinyu Su, Yuande Lin and Dongyang Su
V
Preface
The commercialization of state-owned enterprises and development of stock
markets in China during the past decade have been impressive by any
standards. As of the end of October 1999, stock markets in Shenzhen and
Shanghai had 930 companies listed, with a market capitalization of US$340
billion or about 30% of GDP, approximately the same as that of Singapore,
Thailand, Malaysia and Indonesia put together. There are now 38 million
individual stockholders, second in number only to the US. A relatively
complete securities legal framework has taken its initial shape. In addition,
there are 43 H-shares and 42 Red Chips listed in the Hong Kong stock
exchange, with a market capitalization exceeding US$86 billion, or 17% of
the Hong Kong market capitalization. With the return of Hong Kong in
1997, the mainland inherited a world class financial centre, with deep and
mature banking and securities markets. It is increasingly recognized that
stock markets will play a positive role in China's emerging socialist market
economy and create tremendous amount of investment opportunities for
domestic and international investors.
The extraordinary expansion and rapid growth of Chinese stock markets
have been accompanied by some difficulties. Problems include fuzzy prop-
erty rights and legal protection for investors, large initial public offering
(IPO) underpricing, high costs of capital, ineffective corporate governance,
market segmentation and high stock-market return volatility. In particu-
lar, the following questions provide researchers with an interesting area of
research:
(1) Are returns predictable in Chinese stock markets? How well do the
stock markets in China function relative to other developed and
Vll
Vl l l
Preface
emerging markets?
(2) Are Chinese stock markets volatile? Does stock-market return
volatility change over time? How frequent are "big surprises" in
Chinese stock markets? What are the causes for these "big sur-
prises" ?
(3) What explains differences in prices and expected returns between
the classes of shares that can be bought by Chinese citizens and for-
eign investors? Is the time-series variability in the spread between
domestic and foreign share returns consistent with differences in
risk exposures and expected risk premiums?
(4) Why are domestic shares more volatile than foreign shares? How
are the variation in volatility-related expected intensity of informa-
tion flows and the amount of informed trading related to informa-
tion correlates, such as the number of investors, change in profits,
and firm size?
(5) Was the extraordinarily large IPO underpricing caused by bribery
and insider trading, high level of government policy uncertainty, or
government's desire to encourage wide public participation in the
equity markets? What explains the cross-sectional differences in
underpricing? Why the degree of underpricing is much smaller for
foreign-listed IPOs? Why do Chinese IPOs perform much worse in
the long-run than those in other countries?
(6) What are the critical differences in firm characteristics (liquidity,
leverage, and investment opportunities) between firms that issue
new shares after their IPOs and those that do not? Among the
firms that do issue new equities after their IPOs, do agency costs,
political costs and information asymmetry affect their choices of
different financing methods? How can we explain the variation
in abnormal announcement day returns across different financing
methods?
(7) Is accounting information value-relevant in the Chinese stock mar-
kets? Why do investors over-react to earnings release in domestic
markets? To what extent have analysts' earnings forecasts affected
the market valuation of stocks?
(8) What are some of the debates on privatization and reform of state
enterprises? What was China's past experience? How can listed-
companies deal with owner and manager relationships and organize
efficient corporate governance structure? Will China jump into the
Preface IX
sea of private enterprise or remain suspended in a trance-like state
under the notion that market socialism will solve its problems?
Chinese Stock Markets: A Research Handbook is the first to provide a
comprehensive study of the aforementioned issues in a technical manner
drawing upon rigorous theoretical and quantitative analyses. A salient fea-
ture of this handbook is its breadth of coverage. While individual chapters
may be narrow but deep in scope, the handbook as a whole encompasses
diverse subjects and cuts across a number of fields in finance-investments,
financial markets and institutions, international finance, risk management,
corporate finance, and corporate governance. The handbook is intended
as a specially compiled research monograph and a benchmark of reference
which, I hope, will be of considerable use to academia, practitioners, and
policy-makers interested in issues surrounding state enterprise reform and
equity market development in China.
This handbook has nine chapters. Every one except the first two chap-
ters brings together a set of rigorous academic research studies on issues
that enhance the existing body of knowledge in emerging markets. These
chapters typically begin with a discussion of the economics and finance
theory underlying the application, outline issues in econometric implemen-
tation (including measurement issues), and then detail important empirical
findings. Each chapter engages readers in a carefully designed topic in-
volving replication and extension of cutting-edge research in a coherent
framework.
This handbook has benefited from the comments and insights of a large
number of people, particularly my dissertation advisors at Ohio State Uni-
versity. Numerous colleagues and friends have also provided constructive
and challenging suggestions. I owe special thanks to Yuande Lin, who
read each chapter with great care and offered many terrific suggestions.
I take pleasure to thank the following people in numerous ways: Torben
Andersen (Northwestern University), James Ang (Florida State Univer-
sity), Warren Bailey (Cornell University), Prasad Bidakota (Kansas State
University), David Brasington (Tulane University), Zhiwu Chen (Yale Uni-
versity), Gregory Chow (Princeton University), Georgio DeSantis (Univer-
sity of Southern California), Xinghai Fang (Shanghai Stock Exchange),
Yue Fang (University of Oregon), John Fernald (Board of Governors of
the Federal Reserve System), Belton Fleisher (Ohio State University), Yan
He (San Francisco State University), Gary Jefferson (Brandies University),
X Preface
Pok-sang Lam (Ohio State University), David Li (Hong Kong University of
Science &: Technology), J. Houston McCulloch (Ohio State University),
Barry Naughton (University of California at San Diego), Masao Ogaki
(Ohio State University), Min Qi (Kent State University), Bruce Reynolds
(Cornell University and Union College), Jay Ritter (University of Florida),
Alan Viard (Federal Reserve Bank at Dallas), Xin Zhang (China Securi-
ties Regulatory Commission) and Zhenlong Zheng (Xiamen University and
University of California at Los Angeles).
The experience of transforming a manuscript into final book form is
due in large part to the dedication, thoroughness, and highest professional
standards of the publishing team in World Scientific. In particular, it is a
pleasure for me to thank my editor, Joy Quek, for her meticulous work and
great professionalism.
DS
Contents
Preface vii
Chapter 1 Devel opment of Chinese Stock Markets 1
1.1 Introduction 1
1.2 Setting the Scene: Shanghai and Shenzhen in the Early
Reform Period 3
1.3 The Establishment of Secondary Markets 5
1.4 The Participation of International Investors 11
1.5 The Role of Mutual Funds and Other Institutional Investors . . 19
1.6 After the Fifteenth National Congress: Increasing Reversal to
Capitalism 22
1.7 Problems and Dilemmas 26
Chapter 2 Structural and Institutional Characteristics 33
2.1 Introduction 33
2.2 Regulatory Framework 35
2.2.1 Organizations in Charge 35
2.2.2 Securities Laws and Regulations 38
2.2.2.1 P.R.C. Company Law 39
2.2.2.2 P.R.C. Securities Law 41
2.3 Ownership structure 44
2.4 Listing Standards and Procedures 49
2.4.1 Requirements for Stock Listings 49
2.4.1.1 A Shares 49
2.4.1.2 B-Shares 50
xi
xii Contents
2.4.1.3 Overseas'Listing of Foreign Shares 53
2.4.1.4 Red Chip Listing 54
2.4.2 Procedures for Stock Listing 55
2.5 Market Microstructure 58
2.5.1 Registration of Stock Dealers 58
2.5.2 General Rules for Trading 59
2.5.3 Takeover of Listed Companies 64
2.5.4 Settlement, Clearing and Payment System 66
2.5.5 Accounting System 67
2.5.6 Taxes and Fees 70
2.5.7 Suspension and Termination of Listed Companies . . . 72
Chapter 3 Risk, Return and Regulation in Chinese
Stock Markets 75
3.1 Introduction 75
3.2 Stock-Market Return and Volatility Pattern 76
3.3 Day of the Week Effect 82
3.3.1 Analysis of Variance Approach 82
3.3.2 Moving Average Approach 83
3.4 Market Efficiency Hypothesis 86
3.4.1 Random Walk Hypothesis 86
3.4.2 Cointegration-Based Market Efficiency 90
3.5 GARCH Models 93
3.5.1 Model Specification 94
3.5.2 Characterizing Variance in Chinese Stock Markets . . . 94
3.5.2.1 Normal Distribution 95
3.5.2.2 Standardized t-distribution 95
3.5.2.3 Stable Distribution 96
3.5.3 World Versus Local Factors in Volatility 96
3.6 Estimation and Empirical Results 97
3.6.1 Model Comparison 98
3.6.2 Parameter Estimates 102
3.7 Government Regulation and Market Volatility 105
3.8 Volatility Asymmetry and Spill-over I l l
3.8.1 A Partial Adjustment Model with Asymmetries 113
3.8.2 Asymmetric Behavior on Returns and Volatility . . . . 116
3.8.3 Stock-Market Volatility Spill-over Between Mainland
China and Hong Kong 118
Contents xiii
3.9 Summary 121
Chapter 4 Ownership Restrictions and Foreign
Shares Discount 123
4.1 Introduction 123
4.2 Ownership Restrictions in Chinese Stock Markets 125
4.3 A Price Discrimination Model 135
4.3.1 Modeling Ownership Restrictions 135
4.3.2 Investors' Maximization Problems 137
4.3.3 Firms' Maximization Problems 141
4.3.4 Corner Solutions for Domestic Firms' Maximization
Problems 144
4.3.5 Relaxing Ownership Restrictions 146
4.4 An Intertemporal Capital Asset Pricing Model 148
4.4.1 Stock Price and Return Dynamics 148
4.4.2 Intertemporal Maximization by Domestic Investors . . . 149
4.4.3 Intertemporal Maximization by Foreign Investors . . . . 151
4.5 Testable Implications From the Models 154
4.6 Empirical Results 156
4.6.1 Estimating Betas 156
4.6.2 Errors-in-Variables and Tests for B-share Premiums . . 159
4.6.3 Idiosyncratic Variance Effect and Firm Size Effect . . . 164
4.7 Summary 165
Chapter 5 Excess Volatility in Domesti c Share Markets 169
5.2 A Modified Mixture of Distribution Approach 170
5.3 Estimation: Generalized Method of Moments 173
5.4 Data Adjustments 175
5.5 Empirical Results 193
5.6 Cross-Sectional Analysis 200
5.7 Time-Series Analysis 209
5.8 Summary 210
Chapter 6 The Underpricing of Initial Public Offerings 215
6.2 The New-issue and Offering Process 217
6.3 The Role of Financial Variables in the Pricing of IPOs 221
xiv Contents
6.3.1 The Chinese Data 221
6.3.2 A Benchmark Regression 224
6.4 The Adverse-Selection Models 227
6.5 The Signaling Models 235
6.5.1 Empirical Implications from the Signaling Models . . . 235
6.5.2 Correlation Between IPO Returns and IPO Proceeds . . 238
6.5.3 Correlations Among IPO Underpricing and Issuer's In-
trinsic Value, Fractional Ownership and Project Variance 240
6.5.4 IPO Underpricing and SEOs 241
6.6 Bribery and Lottery Hypotheses of IPO Underpricing 246
6.6.1 Bribery Hypothesis: Underpricing of Chinese IPOs are
Gifts to Public Officials or Favored Purchasers 246
6.6.2 Lottery Hypothesis: Lottery Mechanism in Share Allo-
cation Contributes to High IPO Underpricing 249
6.7 Underpricing of Foreign-Share IPOs 251
6.8 Long-run Performance of IPOs 256
6.9 Summary 260
Chapter 7 Corporate Governance and Post-IPO Financing 267
7.1 Shareholders' Behavior and Corporate Governance 268
7.1.1 Organizational Structure and Board Composition . . . . 269
7.1.2 The Role of Government 270
7.1.3 Ownership Concentration and Firms' Performance . . . 274
7.2 Political Costs and Agency Costs of Equity Financing 277
7.2.1 The Failure of Management Responsibility Contract Sys-
tem 278
7.2.2 Cooperative Shareholding System 279
7.3 Choices of Post-IPO Financing 285
7.3.1 Soft Budget Constraints and The Role of Debt 285
7.3.2 Costs of Equity 289
7.4 The Information Content of Different Financing Choices . . . . 295
7.5 Institutional Transformation to Improve Corporate Governance 299
7.5.1 Privatization 299
7.5.2 Denationalization 301
7.5.3 Diversification 302
7.6 Summary 304
Contents xv
Chapter 8 Accounting Information and Stock Performance 307
8.2 Roles of Financial Disclosure for Performance Evaluation . . . 309
8.3 Corporate Disclosures Made by Listed Chinese Companies . . . 312
8.3.1 A Brief History of Accounting Development 312
8.3.2 Financial Reporting Practices 314
8.3.3 Capital Market Infrastructure 318
8.4 Stock Returns Around Earnings Releases 320
8.4.1 Information Content Analysis 321
8.4.2 Earnings Announcements and Stock Price Reactions in
China 322
8.5 Why Domestic Investors Over-react to Earnings Release? . . . 331
8.5.1 Government and Management as Informed Investors . . 331
8.5.2 Investors' Sentiments 332
8.5.3 Market Valuation and Transparency 333
Chapter 9 Internationalization of Chinese Stock Markets 335
9.1 Foreign Investment in Domestic B-Share Market 335
9.1.1 Market Microstructure Issues 336
9.1.2 Risks Involved 338
9.2 Overseas Listing of Chinese Stocks 339
9.2.1 Cross-Listing of H-Shares 340
9.2.2 Red Chips Rising 345
9.2.3 Overseas Listing in the U.S., U.K. and Singapore . . . . 347
9.3 Concerns That Emerged 352
9.3.1 Market Segmentation and Foreign Exchange Control . . 354
9.3.2 A Lack of Foreign Participation and Competition in Fi-
nancial Markets 357
9.4 China Moving Toward World Capital Market: Strategic Issues
and Options 359
9.4.1 Enhancing the Interest of Overseas Investors 359
9.4.2 Improving the Role of Financial Intermediaries 361
9.4.3 Introducing Insurance Capital to Securities Market . . . 363
9.4.4 Strengthening China Investment Funds 365
9.4.5 Reforming Pension System 369
9.5 Liberalization of Capital Movements 370
9.5.1 What the Theory Says 371
9.5.2 China's Experience 373
xvi Contents
9.5.3 Policy Issues 376
Appendi x A Securities Law 381
A.l Introduction 381
A.2 The Role of the Stock Market 383
A.3 Regulating the Chinese Stock Market 384
A.4 Capital Formation 387
A.5 Securities Trading and Information Disclosure 388
A.6 Acquisitions 390
A.7 Market Infrastructure 392
A.8 Investor Protection 395
A.9 Overseas Markets 397
A.10 The Role of Lawyers 398
A.11 Summary 398
Appendi x B Summary of P. R. C. Taxes 401
Appendi x C Major Events in Chinese Stock Markets 405
Bibliography 415
Index 425
Chapter 1
Development of Chinese Stock
Markets
1.1 Introduction
China has experienced unprecedented growth in Gross Domestic Product
(GDP) from 1987 to the present, growing at an average rate of 9.5% per
year (see Figure 1.1). Its economy today is more than four times as large
when the reforms were first introduced in 1979. The reform has affected
all sectors of the economy and has lifted a enormous number of people out
of starvation and into the consumer middle class. The commercialization
of the Chinese economy has also opened it up to foreign direct investment.
The continuous market development has been impressive by any standards.
Prior to 1978, the entire financial system in China was controlled by
state owned banks. Investments were channeled through direct grants
from the state budgetary funds or from government allocated bank credits.
Therefore, China's move towards market economy with modern financial
market was a remarkable breakthrough in terms of ways of raising much
needed capital for economic development.
In 1990, there were no stock markets. As of October 31, 1999, 930 com-
panies were listed in the Shanghai Securities Exchange (SHSE) and Shen-
zhen Securities Exchange (SZSE), with a market capitalization of US$340
billion or about 30% of GDP, approximately the same as that of Singapore,
Thailand, Malaysia and Indonesia put together. There were 38 million
individual stockholders, second in number only to the U.S.. In addition,
there are 43 H-shares and 42 Red Chips listed in the Hong Kong stock
exchange, with a market capitalization exceeding US$86 billion, or 17% of
the Hong Kong market capitalization. With the return of Hong Kong in
1
2
Development of Chinese Stock Markets
Real GDP Growth
Percfn!
1987 :
Fig. 1.1 Real GDP Growth
1997, the mainland inherited a world class financial center, with developed
and mature banking and securities markets. Because China continues to
have a high savings rate of 36% of GDP and a high current account surplus,
much domestic savings have been pumped into domestic capital markets.
Bank assets and securities market capitalization (including equity and bond
markets) accounted for 139% and 41% of GDP in 1999, respectively.
Considerable achievements have also been made in the legal and regula-
tory framework for domestic capital markets. In addition to the enactment
of the banking and central bank legislation, and the PRC Company Law,
the latest piece of law is the PRC Securities Law. Furthermore, private
enterprises are now permitted to go public in China, and more state-owned
corporations are allowed to raise funds in international capital markets.
Because these firms have to adhere to more stringent accounting standards
and disclosure requirements, they usually have better corporate governance
practices and are more responsive to market needs. Needlessly to say, such
Setting the Scene: Shanghai and Shenzhen in the Early Reform Period 3
commercialization and corporatization better prepare Chinese firms to meet
global challenge and competition. It is increasingly recognized that stock
markets will play an important role in China's transition towards a market
economy with modern financial system and create tremendous amount of
investment opportunities for domestic and international investors.
1.2 Setting the Scene: Shanghai and Shenzhen in the Early
Reform Period
The development of the stock markets is one of the most important el-
ements of China's reform in the financial system. Before 1949, Shanghai
was a major banking and financial center with a stock market similar in size
to that of Tokyo and considerably larger than that of Hong Kong. Fluctu-
ations in the Shanghai stock exchange over the period 1919-1949 had a lot
of influence on other world-class financial markets. In fact, Shanghai was
the world's third largest after New York and London.
For approximately three decades after the Communist Party took power
in 1949, the government strictly controlled virtually all channels of invest-
ment. All investments made by enterprises were either from direct grants
from the state budgetary funds or from government allocated bank cred-
its. Shanghai's previous role as an international financial and commercial
center was reduced to a domestic Chinese industrial center. During those
three decades, the entire Chinese financial system was dominated by the
state-owned banks, such as the People's Bank of China (PBOC), a few spe-
cialized banks and their local branches. The official process that allocated
investments across regions and industrial sectors was often bureaucratic,
and in many cases, caused inefficient use of resources.
Before the economic reform was embarked, Shenzhen was just a small
remote town close to Hong Kong in southern China. When China began
its open-door economic policy, Shenzhen became one of four Special Eco-
nomic Zones (SEZs), receiving a lot of preferential policies from the central
government and attracting a consistent flow of investment from Hong Kong
and abroad. After many years of economic development, Shenzhen became
one of the major metropolitan cities in China. Its population is close to
3 million and has one of the highest skilled work forces in China. Both
Shanghai and Shenzhen has established themselves as sound investment
places for domestic and international investors.
4 Development of Chinese Stock Markets
As the structural reform in China continued, the central government
experienced a decline in fiscal revenues. This prompted the Ministry of
Finance (MOF) to sell bonds to cover its budget deficit. The issuance of
long-term treasury bonds in 1981 was the government's first experiment
in raising capital and signified the official re-opening of China's securities
markets. The distribution of treasury bond issues was through mandatory
purchase quotas divided among local governments, enterprises and individ-
uals.
1
Since the early 1980s, China's financial sector experienced major in-
stitutional changes. The PBOC ceased its commercial banking functions
and turned into China's central bank. Four major subsidiaries of PBOC,
namely, the Industrial and Commercial Bank of China, the Agricultural
Bank of China, the Peoples Construction Bank of China, and the Bank of
China were turned into commercial state banks, and a number of urban and
rural credit unions were established as subsidiaries of the four commercial
state banks. As the central bank, the PBOC controls the money supply,
determines interest rate, and handles foreign exchange reserves through
its division, the State Administration of Exchange Control. Commercial
state banks passed on PBOC' s policy and replaced the government budget
funds as the main financial channels for enterprise investments. In addi-
tion, China International Trust and Investment Corporation (CITIC) and
its subsidiaries were established throughout the 1980s and became the ma-
jor consulting firm and investment bank in China. CITIC invested and lent
funds raised from international financial institutions, such as the World
Bank and International Finance Corporation. The whole banking system
was decentralized, and inter-bank competition was encouraged. However,
interest rate remained under government's control. Monetary and credit
policy continued to take the form of a credit plan that was implemented
by PBOC through a set of credit quotas for each commercial state bank.
All levels of government still constantly intervene the daily businesses of
virtually all financial institutions.
During the early reform period, government saving has dropped sharply
in response to deteriorating financial position of state-owned enterprises.
However, household saving has increased very rapidly in response to the
1
Although there was no secondary market to trade treasury bonds at t hat time, the
issuance of bonds paved the way for further fund raising activities by state-owned and
private enterprises, eventually leading to the first publicly issued stock for Beijing's
Tianqiao Department Store in 1984.
The Establishment of Secondary Markets 5
new opportunities created by transition. Individual incomes were higher,
but there was a lack of alternative financial instruments. Individuals were
forced to either hold cash or deposit in the state banks where interests were
controlled at levels lower than the market rate. Naughton (1987) found that
total household saving, including both in-kind and financial saving, jumped
rapidly from 7% of household income in 1978 to 17% in 1982, and have
continued to increase steadily since. Even more crucially, financial saving
tripled, increasing from 2.3% of household income in 1978, to an average of
6.8% in the years 1980-83. As of 1995, households were generating 70% of
domestic saving, over 25% of GDP. There was 2194 billion yuan in deposits
in banks, accounting for 38% of the GDP.
In response to the changes in savings pattern, China's financial reform
deepened as more competition was introduced into the banking system.
In 1981, the central government began to issue treasury bonds to finance
deficits. Shortly thereafter, the authorities permitted the issuance of other
types of bonds-including various provincial and local government bonds
and enterprise bonds. Enterprise bonds were strictly controlled in order to
avoid conflicts with the priorities set in the credit plan. As of the end of
1989, the total issue of securities amounted to 166 billion yuan, of which
99% were bonds.
In 1984, as China began to experiment further economic reform, the
issuance of bonds paved the way for stock issues as alternative methods
of raising capital. The first real step was taken in 1984, when 11 state-
owned enterprises (SOEs) became shareholding corporations. This led to
the first publicly issued stocks by Feile Acoustics in November 1984, with
10,000 shares at 50 renminbi yuan per share. Two months later, Yenchung
Industrial Corporation and Beijing Tianqiao Department Store both issued
shares to the public. However, there was no over-the-counter market in
China and virtually no trading of stocks and bonds before 1986.
1.3 The Establishment of Secondary Markets
The secondary markets for securities started when over-the-counter (OTC)
markets were set up in Shanghai, Beijing, Tianjin, Shenyang, Harbin and
Guangzhou to trade corporate bonds and shares in 1986. At that time, the
markets dealt exclusively with bond trading.
2
The turnover was thin and
the liquidity was low because the public had very little knowledge of the
concept of equity markets. Despite that, Shanghai quickly emerged as the
major bond trading center in the nation. By the end of 1987, eight trading
counters had been opened in Shanghai. Four professional brokerage houses,
namely, the International Securities Corporations, the Haitung Securities
Corporation, the Shenyin Securities Corporation, and the Chenshing Secu-
rities Corporation, were established in late 1980s. They helped government
issue bonds and firms issue stocks.
On December 19, 1990, the first government-approved securities market,
Shanghai Securities Exchange (SHSE), was established to allow investors
and enterprises to participate in stock trading. It adopted a non-profit cor-
porate membership system and dealt with spot transactions, not including
derivative securities. It recruited members nationwide so that it might serve
as a nationwide securities transaction center. Most of the listed companies
on SHSE are based locally in the Shanghai area. The establishment of SHSE
not only marked the beginning of organized securities sales, but also her-
alded the first use of computers in stock and bond trading in China, thereby
putting an end to all paper transactions by offering investors the improved
efficiency of computer-aided investing. Quickly after trading started, SHSE
was doing so well that their was excess demand over supply of securities
which led to dramatic increases in share prices. A few issues skyrocketed
to more than 800 percent times their original issuing price. Public interest
on stocks was, to say the least, piqued.
3
In 1987, the Shenzhen Development bank began to issue shares to the
public. In the next three years, five more issues were floated and actively
traded in the OTC market in Shenzhen. The unique role of Shenzhen
in the Chinese economy led to the formal establishment of the Shenzhen
Securities Exchange (SZSE) in July 1991. Most of the companies listed
on SZSE were based in industrial and commercial cities in inland China.
SZSE was subject to identical regulation and trading procedures as the
SHSE with respect to debt and equity issues. For example, both securities
2
Because of a very small number of shares outstanding in a few corporations, stock
transactions only represented 1%, 2%, and 1% of total securities transactions in 1986,
1987, and 1988, respectively.
3
As early as the closing month of 1990, more t han 400,000 residents in Shanghai had
taken part in securities trading (Fang, 1991).
exchange operate under an auction market environment without specialists
or market-makers. The highest authority of both securities exchanges is
the general meeting of members and the board of directors, which consists
of the chairman, vice-chairman, president and three managing directors.
A supervisory committee oversees the day-to-day trading operation. The
president is the legal representative who controls daily operation of the
exchange.
In 1992, the SHSE and SZSE further standardized their transactions
through a sophisticated over-the-counter computer network known as the
Securities Trading Automated Quotation System (STAQS), based on the
U.S. NASDAQ.
4
Under this network, brokerage houses located throughout
the country could purchase and sell stocks and bonds on behalf of investors,
just like how the NASDAQ market worked in the U.S.. Trading is continu-
ously conducted by an order-driven computerized matching system. Buy or
sell orders are put into the central computer either through the terminals
on the floor or the remote terminals at the broker's office. The bid and
offer prices and other transaction information are shown on every terminal
and transmitted through a satellite system to all the member brokers na-
tionwide.
5
In that year, "share fever" really took hold. The trading volume
increased tenfold and the SHSE index quadrupled.
The business scope for SHSE and SZSE includes providing place and
facilities for trading and managing the settlement and delivery for all listed
stocks. The main duties of the exchanges include: (1) to formulate trading
rules; (2) to publish the trading information; (3) to report business and
management to relevant authorities; (4) to take technical and temporary
measures to suspend or stop the trading; (5) to deal with supervision; and
(6) to supervise listed companies and corporate members.
There are mainly two types of stocks traded in SHSE and SZSE: A-
shares, which are bought and sold only by Chinese investors; and B-shares,
which are restricted to foreign investors.
6
There is a complication in B-
4
The system provided members with real-time price quotations, direct and indirect read-
ing facilities, clearance and settlement, trading information, financial news and other
related services.
5
Real-time quotations from SHSE and SZSE can now be picked up anywhere on the
mainland, even in Tibet. Some 800 listed companies, large and small, can be bought
and sold by Chinese citizens in any of the offices of the 700 brokerage firms t hat dot
China' s towns and cities.
6
While the domestic share market has attracted the interest of more t han a few local
shares trading since foreign exchange is under strict control by the State
Administration of Exchange Controls. Subscriptions, dividends, and cash
distributions to foreign shareholders are paid in foreign currencies (in U.S.
dollar at SHSE and in Hong Kong dollar at SZSE), based on the same-day
official exchange rate. Besides A- and B-shares, a company can issue shares
traded in foreign exchanges (subject to relevant authorities' approval, of
course). The following is a list of China-related equity shares:
A-Shares: They are shares solely available to domestic Chinese
investors. A-shares are denominated in the Chinese currency, ren-
minbi (RMB), subscribed for and traded in RMB by domestic in-
vestors. A-shares can be further classified into four categories ac-
cording to the ownership: state shares (held by the government
or its agencies), legal-person shares (held by other companies or
institutions), employee shares, and individual shares (held by the
public). Currently, shares held by the general public are the only
shares that can be traded on the exchanges, although a few em-
ployee and legal-person shares are traded on STAQ. By the end of
1998, there were 842 listed A-shares in SHSE and SZSE with a total
market capitalization of RMB 1,950.6 billion, equivalent to 24.43%
of GDP;
B-Shares: These securities have exactly the same ownership and
dividend rights as do A-shares but can be purchased (and held)
only by holders of foreign passports (including by those who hold a
Hong Kong passport). The separation of A- and B-share markets
reflects the central government's policy of restricting the foreign
control of vital SOEs and its desire to prevent manipulation of
China's fledgling stock markets from abroad. Maintaining a distinct
stock market for foreign investors in which they participate using
their own currency also helps prevent the devaluation of renminbi
from excessive sales. As of December 1998, there were 103 Chinese
companies selling B-shares worth U.S.$4.37 billion
H-Shares: These are the stocks of SOEs incorporated in mainland
but listed on the Hong Kong Stock Exchange (HKSE). H-share
speculators, state treasury bills represent the most viable investment alternative for the
average risk-averse Chinese investor. As a result, China' s bond markets have developed
rapidly since SHSE and SZSE were established. The annual rate of issuance has risen
dramatically from 19 billion yuan (US$2.3 billion) in 1991 to more t han 190 billion
yuan (US$22.9 billion) in 1996.
issuers are subject to more stringent disclosure rules than A-share
issuers. H-shares are denominated in Hong Kong dollars and can
be traded by any investor. By the end of 1998, 43 firms had issued
H-shares.
Red Chips: Similar to H-shares, red chips are listed on the HKSE.
Red chip issuers are typically companies controlled by mainland
government or SOEs but incorporated in Hong Kong. The Hong
Kong entity is usually a shell corporation of the mainland coun-
terpart and is capitalized through the public offering. It is usually
able to purchase assets located on the mainland through what is
known as an "asset injection". The managers of many red chip
companies have excellent political connections with the mainland
central administration.
L-Shares and N-Shares: These are the stocks of Chinese enterprises
that have chosen to list on either the London Stock Exchange (LSE)
or New York Stock Exchange (NYSE), respectively
7
N-shares are shares in Chinese companies that trade in the form
of American Depository Receipts (ADRs), especially those listed
on the NYSE. These companies are the best managed and most
profitable companies in China.
8
Table 1.1 and Figure 1.2 present ownership structure based on different
share types in Chinese stock markets.
As Tables 1.2, 1.3, and 1.4 and Figures 1.3 and 1.4 indicate, the amount
of capital raised, trading volume, turnover ratio and the total market cap-
italization of the Chinese stock markets have increased significantly over
time. By the end of 1998, there were 842 companies listed in the SHSE
and SZSE with market capitalization of RMB1,950.6 billion, equivalent to
24.43% of GDP. There were 1,011 listed securities including equity shares
(A- and B-shares), securities investment funds, government bonds (for both
spot and repo trading), corporate bonds, and corporate convertible bonds.
China's listed companies had issued a total of 74.61 billion shares in the
7
The first N-share listed on t he NYSE was Brilliance China Automotive Holdings, which
went public in 1992. One L-share exists on t he LSE, and is jointly listed in Hong Kong.
8
Ni ne Chinese companies are listed on the NYSE: Beijing Yanhua Petrochemical Co.;
China Eastern Airlines Corp.; China Southern Airlines; Guangshen Railway Co.; Hua-
neng Power International Co.; Jilin Chemical Industrial Co.; Shandong Huaneng Power
Development; Yanahou Coal Mining Co.; and Shanghai Petrochemical Co.. Another
19 N-shares are traded over-the-counter.
D
e
v
e
l
o
p
m
e
n
t

o
f

C
h
i
n
e
s
e

S
t
o
c
k

M
a
r
k
e
t
s

i

.
s

I

g
s

6
?

6
S

o
o

b
0

C
3

J
-
I

C
D

>

I
S

a

o

S
S

l
O

b
o

c
3

t
*

o

>

c
S

c

o

6
?

l
O

s
?

s
s

P
.

r
s

-
5

1
2

|
8

c

c
3

J
3

C

T
3

C

0
3

7
,

3

I
-

J

-
S

g

o

5

g

&

c
S

t
o

J
3



-
H

O

S

'
u

b
O

b
O

a
)

~

i
n

H
I

2
I

a

g

O
T

b
o

S

S

2

o

.
s

3

_

5

2

>

C
D

X

>
>

c

r
i

n
.
; com
c
a

S

a
ubsta
e
n

b
O

e

o

M
l

G

a

-
5

-

3

O

Q
J

8

*
>

>

W

c

O

t

*
s

o
a

o

a

C
D

C
D

b
O

C

g
t
g

a
S

o

J
?

=

3

C
O

i
/
]

M

3

C
D

C
D

c
t
3

J
2

3

B

<
;

m

b
O

a

o

-
P

e
n

c
t
i

J
2

C
D

i
n

.
,

a

2

t
8

b
O

e
n

-
S

<
"

J
-
e
n

J

<

c
t5

C
Q

s

J
3

C
O

C
D

C
D

>
,

2

9
-
C
O

C
D

C
tJ

-
C

H

The Participation of International Investors 11
A Share
24.06%
Other
Unt radabl e
Shares
3.29%
H Share
4.75%
Legal Person
Shares
28.35%
St a t e Owned
Shares
34. 25%
B Share
5.30%
Fig. 1.2 Share Ownership Structure (Source: CSRC)
markets and had raised a total of RMB 355.31 billion. The total amount of
shares can be broken down as follows: (1) A-share market: 34.302 billion
shares and RMB 168.709 billion; (2) B-Share market: 9.598 billion and
US$ 4.745 billion; and (3) H-share markets: 30.719 billion shares and US$
10.02 billion. There were 662 corporate members, 39.11 million investors
accounts, and more than 2,400 interconnected retail branches and trading
centers. (See Tables 1.2, 1.3).
There are a number of differences between the SHSE and SZSE. Many
companies listed in the SZSE are joint ventures with close relations with
firms in Hong Kong, thus they have better international exposure. By con-
trast, the SHSE is dominated by state-owned enterprises and is more closely
monitored by the central government. With regard to foreign investment,
the SHSE requires that all foreign investors must subscribe for B-shares
through a financial institution approved by the exchange while the SZSE
allows foreign investors to subscribe for B-shares either directly or through
a financial intermediary. Figure 1.5 plots the time-series of composite stock
indices for Shanghai and Shenzhen exchanges.
1.4 The Participation of International Investors
A driving force in China's remarkable advances in securities markets has
been the need to access foreign capital used to fuel the operations of SOEs.
D
e
v
e
l
o
p
m
e
n
t

o
f

C
h
i
n
e
s
e

S
t
o
c
k

M
a
r
k
e
t
s

<

i
n

"
-
2
"
2

5

S
4

-
a

o

3

~

C
O

C

C
O

O

C
D

O

J
3

f
l

C
O

w

O
J

u

c
6

c
o

i

<

<
u

*
-
t

6

-
G

c
o

i

f
f
l

C
D

t
-
i

c

4
3

C
O

1

"
C
T
3

O

,

t
-
H

5

P
Q

'

M

i
-
l

-

^

a
j

a
)

<

m

x

T
h
e

P
a
r
t
i
c
i
p
a
t
i
o
n

o
f

I
n
t
e
r
n
a
t
i
o
n
a
l

I
n
v
e
s
t
o
r
s

1

1

^
-
,

E

C
D

O

i
i

*
w
'
W

(
D

u

V
)

.

C
O

<
H

O

d

a

C
D

C

o

c

f

,

^

a

C
D

O

^
H

C
O

p
^

s

^
w
^

T
J

1
1
1

w

H

<
S
>

f
t
f

i
-
H

t

Pit
<
a

o

D

r

.
.
.

1

i

i

I
T

E

i

D

C
D

C
D

C
D

C
D

C
D

C

D

O

C
D

C
D

C
D

C
D

D

U
D

^
J
-

C
-
>

c
M

C
O

e
n

e
n

t

e
n

e
n

^
-
i

<
Q

e
n

2
i 1994 1995
e
o

e
n

2

C
S
J

e
n

e
n

i
i

i
i

e
n

e
n

i
i

C
D

e
n

e
n

i
-
H

e
n

C
D

e
n

i
-
4

1988 1987
D

D
e
v
e
l
o
p
m
e
n
t

o
f

C
h
i
n
e
s
e

S
t
o
c
k

M
a
r
k
e
t
s

J

r

1

O

O

r
1
ion (
F

<

a

S

H

O

^

X

2

B

r
t

Q

O

H

g

C
D

P
.

^

E

E

u

n

t

I

i

Mark
Turn
Volu

n

n

n
r

,

r

u

r

:

'
"

'
~

'

^
^
H

|
^
^
|

-
r

.

i

C
O

C
I

e
n

C
n

e
n

u
s

e
n

e
n

e
n

e
n

e
n

C
O

C
J
i

e
n

C
M

e
n

e
n

o

o

o

o

o

o

o

C
D

O

O

O

O

O

O

o

o

o

o

o

o

o

I
D

O

I
T
3

O

l
>

O

I
T
3

C
O

C
O

C
M

C
M

-
H

-
H

T
h
e

P
a
r
t
i
c
i
p
a
t
i
o
n

o
f

I
n
t
e
r
n
a
t
i
o
n
a
l

I
n
v
e
s
t
o
r
s

1
5

Table 1.3 Market Statistics
Shenzhen and Shanghai Stock Exchanges
Year Trading Volume Turnover Total Market Capitalization
(in 100 million) (RMB 100 million) (RMB 100 million)
1,048
3,531
3,691
3,474
9,842
17,529
19,505
Source: China Economic Information Network
1992
1993
1994
1995
1996
1997
1998
38
234
1,013
705
2,533
2,561
2,154
681
3,667
8,127
4,036
21,332
30,722
23,544
Stimulated by the Chinese government's endorsement and the activities of
institutional investors, over the past ten years China has attracted more
foreign capital than any other Asian country. Although its legal and reg-
ulatory system still presents barriers to foreign capital investing, China's
planned entry into the World Trade Organization (WTO) and the emer-
gence during 1999 of a fully-fledged Internet boom in mainland China and
Hong Kong are likely to stimulate commercial and legal changes which,
for the first time, will make U.S.-style venture capital investing in China
possible.
9
9
Venture capital is one of the principal sources of finance for companies with high growth
potential. Generally it is not a long-term investment. Essentially, venture capital (held
by capital-rich investors) is injected into an entrepreneur' s business in return for equity.
The start-up is nurtured for a period of time, and, once sufficient returns have been
realized, the venture capitalist exits via a trade sale, a sale to private investors, or an
IPO. In recent years, the consistently high returns achieved by venture capitalists have
at t ract ed an increasingly greater amount of investment funds, expanding t he amount
of money available for investment in entrepreneurial start-ups. The dramatic growth
of venture capital investment in hi-tech industries, first in the United States and more
recently in Europe, has spurred significant interest in venture capital investment in
China and other part s of Asia. Many of the leading U.S. venture capital firms have
opened offices outside the United States to take advantage of emerging foreign markets
The Participation of International Investors 17
Table 1.4 GDP and Stock Market Capitalization
Year
1992
1993
1994
1995
1996
1997
1998
GDP
(RMB 100 million)
26,638
34,634
46,759
58,478
67,885
74,772
79,748
Market
Amount
1,048
3,531
3,691
3,474
9,842
17,529
19,506
As
Capitalization
a Percentage of GDP
3.93
10.2
7.89
5.94
14.5
23.44
24.46
Source: China Securities Regulatory Committee
A major player in the markets for foreign shares is international invest-
ment banks, which trade and underwrite B-, H-, L-, N- and red-chip shares
and arrange specific project finance or raise funds for mainland companies
with global debt issues denominated in foreign currency. However, with
the exception of a handful of international investors who possess significant
sales and trading networks in mainland China, most investors have found
the domestic B-share market to be too immature and thinly traded to war-
rant large scale participation. While a number of international investment
banks maintain a presence in mainland China with a view towards future
development, they prefer to focus their efforts on underwriting sizeable eq-
uity and bond offerings in markets like Hong Kong and New York which are
dominated by sophisticated institutional investors and deep-pocketed indi-
viduals. The sheer market capitalization and liquidity such markets can
boast are vital for selling large deals successfully. Only 14 international
and lower-priced opportunities. It comes as no surprise t hat China, recipient of the
world's second highest amount of foreign direct investment, has at t ract ed the atten-
tion of technology-oriented venture capitalists. The emergence of an Internet boom in
Asia, and China' s expected entry into the WTO are stimulating an additional inflow of
venture capital to China' s hi-tech sector.
investment banks currently participate in underwriting B-shares.
10
For the most part, international investors have chosen to concentrate
their efforts on bringing mainland corporations go public through H-share
or red chip listings. The HKSE, in addition to attracting cash flows from
a local population of sophisticated individual investors, boasts capital liq-
uidity supplied by sizeable domestic and overseas institutions. As a result,
buying shares in Chinese companies is not only a good investment, it's good
insurance. Beijing Enterprises, for instance, is the investment arm of the
Beijing municipal government. It also owns some McDonald's restaurants
in Beijing. In 1997, its not-yet-issued H-shares were over-subscribed by a
factor of 1200. In fact, the issue had attracted investment capital of about
Hong Kong Dollar 200 billion, or about twice the money supply in Hong
Kong. People had withdrawn so much cash as a consequence of the issue,
that Hong Kong banks had asked Beijing Enterprises not to cash the checks
it had received-at least not until the banks could deal with their shortage
of vault cash.
Competition among investment banks to lead Hong Kong deals is fierce,
as they battle among each other to gain market share and establish them-
selves as globally recognized leaders in China-related equity issuance. Nev-
ertheless, share issues have been awarded fairly evenly among investment
bankers, as the relevant Chinese authorities have tried to maintain a stable
of investment banks with which to do business. This policy of encouraging
balanced opportunities for as many international investors as possible has
worked well to spread financial risk.
In the minds of the foreign creditors or investors acquiring equity stakes
on mainland Chinese companies, there are a number of important concerns.
One is the degree of institutionalization of the Chinese equity and bond
10
In addition to underwriting, one of the areas of interest by international investment
banks in recent years has been in advising multinationals on their investments within
the mainland China. With increasing frequency, foreign corporations entering the Chi-
nese market have been seeking to form joint ventures with local businesses rather t han
setting up their own operations. This activity has warranted the advisory services of
investment banks with in-house expertise in providing investment analysis and corpo-
rate structuring. Given the operating inefficiencies of Chinese state enterprise sector
and its need for significant restructuring and consolidation, international investors
have been anticipating a wave of merger and acquisitions. A slice of this potentially
lucrative pie could supply a needed revenue stream for investment banks with offices
in the mainland, as well as introducing them to a stable of newly revamped corporate
entities looking for help raising funds in the capital markets.
The Role of Mutual Funds and Other Institutional Investors 19
markets, which determines the range of investment banking services. An-
other is the extent of liquidity flows and the level of investor sophistication.
The most vulnerable sector in this respect is domestic asset management,
where foreign investors yet to capture the investing dollars of the Chinese
population through vehicles similar to U.S. mutual funds. Mutual funds
will likely bring stability to the market by introducing the philosophy of
long-term investing to the public at large, thereby reducing the influence
of short-term speculation that widely characterizes the market.
International investors also wish that Chinese government would open
bond markets to further underwriting business, issuing debt for private
enterprises in both the domestic and global arenas. The most obvious
regulatory problem in this regard is the convertibility of the renminbi. As
some economists have argued, making the renminbi fully convertible in the
capital account will result in a merger of A- and B-shares, thereby leveling
the playing field for participation by all local and overseas investors.
In addition to these equity and bond market openings, international in-
vestment professionals will love to develop businesses for the more sophisti-
cated financial services such as currency hedging and financial derivatives.
The demand for these capabilities within China also seems pinned to cur-
rency reform, as only when the renminbi is fully convertible and allowed
to float freely will companies need to worry about hedging their foreign
currency risk. Nonetheless, this promises to be a profitable sector of the
investment industry for international investors, as only they have the risk
management systems and expertise to manage these types of contracts on
a large scale.
1.5 The Rol e of Mutual Funds and Other Institutional In-
vestors
Institutional investors are financial institutions, such as pension funds, life
and causality insurance companies, and mutual funds, that invest money
on behalf of others.
The role of institutional investors in financial markets is likely to be-
come more important in the coming decade. For this reason, institutional
investors will increasingly exert an impact on the functioning of financial
markets. The inter-relations between institutional investors and the func-
tioning of financial markets are multi-dimensional and complex, encompass-
ing a number of distinct policy issues.
The increasing importance of institutional investors means that their
impact on the functioning of financial markets is steadily growing. Institu-
tional investors have been growing in size dramatically over the past decade.
In most OECD (Organization for Economic Cooperation and Development)
countries, institutional saving institutions play a key role in domestic finan-
cial systems. For example, U.S. pension funds as a group control over $5
trillion in assets. Using a sample of OECD countries, Merton (1995) finds
that institutional savings institutions represent, on average, 16% of total
financial assets in those countries.
Three basic activities can be distinguished to define the growing role of
institutional investors on financial markets. There are: (1) Institutional in-
vestors are collectors of savings from households. (2) Institutional investors
are suppliers of funds in the market for securities and other financial as-
sets. (3) Institutional investors are participants in primary and secondary
markets for bonds, equities, foreign exchanges, and financial derivatives.
The involvement of institutional investors in capital market transactions
is increasing in tandem with their growing financial clout. A strong commu-
nity of institutional investors seems to be a recondition for the development
of liquid securities markets with sophisticated financial vehicles. In view of
the growing influence that institutional investors exert on the structure of
capital markets, it is generally recognized that policy-makers need to take a
closer look at both the functioning and the regulation of these institutions.
The regulation includes the functioning of the fund management profession,
the impact of changes in demography of pension schemes on the efficiency
of the allocation of savings, and the influence of the international portfolio
diversification.
The participation of institutional investors in the Chinese stock mar-
kets began in 1992 when the Zi Bo Township Enterprise Investment Fund
was established as the first company-style close-end fund. In the same
year, Shenzhen Investment Fund Regulatory Provisions was enacted as the
symbol of standardization of China's mutual fund markets. The Shanghai
government formulated Managerial Methods of Investment Trust Fund in
Shanghai. In August 1993, the SHSE published the Experimental Meth-
ods of Fund Stock Listing on the Shanghai Stock Exchange. In 1994, Seng
Yian and Nan Fang Securities Exchange centers were connected with the
SHSE and SZSE. This led to the formation of two nation-wide mutual fund
exchange markets in Shanghai and Shenzhen. At Shanghai's fund mar-
The Role of Mutual Funds and Other Institutional Investors 21
ket, there were one listed fund (i.e., Zi Bo Fund) and eleven net listed
funds, among which eight were Seng Yan's funds and three are Guang-
dong's funds, with a total market value of 662 million renminbi yuan. The
total trading volume was 11,740 million renminbi yuan and 30,567 million
renminbi yuan in 1994 and 1995, respectively. At Shenzhen's fund market,
there were ten listed funds with a total market value of 2,050 million ren-
minbi yuan. The total trading volume was 24,015 million renminbi yuan
and 30,452 million renminbi yuan in 1994 and 1995, respectively. Looking
at the trading volume and market performance, mutual fund markets in
Shanghai and Shenzhen had become an important and integrated part of
China's domestic equity markets.
To establish a mutual fund, an application needs to be filed with the
PBOC, which is required to review the application within 90 days from the
day it receives all required documents, and approve the application with the
State Council's relevant division. The SZSE stipulates that fund managers
and trustees shall obey the following rules:
cannot take the business of securities underwriting;
cannot take the business of trust loan or provide loan guarantee;
cannot make investment with unlimited liability;
cannot participate in trust trading;
cannot trade securities with the funds from the same manager;
cannot invest on beneficiary receipts of the other funds;
cannot invest in company stocks which have relations to the fund
managers or trustees;
the investment in a listed stock shall be no more than 10% of net
value of the fund;
the investment in a listed stock shall be no more than 10% of the
total shares of the company; and
other investment limitation requirements ruled by government or-
gans in charge.
Because mutual funds in China are listed and traded in the securities
market, they have to meet the information disclosure requirements on se-
curities. The requirements on listing documents and public announcement,
the publication of the fund's net capital, and the semi-annual and annual
reports are similar to that of a listed stock. The income of a fund mainly
includes share dividend, interest, and capital appreciation. There is no tax
imposed on fund investment in China. However, investors have to pay 1%
of managerial fee based on the market value of their investment. In the
SHSE and SZSE, the service fee for trading one fund unit is one renminbi
yuan, the commission fee is 0.4% of the trading value, and the settlement
fee is 0.1% of the trading value.
1.6 After the Fifteenth National Congress: Increasing Re-
versal to Capitalism
National congresses are among the most important political events in China.
The 15th National Congress of the Communist Party of China (CPC), held
in September 1997, saw the Jiang Zemin administration take new steps in
preparation for the coming 21st century. A number of historical events, es-
pecially the death of Deng Xiaoping and the transfer of sovereignty of Hong
Kong from the Great Britain to China, kept the Chinese people focused on
the a number of crucial issues. The 15th National Congress formally des-
ignated Deng Xiaoping Theory as the Communist Party' s guiding thought
and pushed the cause of building socialism with Chinese characteristics
comprehensively to the 21st century.
In his report to the Party' s 15th National Congress, President Jiang
Zemin made a comprehensive and profound exposition on China's economic
and political problems and set forth the task of deepening economic and
political reforms, with emphasis on balancing reform, development and sta-
bility for China's economy.
During the 15th National Congress, Premier Li Peng proposed the fol-
lowing plan to restructure China's SOEs, nearly 70 percent of which are
profit-losing ones:
Taking reform of SOEs as the central point in economic restructur-
ing, the government will take practical measures to change the way
enterprises operate, with the establishment of a modern enterprise
system as its orientation.
Guiding enterprises according to their type and aiming at improv-
ing the state sector of the economy as a whole, the government will
carry out a strategic reorganization of SOEs by learning to manage
large enterprises well while relaxing control over small ones.
The government will explore and developing various forms of public
ownership of firms.
After the Fifteenth National Congress: Increasing Reversal to Capitalism 23
The government will reform SOEs through internal reorganization
and the establishment of more effective corporate governance struc-
ture.
The government will encourage mergers, standardize bankruptcy
procedures, retrain laid-off workers, and increase government effi-
ciency by reducing staff employees.
The government will promote a social security system.
Since September 1997, China has been busy carrying out strategic cor-
poratization and reorganization of SOEs with full steam. This was mani-
fested in the adoption of a major plan for reorganizing the State Council
in March 1998 and the acceleration of SOE restructuring. Under Li Peng's
program, all but 3,000 SOEs will be corporatized in an effort reduce the
over-sized state sector. China will also focus on strengthening approxi-
mately 1000 strategic and key SOEs that are critical to the country's eco-
nomic interests. Efforts have been made to support the development of
hundreds of major enterprise groups that are pivotal to the national econ-
omy. This strategy will likely entail bankruptcies, mergers and acquisitions,
and public stock offerings, an ownership diversification process known as
privatization but usually termed corporatization in China.
The vast majority of China's 300,000 SOEs are small, inefficient, and
severe threat to nation's economy. They are typically unprofitable and
have been subsidized by the state banks for years. As a result, the banking
system is crippled by a gridlock of non-performing loans to unprofitable
state enterprises, resulting in capital starvation among more viable state
and non-state firms. The strategic corporatization and reorganization of
SOEs will not weaken the state sector, but rather concentrate limited state
funds in the areas of real need. Reducing government control over thousands
of enterprises in China will help to create a more efficient and modern
financial system. This in turn will help to develop a more competitive
economy, bolster fledgling capital markets and speed China's integration
into the global economy.
In revamping the ownership structure of SOEs and achieving success
in enterprise reform, the government has been developing a flexible and
market-driven financial system to ensure that capital resources are chan-
neled toward efficient uses, and away from hopeless causes backed by polit-
ical clout. Liberalizing interest rates, establishing full currency convertibil-
ity, and opening the securities industry to private investors-both domestic
and foreign-are among the strategies proposed to spur the development
of more efficient capital markets for the 21st century. (See Figure 1.6.)
A deregulated financial system would enable Chinese firms to compete ac-
cording to their efficiency, and price and quality of their goods and services.
Such a market-based environment would help ease trade friction between
China and other countries and pave the way for its accession to the WTO.
Many experts have suggested that for those SOEs that play a controlling
role in economic fields touching upon national security and social stability,
they need to maintain a dominant role in infrastructure and vital basic
industries. In the backbone industries, such as machinery, electronics and
petrochemicals, a group of large conglomerates should be supported to
promote the competitiveness of China's economy in the world economy. The
re-shuffle of the state sector does not mean selling off all the SOEs because
it could lead to the loss of state assets and encroachment of employees'
interests.
11
This tendency has somewhat been curbed under the guidance
of policies adopted by the central government. As long as a complete legal
system and its effective enforcement are in place, the strategic re-shuffle
proposed by the Fifteen National Congress will not weaken the state sector,
but rather concentrate limited state funds in the areas of real need.
To step up the reform of the financial system to meet the needs of
the establishment of a socialist market economic system, the core are to
strengthen the currency regulation function of the central bank and the
financial supervisory system, to promote the commercialization of state-
owned banks, to improve the capital structure and quality of the commercial
banks, to vigorously but steadily develop and regulate the financial market,
to improve and enact financial laws and regulations, and to learn modem
financial management expertise to improve corporate governance.
In the spirit of the Fifteenth National Congress, China launched a series
of reform measures in order to further standardize and develop the securities
market and to adapt to economic globalization and internationalization of
the securities market.
Reforming the stock-issuing system. Enterprises are free from the
restriction of the form of ownership in the issuing and listing of
11
Plant closures have begun throughout China. Wang Zhongyu, Minister of the State
Economic and Trade Commission, said t hat only the strong would survive. China had
some 100,000 jobs t hat will be lost as the result of closing all or most of the weak
companies. Wang said t hat another worry was the corruption at the managerial level.
Managers often failed to pay workers and absconded with a company' s assets.
A
f
t
e
r

t
h
e

F
i
f
t
e
e
n
t
h

N
a
t
i
o
n
a
l

C
o
n
g
r
e
s
s
:

I
n
c
r
e
a
s
i
n
g

R
e
v
e
r
s
a
l

t
o

C
a
p
i
t
a
l
i
s
m

2
5

C
O

"
J

<
D

Q
.

C
O

I
s

p

X
3

Q
.

3

C
O

-

=

t
o
.
2

C
O

_

m

+
a

c
o

b

u
i

c

2

c
n
^
-
5

=

a
>

4
-
.

c
o

T
3

'
o

_

.
^
z

T
J

-
.

(
J

r
o

o

C
O

*

_
>
*

_

C
O

<
j

E

t

c
o

Q
_

o

K
-

>
H

{
=
!

0
)

.
Q

E

=
3

a

C
O

D
l

C
O

4

C
O

C
D

3

<
D

t
J
T
3

Z
5

C
O

g

U

O

F

d
>

^

4
1

O

C
O

o

O

C
O

c
l
^
T

u

8
1

C
O

o

C
O

u

c
:

>

C
L

Q
.

s

c
o

o

>
2

E

4
-
1

o

o

C
O

^

Q
.

4
-
J

C
O

4
1

G
O

4
Z

C
O

o

0
>
-
0

Q
.

o

E

U

i

a
.

Q
)

C
O

C
O

o

'4
1

C
O

<
.
i
=

o

C
O

C
O

C
O

o

C
O

T
3

"
C

-
C

t
t

C
O

*
-

0
)

O

U

=

H
-

c

(
S
3

4
_
;

C
O

S
Z

o

C

+
J

5
.

*
-

C
O

o

i
;
r

i
-
o
-
t
:

=
J

E

=
3

S

5

U

0
>

C
O

C
O

*
*

o

E

o

4

0
)

-
=

C
O

N

;
2

o

t

B

.

E

C
O

C
O

"
O

>
,

c

5
^

-
o

<
D

O

<
=

Q
.

C

x
:

.
E

4
1

I
-
C
O

H
E
-
E

&
o

E
2

^

C
L

-
Q

c
o

r
o

t
S

c
o

S

5

C
O

(
J

M

o

u

C
L

C
O

I

C
L

o

C
L

Z

o

X

N

Z
)

1

^

(
)

4
1

a

(
-
C
O

L
T
D

4
-
J

4

o

i

o

c

^

C
O

o

'
-
,

T
;

C
O
O

stocks. SOEs, collective enterprises and joint ventures can all apply
for the issuance and listing of stocks.
Encouraging online securities commission deals. The China Secu-
rities Regulatory Commission (CSRC) published Temporary Man-
agement Method for Online Securities Commission Deals and Ap-
proval Procedures for Securities Companies Online Commission Deals
in the first half of this year, thus providing necessary conditions for
the standardized development of online commission deals. As a re-
sult, over 40 securities companies have submitted their applications
for online commission deals to the CSRC.
Establishing the Growth Enterprise Market (GEM). This project
helps to create a favorable environment for direct finance by hi-
tech enterprises and growing medium and small enterprises as well
as provide investment opportunities for venture capital. In the
meanwhile, related preparatory work concerning the formulation of
regulations and systematic development of GEM is being stepped
up.
Fostering institutional investors. China will carry out pilot activ-
ities of sino-foreign joint venture mutual funds. In addition, the
CSRC and the China Insurance Regulatory Commission (CIRC)
will cooperate in establishing insurance and pension funds by the
pilot mutual fund companies. If successful, foreign mutual funds
can have a chance to enter China's securities market.
In short, there has been increasing reversal to capitalism after the Fif-
teenth National Congress. Securities markets will play an important role in
China's transformation into a market economy with modern capital mar-
kets.
1.7 Problems and Dilemmas
So far China's financial and economic reform has been successful, espe-
cially in comparison with Russia and other former Soviet Union countries.
However, a number of problems are still outstanding:
First of all, The economic importance of SOEs is waning. Many SOEs
cannot run in full capacity due to an over-supply of the products and repet-
itive construction, a result of government interference in the many years of
Problems and Dilemmas 27
the past. Moreover, many enterprises are overburdened by debts, as they
blindly sought to develop big and comprehensive projects without adequate
funding; the banks were forced to issue excessive amount of currency and
grant loans. In addition, many enterprises are overstaffed under a big em-
ployment pressure. Local governments often pressed enterprises to recruit
more people than they needed. As a consequence, SOEs now contribute
just a third of the industrial output (against 75 percent in 1981). It is in-
creasingly difficult for the central government to cover the loss of SOEs and
maintain a bloated employment. In 1996 the number of people laid off by
SOEs increased sharply to 10 million. In 1986, SOEs produced two-thirds
of all manufacturing goods and accounted for four-fifths of export growth.
In 1996, propped up with subsidies and hiding behind tariffs and other
protection, SOEs can still account for half of all manufacturing goods, but
barely a fifth of the growth in exports. Losses in the state sector now ex-
ceed profits. About 70 percent of the 100,000 state firms are money-losing
enterprises. The state sector has imposed huge costs on the rest of the
economy.
Second, Chinese stock market is overwhelmed with individual investors,
most with limited understanding of financial risk and investment, but is lack
of long-term institutional investors such as mutual funds, pension funds and
insurance companies. In stark contrast with world-class financial markets,
there are about 45 million individual investors but only 23 investment funds
in China by the end of 1999. Two important decisions were made in August
1999. The first was to allow state-owned enterprises to invest in the stock
markets and if they held stocks for a period of at least six months. The
second was to allow life insurance companies to allocate up to 15 percent
of their assets portfolio for investment in the stock market. However, it
is yet to be seen whether the trend of encouraging the establishment of
institutional investor will persist.
Third, although it is recognized that foreign capital is an important
source of funds for domestic firms, foreign investors are not allowed to invest
in domestic equity markets. Foreign investors in B- and H-share markets
tend to be institutional investors, who constitute yet another set of long-
term investors. It is crucial not to turn them away. H-shares are traded
in Hong Kong and are subject to the fluctuation of international financial
markets. They are regulated by the Hong Kong authorities who have a
long track record that is up to international standards. But B-shares are
regulated by mainland Chinese authorities who yet need to establish a track
record. H-shares are reasonably liquid while B-shares are plagued with
illiquidity problem. In addition, the pricing of H-shares still lags behind
their counterpart A-shares. The pricing of B-shares is even wider apart from
their counterpart A-shares. An appropriate difference between the prices
of the stocks of the same quality in different markets is understandable,
but a wide gap is obviously abnormal. Current A-stock prices are two to
six times the price of equivalent H-shares. And the price of A-shares are
around 2.5 to three times the price of equivalent B-shares. When the A-
share stock prices are so much higher than the B- or H-share prices for the
same company, the market price is obviously too high for A-shares.
Fourth, non-tradable shares (state and legal person shares) have become
the rot at the heart of China's stock markets. They put the characteristic
in that slogan "stock markets with Chinese characteristics". They enable
the government to claim that share issuance is not akin to privatization.
This strategy has been useful, but the problem is that the state is that legal
person and state shares afflict the markets in different ways. Legal person
shares are held by approved state institutions, usually firms under the con-
trol of the same local government. They typically hold about 30 percent of
a company. They receive dividends and tend to be active shareholders. A
World Bank study in 1997 found that legal person shareholders are often
elected to their companies' board of directors. Through that, legal person
shareholders gained voting power on major decisions, as well as inside in-
formation on the firm. To crown it all, legal person shareholders face little
threat of a hostile takeover. That puts them in a powerful position. How-
ever, legal-person shareholders often abuse their powers. They frequently
"borrow" assets, usually cash, from their listed companies, and it's rare for
firms to be paid back. Whatever harm it may do the company, small share-
holders can only watch and weep. Of the 841 companies listed at end of
1998, 467 of them had their funds used by their majority shareholder. State
shareholders, in contrast, often don't have enough power. State shares are
usually managed by the State Asset Management Bureau (SAMB), from its
headquarters in Beijing or local offices. Their ultimate owner is the State
Council. The SAMB passes dividends onto the Ministry of Finance, for
inclusion in the national budget. It therefore lacks an important monetary
incentive to supervise the firms it owns.
Fifth, Chinese stock market is much more volatile than matured mar-
kets. For example, the Shanghai composite index is about 800 times as
volatile as the NYSE composite index, and 400 times more volatile than
Problems and Dilemmas
29
the Hong Kong Hang Seng index. There are two possible explanations for
the high stock-market volatility in China. One is that the fundamentals
for Chinese stocks are more volatile than industrialized countries. For in-
stance, China's monetary policy is more unpredictable and subjective than
that of the U.S. monetary authority. It is common to see the Chinese cen-
tral bank loosens the money supply, the economy over-reacts to easy credit,
and the central bank suddenly intervenes and tightens the credit. On the
contrary, the U.S. Federal Reserve (as well as the monetary authorities in
Germany and Japan) typically announces its monetary policy changes well
in advance. In this sense, the macroeconomic policy is more volatile in
China.
Another reason for high stock-market volatility is the strong presence of
noise traders. When there is more news (including gossips) flows into the
market, there is more extreme movements in stock prices. There is more
suspicion of insider trading and other securities violations. People often fail
to incorporate a company's fundamentals into stock prices. Their trading
is more likely to be based on mood swings and things that they thought to
be informative but are not information in reality.
Sixth, the development of financial instruments has been limited to
the capital market. No nationally integrated money market has yet been
developed, and most banks lack the skills to develop new products that
could create competition in the financial sector. Although local inter-bank
centers have been emerged since 1986 and have played a useful role in
the local redistribution of surplus funds, they have not operated as inter-
bank markets in the traditional sense. Since non-bank financial institutions
and even some manufacturing enterprises can participate, these centers
often serve as long-term financing channels for non-bank and non-financial
institutions, thereby circumventing the credit plan.
The absence of nationally organized money markets-one of the salient
features of China's financial sector in the early 1990s-is related to its slow
start on interest rate liberalization, its lack of a modern payment and settle-
ment system, and the state banks' inefficient administrative organization.
The lack of nationally integrated inter-bank and money markets has, in
turn, hampered efforts to achieve the transition to indirect instruments of
monetary policy. In the absence of an inter-bank market where surplus
funds could be redistributed, the existing indirect instruments have been
used to channel liquidity from one region or entity to another. The author-
ities have regulated liquidity by adjusting reserve requirements and using
the PBOC' s lending facilities for banks while maintaining relatively stable
interest rates.
The present economic and financial problems require, therefore, further
active steps in reforming the entire banking system and the state-owned
enterprises. There are more than 300,000 SOEs, a third of which are in-
dustrial. Most of SOEs employ urban workers. They also account for up
to 90 percent of all loans granted by state banks. However, SOEs produce
less than 40 percent of the nation's industrial output, and their share has
been falling relative to firms of other types of ownership.
12
Despite some progress in the development of property rights in China,
it is still rather difficult to draw the line between the state as owner and
as manager. The latter still has been delegated excessive control from the
government and the party. The absence of clear definitions of ownership
has also been an obstacle to foreign investment. The obscure separation of
ownership and management has led to a process of "spontaneous privatiza-
tion" which is taking place in many small state firms, a growing number of
which are being stripped bare by managers and local government officials
who seize assets and leave debts behind.
To reduce spontaneous privatization, the central government is deter-
mined to keep the 1,000 or so largest SOEs of vital national interest under
its control. This is sensible for the time being, but will be costly in the
long-run. The enormous increase in state bank lending (the so called "pol-
icy loans") to support the survival of these SOEs represents the price of
delaying serious state enterprise reform.
13
Bad debts have amounted 22
percent of the total credit extended by the four major state banks and up
to 80 percent of the portfolios of other commercial banks. In short, China's
banking system is almost insolvent, i.e., its bad debts exceed its capital.
14
12
Section 2.3 provides more detailed analysis on ownership structure and firm's perfor-
mance. Figure 2.1 presents shares of industrial output according to different types of
ownership of firms.
13
Thi s is also referred to as "soft budget constraint". A firm is said to have soft bud-
get constraint if it expects to be bailed out in case of financial trouble. Soft-budget
constraint can lead to failure t o abide financial principles by t he management and is
therefore a source of inefficiency.
1 4
The t rut h is t hat state-owned industries owe the state banking system a great deal
of money due to the soft budget constraint. About US$600 billion in bank loans are
outstanding in China. (90 percent of the loans are given to SOEs.) These loans account
for an unusually high proportion of all financing equivalent to about 70 percent of the
GDP. China has only one private bank and its capital markets are immature, so t he
Problems and Dilemmas 31
Over the long-term, the costs of the financial system for the state sector
will surely mount. Lending is growing faster than state output while bank
profits shrink steadily as a proportion of assets.
In summary, after years of reform, no other emerging market economy
faces greater challenges than the mainland Chinese economy. Ultimately,
the future of China's capital markets does not rely on the success or failure
of one specific reform. Rather, it depends on a host of policies yet to be
implemented by the central government. With an economy that is increas-
ingly linked to the globe (external trade has accounted for approximately
38% of GDP in 1999) and on the brink of joining the WTO, China must
rapidly reform its state enterprises and banking sectors to enable these to
compete with foreign firms on a global footing. State-owned enterprises
still accounts for roughly one-third of GDP and employment. State-owned
banks also account for 75% of total banking assets. What are urgent now
are to set up a new investment mechanism in which the roles of the govern-
ment and the enterprises are separated, to reform and improve the financial
system to make the cost of capital small, and to establish and improve the
social security system.
job of financing investment rests almost entirely on the shoulders of state banks. Those
banks also have an unusually high proportion of bad loans (about 30 percent of the
GDP) which they have small hope of ever having repaid. Bad loans in China would
swallow up banks' capital several times over if banks had to account for them. By any
conventional calculation the banking system is insolvent.
Chapter 2
Structural and Institutional
Characteristics
2.1 Introduction
The most salient feature of China's securities market is the emphasis on
state control of the financial markets. When stock exchanges were first
established in Shanghai and Shenzhen, legislation on securities came under
the jurisdiction of a number of ministries and commissions under the State
Council such as the China Securities Regulatory Commission (CSRC), the
Ministry of Finance (MOF), the People's Bank of China (PBOC) and the
State Asset Management Bureau (SAMB). In addition, the Shanghai and
Shenzhen municipal governments, the local branches of PBOC, and the
local divisions of the various ministries and commissions were also involved
in exercising the regulatory, supervisory and administrative functions for
the securities markets in Shanghai and Shenzhen. During the early stage
of development of securities markets, there was not only an overlapping of
legislative power but also a lack of demarcation of the administrative and
supervisory functions at different levels and in different departments.
Since 1992, the CSRC supervises and regulates securities markets in
China. The PBOC acts as the approval authority for traders and securities
firms while the State Council Securities Policy Commission determines the
annual number of listing and volume of issues. The state planning com-
mission sets quotas for stock and debt listings during a given year. These
quotas are then divided among provinces. In the bond quota allocation,
preference is given to infrastructure projects and 300 designated SOEs.
Announced quotas appear to change according to market conditions. Ad-
ditional regulatory constraints include outlawing share issues for financial
33
34 Structural and Institutional Characteristics
and real estate companies and forbidding mutual holdings between securi-
ties institutions and non-securities financial institutions.
Enterprises requesting a foreign share listing are required to obtain ap-
proval from the Ministry of Foreign Trade and Economic Cooperation. In-
dividuals are limited to a maximum of 25% of a firms B-Shares and total
foreign ownership is limited to 49% of public shares. Foreign firms may
establish representative offices but not local securities branches or sub-
sidiaries. All transactions involving foreign investors require the partner-
ship of a local broker.
Foreign banks are prohibited from selling mutual funds to Chinese cit-
izens or institutions and must obtain government approval to participate
in underwriting. Foreign securities firms are prohibited from introducing
new domestic financial products, although they may offer off-shore deriva-
tives. However, repatriation of profits requires government approval. Fur-
thermore, Chinese residents may not invest overseas without government
approval.
It is widely recognized that a well-functioning securities market can pro-
vide an opportunity to raise capital for investments, enhance the efficiency
of state assets and social resources, reduce the debt-servicing burden, mobi-
lize savings deposits through facilitating an increased availability of assets,
and create a higher yield with a lower risk, hence a higher return on savings.
In contrast, central planning does not provide the most efficient means to
channel funds from savers to borrowers. Major structural problems of the
emerging Chinese security markets include its artificial pricing mechanism,
the limited number of listed securities and a lack of modern communication
facilities and training.
Arbitrary Pricing: Arbitrary government intervention can lead to
mispricing of capital assets. China currently operates under a dual
interest rate system, with an official rate and a market rate. When
rates are high, state enterprises experience higher cost of capital.
State-ownership can be somewhat of a bottleneck to stock market
development because the use of funds raised through issuing stocks
remains subject to government control. Furthermore, the govern-
ment has self-imposed ill-liquidity through prohibiting employees
to sell employee shares. Limited domestic mobility of capital also
constricts market development.
Limited Number of Listed Securities: The securities markets pro-
Regulatory Framework 35
vide an opportunity to raise capital through both promoters sub-
scription and public flotation. Without securities markets, firms
have to rely on retained earnings. However, experimentation with
stock issues in China has remained limited and the amount of funds
raised has been small.
Over Dependence on Household Savings: Although China benefits
from a high savings rate, one of the highest in the world, the Chi-
nese banking system have become overly dependent on household
savings. (Household savings are estimated to account for 60% of
deposits.)
Lack of External Capital: The availability of external capital is an
important element in creating a viable securities market. External
capital is a means to supplement but not replace internal source
of funding. Foreign exchange restrictions can inhibit market devel-
opment. The current exchange controls have led to an over-valued
renminbi and the a black market for foreign currency that further
thwarts the ability of the government to effectively implement strict
monetary policies.
These weaknesses in the current system lead to increases in cost of cap-
ital for Chinese firms which, in turn, makes it more difficult for them to
compete with foreign firms. One could argue that Chinas has used its secu-
rity market more as a means to shift funding of inefficient state enterprises
from the government to the public than as a means to fairly value com-
panies. Under the present heavy regulation and strict control of financial
markets, China has the potential to benefit from a further opening of mar-
kets. China's financial reform needs to include a greater use of interest
rates as an opportunity cost to determine cost of capital.
2.2 Regulatory Framework
2.2.1 Organizations in Charge
According to Managerial Regulation of the People's Bank in P. R. China
announced in 1986 and other relevant regulations, the security organs in
charge at Shanghai and Shenzhen during the period of the second half of
1980s to the beginning of 1990s are PBOC Shanghai Branch and Shenzhen
Branch. Since the beginning of 1990s, because of the security regulation
was increasingly complex, the State Planning Commission, the State Re-
construction Committee and the Department of Finance are involved in the
security management.
In 1992, the State Council decided to establish the Chinese Security
Exchange Commission (CSEC) and its enforcement agency, the CSRC. In
April 1993, the State Council published Provision of Security Issuing and
Trading, which stipulates that the CSEC is the organ in charge of the na-
tion's security market and is entitled to manage the nation's security market
according to the law; and that CSRC is CSEC's enforcement agency which
exercises the power of management and supervision of security issuing and
trading according to relevant laws and regulations. The chairman of the
CSEC must be one of the vice premier and the chairman of the CSRC
must be one of the vice chairmen of the CSEC. The CSRC has its routine
working departments such as issuing department, law department and oth-
ers. If necessary, the State's security organs in charge may establish branch
organs at local administration, exercising the power of management and
supervision. However, up to now, all the local security organs in charge are
established by the local governments. Since then, China's security organs
in charge have gone through a transitional period from being concurrently
in charge to specialized in charge.
The CSRC, whose members are high-level government officials from
various state ministries, exists to create policies and guidelines for regulat-
ing the exchanges. The CSRC exists to execute and enforce these policies
through day-to-day direct market supervision by its staff of trained securi-
ties professionals. Though the recent passage of securities regulations has
further empowered both CSRC and PBOC with the authority to oversee
market development, the PBOC has retained the ability to license and su-
pervise certain securities issues, thereby complicating the power structure.
Before the enactment of the P.R.C. Securities Law, the effectiveness of the
CSRC and PBOC has been limited due to unclear definition and delineation
of the exact roles of the various supervisory authorities. They must rely
upon a complex series of overlapping guidelines to do their job. When the
Securities Law came into effect in July 1999, the power of the CSRC was
enhanced.
According to China's Company Law and security regulations, the CSRC,
whose functions are similar to that of the U.S. Securities Exchange Com-
mission, has the following responsibilities:
Formulating policies, plans, laws and regulations concerning secu-
rities and futures markets;
Supervising securities and futures markets and regional/provincial
supervisory institutions;
Overseeing the issuance, trading, custody and settlement of equity
shares, convertible bonds and securities investment funds; approv-
ing the listing of corporate bonds; supervising the trading activities
of listed government and corporate bonds;
Supervising the listing, trading and settlement of domestic futures
contracts; overseeing domestic institutions involved in futures busi-
ness abroad;
Supervising the activities of listed firms and their shareholders who
are liable for information disclosure in securities markets;
Supervising securities and futures exchanges, their senior manage-
ment and securities associations;
Supervising securities and futures firms, securities investment fund
managers, securities registration and settlement firms, futures set-
tlement institutions and securities and futures investment consult-
ing institutions; approving the qualification of fund custody insti-
tutions and supervising their business; drafting rules on the quali-
fication of senior management for the above institutions; granting
the qualification of those involved in securities- and futures-related
business;
Supervising direct or indirect issuance and listing of shares abroad
by domestic firms; supervising the establishment of securities insti-
tutions abroad by domestic institutions; and supervising the estab-
lishment of domestic securities institutions by foreign organizations;
Supervising information disclosure and proliferation related to se-
curities and futures and overseeing statistics and information re-
sources management for securities and futures markets;
Helping grant the qualification of law firms, accounting firms, asset-
appraisal firms (and their professional staff) engaged in securities
and futures intermediary businesses, and supervising their activi-
ties.
In the beginning of 1996, The State's Planning Commission and the
Ministry of Finance formulated Rules of Collecting Fees for Management
and Supervision on Security and Futures Market. According to the Rules,
CSRC may collect management and supervision fees in the amount
of 0.25% and 0.04% of the annual trading value from the relevant
stock exchanges and futures exchanges.
CSRC may collect management and supervision fees in the amount
of 0.1% and 5% of the registered capital from the relevant security
corporations and companies which are concurrently undertaking
security business.
CSRC may charge 30,000 renminbi yuan application fee from com-
panies which apply to issue stocks.
2.2.2 Securities Laws and Regulations
Because China's securities market was established in an economic environ-
ment that bears little resemblance to a typical market-oriented economy.
There was not even a complete set of nationally applicable commercial laws
for people to comply with. When the two stock exchanges in Shanghai and
Shenzhen were established in China's most dynamic economic zones, in
the wake of their birth with only 13 local companies listed, the two local
governments started to supervise/regulate the exchanges according to lo-
cal legislation. In the rocky course of the exchanges' evolution from local
markets into national markets, the inconsistency of local legislation and
inadequacy of supervision led to serious disorder and low efficiency of the
markets. The infamous "August 10th" incident in Shenzhen in 1992, an
ill-fated initial public offering (IPO) through a lottery distribution method
going awry, was a harsh lesson which triggered the establishment of China's
national securities regulatory agency and reform of securities-related legal
system. The setting up of the CSRC in 1992 started a new era of securities
legislation and enforcement in China.
With several years of painstaking efforts, a series of laws and regulations
governing the securities market have been enacted. At a time, there were
over 250 pieces of national securities-related laws, regulations and rules
covering all aspects of the securities market, from public offering to listing
and trading of shares, from information disclosure of listed companies to
supervision of financial intermediary, from clearing and settlement to legal
liabilities and securities fraud, and from corporate governance to overseas
listing of Chinese companies. All existing market products including eq-
uity shares, securities investment funds, convertible corporate bonds and
commodities futures contracts are also legislated. Administration of Issuing
and Trading of Shares Tentative Regulations (1993), P.R.C. Company Law
(1994) and Overseas Listing of Chinese Companies Limited by Shares Spe-
cial Rules (1994) marked important legislative milestones before the P.R.C
Securities Law was eventually enacted in December 1998. The criminal-
ization of securities fraud and insider trading in 1997's P.R.C. Criminal
Law Amendments was also an important step to secure orderly operation
of the market and combat securities-related crimes. Rules such as Merger
& Acquisition Rules and Corporate Governance Rules have been put on
top of the CSRC's rule-making agenda.
2.2.2.1 P.R.C. Company Law
On December 29, 1993, the Standing Committee of the Eighth National
People's Congress adopted the P.R.C. Company Law, which came into ef-
fect on July 1, 1994 and was amended on December 25, 1999. In addition
to state enterprises, collective enterprises, township and village enterprises,
and private enterprises, the Company Law recognizes two additional kinds
of corporate legal persons in China: limited liability companies and joint
stock companies. The law restricts the number of shareholders for limited
liability companies to no more than 50 or will be otherwise unsuitable for
listing.
Under the Company Law, approval from the relevant State Council de-
partments or from provincial government departments must be obtained
before a company can be incorporated. The law provides two methods of
incorporation for joint stock companies. Such entities may be incorporated
by promotion or by public subscription. If a company is incorporated by
promotion, the entire registered capital of the company must be subscribed
for by at least five promoters. More than half of the promoters must re-
side within mainland China. According to the Special Regulations on the
Overseas Offering and Listing of Shares by Joint Stock Limited Compa-
nies, SOEs or enterprises with the majority of their assets owned by the
government can be restructured as joint stock limited companies, and un-
der some special circumstance, they may issue shares to overseas investors.
These companies, if incorporated by promotion, may have fewer than five
promoters and can issue new shares once incorporated. If a company is
incorporated by public subscription, at least 35 percent of its total shares
must be subscribed for by the promoter and the remaining shares must be
offered to the public.
The registered capital of a joint stock limited company is divided into
shares of equal par value. According to the Company Law, the registered
capital of a company is the amount of its total paid up capital as registered
with the relevant Bureau for Industry and Commerce. The minimum reg-
istered capital of a joint stock company is renminbi 10 million. The total
capital of a joint stock company which proposes to apply for its shares to be
listed on a stock exchange must not be less than renminbi 50 million. The
liability of shareholders of a joint stock company is limited to the extent of
the shares held by them, and the liability of the company is limited to the
full amount of all the assets owned by it.
Under the Company Law, joint stock companies are permitted to invest
in other joint stock companies and limited liability companies. However,
apart from investment companies and holding companies specified by the
State Council, the amount of a joint stock company's aggregate investment
in other companies may not exceed 50 per cent, of its net assets and a joint
stock company's liabilities with respect to such holdings is limited to the
amount invested.
1
Among numerous tasks to develop and strengthen the securities mar-
ket, the CSRC needs to tackle with several pressing legal and structural
issues. The P.R.C. Company Law (1993), China's prime corporate legisla-
tion enacted when modern corporation system were just introduced to the
country as a brand-new vehicle on the road of SOE reforms, has increas-
ingly shown its many inadequacies and some innate problems. The current
public offering system for equity and debt instruments needs to undergo
further conceptual and procedural changes to facilitate the market-based
pricing of securities and to make the corporate mechanism function in its
real sense. Corporate governance can never be overemphasized to upgrade
the accountability of listed companies and to enhance the roles played by
their directors and management. The development of the B-share market
and overseas listing of Chinese companies are calling for improvement of
transparency and removal of policy and technical barriers for higher liquid-
ity. Acknowledging these issues and many more, CSRC is doing everything
l r
The Company Law stipulates t hat companies which had been established previously
under various laws and regulations prior to the implementation of the law would not be
affected by the Company Law, but companies which had not wholly complied with the
provisions of the Company Law would have to satisfy the relevant requirements within
a specified period of time. The State Council has subsequently promulgated detailed
implementing measures to bring the Company Law into force.
within its power to persuade the country's legislators to amend, or better
yet, to overhaul the P.R.C. Company Law, to reform the IPO system, to
gradually abolish the political approval system and to allow market force to
come into play, and to introduce educational programs to train and exam-
ine senior management and directors of listed companies. All these efforts,
hopefully, will pay off to make China's corporate legal framework and prac-
tice more conducive to the new environment and to the fast growth of the
securities market.
2.2.2.2 P.R.C. Securities Law
Securities enforcement was developed along with the securities legislation
and market development in China. When the CSRC was established in
1992, the government and the regulators were totally inexperienced in secu-
rities market enforcement due to the fact that China's securities market had
emerged from a long-lasted planned economy. Without necessary power to
fulfill its regulatory duties, CSRC's enforcement capability was very much
limited. Chinese stock markets was described by many as a "gambling
place" characterized by the unsophisticated, short-term speculators and
the manipulation of institutions. Little was there for the long-term market
growth and confidence of investors at home and abroad.
With rules and regulations coming gradually into place and regulators
becoming more experienced in practice, the general securities enforcement
capacity has been greatly enhanced. In 1998, over 100 cases of complaints
were lodged and investigated by the CSRC. 56 of those were concluded with
administrative sanctions imposed and three suspects were transferred to
judicial authorities for criminal prosecution. The number of prosecutions
in 1998 was more than the total number of cases handled by the CSRC
between 1995 and 1997. These cases involved various malpractices such as
market manipulation, false financial statement, fraudulent conducts, insider
trading, and illegal IPO and overseas futures transactions. The treatment
of serious cases such "Qiong Minyuan" fraudulent accounting statements
and practice, "Hong Guang" false financial statements and "9703 Polly-
wood Contract" price manipulation have won extensive public acclaim and
served to build up investors' confidence. In the course of building chan-
nels for foreign investment in China through overseas listing of Chinese
companies and development of B-share markets, international cooperation
in terms of enforcement, information exchange, and mutual assistance have
become an important field for both the CSRC and its overseas counterparts.
The CSRC has so far established bilateral Memorandum of Understandings
(MoUs) and similar arrangements with 13 regulatory counterparts in the
world.
2
On December 29, 1998, the P.R.C. Securities Law was passed by the
Standing Committee of the National People's Congress. The law took effect
on July 1, 1999. This is the first national securities law in mainland China,
and it is divided into 12 chapters and 214 articles regulating, among other
things, the issue and trading of securities, takeovers by listed companies,
securities exchanges, securities companies and the duties and responsibili-
ties of the State Council's securities regulatory authorities.
3
The Securities
Law is the fundamental law which comprehensively regulates activities in
the Chinese securities market. Perhaps the most important accomplish-
ment of the making of the law is to indicate a consensus view among the
Chinese authorities that:
The nature of the stock market and its role and function within the
Chinese socialist market economy has now been firmly entrenched;
any effective regulation of the stock market must have a strong legal
foundation; and
given the existing realities of the stock market against the back-
ground of the Asian financial crisis in 1998, the authorities believe
that legal regulation of the stock market cannot be further delayed.
The Law thus arguably indicates that, since the stock market is now an
integral part of the Chinese socialist economy, the use of securities as a tool
of capital formation will be encouraged and the quality of the stock market
will thereby improve with time. In other words, it provides a firm basis
for the belief that greater business opportunities, both domestically and
overseas, will be created in the Chinese capital market. For example, Article
2
According to an article published by China Online on November 5, 1999, t he CSRC
has already signed MoUs on the supervision of securities markets with the regulatory
agencies in Hong Kong, U. S., Singapore, Australia, U. K., Italy, Japan, Malaysia,
Brazil, Ukraine, France, Luxemburg and Germany. Under the terms of the MoUs,
China and the above countries and regions plan to share information, exchange and
train staff, thus paving the way for Chinese companies to t ap funds from international
capital markets.
3
See Appendix A, which excerpts from "The PRC Securities Law and t he Development
of the Mainland Securities Market" by Alexandra Lo, for a more detailed summary of
the P.R.C. Securities Law.
29 of the Securities Law provides that Chinese enterprises which intend to
directly or indirectly issue or list their securities outside the mainland China
must first obtain approval from the State Council's regulatory authorities.
Article 213 of the Securities Law provides that specific measures shall be
formulated separately by the State Council if specific shares of Chinese
companies are to be subscribed and traded in foreign currencies by person
and organization outside mainland China.
4
In addition, the law provides strong legislative support for the CSRC
in the discharge of its enforcement functions, giving CSRC the power to
investigate, to impose administrative sanctions, and to transfer cases to
administrative and judicial authorities. To ensure all securities laws and
regulations are effectively and consistently enforced, the securities regula-
tory system has been restructured to form a centralized and unified regula-
tory system, with CSRC as the prime regulator over the securities market
and the local regulatory agencies transformed into regional offices of the
CSRC. The enforcement capacity of CSRC will be further strengthened af-
ter additional enforcement branches of CSRC are established in the near
future.
5
As the securities markets in Shanghai and Shenzhen grow rapidly to
have listed today about 1000 companies with a market capitalization of
about 3 trillion renminbi accounting for 30% of the country's GDP, a com-
plete securities legal framework has to be developed over time. The enacted
P.R. C. Securities Law is only a foundation of an ambitious reform project.
A great amount of work is still ahead to solve the inconsistencies and con-
tradictions amongst laws and rules, and to clearly define the essential terms
and gradually refine the legal framework. The current task of the CSRC
is to make the legal system work more effectively through updating the
P.R.C. Securities Law and make the law fully operational. No longer "a
tiger without teeth", the Chinese regulator is sharpening its teeth to meet
4
I n the meanwhile, the issue and trading of foreign shares (including H shares and B
shares) are still mainly governed by the rules and regulations promulgated by the State
Council and the CSRC.
5
An efficient operation of the present regulatory and enforcement system will, of course,
rely on people who are running it. Proper regulation and enforcement is an "art" t hat
all securities regulators must grasp. Therefore, in the process of consolidating China' s
regulatory system, it is of paramount importance to equip regulatory personnel with
enforcement techniques, procedural rules and administrative skills. Wi t h the fast devel-
opment of the markets, the regulator is still facing great challenges in its enforcement
work.
the challenges of a rapidly growing market. To this end, the Chinese gov-
ernment and the regulators are committed to establishing a complete, fair
and balanced legal system in the next few years.
At the turn of the new millennium, China is at the most critical moment
of her economic development and reforms. To achieve her development
goals in the next millenium, China needs to attach great importance to the
role to be played by her securities market at this transitional stage. On
the basis of the achievement in the securities legislation and enforcement,
the regulators will have to work harder to regulate the market according
the rules of the market itself. The CSRC will experience its own transition
from regulating the market through administrative means to doing that
through market-oriented means and self-regulatory mechanisms. Only in
such a way can the CSRC become a true 'policeman' of the market instead
of an administrator. It is encouraging that the regulator is already on such
a track. A developed securities law and matured regulators will surely be
needed to safeguard a fair, efficient, orderly and sustained development of
China's securities markets.
2.3 Ownership Structure
The dilemma for designing any good corporate governance structure is how
to make corporate executives accountable to the other contributors to the
enterprise whose investments are at risk while still giving those executives
the freedom, incentives, and control over resources for competitive purposes.
Ownership structure reforms in China can be characterized, similar to other
recent economic reforms, as gradual and pragmatic. Effectively separating
ownership from management for SOEs is complicated by the fact that the
State, or its agents, carry out shareholder functions otherwise performed
by private owners in market economic systems. A further complication is
that political reasons foster management's dependence upon, rather than
its separation from, the ownership. For example, many industrial managers
of the mid 1980s owed their jobs to the personal connections they held with
bureaucratic superiors and depended on these connections for bail-out from
serious economic problems that plagued their firms. Both commercializa-
tion and corporatization move the status of SOEs from political institutions
more towards economic entities. In order for the commercialization and cor-
poratization to be effective, a mechanism must be designed to ensure that
Ownership Structure 45
state' s ownership is one of limited liability so as to hold the management
accountable for its losses. The Chinese government has been working on
this important ownership aspect of SOE reform. Table 2.1 presents the
time-line of measures and legislation undertaken by the State to enforce
ownership structure reform. Figure 2.1 shows shares of industrial output
by ownership between 1978 and 1996.
When enterprise reform was launched in 1984, the main barriers to Chi-
nese economic development were seen as bureaucratic rigidity, inefficiency,
and the irrationality of central planning. The ownership reform was not
meant to change the fundamental rules of the game but merely to provide
more incentives for enterprises to maximize their profits and increase their
efficiency. It was not anticipated that when price liberalization would be
undertaken, that fundamental structural changes would also take place, i.e.
the allocation of resources would no longer be dominated by the state. The
command and control system was fundamentally dismantled and the reform
took on a life of its own; it was not foreseen by the original designers and
that this would require further change in government policies to accommo-
date such an evolution. The wide gap between intentions and outcomes
imposed pressures on the existing institutional structures and demanded
further institutional adjustment.
Granting SOEs more autonomy without separating ownership and man-
agerial rights seemed to improve productivity and economic growth but
simultaneously triggered social and economic problems which devastated
the reform and the economy as a whole. The undesirable outcomes, in-
cluding corruption, increased predatory activities, inflation, and the near
bankruptcy of the state, have become the main stimulus for further reform
in China since the early 1990s. The commercialization and corporatiza-
tion of existing SOEs were centered on the arbitrariness in distributional
relations between the State and the enterprise and their related incentive
problems. There were done by clarifying property rights and operational
rights to achieve a higher degree of accountability and to form competitive
and independently-managed industrial conglomerates integrating industry,
business, scientific research, and finance into one entity. This entity would
be able to utilize market forces to promote optimum use of existing resources
and to rationalize internal organization and structure. By early 1993, more
than 10,000 companies had been merged into such conglomerates.
Within the share holding restrictions imposed by the CSRC are specific
rules ensuring that no individual Chinese or foreigner can purchase a ma-
S
t
r
u
c
t
u
r
a
l

a
n
d

I
n
s
t
i
t
u
t
i
o
n
a
l

C
h
a
r
a
c
t
e
r
i
s
t
i
c
s

h
o

J

>
-
m

t
o

S
I

I

o

=
c

o

W

n

<
S
>

T
O

N

h
o

r
S

B

2
1

5

-
a

t
o

.
2

0

C
D

S

I
s

c

o

I
D

1

2

a
>

o

c

c
8

-
u

1

C
D

I
s
.

t
o

a

a
.

o

*

t

'

Q
<

"
-

3

T
3

1
1

-
a

l
l

a
-
a
s
-
G

^
H

-
O

-

s

.
s

o

p

a
j

c

O

b
o

S

>
>

a
s

a
>

h
e

(
8

-
S

'
-
3

0
)

2
,

"
8

S

K

)

y

<
>

B

'
s

a

'
g

"

^

c

D

I

i

4

B

M
l

<

-
a

l
/
l

C
T
>

t
4
_

c

3

?
>
T
t

a
t

M
l

J
-
i

i
n

W

o

C
O

o

O
S

&

9
l

M
l

+
J

T
3

N

t
d

1

J

r
o

>
,

n

u
spo
O
J

f
f
!

2

I

S

+
J

"
2

+
a

a

o

H

O

-
a

c

c
8

E

o

-
o

n
S

J
3

o

e
n

a

0

=

c

h
o

a
i

c

I
f
l

a

2

e

.
2

O
J

C
O

J
f
i

C
O

g
.

C
O

C
O
"

0
)

X

"
S

C
O

&
.
S

o

a
>

o

J
3

a
j

.
2
3

1
e

3

D
.

o

H

O

C
O

c
f
l

0
3

C

J
3

O

C
O

M

I
O

O
J

M

h
j

C
O

C
O

o

C
O

c

0
3

E

o

u

-
a

a

a

a

o
3

a

D
.

S

-
a

o

O
!

J

W

3

I
-
a

"
-
;

h
o

o

C
Q

J
-

0
)

.
;
S

i
s

C

a

.
B

4
)

I
f
l

>

C
O

-
u

c
a

a

I
/
J

O

r
/
-
)

*

E

a
;

"
J

S
,

^

M
>

,

a
;

h
o

^

o

0

O

J
3

O

.
2
2

3

5

s
i

l
-
l

I
S

d

a
j

"
3

a
.

I
"
3

1

>
.

c

0
3

a

S

o

O

O
w
n
e
r
s
h
i
p

S
t
r
u
c
t
u
r
e

4
7

Table 2.2 Ownership of Shares on the Shanghai and Shenzhen Securities Exchanges
1992
1993
1994
1995
1996
1997
Government
63.1%
59.9%
51.3%
56.7%
49.2%
45.7%
Percentage of Shares Held by
Legal Entities
9.9%
10.7%
18.0%
14.9%
17.5%
20.8%
Domestic
Individuals
21.0%
22.4%
9.8%
14.0%
17.6%
17.0%
Overseas Funds
& Individuals
5%
7%
20.9%
14.5%
15.8%
16.5%
jority stake in any listed company. Table 2.2 presents ownership of shares
on the Shanghai and Shenzhen Securities Exchanges. As shown on the ta-
ble, of the outstanding shares for any firm, only 30% can actually be traded
in Shanghai or Shenzhen. Government agencies own a large proportion of
total stock, 46% in 1997, down from 63% in 1992. The category of legal
entities refers to legally constituted autonomous organizations. It is cru-
cial that in the Chinese case these legal entities are generally not mutual
funds, pension funds, or insurance companies, but are generally holding
companies established by government agencies as a management tool for
government-owned stocks. Thus, most Chinese observers consider them as
secondary government ownership of shares. Shares classified as government
of legal-entity owned are not allowed to circulate on the exchange. Thus,
approximately two-thirds of total share value on the securities exchanges
did not circulate.
6
8
It should be noted t hat after the Fifteenth National Congress, restructuring of SOEs
involved a much more rapid rate of ownership conversion, increased privatization, and
much greater use of joint stock corporations. Government limitations requiring t hat
the state maintain majority or controlling interests were repealed.
Listing Standards and Procedures 49
2.4 Listing Standards and Procedures
2.4.1 Requirements for Stock Listings
Options available to a company for a stock exchange listing include:
Domestic listing of A-shares;
Domestic listing of B-shares;
Overseas listing of Foreign Investment Shares;
Red-chip listing which can take various forms.
A company could list its domestic investment shares (in the form of A-
shares) or foreign investment shares (in the form of B-shares) in Shanghai or
Shenzhen or its foreign investment shares on exchanges which have signed
a Memorandum of Understanding with the CSRC or undertake a red-chip
listing. The relevant laws and regulations governing stock listings are the
Administration of the Issuing and Trading of Shares Tentative Regulations,
P.R.C. Company Law, and P.R.C. Securities Law.
2.4.1.1 A Shares
A-shares are different from other categories of domestic investment shares
such as state-owned shares. A-shares are domestic investment shares issued
by Chinese companies which are listed on the SHSE and SZSE. A-shares
may only be subscribed by and traded among Chinese citizens and/or en-
tities.
Under the Administration of the Issuing and Trading of Shares Tentative
Regulations promulgated on April 22, 1993, before listing on the SHSE or
SZSE, a shareholding company (also referred to as a joint stock limited
company) must first be established. The procedure and requirements for
establishing a shareholding company are set out in the Company Law. Note
that the minimum amount of registered capital is renminbi 10 million which
has to be paid up in cash at the time of filing application and supported
by capital verification certificates.
A shareholding company must comply with the following criteria in
order to apply for listing:
The shares must be issued to the public (as opposed to a private
placement);
The total share capital after a public offering must not be less than
renminbi 50 million;
The amount of share capital subscribed by sponsors must be at
least renminbi 30 million and must represent at least 35% of the
total share capital issued;
At least 25% of the shares must be issued to the public and not more
than 10% of this amount may be subscribed by the staff and work-
ers. For Chinese companies with an issued share capital totaling a
nominal value of more than renminbi 400 million, the percentage
of shares issued to the public must account for not less than 10%
of the total shares of the company;
There must be more than 1,000 individual shareholders holding
shares with a nominal value of over renminbi 1,000 and the total
nominal value of shares held by those individuals must be at least
renminbi 10 million;
Except for a newly-formed company limited by shares, the Chinese
company must have made profits in each of the last three years;
If a shareholding is established by restructuring an existing enter-
prise, the existing enterprise must have a three-year track record of
earning profits;
Any other conditions imposed by the CSRC. (See the P.R.C. Se-
curities Law.)
The Securities Law stipulates that Chinese companies proposing to is-
sue and list their A-shares must first be approved by the CSRC. The CSRC
must make a decision within three months from the date on which an ap-
plication is received. The stock exchange from which a company is seeking
a listing is required to verify applications and make a final decision within
six months from the date on which an application is received. The com-
pany and the stock exchange should arrange for the procedure regarding
the listing of the shares once an application is approved. The Securities Law
also sets out general regulation on securities trading, continuing disclosure,
insider dealing and other prohibited trading acts on A shares.
2.4.1.2 B-Shares
Foreign investment shares may be listed as B-shares on the SHSE or SZSE.
The term "B-shares", "domestically listed foreign investment shares" and
Listing Standards and Procedures
51
"special renmOT&z-denominated shares" all refer to the same thing-ordinary
shares of Chinese shareholding companies that are denominated in renminbi
but traded in foreign currencies, such as U.S. dollars, on a Chinese securities
exchange. B-shares can only be subscribed by and traded among foreign
legal and natural persons and other entities, legal and natural persons from
Hong Kong, Macau and Taiwan, and Chinese citizens who are resident
abroad.
The regulatory framework surrounding the issue of B-shares was sim-
plified by the State Council in December 1995 by the introduction of Pro-
visions on Listing of Foreign Investment Shares Inside China by a Share-
holding Company on December 25, 1995 and the Provisions on Listing of
Foreign Investment Shares Inside China by a Shareholding Company: Im-
plementing Rules on May 3, 1996. These provisions clarified the procedures
involved in applying for approval to issue B-shares. They also set out the
application procedures and approval requirements for companies seeking
to issue B-shares in order to increase their share capital. The provisions
contain important matters such as information disclosure, and the trading
of B-shares by stock brokers and agents.
The key contents of the provisions on B-Share listings include the fol-
lowing:
The CSRC is responsible for the regulation and supervision of the
issuing and trading of B-shares and related activities in relation to
B-shares;
In addition to the directors, supervisors and managers of a B-share
company, other senior management personnel, including a share-
holding company's chief financial officer, secretary and other exec-
utives specified in the company's articles of association, owe duties
of good faith and diligence to the company;
Chinese citizens residing outside mainland China may purchase B-
shares;
The derivative forms of B-shares, including warrants and depository
receipts, may be circulated and traded outside China.
The B-Share listing implementing rules set out further detailed provi-
sions governing the issue and trading of B-shares. They expand upon the
application procedures set out in the B-share listing provisions to gain ap-
proval for the issuance of B-shares and list the documents to be submitted
to the CSRC in support of an application. The key contents of the B-share
listing implementing rules are as follows:
An over-allotment option (commonly known as a "Greenshoe") may
be granted by a Chinese shareholding company to the underwriters.
With the approval of the CSRC, a company may set aside up to 15%
of the total amount of the proposed B-share issue which constitutes
the option. Such reserved shares will be considered as part of the
issue;
The distribution period for B-shares may not exceed 90 days;
The governing law of the underwriting agreement must be Chinese
laws;
Within 15 days after the closing date of the initial share distribu-
tion, the lead underwriter must submit to the CSRC a distribution
report and a list of the 10 largest holders of its B-shares, and details
of their holdings. The distribution report must contain details of
the distribution process;
Domestic brokerage houses are required to report to the CSRC
details of the number of B-shares held by them as a result of par-
ticipating in the underwriting of a B-share issue;
In addition to appointing Chinese accounting and auditing firms, B-
share companies may also appoint foreign accounting and auditing
firms that comply with the Chinese regulations to audit or review
their financial statements;
B-share companies must give prior notice to their auditors of dis-
missal or non-renewal of appointment and the auditors are entitled
to present their views on any matter concerning the company's
financial situation before the company's shareholders in the an-
nual meeting. The P.R.C. Securities Law (Article 213) states that
shares of Chinese companies designated for subscription and trad-
ing by foreign investors (B-shares an H-shares) are governed by
measures separately formulated by the State Council.
B shares are subject to a strict annual quota system. Each year, the
State Council decides the amount of B-share quota in U.S. dollars for that
year. For example, the quota was US$1 billion in 1993. The B-share listing
provisions did not stipulate listing venue. In practice, the Shanghai Securi-
ties Exchange attracted bigger and more reputable companies to list their
shares.
2.4.1.3 Overseas' Listing of Foreign Shares
All foreign listings must be approved by the CSRC and the foreign stock
exchange (and regulatory authority, such as the SEC in the case of a U.S.
listing). Foreign shares (such as H-shares) must be issued in registered
form and denominated in renminbi even though they are traded in foreign
currencies. Depository receipts issued over H-shares are also treated as
foreign shares.
Before a Chinese company can undertake an overseas listing at a desired
overseas exchange by means of a direct listing, the overseas exchange has to
sign a Memorandum of Understanding (MoU) with the CSRC. The MoU
deals with cross-border regulatory issues such as supervision, disclosure
requirements and securities enforcement.
In addition to complying with the requirements prescribed by the stock
exchange on which the shares are listed, a company seeking to list overseas
must also comply with the Articles of Association of Companies Seeking
a Listing Outside the P.R.C. Prerequisite Clauses issued by the Securities
Office of the Restructuring Commission and effective as of September 19,
1994, P.R.C. Company Law, and P.R.C. Securities Law. Key contents of
the Prerequisite Clauses include the following:
State enterprises to be restructured into shareholding companies
may have fewer than five promoters;
The period between an overseas listing and a subsequent issue may
be less than 12 months;
A 45-day written notice is required to convene shareholders' meet-
ings;
A quorum of 50% of voting shares is required to convene a gen-
eral meeting and shareholders must give a 20-day notice of their
intention to attend a general meeting;
The articles of association are binding not only on a company and
its shareholders, directors, supervisors and general managers, but
also on other senior officers including the chief financial officer and
secretary to the board;
Shares issued outside China may be in the form of warrants or other
derivatives subject to the approval of the State Council Securities
Commission. An over-allotment option of up to 15% of the total
issue may be granted by an issuer to the underwriters;
Dividends on overseas foreign shares should be declared in renminbi
and paid in foreign currency;
The register of holders of foreign invested shares listed outside
China may be kept abroad and maintained by an agent.
Cooperation between Hong Kong and mainland stock market regulators
started in 1993 when the issuance of H-shares on Hang Seng Stock Exchange
was first proposed. By 1994 Hong Kong was able to update its legislation on
listing requirements of PRC issuers. In November 1994, legislators added to
Chapter 19 of Hong Kong Stock Exchange Ordinance a sub-clause known
as Chapter 19a, which was solely devoted to PRC issuers. Once listed on
the Hang Seng Stock Exchange, a PRC issuer is subject to all relevant Hong
Kong laws and requirements, including the Hong Kong code on takeovers
and mergers.
2.4.1.4 Red Chip Listing
Red-chip companies are those incorporated in Hong Kong and listed on the
Hong Kong Stock Exchange but with controlling shareholders from main-
land Chinese entities. In the early stage of development in Chinese stock
markets, some companies had successfully bypassed the official listing chan-
nels and gotten listed either through a backdoor listing or by acquiring a
Hong Kong "shell" company. After the CSRC was established, mainland
and Hong Kong regulators started to cooperate and coordinate on red chip
issues. Both sides agreed that before granting listing to a mainland Chi-
nese company, each side would inform the other of the nature of the listing,
company type and other related information. Nowadays red chip compa-
nies are generally diversified conglomerates which have grown rapidly by
the injection of assets from their parent companies in mainland China.
Red-chip companies include China Telecom, Beijing Enterprises Holdings,
China Everbright, Shanghai Industrial Holdings, China Resources Enter-
prises and the "window-companies" or "International Trust and Investment
Corporations (ITICs)" of provincial governments.
Until very recently, red-chip companies did not, strictly speaking, fall
within the supervision of mainland Chinese authorities. Therefore, they
were able to conduct restructuring and raise funds from overseas for new
investments easily. In June 1997, the CSRC in conjunction with the State
Council introduced the Notice on Further Strengthening the Administration
of the Listing and Issuing of Shares Overseas (also called the "Red Chip
Notice"). The purpose of the Red Chip Notice was to protect domestic
assets from being channeled overseas and from being sold off indirectly to
overseas investors at a discount. This is accomplished by purporting to
more stringent requirements for reporting to, and obtaining approval from,
Chinese authorities. The Red Chip Notice targets both listed and unlisted
companies registered outside mainland China.
The main content of the Red Chip Notice is as follows:
If a foreign listed company is registered and controlled by Chinese
shareholders and is undertaking a spin-off listing or additional issue
of shares, it is subject to the supervision of the CSRC and the
majority shareholder must report to the CSRC.
Domestic shareholders that have held foreign and domestic assets
for three years or more who seek to inject their assets to a red chip
company may do so provided that prior consent is obtained from
the local provincial government or relevant department of the State
Council. Domestic assets held for less than three years may not be
used in connection with foreign share issues unless there are special
circumstances that the CSRC deems to be appropriate.
Consent on restructuring and subsequent equity offerings must be
obtained from the provincial government or relevant State Council
department and a report must be made to the CSRC for examina-
tion and final approval.
Acquisition, share swap or other methods of injecting assets into a
foreign Chinese holding company require similar relevant consent
from the CSRC.
7
2.4.2 Procedures for Stock Listing
Although the capital markets in China has developed rapidly over the past
seven years, the central government still commands considerable authority
over many structural aspects of market operations. A key element of this
authority is the listing requirements and procedures which the CSRC con-
trols. Together with the State Planning Commission, the State Council,
the State Commission for Restructuring the Economy, and the Ministry of
Finance, the CSRC sets an annual quota for the total value of shares to
be brought to market. Once the quota has been established, a share of the
7
The CSRC has made it clear publicly t hat its considers asset injections as an unautho-
rized sale of state assets, and approvals will only be issued in special cases.
total allowance is allocated to each province, which then make their own
selections among local companies who have applied for stock issues. The
decision of the provincial authorities rests mainly on two factors: the size of
the share-capital quota allocated to them and whether or not the company
is in a "key industry" which the government has chosen to promote. Hav-
ing been selected, the selected enterprises are recommended to the CSRC
which will then review the selections made by individual provinces for final
approval.
This heavily-regulated method of selecting firms that go public has been
criticized by many experts as not being in the best interest of the public
investors. Critics point out that SOEs with capable management and fun-
damentally sound balance sheets might be passed over in favor of other less
commercially viable firms which happen to be in the government's latest
key industry category. Although companies to be listed are required to
have earned annual profits of 10% per year for the past three years prior to
their initial public offerings and are supposedly the most promising SOEs,
those receive permission to issue shares are actually part of the slowest-
growing sector of the economy-a sector which accounts for only about 30%
of the country's industrial output. Nevertheless, the CSRC has found its
control over the supply of shares useful in influencing market conditions
and investor sentiment. For example, when key market indices have gone
up sharply in June 1996, authorities began voicing their concern over the
formation of a bubble in stock valuations. When their warnings did not
serve to slow the market's rise towards new highs, the CSRC announced it
would double the issue volume in 1997, allowing 30 billion renminbi (US$3.6
billion) worth of A and B-shares to be placed. By increasing the supply
of stock available for purchase, the CSRC hoped to soak up the specula-
tive demand rampaging in the market, thereby promoting a more stable
investing environment.
There are two steps involved in selecting an issuer for A shares. The
first stage involves provision of share issue quotas by the CSRC to each
provincial government, autonomous zone, municipality directly controlled
by the central government, and municipality with a separate economic plan.
Factors taken into consideration when allocating share issue quotas include
(1) the development of the economy of a particular province or a specific in-
dustry, which is typically a result of the macroeconomic conditions at that
time; (2) applications by prospective A-share issuers to the local municipal
government or the relevant industrial department of the State Council; (3)
recommendation by local municipal government or the relevant industrial
department of the State Council to the CSRC; (4) preliminary examina-
tion by the relevant departments of the CSRC; (5) opinion of the relevant
ministry which governs the industry in which the prospective A-share is-
suer has made the petition; and (6) issues exceeding US$30 million must
be approved directly by the State Council.
The second stage involves gaining initial approval from the relevant local
government or State Council ministry, submission of formal application
materials to the CSRC, review and approval by the issuing department
of the CSRC, and examination by the issuing examination committee of
the CSRC. Once the issuing examination committee approves the A-share
issue, the CSRC decides the specific timing of the IPOs and will issue a
formal approval certificate.
The selection and approval process for B-share issuers is pretty much the
same as that for A-shares. The selection process involves recommendation
by the local municipal government or relevant industrial department of the
State Council and pre-selection by the CSRC. The approval process involves
initial examination by local municipal government or the relevant industrial
department of the State Council, submission of application materials to the
international business department of the CSRC, examination and approval
by the international business department of the CSRC, and examination
by the issuing examination committee of the CSRC.
All foreign listings must be approved by the CSRC and the foreign
stock exchanges on which listings are sought before any listing application
is made. In the past, Chinese companies have been approved by the CSRC
in batches. The latest (the 5th) batch of Chinese companies when this
book is written included Wenzhou Infrastructure, Ningbo Port, Zhejiang
Tourism, Hebei Expressway, Shandong Expressway, Guangdong Agricul-
ture, Heilongjiang Agriculture, and Beijing Capital Airport.
In addition to the CSRC approval, any Chinese company seeking to
list its shares on an overseas stock exchange may have to receive approvals
from the State Council, the State Assets Bureau, the Ministry of State
Land and Resources, the State Development and Planning Commission, the
State Economic and Trade Commission, the prospective issuer's supervising
ministry, and the overseas stock exchange on which the securities will be
listed. Typically the issuers' advisors will undertake the preparation work
for the public offering before all of the aforementioned approvals have been
issued.
A Chinese company seeking to issue and list foreign investment shares
outside China must complete a three stage procedure. The first stage in-
volves making an application to the local municipal government or the min-
istry or department-in-charge of the relevant industry, which will review the
application and if the company is deemed appropriate, its application will
be sent to the CSRC. The second stage will involve a selection process
conducted by the CSRC in consultation with the relevant local municipal
governments or ministry or department-in-charge. A short list of success-
ful companies will be notified and then released to the public. The final
stage comprises the examination and approval by the CSRC during which
the CSRC will review prescribed documents in accordance with applicable
securities laws and regulations.
2.5 Market Microstructure
2.5.1 Registration of Stock Dealers
One of the most successful pieces of reform recently enacted by the State
Council included regulations to establish standards within the brokerage
industry. In addition to formally prohibiting insider trading and requiring
that all companies publicly disclose news which could affect share prices,
the Management Regulations for Securities Industry imposed the following
provisions on all securities firms dealing with A-shares trading:
5% of all profits must be kept as reserves or margins.
Adequate risk-monitoring systems must be developed.
No more than 80% of a firm's assets can be held in stocks.
Two-thirds of a securities firm's managers must be certified by the
CSRC.
No securities firm can buy more than 20% of the outstanding float
of a listed company during any given trading day.
Regulations for securities firms doing business in B-shares and H-shares
have also been issued recently. An important part of the regulation is
to require companies that have already listed their B-shares to undergo
an annual financial auditing based on International Accounting Standards
(IAS). Under these regulations, both foreign and domestic brokerage firms
must meet the following criteria to do business in China:
Market Microstructure 59
Stockbrokers must have assets of at least 50 million renminbi (or
U.S. $6.0 million).
They must have suitable trading sites and an extensive network of
sales and research staff.
Overseas brokers must be licensed by the CSRC every two years.
Domestic brokers must be authorized by the State Administration
of Foreign Exchange (SAFE) to do business involving foreign cur-
rency transactions.
In addition to regulating the market by imposing rules on brokerage
houses, securities authorities have recently launched a crack-down on il-
legal practices by investors, particularly at the Shenzhen exchange. The
CSRC now officially prohibits the ownership of B-shares by P.R.C. citizens,
although many continue to mask such purchases by routing them through
a series of accounts at different banks. In a further attempt to prevent
the rampant speculation in A-share markets, the PBOC has prohibited all
forms of margin trading.
8
2.5.2 General Rules for Trading
Securities that have been approved for listing and trading must be quoted
and traded on stock exchanges, and trading must be conducted by a public,
centralized post at competing prices. Securities firms cannot provide loans
or lend securities to their customers. Officers and employees of stock ex-
changes, securities firms, securities registration and clearing companies, and
securities regulatory authorities may not engage in stock trading, whether
directly or indirectly.
9
Similarly, professionals who issue documents in con-
nection with a share issue (e.g., audit reports, legal opinions, and so forth)
may not purchase or sell shares during and six months after the offering
period.
A company applying to list its shares must first be approved by the
CSRC, and the CSRC may authorize the stock exchanges to verify and
8
Under the current P.R.C. law, margin trading is a criminal offense punishable as em-
bezzlement. Although the Chinese securities industry is not really governed by criminal
law at present, this ruling is particularly harsh as the embezzlement of more t han 20,000
renminbi (US$2,400) carries the death penalty.
9
The P.R.C. Securities Law may appear draconian to have imposed such a restriction,
but in fact similar regulations had appeared in various regulations enacted in 1993.
approve applications for listing of shares in accordance with statutory con-
ditions and procedures. The stock exchanges will verify and decide on
applications and will, within six months after the applications have been
received, arrange for the shares in successful applications to be listed. The
government may encourage new listings of qualified enterprises in key in-
dustrial sectors it deems appropriate. If a listed company ceases to meet
the conditions for listing prescribed in the P.R. C. Company Law, the listing
of its shares will be suspended or terminated. Furthermore, the CSRC may
authorize the stock exchanges to suspend or terminate listings in accordance
with the law.
After a company is listed in a stock exchange, it is required to prepare
and announce interim reports and annual reports. If a major event not
known to investors has occurred that may have serious impact on the price
of shares of a listed company, the listed company must immediately sub-
mit a report to the CSRC and the relevant stock exchange detailing the
occurrence of the event. Examples of major events include:
a major change in the companys business policies or scope of busi-
ness;
a decision by the company on a major investment or asset purchase;
or
conclusion of a material contract that may have a major effect on
the companys assets, liabilities, rights, interests or business results.
There is an exception under which a listed company having sufficient
reason to believe that the disclosure may harm its interests and that non-
disclosure will not lead to major fluctuations in the share price, is not
required to make disclosure subject to the consent of the relevant stock
exchange.
The P.R.C. Securities Law strictly prohibits people with knowledge of
inside information on securities trading from carrying out securities trans-
actions by making use of that information.
10
No person with knowledge
of inside information on securities trading, and no person who has illegally
1 0
The term "inside information" means information, not known to the public, t hat
relates to the companys business or financial affairs or t hat may have a major effect
on the price of the companys securities. The description "people with knowledge of
inside information on securities trading" refers to: (1) directors, supervisors, managers,
deputy managers and other senior management personnel of a listed company; (2)
shareholders holding 5% or more of the shares in a company; (3) people who are able
to obtain company information concerning the trading of its securities by virtue of their
obtained inside information, can purchase or sell securities of the relevant
company, divulge such information or counsel another person to purchase
or sell such securities.
Under the present Securities Law, the actual knowledge of inside in-
formation is irrelevant in determining whether a person is a "person with
knowledge of inside information". In other words, a person is deemed a
"person with knowledge of inside information" by virtue of the positions
and offices the person holds, and the duties and functions the person per-
forms, in listed companies, professional intermediary firms and securities
regulatory authorities. However, this interpretation of the definition cre-
ates an anomalous result when applied to the article of the Securities Law
that imposes an absolute prohibition on sale and purchase of securities by
a "person with knowledge of inside information". Thus, a person who is
able to obtain company information concerning the trading of its securities
by virtue of the position held in the company but who has not actually
obtained such information is prohibited to purchase or sell securities of the
company. It is doubtful whether this anomalous result is indeed intended
by the legislators.
11
Apart from the prohibition on insider dealings, the P.R.C. Securities
Law also prohibits market manipulation activities, the making of false state-
ments in the trading of securities, and activities amounting to deception of
clients. Further, the new law prohibits state-owned and state-controlled
enterprises from engaging in share speculation.
Because the SHSE and SZSE are located in the same time zone, trading
takes place in more or less the same hours, with only little differences in
each other's opening, mid-session and closing times. The official trading
hours of the SHSE are from 9:30 AM to 11:30 AM and 13:00 PM to 15:00
PM for stock trading; 10:15 AM to 11:45 AM and 13:00 PM to 15:30 PM
positions in the company; (4) personnel of the CSRC and other regulatory bodies; and
(5) relevant personnel of professional intermediary firms and securities registration and
clearing institutions who participate in or oversee securities trading.
u
A t present, insider dealing activities are regulated by the Provisional Measures on the
Prohibition of Deceptive Securities Dealing Activities (the Deceptive Dealing Mea-
sures) of 1993. All firms and individuals are prohibited from engaging in t he issue and
trading of securities by making use of inside information with a view to making profits
or avoiding losses. Actual knowledge and use of inside information must be present for
an act of insider dealing to occur. The Deceptive Dealing Measures does not contain
an absolute prohibition on sale and purchase of securities in terms as broadly defined
as those of theP.R.C. Securities Law.
for treasury bond; and 13:00 PM to 15:30 PM for treasury bond futures.
Trading holidays include January 1, the New Year's Day; Chinese lunar
New Year's week-long holidays; May 1, the International Labor Day; and
October 1-2, the National Day holidays. The official trading hours of the
SZSE are from 9:30 AM to 11:30 AM and 13:00 PM to 15:00 PM. The
SZSE has the same holiday schedule as the SHSE.
The trading cost is approximately 1.05% of the total value of the trans-
action on the SHSE. This includes a brokerage fee of 0.7% of the gross
consideration (with a minimum trading fee of renminbi 5), a transfer fee
of 0.1% of face value (minimum US$1), stamp duty of 0.3% of the gross
consideration, and a clearing fee of US$4 per execution for individuals and
corporations (US$8 for custodians). There is no registration fee. The trad-
ing cost is 1.00% of the total value of the transaction on the SZSE, a bit
less than the SHSE.
According to China's securities regulations, the listed securities in the
SHSE and SZSE must be traded through the centralized competitive method.
The securities exchanges are entitled to formulate trading rules which be-
come effective after approved by the State Council's. Under these rules,
investors who want to participate in the centralized competitive trading
must follow the procedure of opening accounts, entrustment, conclusions,
clearing and delivery, and transfer of accounts.
12
To trade in the SHSE and SZSE one must open separate securities
and cash accounts. At the end of a trade, adjustments on the amount of
securities and cash will be made in those accounts. Investors must apply
for accounts directly through the registration department of the securities
exchanges or their securities dealers. However, both domestic and foreign
investors may entrust domestic and foreign securities dealers or agencies to
open an account of renminbi special share. The capital account is issued
for deposit of cash, clearing and receiving dividend. It must be opened by
the investors at the selected securities dealer's business place. The cash
in the account is deposited in a bank by the dealers as current deposit,
and the interests will be automatically put in the account. Investors have
a magnetic card and a deposit card for their securities account and cash
1 2
The method of non-centralized competitive trading refers to the process under which
investors trade their securities through buying and selling, bid or auction in the rec-
ognized markets outside the exchanges. Non-centralized competitive trading markets
include securities dealer's business counters, local securities exchange centre and secu-
rities offering price system.
Market Micro structure 63
account.
In buying and selling securities, the commission principal (investor) shall
fill in the commission sheet. The commission sheet is printed by the dealers
according to the requirements set forth by the exchanges. The commission
sheet shall list the name of the commission principal, shareholder's code
number, the date and time of commission, the kind of securities, the volume
and face price, the commission price, effective date, etc.. The commission
may be made through written form, by telephone, fax, cable, etc.. In
Shanghai and Shenzhen, securities dealers are prohibited from accepting
the full commission which delegates the dealers with the power to decide
the kind, volume and price of the traded securities. Securities dealers are
also prohibited from accepting the commission of trust trading.
Although both the SHSE and SZSE use computer systems to reach con-
clusion through centralized competitive price, they have some differences
carrying out orders. On the SHSE, the representatives of the dealers, who
sit in the exchange hall, are the actual operators who make the conclusion
through the computer system. On the SZSE, despite that there are rep-
resentatives of the dealers who sit in the exchange, the actual operators
are the dealers themselves who may enter the orders from their business
place outside the exchange through the computer system. The centralized
competitive price method adopts the principle of price priority and time
priority.
After conclusion, the buyer and seller shall pay cash and securities
through their dealers. This process includes two steps: In the first step, the
volume and value of the same stock will be calculated through an affiliated
or independent clearing house. The dealers will only pay the net different
value. In the second step, stocks are delivered and payments are cleared
between investors and their dealers. Because of the centralized preserva-
tion and non-paper trade adopted by the SHSE and SZSE, the transfer of
accounts on the underlying securities is made through keeping changes in
accounts during the day and completing the transfer automatically via the
computer system at the end of the trading day. In August 1995, the SZSE
and the Securities Registration Corporation have adopted a unified securi-
ties account card which preserved transfer, distribution and subscription of
shares, unified share statements, and reduced transaction costs.
13
1 3
On the SHSE and SZSE, B-shares are traded in paperless form using an automatic
transfer and computerized system. Buy/sell orders are matched automatically by
2.5.3 Takeov er of Listed Companies
The P.R.C. Securities Law provides that listed companies may be taken
over by means of takeover by offer or takeover by agreement.
When an investor has acquired, through securities trading at a stock
exchange, five percent of the issued shares of a listed company, the investor
must, within three days after the five percent threshold shareholding is
reached, submit a written report to the CSRC and the stock exchange,
notify the listed company, and make a public announcement. During the
initial three-day period, the investor may not purchase or sell shares of
the listed company being acquired. Thereafter, this reporting obligation is
triggered every time the shareholding of the investor increases or decreases
by an amount equivalent to five percent of the issued shares of the listed
company.
After the investor has acquired, through securities trading at a stock
exchange, thirty percent of the issued shares of a listed company, and if
the investor continues to acquire shares of that company, the investor must
make a takeover offer to all the shareholders of the listed company, unless
the CSRC grants an exemption. The offer period must last for at least
thirty days but not more than sixty days. During the offer period the offer
must remain irrevocable.
If upon expiration of the offer period the number of shares in the target
company that has been acquired by an investor exceeds seventy-five percent
of the total shares outstanding of the company, the listing of the target
company must be terminated. If this number has exceeded ninety percent
of the total shares outstanding of the target company, the shareholders who
have not yet accepted the offer will have the right to sell their shares to the
investor on the same terms as those of the offer and the investor must take
up those shares.
Where a takeover involves shares held by state-controlled enterprises,
approval must be obtained from the relevant authority in charge. In the
case of takeover by agreement, the investor may acquire shares from share-
holders of the target company by entering into an agreement with them.
The investor must, within three days after the agreement is reached, submit
computer, based on price priority followed by time priority, and confirmation is relayed
to investors on the same day. Foreign brokers with special exchange seats are permitted
to execute their orders directly, without passing through local brokers as they used to
before 1993.
Market Microstructure
65
a written report on the agreement to the CSRC and the relevant stock ex-
change, and make a public announcement. The takeover agreement cannot
be completed until the announcement has been made.
It is important to realize that determining the extent of application of
the "takeover by agreement" option under the current Securities Law is
difficult and involves at least the following problems. The Securities Law
seems to suggest that the initial thirty percent stake of an investor in a
target company must be built up by buying shares incrementally at the
relevant stock exchange. This interpretation of the law is consistent with
the more fundamental requirement that securities that have been approved
for listing and trading must be quoted and traded on stock exchanges. This
begs the question: Does the "takeover by agreement" option override this
requirement and make it possible for shares of a target company normally
required to be traded on a stock exchange to be acquired by agreement by
an investor intending to take over the company? If the answer is no, does
it then mean that the option extends only to the acquisition of domestic
shares that are not required to be publicly traded, e.g. the Chinese legal
person shares? If indeed the application of the option is restricted to non-
publicly traded shares, is the option available from the very beginning when
the investor starts to build up its stake, or is it available only after the
investor has acquired at least a thirty percent shareholding by buying shares
at the relevant stock exchange and has then obtained from the CSRC an
exemption from the obligation to make a general takeover offer? These
questions await to be resolved in the Securities Law.
With regard to the takeover of a company that has, in addition to
domestic shares, issued B-shares or H-shares, the Securities Law does not
provide an explanation on how enforce the takeover provisions. Because
of ownership restrictions, an investor cannot buy and hold both domestic
shares and B- or H-shares simultaneously, the specific requirement on an
investor to make a takeover offer to acquire all the shares in the largest
company is simply not feasible. A similar problem exists with regard to the
state-owned shares of a target company as those shares cannot be purchased
by private investors. The detailed Implementing Securities Regulations to
be formulated under the P.R.C. Securities Law will need to address these
problems.
2.5.4 Settlement, Clearing and Payment System
The securities exchanges and trading centers do not utilize a system of spe-
cialists as is done on the New York and American Stock Exchanges. Rather,
orders are transmitted through a broker to a clerk on the exchange floor.
The clerk then inputs the order into a central computer which matches
buy and sell orders. Customer accounts are held by the stock exchange,
rather than brokerage house. While the brokerage house opens the account
and executes order for customers, the exchange physically maintains the
account and ensure settlement and clearing of accounts.
14
Seeking to comply with the Group of Thirty (G30) Standards, Hong
Kong Securities Clearing Company Limited now operates the Central Clear-
ing and Settlement System (CCASS) as the central depository body for the
settlement of trading. The CCASS provides T+l matching for brokers who
trade and requires T+2 affirmation of settlements by custodian banks. In
the SHSE, settlement time for A shares is now three days (T+3) but for B
shares remains T+l . Turnaround trading time for A-shares on the SHSE
is the same day. The SHSE adopts only the qualified DVP/RVP (delivery
versus payment/receipt versus payment) model. It uses a central depository
system through a company called the Shanghai Securities Central Clearing
and Registration Corporation (SSCCRC), which also serves as a clearing
house. However, there is so far no payment in same-day funds. The SHSE
is now updating its payment system to keep its technological superiority.
The new system, known as the central Real Time Gross Settlement pay-
ment system (RTGS), will promise more risk reduction possibilities for the
clearing houses, such as payment clearing facilities through RTGS system.
The SHSE promises to be more technically capable in securities trading in
the future.
The SZSE is more or less in line with its big brother exchange in Shang-
hai. The clearing and depository house is called the Shenzhen Securities
Central Clearing Company Ltd.(SSCC). The settlement period is T+3 for
both A- and B-shares. The securities settlement and income, paid in Hong
Kong dollars, can be freely repatriated. Short-selling and over-the-counter
14
Thi s is no margin trading or short selling in Chinese securities markets. It normally
takes 5,000 renminbi to open an account. Large investors, with accounts generally over
500,000 renminbi, have greater access to information available through the brokerage
houses and have greater ability to execute trades. Small investors, who constitute the
bulk of the market in China, generally trade in one hundred and two hundred share
lots.
trading are illegal. Market entry requires obtaining a unique foreign in-
vestor code prior to trading.
2.5.5 Accounting System
China employed an accounting system established in the centrally planned
economy for more than 40 years. This old accounting system adopted
different accounting rules for different types of ownership (state-owned,
collectively-owned and privately owned) since the 1950s. There was no law
on accounting rules until 1985 when the National Congress passed its first
Accounting Law, which unified different accounting practices.
15
Since then,
several major modifications have been updated to adapt to the market-
based economy in the country but there are still a number of significant
differences with the standard international accounting practice.
Because securities markets had developed rapidly from the beginning of
the 1990s, especially from 1992 when the government decided to establish
the socialist market economy and transform the SOEs. The influence of
stock markets on Chinese accounting practices was manifested by changes
in accounting objectives stated by the Ministry of Finance. More and more
people from both the accounting profession and outside support the idea
of providing accounting information to investors, creditors and the public.
In order to provide better information about profitability, financial perfor-
mance and financial stability of listed companies, the SHSE transformed
the 1992 annual financial reports of the earliest 8 listed companies into re-
ports in accordance with international reporting standards. The financial
reports of these 8 listed companies were originally produced from different
Chinese accounting regulations, because the companies belonged to differ-
ent industries. This provided practical experience for the standard financial
reporting that took place later in 1993. At the same time, China issued the
first B-share stock on the SHSE in 1992 and the first H-share stock in the
Hong Kong Security Exchange in 1993. Companies which issued B- and
H-shares were required to publish accounting reports which were not only
15
For instance, the Accounting Law unified three methods of book-keeping. The
debit/credit system was typically used in state owned industrial enterprises, joint ven-
tures and international trade companies. During the Cultural Revolution this system
had often been replaced by the increase/ decrease method and this continued to be
used by many commercial enterprises. Many banks, some collective enterprises and
budgetary government units used the receipt/payment method.
in accordance with Chinese accounting rules, but also in accordance with
international accounting standards. This gave Chinese accountants more
opportunity to exchange ideas with accountants from outside the country.
Against this background, the Ministry of Finance issued An Accounting
Regulation for the Experimental Listed Company in July 1992. The accoun-
tants in listed companies were required to follow international accounting
practice and to use the double-entry method in book-keeping. The implied
premises and principles in the named regulation provided a basis for the
Basic Accounting Standards for Enterprise issued in November 1992 fol-
lowed by 13 new accounting regulations for different industries. This was
a remarkable change to unify accounting systems used by different indus-
tries and to move financial accounting forward in line with international
accounting practice.
16
The Basic Accounting Standards is the first accounting standard in Chinese account-
ing history. It consists of 10 chapters: general standards; general principles; assets;
liabilities; owners equity; revenues; expenses; profit; financial reports and appendix.
The basic accounting standards have simplified over 40 old regulations for different
ownership and different industries. Particular aspects of the new accounting system
include: (1) The government sets the basic and special accounting standards, then sets
the practical accounting regulations on the basis of the standards. (2) Accounting en-
tity, going concern, periodicity, and monetary cost measurement are the underlying ac-
counting premises (concepts) in the basic accounting standards. (3) A series of general
accounting principles have been established in the standards. These are objectivity,
relevance, comparability, consistency, timeliness, clarity, materialism, accrual, match-
ing, prudence, historical cost and differentiation of revenue and capital expenditure.
These 12 principles are used as the criteria for guaranteeing the quality of accounting
information and provide the means for determining and measuring economic trans-
actions. (4) The terms: assets, liabilities, owners equities, revenues, expenses, and
profits are defined as accounting elements. Consequently, the old accounting equation
has been replaced by the new accounting equation: assets = liabilities + owners equi-
ties. The new accounting system is able to provide more relevant information about
enterprises proprietorship, solvency, financial adaptability and profitability. (5) The
formats and components of financial statements have been standardized. The new
external financial reporting system includes the balance sheet, the income statement
and the statement of financial position. Therefore, the financial reports are able to
provide more relevant, comparable, understandable and complete financial informa-
tion to the readers. (6) The capital maintenance principle has been stressed to protect
the investors equity. Capital accounts including issued capital stock have been set
up to record and protect investors equity. Realized gains or losses on assets and de-
preciation of long-term fixed assets are included in the profit and loss account but
these no longer increase or reduce investment (different kinds of funds provided by the
government under the planned economy) as before. This makes property rights and
investors equity clear in the accounts. (7) The total cost method has been changed into
the manufacturing costing method in accordance with internationally applied costing
Market Micro structure
69
China's accounting reforms have been going deeper since 1993 through
the efforts of the Ministry of Finance and accounting professionals. The
new accounting standards are being improved and developed into a fully-
fledged accounting system that meets the increasing needs of the socialist
market economy. The transformation of SOEs and the development of se-
curities markets have further influenced the accounting reform. Since 1992,
more and more SOEs have been transformed into joint-stock companies and
listed on the stock market. Many practical accounting issues such as how to
account for mergers and reorganization, valuing SOEs, goodwill and other
intangible assets have arisen and required research. On the other hand, the
increase in public stock holders and the growth of the stock market require
the listed companies to provide true, fair and reliable accounting informa-
tion to the users of financial reports. Thus the state Security Supervision
and Administration Commission issued the Detailed Practical Rules and
Regulations of Disclosing Information for Listed Company in June 1993
and the Format and Content of Annual Reports and Mid-term Reports in
1994. These regulations further standardized accounting practices and im-
proved the quality of financial information.
By the end of 1998, the Ministry of Finance had issued 8 specific ac-
counting standards, including standards for cash flow statement, enterprise
investment, rearrangement of liabilities, revenues, construction contracts,
post-balance sheet transactions, income statement, and accounting policy,
accounting estimation and correction of accounting errors. The Account-
ing Standards for Enterprises: Cash Flow Statement has been put into
practice in the listed companies from the accounting year of 1998 and the
others were implemented in listed companies from January 1, 1999. All
these have firmly proved the determination of accounting reform by the
government. Along with the diversity of investors and the increase in the
number of shareholders, transparent, open, complete and comparative ac-
counting information together with external auditing have become more
and more important.
Accounting reform and improvements in accounting information have
provided favorable conditions for economic transformation and develop-
ment of the capital market in China. Disclosing accounting information on
practice. (8) The three methods of book-keeping have been unified as the international
double-entry method. (9) Enterprises have been allowed to use the prudence principle.
This permits enterprises to use accounts receivable and the last-in-first-out method in
inventory pricing.
stock market listed companies has become an important issue because of
its significant influence on the security market. There is strong evidence
that disclosed accounting profits have conveyed information to the market
with subsequent effects, which indicates a kind of rational investment power
based on the nature of capital in the Chinese securities markets.
2.5.6 Taxes and Fees
The main goals of the financial reform include expanding the share of cen-
tral government's fiscal income, transforming state specialized banks into
profit-making and risk-taking institutions, and strengthening the role of
the central bank in issuing money, managing basic monetary instruments,
and regulating and controlling total credits and interest rates. The new
tax reform system that became effective in 1994 aimed at unifying the tax
code, simplifying the tax system and its administration, and strengthen-
ing the overall control capability of central Government by enlarging the
central-local tax revenue ratio of 45:55 in 1992 to 60:40, a pattern more in
line with that in other countries.
17
The old tax (consolidated industrial and commercial tax) was replaced
by a new value-added tax at 17 percent standard rate supplemented by a
consumption tax (mainly for luxury goods) and a business tax (mainly for
the service sector). A single personal income tax was also introduced. The
tax rate for all enterprises was equalized at 33 percent with some special
treatment allowed-27 and 18 percent respectively for the first two years for
high-tech companies. Along with tax reforms there were also reforms in
accounting and auditing systems as mentioned above.
There are two types of securities taxes for Chinese investors: the with-
holding tax on dividends and the capital gains tax on securities traded.
Withholding tax is imposed on interest, rent, royalties and capital gains
received by non-residents. The statutory rate of withholding tax is 20%.
17
Chi na has implemented a new tax law and regulation system from January 1, 1994.
The new tax law has made big changes in major taxes such as value-added-tax and
income tax. The related accounting regulations on purchase costs, revenue recording,
accounting for tax payable, income tax accounting and the contents of the income
statement were adjusted in response to the new tax law system. Accounting practice
adapts to tax changes but accounting keeps its objective as providing information
to investors, creditors and the public rather t han serving t he government only. The
accounting st andards and regulations have maintained some independence from the
t ax law system, though accounting practice submitted to taxation in t he past decades.
It is reduced to 10% for income derived from specially designated zones in
China. It could also be reduced under double tax treaties. Dividends from
a foreign investment enterprise to a foreign shareholder are exempt from
tax. In Hong Kong, dividends are treated as having been paid out of taxed
profits and are thus not subject to further taxation. In particular, there
are no Hong Kong taxes imposed on foreign shareholders receiving Hong
Kong dividends, and there are no withholding taxes on dividends paid by
Hong Kong companies.
In Shanghai and Shenzhen, buyers and sellers are subject to as much
as 0.3% securities trading tax based on the value of trade. The securities
trading tax must be paid or deducted by the clearing and settlement firm.
In addition to securities trading tax, individuals have to pay 20% income
tax for securities dividend they received according to the Individual Income
Tax Law of the P.R.C.. Enterprises have to pay additional income tax for
the dividend they receive according to the Income Tax Law of the P.R.C.
Concerning Enterprises with Foreign Investment and Foreign Enterprises,
and to the Enterprises Income Tax Provisions. However, there is no capital
income tax in Shanghai and Shenzhen.
Securities trading fees are the fees the trade participants shall pay in
accordance with securities regulations and rules made by the exchanges.
There are two types of securities trading fees: listing fees and service fees.
Listing fees include an initial listing fee payable in sum and a monthly fee.
Both types of fees are paid by the issuer to the exchanges. According to
the rule of the SHSE, 0.03% and 0.1% of the issued value is charged for A-
and B-share listings, respectively. Initial listing fees must be paid no later
than 3 days before listing. The monthly fee is 0.001% of the issued value
for A- and B-shares, and must be paid from the next month of the listing
day until the monthly listing is terminated.
Service fees are paid by investors and consist primarily the commission
paid to dealers. In the SHSE, the standard commission is 0.4% and 0.6% of
the value of trade with a minimum of 10 renminbi and 20 U.S. dollars for
A- and B-shares, respectively. The commission may be reduced for B-share
transactions if the trading volume is large. In the SZSE, both the buyer
and the seller must pay 0.35% of the value of trade to their dealers as the
commission with a minimum of 5 renminbi for A-shares. The commission
is 0.6% of the value of trade with no minimum service fee for B-shares.
There is no commission if the stocks are not trade according to the buyers'
or sellers' orders.
18
In addition, member dealers of an exchange have to pay an entry-place
fee which includes the fees paid by both dealers to the exchange according
to a certain rate after the conclusion of the trading. The fee is generally
0.1% of the value of trade for A- and B-shares plus an account transfer fee
in the SZSE. The fee is 0.03% of the value of trade in the SHSE.
Appendix B summarizes different types of taxes in China.
2.5.7 Suspension and Termination of Listed Companies
After a stock is listed, it is supervised by the securities organs in charge and
the exchange. When a listed company fails to meet the listing requirement
because of the change of social economic conditions or its own business
difficulty, its qualification for listing will be cancelled and all future trades
will be terminated. According to Article 157 of the P.R.C. Company Law,
the events which cause suspension of a listed stock include:
(1) a company's total capital and share distribution has been changed
to an extent below the standard of listing requirement;
(2) a company fails to disclose its financial situation or produces a false
accounting statement;
(3) a company is involved in some serious illegal activity;
(4) a company has made continuous losses for the previous three years.
Article 158 of the Company Law stipulates that when either condition (2)
or (3) occurs and the investigation proves, the circumstances are especially
serious. When either condition (1) or (4) lasts longer than the required
period of time, the listed stocks shall be terminated.
In addition to the above regulations, the SHSE and SZSE require that
if a listed company plans to reorganize, to replace old stocks with new
ones, to provide additional shares for raising more capital, or to distribute
special share dividends, it must inform the exchange for suspension of the
listed stocks. Suspension of a listed stock cannot exceed a certain period
18
Securities trading fees cover the fees paid by the listed companies, dealers and investors.
In the past, securities trading fees and service fees charged by the SZSE were higher
t han those by t he SHSE. But since October 4, 1995, t he fees have been reduced by
approximately 60% on the SZSE. In fact, the service fees charged by the SZSE is now
12.5% lower t han t hat by the SHSE. Moreover, the SZSE has adopted a number of
measures to reduce the securities fees for B-share transactions.
Market Micro structure 73
of time.
19
When the aforementioned activities have been concluded, the
listing for the company must be renewed.
Termination of a listed stock happens when the listed stock fails to
meet the minimum listing requirement or when suspension time is due but
the causes of suspension have not disappeared. Termination of a company
is determined and approved by the relevant securities authorities. The
following events may cause suspension and termination of a listed stock:
failure to meet the minimum requirement to maintain its listing,
e.g., grossly violating the share distribution procedures, keeping
very little capital, or having a very small trading volume in the
listed exchange;
the listed company is involved in some illegal activity;
a listed company has serious business problem, e.g., a listed com-
pany incurs continuous losses and is faced with bankruptcy because
of high financial distress or a listed company is declared bankrupted
by the court which leads to the termination of the listed stock;
other illegal activities such as failure to pay the listing fee.
The Shanghai and Shenzhen Exchanges are entitled to make decision of
suspension and termination, and the decisions shall be filed or approved by
the securities organ in charge. Since the Company Law became effective
in 1994, suspension and termination of a listed stock based on legal events
shall be made by the securities organ in charge, and be enforced by the
stock exchanges. Suspension and termination of the listed stocks based on
the situation under which the companies are dismissed, ordered to close,
or declared bankrupt shall be decided by the security organ in charge.
After termination of a listed stock, the stock may still be traded at a non-
concentrated competitive trading place authorized by the law.
According to the rules and regulations of the SHSE, the suspension for a listed company
cannot exceed nine months.
Chapter 3
Risk, Ret urn and Regulation in
Chinese Stock Markets
3.1 Introduction
In the realm of financial studies, the concept of risk and return as well
as the relationship between risk and return is of primary interest. When
studying financial markets, the notion of risk and volatility are synonymous.
Risk (volatility) is the key parameter of interest in studying option pricing,
risk management, portfolio management, the Capital Asset Pricing Model
(CAPM), and the Arbitrage Pricing Model (APT). Empirically, it has been
observed that the level of risk of a financial asset changes over time. In this
chapter, we follow Su and Fleisher (1998a, 1999b) in examining the risk and
return behavior in Chinese stock markets. We will pay particular attention
on how changes in government regulations affect stock-market volatility.
Understanding stock-market risk and return behavior is very impor-
tant, because in a market consisting of risk-averse investors, greater volatil-
ity (risk) would lead investors to demand a higher risk premium, creating
higher cost of capital, which impedes investment and slows economic de-
velopment. We investigate the following questions:
(1) What characterize the risk and return behavior in Chinese stock
markets?
(2) How well do Chinese stock markets function relative to other de-
veloped and emerging markets?
(3) Can we predict stock-market returns using local and global infor-
mation variables?
(4) What are some important causes of the high stock-market return
volatility?
75
76 Risk, Return and Regulation in Chinese Stock Markets
(5) Are there causality relationship and volatility spill-over effects be-
tween Chinese stock markets and stock market of similar economic
and geographical conditions?
In Section 3.2, we lay out the basic pattern of returns and volatility in
Chinese stock markets. In Section 3.3, we explore whether the distribution
of stock returns is identical for all days of the week. In Section 3.4, we
investigate the market efficiency hypothesis using variance ratio tests and
residual-based cointegration tests. In Section 3.5, we establish an empirical
model that captures the effects of local and global information variables
on the conditional mean of weekly stock-market returns. We characterize
the conditional variance of returns using three alternative formulations of
the error-generating process. In Section 3.6, we select the best formulation
of the empirical model and report the econometric results. We then apply
the best-fitted model to examine the effect of exogenous government policy
variables on stock-market return volatility in Section 3.7. In Section 3.8,
we examine volatility asymmetry and how the volatility of Chinese stock
markets is related to Hong Kong market. We conclude the chapter with a
summary of findings in Section 3.9.
3.2 Stock-Market Return and Volatility Pattern
As we have mentioned earlier, there are two categories of shares traded
in the stock markets in mainland China: A-shares are available only for
Chinese citizens and B-shares can be purchased only by foreign investors.
Both A- and B-shares are traded in the two official exchanges in China,
Shanghai and Shenzhen. A-shares dominate Chinese stock markets in terms
of market capitalization and level of activity. (The number of B-shares
listed on the two exchanges is less than one-third of the number of A-shares.
B-shares amount to less than 5% of the A-shares' market capitalization and
4% of the A-shares' annual trading value from 1992 to 1998.) Additional
categories of shares include H-shares and N-shares, which are available only
to foreign investors and are traded in the Hong Kong Stock Exchange and
New York Stock Exchange, respectively. The analysis in this chapter is only
based on the A- and B-shares data from the first-day of market trading until
December 31, 1996.
We first analyze stock-market volatility and returns to investors using
Stock-Market Return and Volatility Pattern 77
170*
w
C
% 140K
+
not
11
"m 80*
50V
- 1 0 *
- 4 OK
"H^V^fM^
- Shanchai
Shenzhen
WAv
910430 920330 930227 930930 940430 941031 950930 960930
Fig. 3.1 Weekly Returns for A-share Indices in Shanghai and Shenzhen
data of daily stock market indices for A-and B-shares for both the Shanghai
and Shenzhen securities exchanges. We relate trading in these shares to
daily market indices representing more mature stock markets, including
indices of the MSCI world market, the NYSE and the Hong Kong Stock
Exchange. The daily market indices are based on value-weighted portfolios
of securities and do not reflect dividends.
Figures 3.1 and 3.2 plot time-series of daily stock-market returns for
A- and B-shares in Shanghai and Shenzhen. Table 3.1 presents some sam-
ple distributional statistics for the stock-market indices included in this
chapter. Statistics include:
(i) daily and weekly risk-unadjusted sample mean returns, which are
computed as
r
i,t
ht
Ij,t-i
h
t-i
where Ij
tt
is the value of jth market index at time t and Ij,t-i is
the value of j t h market index at time t 1;
(ii) Sharpe ratio, which is the mean excess stock return divided by the
sample standard deviation;
3 OK
Shanghai
Shenzhen
-15K
921210 930630 940215 940930 950515 951231 960815 970331
Fig. 3.2 Weekly Returns for B-share Indices in Shanghai and Shenzhen
(iii) coefficients of skewness, which is
T
2
T ^-it=i\
r
Ji
t r
i)
( r- i ) ( r- 2)
3/ 2'
T-l Z-*t=lVJ,t Tj)
(iv) coefficients of kurtosis, which is
T(T + 1)
t
r
=
i f a, t - rj)
4
~ ^ \TLi(r
jlt
- fj)
2
(T - 1)(T - 2)(T - 3) [ ^ Ei=i(rj,t - ftf
(v) Ljung-Box portmanteau statistics, which is
-; and
T(T + 2)
M 2
T
,m~l
T-m
where j
>m
is the m-th order autocorrelation of the residuals for the
j - t h time series. M is the number of autocorrelation used.
The coefficients of skewness and kurtosis are jointly estimated with the
mean and variance. The Ljung-Box Q(12) statistic is used to test the
significance of serial correlation up to lag 12.
We find that:
(1) Mean returns on A-shares in Shanghai and Shenzhen securities ex-
changes are higher than those of U.S., Hong Kong and MSCI world
indices, but the standard deviations of returns are also higher. Risk-
adjusted mean returns (Shape ratios) are lower in Chinese stock
markets.
(2) Mean returns for A-shares in the Shanghai Securities Exchange
have been higher than in the Shenzhen Securities Exchange.
(3) Returns on B-shares have been lower than those on A-shares for
both exchanges.
1
(4) The skewness parameters are significantly positive, indicating that
the stock-market returns are not symmetrically distributed.
(5) Coefficients of kurtosis are generally higher in Chinese stock mar-
kets than those in more developed equity markets, suggesting that
"big surprises" are more frequently observed in Chinese markets.
Next, we characterize volatility and risk patterns in Chinese stock mar-
kets. We calculate estimates of the variance of daily stock returns over the
entire sample period. Important features of the markets' volatility (as can
be shown in Figures 3.1 and 3.2) include:
(1) The volatility of stock-market returns is higher in China than in
other developed markets.
(2) A-share markets in Shanghai and Shenzhen exhibit far greater volatil-
ity than B-share markets.
(3) Extreme price volatility exists in both exchanges. For example, the
Shanghai A-share index more than doubled in a single day, from
636.56 on May 22, 1992 to 1341.11 on May 23, 1992; by Novem-
ber 17, 1992, it had fallen by over 70%, to 369.94. The Shanghai
A-share index rose from about 400 in July 1994 to more than 1,000
in September.
(4) The size of price "jumps", measured by the percentage change in the
price index, are smaller for Shenzhen A-shares than for Shanghai
A-shares.
1
International investors who bought heavily into B shares in the first two years after
B-shares were listed and traded in 1992 have become disillusioned by t he low returns
and have started moving their investment elsewhere.
R
i
s
k
,

R
e
t
u
r
n

a
n
d

R
e
g
u
l
a
t
i
o
n

i
n

C
h
i
n
e
s
e

S
t
o
c
k

M
a
r
k
e
t
s

Kong
b
O

a

o

Seng
b
O

C

o
S

K

K

H

C
O

^

U

C
O

c

c
o

J
3

I
S
I

c

C
D

_
C

c
o

c

C
O

J
3

N

G

C
o

C
O

"
3

J
S

b
O

c

J
3

C
O

'
3

M
l

C

c
3

J
3

C
O

x

c
o

-
a

a

2

C
D

c
f
l

4
3

w

P
Q

C
D

[
-
1

c
6

X
!

C
O

<

0
3

c
<
3

e
n

m

0
)

c
3

J
3

w

<

K

v
o

O
*
*

O
"
*

Q

r
H

o

r
t

"
^
.

i
n

_
:

0
0

j

O
O
I
O

1

C
D

1

C
O

C
N

&

v
p

^
O

t
>
*

v
p

e
n

-

C
O

^

B
^

i
n

i
N

S

r
H

O

C
I

t
o

"
*

-

T
J
.

r
t

"

-
o

c
^

-
^

^

*

0
5

*

0
1

C
D

m

1
0

?

.
-
i

6

r
H

r
-
t

C
N

0
}

o

H

n

H

N

q

6

n

c
i

6
S

H

O
^

*

*

S
o

0
0

*

r
H

b
-
.

i
-
H

t
-

O

i
-
H

?

*
=

6
?

&

6
?

K
6
S

C
O

I
O

O

C
O

2

2

r
t

S
;

1
I

1

1

1
I

r
H

C
O

O

t
-
H

C
O

C
M

^

O

*

i
-
H

1

0
0

*

*

.
-
H

I
*
-

C
N

?

C
O
-
C
O

C
O

74*
C
O

I
O

C
N

n
c
a
e
V

r
t

C
O
C
O

"
?

0

i

O

i
-
<

C
O

>
1
1
1

c
p

K

3

3

T
3

6

C

3
2

O
1

x

o

C
O

U
C

C

3

Q
c
o
c
o
^
i
S
c
o
^
h
-
T

l
O
t
>
.
C
O
X
j
C
O
N
n
H

O
O
^
O
O
^
C
O
O
O
^
O
l

"

e
n

o

.
1
0

O

C
O

T
-
H

T

r
H

O

O

-
-
I

c
o

i
>

S
o

0

c
o

*

*
-
C
M
C
M
t
N
.
l
C
.
C
S
I
*

C
O

*

O

-
C
O

O
f
-
t
O
i
i
^
i
i
-
H
i
-
f

c
o
c
i
n
r
i
!
c
o
3
i
n
o
o

<
M

<
-
H

4
.

r
f

.

o

-
:

^
J
*

C
M

O
r
H
r
H
|
C
0
l
O
r
H

f
t
?

^

f
c
?

^

V
.
O

>
.
0

|
>
_

O
^

1
0

S
^

6
s
-

r
n

c
o

*

*

o

0
1

.

c
s

0
1

H

?

^

*
.

^

^

q

Q

O

I
O

C
N

1

C
N

C
M

r
H

0
s
-

v
Q

v
O

|
s
_

t
o
"
"
"
-
C
N
S
.
^
>
-
,
C
O
*

*
-
-
m

m

N

~

-

'
-
-
o

s

t

C
N

t
~

v
i

t
o

T
?

I
O

C
N

r
H

r
H

H
t
>
r
t

H

(
D

f
f
l

0
0

C
N
C
O
Q
A
.
C
O
r
H
O
O
C
O

r
H

O
S

^

-
^

^

C
N

K
_

O

*

C
O

C
N

:

c
o

c
o

1
0

g
S
C
O

1

r
H

1
C

C
N

C
O

o

E

E

S

3

3

.
S

f
f
l

9

a

c
o

'
T

h
o

C

3

1
1

'
r
t

-
*

Q
,

.

C
D

*

S
*

S

T
J

2

I
!

S
S
c
S

*

Third, we use the daily values of A- and B-share market indices to
estimate the monthly variance of stock market returns in Shanghai and
Shenzhen. The monthly variance estimates use only non-overlapping sam-
ple returns in that month, and allow time-varying volatility change.
Because of the existence of autocorrelation, we follow Merton (1980) in
estimating the variance of the monthly return for the A and B share market
indices as the sum of the squared daily returns plus twice the sum of the
products of the adjacent returns,
N
t
N
t-i
a
lt = Y,
r
lt +
2
J2
ri
't
ri
+i>
t
> (
3
-
x
)
=i t = i
where there are iVj daily returns, r ^ , in month t. I do not subtract the
sample mean from each daily return in calculating the variance because the
adjustment is very small.
The monthly standard deviation estimates for A- and B-share returns
in Shanghai and Shenzhen highlight the variation of estimated volatility
in China's stock markets. The major features of the monthly standard
deviation estimates include:
(1) There is strong evidence of time-varying volatility change in both
share categories in both exchanges.
(2) Periods of high and low volatility tend to cluster, and volatility
shows mild persistence.
(3) "Big surprises", or high return volatility, are often observed in Chi-
nese stock markets.
(4) The mean and standard deviation of the stock-market return stan-
dard deviation estimates are higher for A-shares than for B-shares.
We briefly summarize the price and return behavior of Chinese stock
markets as follows:
Risk-adjusted returns to investors are low;
The degree of variability of these returns as evidenced by a conven-
tional measure of volatility is very high;
Volatility is time-varying, leptokurtotic and is mildly persistent.
Volatility of A-share markets is far greater than that of B-share
markets in both Shanghai and Shenzhen;
82
Risk, Return and Regulation in Chinese Stock Markets
3.3 Day-of-the-Week Effect
Because Monday's return is calculated over three calendar days rather than
one single trading day and unfavorable information tends to be released over
the weekend, French (1980) and Rogalski (1984) found significant negative
average returns for Monday using the S&P 500 and Dow Jones Industrial
Average (DJIA) index, respectively.
2
Since then, much research has uncov-
ered day-of-the-week effect for a number of countries in Europe and Asia.
For example, Japanese indices show negative returns on Tuesday, London
Stock Exchange and the Paris Bourse exhibit negative returns on Mon-
day, and Canada, Hong Kong, and Switzerland display negative returns on
Monday and Tuesday. It appears that the day-of-the-week effect is related
to market institutions and trading mechanisms (e.g., the extent of non-
synchronous trading), and therefore is likely to differ across countries. In
this section, we examine whether the distribution of stock returns is identi-
cal for all days of the week for Chinese markets. We do so with and without
adjusting for the seasonal component of each day-of-the-week return.
3.3.1 Analysis of Variance Approach
Assuming that daily stock returns are normal and identically and indepen-
dently distributed (i.i.d. normal), we can test the day-of-the-week effect us-
ing simple analysis of variance approach. Our data consist of daily values of
A- and B-share market indices in Shanghai and Shenzhen from December 6,
1992 to December 6, 1996. Daily returns are computed as follows:
In
r
jtt
= 100 x In ^ ~
Ij,t-\
where Ijj is the value of j t h stock-market index (j=Shanghai A-, Shanghai
B-, Shenzhen A-, and Shenzhen B-share market) at time day t. Monday re-
2
Cross (1973) first documented t hat the performance of stocks, as measured by mean
returns, was significantly higher on Fridays t han on Mondays. More specifically, t he
average return for Monday, calculated from Friday' s closing price to Monday' s closing
price, was negative. French (1980) examined t he day-of-the-week effect using both
calendar-time and trading-time models. He concluded t hat the negative returns from
Friday close t o Monday close were primarily incurred during t he non-trading period,
i.e., weekend effect. Intraday studies of stock price behavior, such as Harris (1986a),
show t hat aggregation, while adding to generality of t he results, reduces the precision
necessary to infer a particular price behavior.
Day-of-the-Week Effect 83
turns are calculated as the difference in logarithmic prices between Tuesday
opening and Monday opening.
The equality of daily mean returns is tested with Hotelling's T
2
-statistic
where // is the multivariate normal vector of daily returns. Hotelling's
T
2
test is the most common, traditional test where there are two groups
formed by the variables that are homoskedastic and linearly independent.
If the overall F-test shows the centroid (vector) of means of the dependent
variables is not the same for all the groups formed by the categories of the
independent variables, tests of group differences are used to explore the
nature of the group differences. Significance tests for multiple dependents
(e.g., Hotelling, Wilks, or Pillai tests) all follow the F distribution and so
an F value and corresponding significance level are printed out for each
of these tests in SPSS, in the section labeled "Analysis of Variance". The
following joint hypothesis is tested:
H
0
: Hi = ^2 = A*3 = Hi = A*5
H
A
: Hi ^ Hj,Vi,j 1,2,- , 5 and i j
where fii is the expected return on day i.
As shown in Table 3.2, the day-of-the-week effect is significant at the
5% level for all four stock indices. In particular, all four share indices have
negative returns on Monday and positive returns on Tuesday, Wednesday
and Thursday. Shanghai A-share and Shenzhen A- and B-share markets
show positive returns on Friday while Shanghai B-share market exhibits
negative returns on Friday. The day-of-the-week effect is statistically sig-
nificant at the 1% level for A-share markets in Shanghai and Shenzhen and
at the 5% level for B-share markets. These findings strongly support the
hypothesis of a Monday's effect on Chinese stock markets.
3.3.2 Mov ing Av erage Approach
Although analysis of variance can be employed to detect any abnormal re-
turns in a given trading day, this methodology is not satisfactory because of
i.i.d. (i.e., identically and independently distributed) and normal assump-
tions on the return generation process. A better approach is to treat daily
returns as time series rather than independent observations selected from
Table 3.2 Day-of-the-Week Effect (Analysis of Variance)
Shanghai Shanghai Shenzhen Shenzhen
A Share
a
B Share A Share B Share
Monday -0.116 -0.129 -0.155 -0.102
Tuesday 0.313 0.296 0.283 0.194
Wednesday 0.941 0.117 0.586 0.301
Thursday 0.578 0.092 0.731 0.136
Friday 0.207 -0.075 0.197 0.048
T
2b
21.96* 11.27* 24.16* 9.35*
"All figures are in percentage terms.
b
* and t denote HQ (daily mean returns are equal) is rejected at the 1% and 5% level,
respectively.
a random sample.
3
Let
r
m,n t>
e
the logarithmic return of an index on day m of week n.
Assume that the data generation process for r
m
,
n
is as follows:
Tm,n ~ U&m ~r iilrn,n "T 0'm^m,n {^^)
where
m = Monday, Tuesday, Wednesday, Thursday, and Friday;
DS
m
is the seasonal component of daily returns and
~ DS
m+s
- 0;
2q
s = -q
RT
mtn
= RT
t
is the random trend component of daily returns;
<7
m
is the standard deviation for all day m returns; and
e
m
,n = e
t
is a i.i.d. N(0,1).
Because negative average returns on Monday could arise from a negative
market trend, we calculate daily returns using the moving average approach.
3
Using a sphericity-test (Andersen, 1984, p. 427), we strongly reject independence and
homoskedasticity for all market indices during the sample period.
Day-of-the- Week Effect 85
Table 3.3 Day-of-the-Week Effect (Moving Average Seasonal Component)
Monday
Tuesday
Wednesday
Thursday
Friday
rp2b
Shanghai
A Share"
-0.139
0.261
1.030
0.627
-0.036
16.05*
Shanghai
B Share
-0.101
0.255
0.164
0.128
-0.104
14.82*
Shenzhen
A Share
-0.128
0.202
0.362
0.685
0.138
10.72+
Shenzhen
B Share
-0.083
0.217
0.515
0.073
0.021
8.16
"All figures are in percentage terms.
b
* and t denote Ho (daily mean returns are equal) is rejected at t he 1% and 5% level,
respectively.
The moving average smoothes the trend of daily returns which are not
assumed to be constant over time.
j
2
MR
m
,n =M
t
= - ^2
r
*+
1
2
1
2
= T 2-~i RTt+s + 3 2__,
a
m+s^t+s (3-3)
s=-2 s = - 2
After eliminating the trend from the daily return data, we obtain the
moving average return for day t (MR
t
). MR
t
consists of a seasonal com-
ponent and a white noise.
Table 3.3 presents empirical results of mean daily returns and Hotelling's
T
2
statistics for seasonal components of each indices. The main conclusion
from Section 3.3.1 is by and large still valid. All stock indices, except
Shenzhen B-share index, exhibit significant Monday effect. The values of
Hotelling's T
2
statistics, except that of Shanghai B-share index, are smaller
under the moving average approach than those under the analysis of vari-
ance approach.
3.4 Market Efficiency Hypothesi s
The fundamental insight of the market efficiency hypothesis is that security
prices reflect optimal use of all available information. It is well known that
the Chinese government intervenes its domestic stock markets in an un-
predictable way from time to time. For example, the Chinese government
suddenly removed the 1% daily price change limit on May 2, 1992 and re-
moved the 5% daily price change limit on May 22, 1992. The Shanghai
A-share index doubled on May 23, 1992. In July 1994, the China Securities
Regulatory Committee (CSRC) announced a series of "market support"
policies to restrict the new supply of shares. The Shanghai A-share index
nearly tripled in two months. More recently, the CSRC announced the im-
position of a 10% limit in the daily movement of any individual share price
on December 13, 1996. If stock markets in China are efficient now, then
there is probably no reason to change current security market regulations.
But if inefficiencies exist, then policies designed to influence the stock mar-
kets may be inappropriate and the government should change or reduce its
control or intervention in the stock markets.
The following two conditions describe efficient stock markets:
(1) Stock prices (in logarithmic form) follow a random walk with drift,
i.e., any new information arriving between this and next period
creates only a random deviation from this period's best forecast so
that price will vary randomly around trend.
(2) Stock prices (in logarithmic form) between markets (e.g., Shanghai
and Shenzhen security exchanges) are not cointegrated, i.e., no his-
torical information on past stock prices at one market can improve
the forecastability of stock prices in the other market.
3.4.1 Random Walk Hypothesis
The random walk hypothesis says that stock price sequences vary randomly
over time around trend, and new information has no predictable compo-
nents.
In P
jtt+1
= otj + In P
jit
+ e
jit
(3.4)
or equivalently,
Market Efficiency Hypothesis 87
r
J,t = Qj + j,t
(3.5)
where r
J)t
is the rate of return on j t h market index. Equation (3.5) implies
that the increments of the variance of stock-market returns are proportional
to the observation interval. That is, the variance of r
t
r
t
_
5
is s times as
much as the variance of rt f"t-i- Therefore, we can test the random
walk hypothesis using a variance ratio test. The null hypothesis is that
the disturbances 6j
t
t are uncorrelated, allowing for quite general forms of
heteroskedasticity.
Lo and MacKinlay (1988) have a detailed discussion of the statistical
properties of the heteroskedastic increments null hypothesis. They show
that the variance ratio statistic is asymptotically equivalent to the following
expression:
VKjiq) = 1 + M
3
{q) l + Y,
2
^"ft(0 (3-6)
l=i
q
where q is the observation interval of alternative variance estimators and is
any integer greater than 1 and Pj(l) denotes the Ith-order autocorrelation
coefficient estimator of r^
t
-
7 J ( 0 = Cov(r
jt
t,rj,
t
+i).
n m =
Cov
(rj,t,r
jt
t+i) _ 7^(0.
PjW
~ Var{r
ht
) " ^ ( O ) '
Tl
7i(0 = ^ ! > * * ~
f
i,T)(r
jtt+
i ~ f
jiT
), 0<KT.
T
t=i
Pj(l)
7j ( 0) '
1
T
s
rE
r
J
Mj(q) is an unbiased variance ratio estimator and is asymptotically zero
under the random walk null hypothesis. Therefore, we only need to com-
pute the standardized asymptotic variance of Mj(q) to derive statistical
inference.
The heteroskedastic-consistent standardized test statistic is:
ZMs
^MM
(37)
where 6j(q) is the asymptotic variance of Mj(q). Lo and MacKinlay (1988)
show that
9-1 /
A
2
? i Er =i +i f e, i -Pj,t~i - H)
2
{pj,t-i -Pj,
t
-i-i - H)
2
01 = 2 j
[T,t=APj,t - Pj,t-i - Aj )
2
]
and
nq
q
t=i
Despite the presence of general heteroskedasticity, the standardized test
statistic Zj(q) is still asymptotically standard normal.
Zjiq) Z
N
(0,l).
Table 3.4 contains values of the variance ratio test statistics VTZj(q) with
heteroskedasticity robust standardized test statistics Zj(q) in parentheses.
For Chinese stock markets, the random walk null hypothesis is strongly
rejected at the 5% level of significance when daily observations are used
and at the 10% level when weekly observations are used. Strong rejection
of the random walk hypothesis using variance ratio tests suggests that Chi-
nese stock markets are inefficient.
4
By comparison, for major world equity
market indices, such as the MSCI, NYSE and Hong Kong indices, random
4
The rejection of random walk hypothesis using variance ratio test also indicates t hat
stock-market returns in China are mean-reverting, which is another piece of evidence
of market inefficiency. See Fama and French (1988), Poterba and Summers (1988).
Table 3.4 Variance Ratio Tests for Random Walk Hypothesis for the Logarithm of
Stock Market Indices
Shanghai A
Shares
Shanghai B
Shares
Shenzhen A
Shares
Shenzhen B
Shares
MSCI World
Index
NYSE
Composite
Hong Kong
Hang Seng
Shanghai A
Shares
Shanghai B
Shares
Shenzhen A
Shares
Shenzhen B
Shares
MSCI World
Index
NYSE
Composite
Hong Kong
Hang Seng
nq
D
1761
1466
1720
1260
1310
1750
1807
Number q of base observations
aggregated to form variance ratio
2 4 8
aily stock market indices
1.069*
(2.4991)
1.2519*
(8.2841)
1.044
(1.5605)
1.2083*
(6.0101)
1.0929*
(3.376)
1.0166
(0.1345)
0.9168*
(-3.0686)
1.167*
(3.2257)
1.3761*
(6.3498)
1.1207*
(2.2811)
1.3827*
(5.7386)
1.0893+
(1.7217)
1.0196
(0.7047)
0.8864*
(-2.227)
1.2696*
(3.278)
1.4872*
(5.1205)
1.2375*
(2.8009)
1.5626*
(5.24)
1.101
(1.2261)
1.0453
(0.8705)
0.8976
(-1.2637)
Weekly stock market indices
374
312
362
274
373
372
377
1.1009*
(1.6689)
1.1486*
(2.1798)
1.1185t
(1.9181)
1.2889*
(3.9365)
1.0139
(0.2275)
1.0134
(0.2186)
0.9808
(-0.3259)
1.1227
(1.0684)
1.2921*
(2.2277)
1.196+
(1.6674)
1.4496*
(3.0685)
1.0397
(0.3478)
1.1013
(0.8865)
1.0755
(0.6615)
1.1669
(0.9001)
1.4022+
(1.8939)
1.323+
(1.7095)
1.6539*
(2.7414)
1.0234
(0.1313)
1.1256
(0.6947)
1.0978
(0.5365)
16
1.3277*
(2.6604)
1.6016*
(4.1895)
1.3523*
(2.7775)
1.7282*
(4.4433)
1.0941 +
(1.7649)
1.0394
(0.477)
0.9549
(-0.371)
1.1671
(0.5866)
1.5735+
(1.8273)
1.2702
(0.9299)
1.8571*
(2.3965)
0.9978
(-0.0079)
1.2632
(0.9278)
1.1388
(0.492)
* and + denote statistically significant at the 5% and 10%
level, respectively.
walk is rejected, but not strongly, for daily observations and can not be
rejected at any reasonably level of significance for weekly observations.
3.4.2 Cointegration-Based Market Efficiency
Central to the cointegration-based test of market efficiency is the relation-
ship between cointegration and error correction models. Denote the loga-
rithm of stock prices for Shanghai A-shares market and Shenzhen A-shares
market as p^
h
t
and p^
z
t
, respectively. K p
a
h
t
and p^
z
t
are cointegrated, then
there must exist an error correction representation of the following form
(Engle and Granger, 1987, and MacDonald and Kearney, 1987):
()-(:)^-^>l>(&M)
(3.8)
and ai + 0.2 ^ 0.
The intuition behind the error-correction equations is that if stock prices
in Shanghai and Shenzhen are cointegrated, knowledge of the lagged linear
combination of price levels in Shanghai and Shenzhen must improve the
forecastability of returns in at least one of the markets. If stock prices
are determined in efficient markets, then they should have incorporated
all available information. Hence, given the past stock prices in Shanghai
(Shenzhen), no other information should be of use in explaining the changes
in its stock prices. In particular,
E(p
s
a
h
t I t
s
-i) = E(pZ I fi&")
where
Qsh _ f jft sh sh \
" t - l

\Pa,t-l' Pa,t-2> Pa,t-3> J-
rysh,sz _ r
S
z sz sz sh sh sh 1
" t - l

\Pa,t-l'Pa,t-2'â,t-3' > Pa,t-H Pa,t-2' Pa,t-3> }
We test for cointegration as follows: first, We test the null hypothesis
of a unit root in the time series of pj
t
(i = sh,sz;j = a,b) against the
stationary alternative for the following three specifications:
Pj,t = PjPj,t-i+j,t (
3
-
9
)
p
jt
t = a) + p*p
Jtt
-i + e*
t
(3-10)
p
jtt
= dj + pj{t - T/2) + PjPj,t-i + i
jtt
(3.11)
Table 3.5 Tests for Unit Roots for t he Logarithm of Daily Stock Market Prices
Ho
Pj =
P'j =
P) =
a* =
Pj =
Pi =
CXj =
Pj =
i
= i
= 0
1
1,
= 0,
= 0
1,
0
Test
statistic"
z(t
h
)
z(t
Pl
)
Z{$i)
z{t~
Pl
)
Z{$2)
Z($
3
)
Shanghai
A shares
-0.745
-2.2641
3.0678
-1.1316
2.019
2.523
Shanghai
B shares
-1.3166
-2.2262
3.1914
-2.5849
2.8259
3.5279
Shenzhen
A shares
-0.036
-1.3089
0.8617
-1.1377
0.7605
1.1357
Shenzhen
B shares
-1.2064
-0.2084
0.7357
-2.0043
2.2132
2.6017
"Source for critical values: Fuller (1976, p. 373) and Dickey-Fuller (1981, p. 1063).
Critical values
1%
5%
10%
z(t
h
)
-2.58
-1.95
n/ a
Z(tpl)
-3.43
-3.12
-2.57
Z( * i )
6.43
4.59
3.78
z(t-
Pj
)
-3.96
-3.41
-3.12
Z($
2
)
6.09
4.68
4.03
Z( *
3
)
8.27
6.25
5.34
All tests are based on Phillips-Perron Unit Root Tests, with size 4 Bar-
lett Window. The Phillips-Perron test statistics share the same limiting
distributions as those of Dickey-Fuller test statistics. The test results in
Table 3.5 show that unit root hypothesis can not be rejected for any of the
daily price time-series. Therefore p*-
>t
is 1(1), Vi and j .
Next, we test the null hypothesis of no pairwise cointegration in stock
prices between Shanghai and Shenzhen security exchanges for both A- and
B-shares. The first step involves estimating the following cointegration
regression by OLS:
m
Pft = ajfi + aj,ip"
t
+ Y,
a
i,s
A
Pj*t-
s
+ H,t (3.12)
s = m
where Ap^
t
_
a
are leads and lags of Ap?*
t
. The estimation results in Table
3.6, Panel (I) indicate that the slope coefficient for the cointegration re-
Table 3.6 Tests for Cointegration
Panel (I)
n
sh .
A shares
B shares
OLS estimation for the cointegration
Oijfi
0.0002
(0.0253)
-0.002
(-0.3533)
&
j,i
0.9126
(14.19)
0.125
(1.4792)
Panel (II) Residual-based cointegration test (ADF
A
Aj,< = Ijfi + 7j,iAj,t-i + Ef =i 7jM^fa,t-
ADF test statistic Critical value
0
regression"
R
2
0.4246
0.0218
test) for jijj
b
k + Vj,t
Q(K)
A shares -18.2696 -2.8649 Q(2) = 16.3798
[0.1744]
B shares ADF test fails to find the value K in which v^
t
is
serially uncorrelated.
"Figures in parentheses are t-statistics and figures in brackets are p-values.
' ADF test determines the appropriate number of lagged differences K by adding lags
until a Lagrange Multiplier test fails to reject no serial correlation in v^
t
at 5% level of
significance.
c
The critical values are computed by MacKinnon (1991, p. 267-276).
gression is close to one and significantly positive for A-share markets, but
is insignificantly different from zero for B-share markets.
The second step is to test for a unit root on the estimated residuals,
p.j
t
t, as a proxy for the true residuals. Since OLS estimation procedure
tends to make the estimated residuals become stationary, an augmented
Dickey-Fuller unit root test should be applied:
K
Ap,
jtt
= 7j,o + 7j,iAj,t-i + 531j,k&frj,t-k +
v
i,t (3.13)
fc=i
GARCH Models 93
The specific hypothesis is:
Ho 7i,i = 0
HA 7 i , i ^ 0
Table 3.6, Panel (II) gives the ADF test statistic with K = 5 for A-
share markets. The null hypothesis of no cointegration between Shanghai
and Shenzhen A-share prices is rejected at 5% level of significance. The
Ljung-Box portmanteau statistic Q(5) is not significant at the 10% level,
indicating that v
a
^
t
is white noise. Therefore, we can safely conclude that
Chinese A-share markets are inefficient. For B-share markets, ADF fails to
generate a i-statistic with white noise Vb,t- Therefore, there is no evidence
that Shanghai and Shenzhen B-share markets are cointegrated. Interna-
tional investors can not use past B-share price information on one exchange
to predict B-share price movement on the other. According to market ef-
ficiency concepts developed by Fama (1991), Chinese B-share markets are
at least weak-form efficient.
3.5 GARCH Models
In this section, we follow Su and Fleisher (1998a) and explore the distribu-
tional characteristics of the stock-market variance process and analyze the
predictability of the stock-market returns using a set of local and global
information variables. In particular, we establish an empirical model that
captures the deterministic components of the variation in the stock-market
returns and characterize the second order conditional moments using three
error generation processes. We first specify a baseline Generalized Au-
toregressive Conditional Heteroskedasticity model, or GARCH(1,1) model,
that expresses the time series of weekly stock-market returns, r
Jj t
on a set
of lagged local and global information variables Z
t
_i . Then we combine
the baseline model with three types of error distributions in formulating a
conditional variance process for Chinese stock markets. Finally, we specify
a set of local and global information variables to explain the variation of
the conditional mean in the baseline model.
3.5.1 Model Specification
Let rj
tt
denote the return on a market index at time t and 7ij
t
t-i repre-
sent a set of lagged local and global information variables that affect the
conditional mean,
r
jtt
= 6jZj,t-i + t
jlU
(3.14)
e
i,t =
a
j,tZj,t (3-15)
z
J
-
t
|n
t
- i ~ 0(0,1,1/), (3.16)
a
),t = "3 +
a
3
a
lt-i + /Vj,i-i> (
3
-
17
)
where u?
t
is the conditional variance, Zj
t
t is the standardized residual
formed by dividing the residual, tj
<t
by the standard deviation, a^
t
, ftt-i
is the set of information available at the beginning of time t, </>() denotes a
conditional density function, and v denotes a vector of parameters besides
conditional mean and conditional variance that may be needed to fully
characterize the probability distribution. Since equation (3.17) defines a
variance, a nonnegativity constraint must be imposed on a and /?, and the
sum (a + j3) must be less than 1 for the volatility process to be covariance
stationary (Bollerslev, 1986).
The GARCH(1,1) model has a distinctive feature, i.e., a set of lagged
local and global information is used to explain the returns on a market
index, and at the same time, conditional variance is modeled to capture
volatility clustering and temporal dependence of market returns. As Nelson
and Cao (1992) point out, a possible misspecification of the mean equation
(3.14) is not of great concern, because the conditional variance estimates
obtained from a GARCH model are robust to an incorrect specification of
conditional mean.
3.5.2 Characterizing Variance in Chinese Stock Markets
The exact form of the error distribution plays an important role in estimat-
ing the GARCH(1,1) formulation. Our findings in Section 3.2 indicate that
the time series of stock returns in China are not independent processes;
large changes tend to be followed by large changes of either sign, and small
GARCH Models 95
changes tend to be followed by small changes. To capture this type of
volatility clustering and temporal dependence of stock returns, we consider
three different functional forms of the conditional density </>(). They are:
Gaussian normal distribution, standardized Student ^-distribution, and Sta-
ble distribution.
5
3.5.2.1 Normal Distribution
Under a Gaussian standard normality assumption, (3.16) becomes:
z
3,t
|n
t
_
1
~J V( 0, l ) , (3.18)
The Gaussian GARCH model can accommodate volatility clustering, but
it is not sufficient to account for all the leptokurtosis, or fat tail in the
return distribution, that appears in the Chinese data. The large number
of very high and very low returns observed suggests that a fatter-tailed
distribution than normal distribution might better characterize the error
process for Chinese stock-market returns.
3.5.2.2 Standardized t-distribution
The conditional density function for Zj
t
t, under the standardized Student
t-distribution with mean 0, variance 1 and degrees of freedom v, can be
written as:
Zj,t\^t-l
v + l
2
-\> -> r
2 /,, o ^ n - ^
1
) /
2
1/2
(l + z ^ - 2 ) -
1
)
u>2. (3.19)
It is well known that the parameter v can be interpreted as the degree of
leptokurtosis. Large values of v are associated with the absence of leptokur-
tosis while small values are associated with some degree of leptokurtosis.
5
Ot her possible parametric densities include normal-Poisson mixture distribution in Jo-
rion (1988), the power exponential distribution in Baillie and Bollerslev (1989), the
normal-lognormal mixture distribution in Hsieh (1989), and the generalized error dis-
tribution, or GED, in Nelson (1991). The application of these densities in t he above
empirical model is beyond the scope of this chapter and can be further exercise for
researchers.
As \jv approaches 0, the Student t -distribution approaches a standard
normal distribution, but when 1/v > 0, the ^-distribution has "fatter tails"
than the corresponding normal distribution.
3.5.2.3 Stable Distribution
Under the Gaussian normal distribution, the length and height of the ex-
treme tails are fixed. Assuming a stable distribution density function for
the error term in the baseline model allows us to explicitly estimate the
tail length and height. Moreover, McCulloch (1997) argues that the stable
non-Gaussian distribution generalizes the Central Limit Theorem to cases
where the second moments of the underlying variables do not exist. It
is especially useful in modeling financial asset returns that exhibit strong
leptokurtosis.
We standardize the residual by using the following transformation:
^, t = ( 4 t )
V
X t , (3-20)
where Zj
t
t is specified to have a symmetric standard stable density with
both a skewness parameter and location parameter equal to zero. Since the
standard Stable distribution does not have a simple mathematical descrip-
tion, the log of the characteristic function is used instead:
\og[E(e
iXt
)] = |*|", (3.21)
where v {v e (0, 2]) is the characteristic exponent to be estimated from the
model. When u = 2, the standard stable distribution becomes the stable
standard normal distribution. When v < 2, absolute moments of order
less than v exist, but those of order greater than or equal to v do not, so
that variance is infinite. For example, when v = 1, the stable distribution
becomes the Cauchy distribution, for which the mean is also infinite.
3.5.3 World Versus Local Factors in Volatility
In the baseline model, r^
t
= <^Zj,t-i -I- tj
tt
, weekly market returns, r
jtt
,
are regressed on a set of lagged local and global information variables that
may influence the conditional mean of stock-market index returns.
Local information variables include:
Estimation and Empirical Results 97
lagged one-week market return, rj
t
t-i',
lagged change in the exchange rate between Chinese renminbi yuan
and U.S. dollar, ARMBUS
t
-i;
lagged change in exchange rate between Hong Kong dollar and U.S.
dollar, AHKUSt-i;
lagged change in weekly turnover rate, AT0j
t
t-i
m
,
World information variables include:
lagged MSCI world index return in excess of the 30-day U.S. Trea-
sury Bill rate, MSCI
t
-V,
lagged Hong Kong Hang Seng index return in excess of the 30-day
U.S. Treasury Bill rate, HS
t
-i;
lagged NYSE index return in excess of the 30-day U.S. Treasury
Bill rate, NYSE^;
lagged change in term structure spread (The yield on Long-term
U.S. Government Security minus the 30-day U.S. Treasury Bill
rate), TS
t
- i ;
lagged change in the 30-day U.S. Treasury Bill rate, ATB
t
-\.
We initially include only lagged local information variables in the base-
line model, assuming GARCH(1,1), TGARCH(1,1) and SGARCH(1,1) er-
ror generation processes in estimating the parameters. In doing this, we
treat the market as fully segmented in the sense that common shocks to
world equity markets do not influence the stock markets in China and there
are no covariance dynamics. To test the hypothesis that global as well as
local factors influence the conditional mean returns in Chinese stock mar-
kets, we subsequently include both lagged local and lagged world informa-
tion variables in the baseline model and estimate all the parameters again,
we use constrained maximum likelihood (CMLE) application module with
Sequential Quadratic Programming method in estimating the model. Both
BFGS (Broyden, Fletcher, Goldfarb and Shanno) and NR (Newton and
Raphson) algorithms have been tried to obtain the maximum likelihood
estimates.
3.6 Estimation and Empirical Results
We first carefully estimate the GARCH(1,1) model under three alternative
formulations of error-generation process. Then we perform a battery of
residual diagnosis tests using the Ljung-Box Q(8) portmanteau statistics
and the likelihood ratio test statistics and choose the error distribution
that is best supported by data. Finally we analyze parameter estimates of
the best-fitted model. Our objective is to seek the best statistical model
through rigorous econometric analysis and use the model to study the effect
of exogenous government policy changes have on the stock-market return
volatility.
3.6.1 Model Comparison
We estimate the Gaussian GARCH(1,1), TGARCH(1,1) and SGARCH(1,1)
models, with and without global information variables.
6
To choose among
the alternative error-distribution formulations for the best fit to Chinese
stock-market data, we first compare the Ljung-Box portmanteau test statis-
tics for serial correlation with 8 lags on standardized residuals (zj
tt
=
j
t
t<J~
t
) for each time series and test the null hypothesis of no autocor-
relation up to lag 8. The purpose of this test is to evaluate whether each
model fully accounts for all autocorrelation of stock-market returns.
As the results in Tables 3.7-3.9 show, there is evidence of serial corre-
lation for all time series under the Gaussian normal GARCH assumption,
as the Ljung-Box Q(8) statistics are significant at the 5% level. The Bera-
Jarque test for normality strongly rejects the null hypothesis of normally
distributed standardized residuals under the Gaussian normal GARCH
model. The skewness and kurtosis in the standardized residuals indicate
the inappropriateness of the assumption of conditional normality in the
error distribution. Under TGARCH and SGARCH assumptions, there is
no evidence of serial correlation in the residuals, as the Ljung-Box Q(8)
statistics are not significant even at the 10% level for most of the return
series. Moreover, the Ljung-Box Q(8) statistics are not significant at the
5% level for six out of eight time series under the SGARCH formulation,
while they meet this level of significance for only three out of eight time
series under the TGARCH formulation. This indicates that the SGARCH
model outperforms the TGARCH model.
The second phase of residual diagnosis is to determine whether the re-
sults under the assumption of the t- and stable distributions are statistically
6
For brevity, we only include estimation results for the best-fitted model. Estimation
results on other models are available upon request.
Estimation and Empirical Results
Table 3.7 Residual Diagnostic Tests for Gaussian GARCH Model
99
Qaiej/aj)
Skewness
Kurtosis
Bera-Jarque
Qsiej/cTj)
Skewness
Kurtosis
Bera-Jarque
Shanghai
A-Shares
Shanghai
B-Shares
Shenzhen
A-Shares
Shenzhen
B-Shares
with only local information variables
0
3.9437
[0.8522]
2.3545
(0.825)
8.6215
(2.1323)
2159.25
7.1712
[0.5183]
3.4975
(1.3)
14.6961
(2.7522)
14084.7
5.2187
[0.734]
6.6258
(2.0366)
54.6961
(10.3288)
1469785
6.9842
[0.5383]
2.1448
(0.7225)
7.6324
(2.0119)
868.83
with both local and global information variables
4.0904
[0.8489]
2.4393
(0.8626)
9.0803
(4.1425)
2754.83
4.6341
[0.7959]
2.1461
(0.6984)
7.2489
(2.588)
812.56
3.5229
[0.8974]
1.7991
(0.3644)
4.661
(1.5642)
186.25
4.1608
[0.8423]
1.9583
(0.4487)
6.799
(2.1139)
486.98
"Figures inside the parentheses are heteroskedasticity-consistent standard errors and
figures inside the brackets are p-values. Qsitj/vj) denotes the Ljung-Box Q statistic
for serial correlation tests with 8 lags on standardized residuals. Bera-Jarque (1982)
statistic is used to test for normality, and is calculated as: ? Skewness
2
+^7 (Kurtosis
3)
3
, which is distributed as Chi-square with 2 degrees of freedom. The critical value
for *
2
( 2) at 1%, 5% and 10% levels are 9.21, 5.991 and 4.605.
different from those obtained under the normal distribution. To do this, we
compute likelihood ratio statistics to test the null hypothesis that the tail-
thickness parameter 1/0 > 0 against the one-sided alternative that 1/0 > 0
for the TGARCH formulation, and I test the null hypothesis that the tail
thickness and length parameter 0 = 2 against the one sided alternative that
0 < 2 for the SGARCH formulation. The values for the likelihood ratio
test statistics are reported in Tables 3.8 and 3.9.
Table 3.8 Residual Diagnostic Tests for TGARCH Model
Qs(ej/(Tj)
Skewness
Kurtosis
LRi/v^tQ
Qsiej/crj)
Skewness
Kurtosis
LR\/vô
Shanghai
A-Shares
Shanghai
B-Shares
Shenzhen
A-Shares
Shenzhen
B-Shares
with only local information variables
0
3.6026
[0.8911]
1.4621
(0.4852)
4.2055
(2.004)
73.43
2.7382
[0.9497]
1.4037
(0.4755)
20.2045
(5.3781)
33.78
5.8249
[0.6663]
2.0193
(0.6331)
7.2603
(2.1141)
57.6
5.4158
[0.7124]
1.4501
(0.4751)
4.4762
(1.8775)
39.14
with both local and global information variables
2.4277
[0.965]
1.054
(0.3572)
1.6593
(1.2108)
65.93
2.3334
[0.969]
1.2572
(0.4083)
2.1504
(1.3253)
39.14
4.6341
[0.7959]
1.6154
(0.4833)
4.623
(1.6774)
54.78
3.8026
[0.8745]
1.46
(0.4755)
3.7825
(1.621)
35.11
figures inside t he brackets are p-values. Qs(ej/oj) denotes t he Ljung-Box Q statistic
statistic is used to test for normality, and is calculated as: ^ Skewness
2
+ ^ (Kurtosis
3)
3
for x
2
(2) at 1%, 5% and 10% levels are 9.21, 5.991 and 4.605.
Under the null hypothesis that 1/0 > 0 in the TGARCH formulation
and ;> = 2 in the SGARCH formulation, neither the LR-^/^^Q nor the LRc,
=2
statistic is distributed as x
2
{i) because they do not meet the standard
regularity conditions for asymptotic x
2
distribution. McCulloch (1997)
calibrates the Monte Carlo critical values for the symmetric stable likelihood
ratio statistic when v = 2. Since all the LRp
=
2 statistics for the Chinese
stock-market indices are highly significant compared with the Monte Carlo
Estimation and Empirical Results
Table 3.9 Residual Diagnostic Tests for SGARCH Model
101
Qsitj/Vj)
Skewness
Kurtosis
- Rl / i / =2
Qs{ej/(Tj)
Skewness
Kurtosis
LR
l
/
l/=2
Shanghai
A-Shares
Shanghai
B-Shares
Shenzhen
A-Shares
Shenzhen
B-Shares
with only local information variables"
3.436
[0.9041]
1.1663
(0.401)
2.5621
(0.9888)
59.76
1.8537
[0.9852]
1.4037
(0.7359)
2.2952
(1.478)
39.76
5.4158
[0.7124]
1.3532
(0.4752)
3.1783
(1.3528)
48.64
with both local and global informati
2.3438
[0.9686]
0.5988
(0.3427)
0.4073
(0.4899)
53.33
0.8201
[0.9991]
1.2258
(0.5107)
2.1443
(1.3888)
44.08
6.6658
[0.5731]
1.4946
(0.5158)
3.8174
(1.4282)
50.43
1.4501
[0.9935]
1.5609
(1.0622)
3.9651
(1.6742)
2.352
on variables
2.361
[0.9679]
1.7991
(0.3667)
5.4245
(2.0366)
2.352
figures inside the brackets are p-values. Qs(tj/o-j) denotes the Ljung-Box Q statistic
statistic is used to test for normality, and is calculated as: ? Skewness
2
+ (Kurtosis
3)
3
for x
2
(2) at 1%, 5% and 10% levels are 9.21, 5.991 and 4.605.
critical values, the null hypothesis of normality under SGARCH model is
strongly rejected.
7
McCulloch also points out that the Monte Carlo critical values are
smaller than the critical values under \
2
distribution for the null hypothe-
7
The Monte Carlo critical values are 0.243, 1.12 and 4.764 for 10%, 5% and 1% level of
significance, respectively.
sis of normality. LRi/
0
^
0
statistics are all highly significant using critical
values under x
2
( l ) distribution. Therefore, the null hypothesis of normality
under TGARCH model is also rejected.
Since the TGARCH and SGARCH models are not nested, a likelihood
ratio test can not be used to distinguish these two models. However, the
Ljung-Box portmanteau statistics seem to support SGARCH formulation
better than TGARCH formulation.
Summing up, the residual diagnosis tests indicate that the GARCH
model with conditional normal errors does not fully capture the leptokur-
tosis and the serial correlation of the standardized residuals. The GARCH
model with Stable error distributions fits the time-series data the best.
8
3.6.2 Parameter Estimates
The parameter estimates for the SGARCH(1,1) model is presented in Table
3.10. The conditional standard deviations are plotted in Figures 3.3 and 3.4.
When only local variables are included in the model, lagged one-week return
exhibits a strong influence on market return. This is probably because of
nonsynchronization of trading and clearing and thinness of the markets.
The constant terms in the conditional mean equations are all very small
and not statistically significant at any level. The coefficient estimates for
the lagged change in weekly turnover rate are also small and insignificant for
all the series. Furthermore, the coefficient estimates for the lagged change
in the U.S. dollar-Chinese yuan and U.S. dollar-Hong Kong dollar exchange
rates are both statistically insignificant.
When the global information variables are added to the conditional
mean, the coefficient estimates for the lagged change in the U.S. dollar-
Hong Kong dollar exchange rate become statistically significant in three out
of the four series. Since both China and Hong Kong enforce fixed exchange
rate regimes, we are puzzled by this result.
9
Among the global information
variables, the lagged Hong Kong Hang Seng weekly index return variable
8
It remains an open question whether other conditional error distributions would provide
a better fit. Another interesting question is whether higher order GARCH models might
provide an even better description. We leave the answers to all of these questions for
future research.
9
The exchange rate between Hong Kong dollar and U.S. dollar fluctuates within a 1%
narrow range, while the exchange rate between Chinese yuan and U.S. dollar seldom
fluctuates over t he sample period, except t hat a few "jumps" are observed.
Estimation and Empirical Results 103
Table 3.10 Maximum Likelihood Estimates of the SGARCH(1,1) Model
Shanghai Shanghai Shenzhen Shenzhen
A-Shares
a
B-Shares A-Shares B-Shares
With only local information variables
constant
r
3,t-l
ARMBUS
t
-i
AHKUSt-i
ATO
t
_i
Qjj
&
i
Pi
V
Mean log
likelihood
-0.004
0.287*
-0.013
0.266
0.006
0.018*
0.615*
0.107*
1.551*
1.309
-0.006
0.202*
-0.012
0.57
-0.001
0.003
0.318*
0.189*
1.54*
1.769
-0.007
0.15*
-0.037
0.152
0.018
0.024*
0.74*
0.214*
1.676*
1.296
-0.004
0.291*
-0.002
0.237
-0.002
0.017*
0.743*
0.205*
1.58*
1.845
With local and global information variables
constant
r
i,t-i
ARMBUSt-i
AHKUSt-i
ATO
t
-!
MSCIt-i
HSt-i
NYSEt-i
TSt-x
ATB
t
_!
ilij
CCj
Pi
V
Mean log
likelihood
-0.007
0.268*
-0.012
0.342*
0.004
0.03
-0.06
-0.31*
0.433
-5.937
0.033*
0.619*
0.121*
1.506*
1.322
-0.005
0.191*
-0.018
0.071
0.001
0.198
-0.29*
-0.22
-0.094
0.03
0.004
0.552*
0.164*
1.493*
1.828
-0.01
0.153*
-0.036
0.744*
0.014
0.345*
0.094
-0.35*
0.175
-12.22
0.011*
0.688*
0.169*
1.7666*
1.311
0.002
0.242*
-0.036
0.32*
-0.001
-0.002
0.087
-0.074
-0.239
1.233
0.02*
0.732*
0.224*
1.567*
1.887
a
* denotes 5% level of significance according to the Wald Confidence Limits.
is significant for the Shanghai B-share return series. The weekly MSCI
world index return variable is significant for the Shenzhen A-share return
series. The lagged NYSE weekly return variable is significant for both
Shanghai and Shenzhen A-share return series. The coefficient estimates for
the lagged change in the U.S. 30-day Treasury Bill rate are large in two
cases, but none of the estimates is statistically significant. Neither do the
U.S. term structure spread variables have any explanatory power.
o
S
v
i
a

0)
n
r\
a
d
s
-t-
t/J
1
o
ri
*
d
o
O
1. 35
1. 15
0. 95
0. T5
0. 55
0. 35
0. 15
-0. 05
- Shanghai
- Shenzhen
yM^yiyi
V "sr
910430 920731 930630 940430 950131
960531
Fig. 3.3 Conditional Standard Deviation from t he SGARCH Model (Shanghai and
Shenzhen A-Shares)
To test the importance of the set of world factors in the conditional
mean equation, I use likelihood ratio statistics to test the joint hypothesis
that the estimated coefficients for the global information variables are all
zero. For the Shanghai A- and B-share, Shenzhen A- and B-share series,
the LRfi, _. _. _. _. _
0
statistics are 3.48, 12.15, 3.84 and 7.06 under
the SGARCH formulation! The critical values for the 1%, 5% and 10% level
of significance are 15.09, 11.07 and 9.24, respectively. Hence, the joint null
hypothesis that the estimated coefficients for the global information vari-
ables are all zero can not be rejected at the 1% significance level, but it can
be rejected for the Shanghai B share series at the 5% level of significance.
We conclude that local and global information variables do explain some
of the variation in Chinese stock-market returns, but evidence that China's
stock markets are integrated into the world financial network is at present
rather weak.
The empirical results shown in Table 3.10 indicate that A-share returns
in Shanghai and Shenzhen tend to respond similarly to common news and
economic factors, as the coefficient estimates in the empirical models are
very similar in terms of the values and statistical significance. For example,
the coefficient estimates for lagged one-week return, lagged change in the
Government Regulation and Market Volatility 105
3
1
1
1
a
o
0. 65
0. 55
0. 45
0. 35
0. 25
0. 15
0. 05
- 0. 05
921210 930630 940215 940930 950515 951231 960815 970331
Fig. 3.4 Conditional Standard Deviation from the SGARCH Model (Shanghai and
Shenzhen B-Shares)
exchange rate between U.S. dollar and Hong Kong dollar, and the lagged
NYSE excess return are 0.268, 0.342 and -0.31 for Shanghai A shares and
0.153, 0.744 and -0.35 for Shenzhen A shares. These coefficient estimates
are all significant at the 5% level according to the Wald confidence limits.
This implies that:
(1) Even though the Shanghai and Shenzhen securities exchanges are
still largely regional exchanges, there is currently potential for them
to become a fully integrated national stock market.
(2) The dissimilarity of coefficient estimates for the information vari-
ables between the A and B share market return series imply that
domestic and international investors have very different investment
sentiment and risk-aversion.
3.7 Government Regul ati on and Market Volatility
We now apply the best-fitted empirical model to study the high stock-
market return volatility in China. We observe that Chinese stock markets
are very sensitive to government regulations. For example, the Chinese
government has expressed the belief that the volatility of "hot money" in-
troduces "excessive" volatility in stock market prices. It therefore imposed
price ceilings and floors on daily stock price movements in order to insulate
the domestic stock market from "fickle" foreign capital movements prior to
May 5, 1992, when it suddenly removed a 1% daily price change limit. A
5% limit was removed on May 22, 1992, which was followed by a doubling
of the Shanghai A share index in one day.
Another price jump followed the "market support" policies announced
by the CSRC on July 29, 1994. These "market support" policies included:
(1) A ban on new listings of A shares for the rest of 1994; (2) easier credit
availability for brokers in Shanghai through a special line of credit, in par-
ticular, the provision of a 1.15 billion U.S. dollar credit line for qualified se-
curity firms to encourage trading; (3) supporting the establishment of new
mutual funds and possible foreign participation in the domestic A share
market; (4) promised merger of the A and B share categories within five
years. Within two months following the announcement of these policies,
the Shanghai A share index nearly tripled.
We test the hypothesis that the government's stock-market liberaliza-
tion policies adopted in May, 1992 led to increased stock-market volatility.
10
We include a policy dummy variable in the best fitted model-SGARCH(l,l)
formulation, so that the conditional variance equation becomes
<j\
t
= u>i + T)jD
t
+ aj-a?
t
_! + 03*1^, (3.22)
where
_ J 0 : before the removal of the 5% daily price change limits
\ 1 : after the removal of the 5% daily price change limits.
Under the assumption that the conditional volatility process is covari-
ance stationary, i.e., (&j + (3j) < 1, equation (3.17) implies that the uncon-
ditional variance of e,- * is equal to
U]
A
before the market liberalization
" l-<Xj-Pj
10
Because the impact of t he stock-market liberalization policy in May 1992 lasted for
a long time, and we are unable to separate this effect from the impact of the stock-
market support policy in July 1994, we do not conduct a formal test on whether t he
stock-market volatility increased after the July 1994 announcement at this point. The
effect of July 1994 announcement on market volatility will be investigated later in this
section.
Government Regulation and Market Volatility 107
policies were announced. If the unconditional variance of stock returns
changed with the removal of daily price change limits, then the coefficient
r]j should be statistically significant and the unconditional variance should
become "
J+r?3
- , n, > 0.
l-aj-0j
Another test of the hypothesis that a change in policy regimes led
to increased stock-market return volatility is to divide the sample into
two sub-periods, one containing the observations before the removal of
daily price change limits and the other representing the post-removal pe-
riod. We then use an SGARCH(1,1) framework to estimate these two
sub-samples separately. If market liberalization policies contributed to in-
creased stock-market volatility, then the unconditional variance / -
1
" 2
should be smaller than ' ...
The results of the two tests are shown in Tables 3.11 and 3.12. The
coefficient estimates of the dummy variable, r)j are statistically significant
for both the Shanghai and Shenzhen A-share market returns.
11
We note
that f]j is about 28% larger for Shanghai than for Shenzhen, suggesting that
volatility in the Shanghai A-share market was influenced to a greater extent
by the removal of daily price-change limits. Moreover, the likelihood ratio
test results in Tables 3.11 and 3.12 also indicate that the null hypothesis
fjj = 0 is strongly rejected at the 5% level of significance for both Shanghai
and Shenzhen.
When we estimate the SGARCH(1,1) model using two sub-samples, we
find that the unconditional standard deviations are 0.1162 and 0.1237 for
the first sub-sample and 0.1685 and 0.1483 for the second sub-sample, re-
spectively, indicating that stock-market volatility increased by more than
45% for Shanghai but by only 20% for Shenzhen after the removal of
daily price change limits. Moreover, the null hypothesis that -. '_ -
x
<
j Pj
~ 2
-ry5j, is strongly rejected at the 5% level of significance for both the
j Pj
Shanghai and Shenzhen sub-samples. Even though additional factors other
than the removal of daily price change limits may have affected the volatil-
ity of stock-market returns over time, we conclude that a volatility increase
11
We do not use the B-share market return dat a because the Shanghai B-share market
started trading on February 21, 1992, which was too close to May 5, 1992, when the
price limits were removed. B shares were not traded in Shenzhen until December 2,
1992.
Table 3.11 Maximum Likelihood Estimates of the SGARCH(1,1) Model with Policy
Dummy Variable
Shanghai A-Shares
a
Shenzhen A-Shares
constant
r
j,t-\
ARMBUSt-!
AHKUSt-!
ATO
t
_!
MSCI
t
-!
HS
t
-!
NYSEt-!
TS
t
_
x
ATB^
CJj
Vj
&i
h
V
Mean log likelihood
LRf)
j=
o
Qsiej/vjY
-0.0064
0.3574*
0.0482
-0.7565*
0.003
-0.134
0.4154*
0.0108
-0.1505
-4.3147*
0.0013*
0.085*
0.5532*
0.3169*
1.5587*
2.0502
17.061
2.3788
[0.9671]
-0.0091
0.1014*
-0.0199
-0.1944
-0.0024
0.2589
0.0114
0.3399*
0.0045
-2.4379
0.0061*
0.0613*
0.4561*
0.4789*
1.6013*
2.1156
5.3915
4.3003
[0.8291]
"Figures inside the brackets are p-values and * denotes 5% level of significance according
to the Wald Confidence Limits.
''Likelihood ratio statistics is used to test the null hypothesis t hat f/j = 0, and is asymp-
totically distributed as Chi-square with 1 degree of freedom. The critical value for x
2
( l )
at the 5% level of significance is 3.8415.
C
Q%(EJ/OJ) denotes the Ljung-Box Q statistic for serial correlation tests with 8 lags on
standardized residuals.
of the magnitude observed could not have occurred without the observed
exogenous change in government policy. We can only speculate at this point
whether reimposing daily price change limits would improve the function-
ing of Chinese stock markets and enhance their role in the capital market
mobilization process. It is interesting to note that the security regulatory
Government Regulation and Market Volatility
109
Table 3.12 Maximum Likelihood Estimates of the SGARCH(1,1) Model Using Sub-
samples
constant
r
j,t-i
ARMBUS
t
-
AHKUSt-i
ATO
t
_:
MSCI
t
^
HSt-t
NYSEt-!
TS
t
-!
ATB
t
_!
il)j
OLj
h
V
Mean log
likelihood
Ulj
LR
b
Shanghai
First
sample
-0.0158
0.672*
_i -0.2109
-0.7789*
0.002
-0.0691
0.0395
0.0669
0.06119
-4.3269*
0.0051*
0.5113*
0.1109*
1.7576*
3.6207
A-Shares
a
Second
sample
-0.0107
0.2409*
0.0682
-0.4907*
0.0137
-0.674*
0.3973*
-0.0339
-0.08545
3.5378
0.0024*
0.6301*
0.2855*
1.6929*
1.8917
0.0135 0.0284
23.301
Shenzhen A-Shares
First
sample
-0.0089
0.528*
-0.0495
-0.0716
-0.0006
0.146
-0.274
0.253
2.616*
2.537
0.0037*
0.4053*
0.3527*
1.7615*
3.5542
Second
sample
0.0023
0.1126*
0.0223
-0.3576
-0.0023
-0.0963
-0.0222
0.3092*
-0.34
-2.8475
0.0063*
0.4232*
0.2903*
1.5995*
2.0785
0.0153 0.022
11.85
Dtes 5% level of significance according to the Wald Confidence Limits.
l/^/^^-l rr\-t-i/-\ r(+o + i (?t i pc i r nc?ari ^ - ^ t a c t + ri i mi l l VnrT"\rf~*"i"Vi*ic'io <" n o + i JZ.
and is asymptotically distributed as Chi-square with 3 degrees of freedom. The critical
value for x
2
(3) at the 5% level of significance is 7.8147.
authorities reimposed a 10% daily price-change limit on any individual stock
on December 13, 1996 in order to curb volatility.
Four volatility spikes common to Shanghai and Shenzhen securities ex-
changes are observed after the removal of daily price change limits: during
December 1992 and January 1993; in January 1994; during July and Au-
gust 1994; and in June 1995. Each of these spikes can be associated with
major changes in regulation. For example, the Shanghai and Shenzhen
stock-market indices reached their lowest historical values on November 17,
1992. To stimulate the market demand for shares, the central bank eased
the borrowing limits for brokers in both Shanghai and Shenzhen through
a special line of credit. At the same time, newly established closed-end
mutual funds started issuing and trading shares. In December 1992, the
market indices rose to an historical high level in both exchanges. In Jan-
uary 1994, the State Planning Committee announced an annual quota of
700 million U.S. dollars for new issues in that year, which was much lower
than the market had anticipated. In addition, the CSRC temporarily pro-
hibited new issue and trading of legal entity shares, which were held by
many state-owned enterprises and accounted for more than 15% of total
market capitalization. These events were accompanied by a volatility spike
in January 1994. On July 19, 1994, the CSRC announced a series of afore-
mentioned "market support and market liberalization" policies. Within two
months following the announcement of these policies, the Shanghai A share
index nearly tripled. In June 1995, the CSRC suspended market trading of
futures on government bonds and at the same time, the central bank set an
interest rate ceiling for corporate and municipal bonds. As a consequence,
a large amount of funds were transferred from the bond markets into the
stock markets.
In order to examine whether there are structural breaks in the stock-
market volatility process, we introduce the following five dummy variables
to break the time-series into five segments:
12
o f 1
* ~ \ 0
D
t
=
between May 5, 1992 and October 31, 1992
0 : otherwise;
D.
i _ / 1 : between November 1, 1992 and November 31, 1993
Di
0 : otherwise;
1 : between December 1, 1993 and May 31, 1994
0 : otherwise;
12
Due to investors' anticipation, stock-market volatility may be affected prior to the date
of the policy change announcement. We allow about two months for such anticipation.
Volatility Asymmetry and Spill-over 111
3
_ f 1 : between June 1, 1994 and April 30, 1995
* \ 0 : otherwise;
4
_ f 1 : between May 1, 1995 and September 4, 1996
* \ 0 : otherwise.
We add D\, Df, Df, and D\ variables in the conditional variance equa-
tion (3.17) and estimate the SGARCH(1,1) model. The conditional variance
equation becomes
4
a
lt = "o + Y, ^
D
\+<*i<*lt-x+PAt-i, (
3
-
23
)
1=1
The maximum likelihood estimation results are presented in Table 3.13.
None of the estimated coefficients for the above dummy variables is statis-
tically significant, although the changes in the magnitude of the coefficients
suggest that the unconditional variances decreased in Shanghai but in-
creased in Shenzhen after the announcement of stock-market liberalization
policies in July 1994. The null hypothesis that f/ij = 7]2j = fi3j = fJ4j = 0
can be rejected for Shanghai but not for Shenzhen, indicating that volatility
became smaller over time in Shanghai while it remained basically unchanged
in Shenzhen after the daily price-change limits were removed.
3.8 Volatility Asymmetry and Spill-over
The GARCH-type of models delineates the volatility as a symmetric re-
sponse in that information absorption is the same to good news as to bad
news. The partial adjustment model of Amihud and Mendelson (1987), Do-
modaran (1993), and Koutmos (1998) is of the asymmetric volatility type
in that the volatility response to good news is different from the bad news.
Amihud and Mendelson (1987) examine the role of different market
mechanisms in inducing stock return autocorrelation by means of a partial
adjustment price model. They find that stock returns are more persistent,
i.e., they exhibit greater autocorrelation when securities are traded in a
call market as opposed to a continuous auction market. Damodaran (1993)
also uses the partial adjustment model and finds that stocks listed in NAS-
DAQ adjust slower to new information than those listed in NYSE/AMEX.
Koutmos (1998) tests the hypothesis that both the conditional mean and
112
Risk, Return and Regulation in Chinese Stock Markets
Table 3.13 Maximum Likelihood Estimates of the SGARCH(1,1) Model with Dummy
Variables
constant
r
i,t-i
ARMBUSt-i
AHKUS
t
-i
ATOt-i
MSCIt-i
HSt-i
NYSEt-i
TSt-i
AT B
t
_!
Qj
a)
^
q
%
&j
fa
V
Mean log likelihood
LRrjj =o
Qs(ej/a-j)
c
Shanghai A-Shares
a
0.0063
0.4816*
0.0168
-0.1217
-0.002
-0.5788*
0.2856*
-0.0393
-0.0498
-2.1456
0.0064*
0.1946
0.1717
0.1571
0.1253
0.5196*
0.2931*
1.8062*
1.9563
13.3076
2.5536
[0.9592]
Shenzhen A-Shares
0.0044
0.1153*
0.0223
-0.3103
0.0006
-0.1731
-0.0049
0.4052*
-0.3774
-2.6475
0.0063*
0.0774
0.1171
0.1129
0.1292
0.6638*
0.3537*
1.736*
2.1165
7.828
0.0691
[0.9999]
"Figures inside the brackets are p-values and
:
to the Wald Confidence Limits.
denotes 5% level of significance according
''Likelihood ratio statistics is used to test the null hypothesis t hat rjij = 772,3 = 7)3^ =
rjij 0, and is asymptotically distributed as Chi-square with 4 degrees of freedom.
The critical value for x
2
(4) at the 5% level of significance is 9.49.
c
Qs{ej/o-j) denotes the Ljung-Box Q statistic for serial correlation tests with 8 lags on
standardized residuals.
the conditional variance of index stock returns are asymmetric functions
of past information and finds that the conditional mean is an asymmetric
function of past returns and the conditional variance rises more proportion-
ately during market declines (the so-called "leverage effect").
In this section, we test whether the stock-market returns in China adjust
to their intrinsic values through partial adjustment with asymmetries and
study the possible spill-over effects between China and Hong Kong.
3.8.1 A Partial Adjustment Model with Asymmetries
The price adjustment process is assumed to be self-adapted from the in-
trinsic value as follows:
P
t
-P
t
-
X
= {l-Q){V
t
-P
t
_{) (3.24)
where P
t
is the observed equity price and V
t
is the intrinsic value, both
expressed in natural logarithmic form. The adjustment cost gives rise to the
friction parameter, 6. The greater the value of 6, the slower the adjustment
process. As a result, the adjustment speed can be measured by (1 6).
13
The intrinsic value of a stock index is assumed to follow a random walk
with drift,
V
t
= TT + V
(
_, + e
t
(3.25)
where {et} is an i.i.d. process with zero mean and 7r stands for the expected
daily intrinsic value. Equations (3.24) and (3.25) imply that the return
13
As described in Koutmos (1998), an economic justification for equation (3.24) can be
based on the assumption t hat market price, Pt, is arrived at through minimization of
the following cost function:
C = a{Pt-Vt)
2
+b(Pt-P
t
-i)
2
,
where a(Pt Vt)
2
measures the cost of being out of equilibrium, and b(Pt Pt-i)
2
measures the cost of making the adjustment. Minimization of the above equation
implies t hat :
Pt-Pt-i =a{a + b)-
l
{V
t
-P
t
-i),
which is essentially equation (3.24) with speed of adjustment given by a(a + 6)
_1
.
process follows an autoregressive process of order one,
14
i.e.,
r
t
= (3 + Or
t
-i + e
t
(3.26)
where r
t
= P
t
P
t
-\, /? = (1 6)ir, and e
t
= (1 9)e
t
- Estimates of the
friction parameter, 9, can be obtained by simply estimating the AR(1) pro-
cess given in equation (3.26). However, if adjustment costs are asymmetric,
then equation (3.24) has to changed as follows:
p
t
- p
t
_! = (i - e
+
)(v
t
- p
t
_!)+ + (i - e~)(v
t
- p
t
_
x
)- (3.27)
in which (V
t
- P
t
-i)
+
=ma x{( V
t
- Pt - i ) , 0} and (V
t
-P
t
-i)- = min{(V
t
Pt _i ), 0} . For 6
+
= 6~ = 6, equation (3.27) reverts back to the symmetric
model. Equations (3.26) and (3.27) imply that stock returns follow an
autoregressive process of order one, given by:
r
t
= /3 + 0
+
r+_i + 0-r
t
-_
1
+ e
t
(3.28)
where r^Lj = max{r
t
_i , 0} and r ^l
1
= mi n{r
t
_
1
, 0} . As can be seen,
the asymmetric autoregressive process given in equation (3.28) utilizes two
different filters, one for positive and another for negative previous stock
returns. Again, when 9
+
= 9~ = 9, the price process is symmetric and
the {r
t
} sequence follows the usual symmetric autoregressive process. As a
result, the optimal one-step ahead forecast will be E(r
t
\lt-i) = (3 + 9r
t
-\-
However, when 9
+
^ 9", the {r
t
} sequence will exhibit more complex
behavior. The optimal one-step ahead forecast will depend on the sign of
n-i, i.e.,
JSWft-d-IStf!'" f ' - ' ^ (3.29)
[ p + 9 r
t
_
1
for r
t
_i < 0
14
Prom equation (3.24), we get Pt as follows:
P
t
= (l-0)Vt+0P
t
-i.
And then,
= Pt-Pt-i
= 0(Pt-i-p
t
_2)-i-(i-e)(v
t
-v
t
_i)
= e(P
t
-i - Pt-
2
) + (1 - 9)(e
t
+ TT)
= (i - e)n + e(p
t
-i - p
t
-
2
) + (i - e)e
t
= /3 + 0rt-i+e
t
The error process, {e
t
}, is assumed to be conditionally heteroskedastic with
time-varying conditional standard deviation given by:
+^+
u
t
- a
0
+ a
:
e ^ - a
x
e
t
_
x
+ Wt-i
(3.30)
where ef_
1
max(0, et-i) and e
t
_
2
= min(0,e
t
_i). This model is intro-
duced in Zakoian (1994) as the Threshod GARCH or TGARCH model.
The advantage of a TGARCH specification is that the conditional vari-
ance is allowed to be leptokurtotic and the probabilistic properties of a
t
are more complicated. In order to make sure that the conditional volatility
is non-negative, a
0
> 0, a+ > 0, a]" > 0, and 0 > 0 non-negativity con-
straints need to be imposed on equation (3.37). Using Generalized Error
Distribution, we employ the maximum likelihood method to estimate the
specified model while allowing the distribution parameter, u, to be esti-
mated from the data. The probability density function of the Generalized
Error Distribution is given by:
f{fi
u
<rt,v) =
V
2
r
' 3\ 1
-
\v).
2
r
\
' i M
-
W
exp
-(5)
<i)
(3.31)
where T(-) is the gamma function and v is the degree of freedom to be
estimated endogenously. When v = 2, the GED distribution is reduced to
the normal distribution and when v 1 it yields the Laplace or double
exponential distribution. Given initial values for et and of, the parameter
set 0 = (/?,9
+
,8~,ao,a^,a^,cj),v) can be estimated by maximizing the
log-likelihood over the sample period. The log-likelihood can be expressed
L(e) = j>g/(Mt,<7
t)
(3.32)
t = i
where /i
t
represents the conditional mean. When estimated using the model
described above, we may add moving average terms into the return equation
(3.28) to alleviate the problem of error distribution.
Furthermore, to understand whether the volatility equation is well-
specified for the underlying asset returns, we employ a battery of diagnostic
tests proposed by Engle and Ng (1993) to check the fit of the model. These
diagnostic statistics are the sign bias test (SBT), the negative size bias test
(NSBT), the positive size bias test (PSBT), and the joint test (JT):
SBT: ( )
2
= a + &S
t
-i+' fc (3-33)
NSBT: ^ = a + bSr^t-i + Vt (3.34)
PSBT: ( ^ )
2
=
a
+ 6 ( l - S
t
- ^ t - i + % (3.35)
JT:
fe)
2
=
a
+
6
i
5
*-i + b
2
S^
1
e
t
-i
+6
3
( l - 5- _
1
) e
t
_
1
+r
? t
(3.36)
where e
t
/o-
t
is the standard deviation and S^'_
1
= 1 if e
t
_i < 0, and S^~_
1
= 0
otherwise. For SBT, NSBT, and PSBT tests, the t-statistic of b is insignif-
icant if volatility estimation is well-specified. On the other hand, the joint
test is to test the hypothesis that &i = b
2
= 63 = 0. The test statistic for
the joint test is the product of the number of observations and R
2
and is
distributed as X
2
(3)-
3.8.2 Asymmetric Behavior on Returns and Volatility
We estimate the partial adjustment model with asymmetries for daily in-
dices of A- and B-shares in Shanghai and Shenzhen. Table 3.14 presents the
coefficient estimates. As shown on the table, the A- and B-share markets
strongly react to good and bad news in Shanghai. The reaction to negative
news is larger than that to positive news in the A-share market because
8
+
< 8~ and the likelihood ratio tests reject the equality of them. This
indicates that in general, price adjustment to intrinsic value for good news
is a little faster than for bad news in A-share market in Shanghai. How-
ever, B-share market in Shanghai seems to respond to good and bad news
symmetrically during the sample period. This reflects the nature of the
B-share market that it opens only to international investors who probably
have better access to higher quality news and can analyze news better than
domestic Chinese investors.
In contrast to the findings in Shanghai markets, the return reaction
to good or bad news is not statistically significant in Shenzhen A-share
market. This implies that the A-share market in Shenzhen may not be
asymmetric in terms of return responses to information arrival. However,
the coefficient estimates for positive and negative returns are all significant
Table 3.14 Estimates of Partial Adjustment Model with Asymmetries
p
e+
e~
a
0
af
a
i
4>
V
LRo+=0-
LR
at=a
-
Ljung-Box(12)
Shanghai
A-share
0.017*
(0.008)
0.852*
(0.064)
0.729*
(0.093)
0.135*
(0.046)
0.427*
(0.103)
0.271*
(0.066)
0.681*
(0.052)
0.926*
(0.042)
13.97*
363.15*
24.83
Shanghai
B-share
-0.092*
(0.0288)
0.347*
(0.031)
0.186*
(0.051)
0.188*
(0.085)
0.601*
(0.175)
0.492*
(0.108)
0.735*
(0.068)
1.305*
(0.061)
1.75
1.09
76.40*
Shenzhen
A-share
0.010*
(0.006)
0.472*
(0.089)
0.029
(0.033)
0.171*
(0.070)
0.205*
(0.084)
0.266*
(0.058)
0.699*
(0.054)
0.891*
(0.38)
10.06*
496.22*
28.91
Shenzhen
B-share
-0.084*
(0.053)
0.306*
(0.065)
0.074*
(0.024)
0.129*
(0.058)
0.196*
(0.079)
0.148*
(0.040)
0.618*
(0.049)
1.056*
(0.049)
1.04
5.87*
84.35*
* and t denote significance at the 5% and 10% level, respectively.
for the B-share market in Shenzhen, and the coefficient estimate of positive
return is larger than that of the negative return. Therefore, the asymmetric
return response to news is present in the B-share market in Shenzhen.
Furthermore, our empirical results show that the volatility exhibits
asymmetric behavior for both A- and B-share markets in Shanghai and
Shenzhen. This implies that past information, including both good and bad
news, can influence the information assimilation process no matter whether
the returns are eventually affected. The degree of asymmetry, calculated
as a^/a+, varies from 0.5 to 1.3. This suggests that Chinese markets are
very sensitive to negative news.
Table 3.15 Specification Tests for Volatility Asymmetries
SBT
NSBT
PSBT
JT
Shanghai
A-share
-1.188
2.37*
0.625
3.306
Shanghai
B-share
1.472
2.163*
0.891
2.924
Shenzhen
A-share
-1.305
1.286
-1.974*
2.558
Shenzhen
B-share
-2.206
f
1.047
0.835
3.179
Table 3.15 presents specification tests for volatility asymmetries (or the
news impact tests). The SBT, NSBT, PSBT, and J T statistics show that
the model captures the volatility asymmetry very well. Therefore, the par-
tial adjustment model with asymmetries can well-describe the absorption
of information in the Chinese stock markets.
3.8.3 Stock-Market Volatility Spill-ov er Between Mainland
China and Hong Kong
To understand volatility spill-over from one market to another, we need
to modify the partial adjustment model with asymmetries to account for
volatility transmission. In particular,
n,t = 5 + 7r-
i)t
_i + <j>ir
jt
t + 4>2rj,t-i + <t>3
r
j,t-2 + î,t (3-37)
h
itt
= u> + ael
t
_
1
+\el
t
_
1
dt-
1
+ (3h
itt
-i + P\r]
<t
+ P2^
2
,
t
_i+ (w\
t
_^ (3-38)
f 1 if e
M
_i < 0
\ 0 otherwise
a,
t
~ N{0, hi,
t
)
The modified partial adjustment model with asymmetries explicitly con-
siders the impact of current and past innovations of market j have on return
and volatility of market i. If market j indeed has impacts on the current
Table 3.16 Partial Adjustment Model with Asymmetric Volatility Spill-over from
Shanghai and Shenzhen t o Hong Kong
7
4>i
<t>2
4> z
a
A
P
Pi
Pi
Pi
Ljung-Box(12)
Shanghai
A-share
0.067*
(0.031)
0.012
(0.019)
0.007
(0.015)
0.006
(0.011)
0.102+
(0.059)
0.148*
(0.036)
0.427*
(0.129)
-0.096
(0.088)
-0.008
(0.014)
-0.002
(0.003)
14.82
Shanghai
B-share
0.048+
(0.026)
0.128+
(0.038)
0.022
(0.026)
0.002
(0.013)
0.064*
(0.021)
0.382*
(0.087)
0.292*
(0.087)
-0.053
(0.061)
-0.013
(0.026)
-0.001
(0.004)
19.05
Shenzhen
A-share
0.074*
(0.038)
0.022
(0.020)
0.016
(0.014)
0.006
(0.011)
0.048
(0.037)
0.207*
(0.064)
0.338*
(0.081)
-0.108
(0.115)
-0.026
(0.041)
-0.003
(0.005)
13.60
Shenzhen
B-share
0.062*
(0.032)
0.018
(0.017)
0.004
(0.012)
0.009
(0.010)
0.091
(0.088)
0.195*
(0.049)
0.401*
(0.166)
0.004
(0.007)
-0.017
(0.083)
-0.001
(0.002)
15.12
* and + denote significance at the 5% and 10% level, respectively.
return and volatility of market i, the coefficient estimates for cfii, (j>
2
, 4>3,
Pi, pi, and p3 will be statistically significant.
Tables 3.16 and 3.17 present parameter estimates for the partial adjust-
ment model with asymmetric volatility spill-over. As shown in the table,
there is no obvious influence of return and volatility spill-over from Shang-
hai and Shenzhen to Hong Kong except the B-share market in Shanghai.
This is probably because the market capitalization and trading volume of
the B-share market in Shanghai is larger than those of the B-share market
Table 3.17 Partial Adjustment Model with Asymmetric Volatility Spill-over from Hong
Kong to Shanghai and Shenzhen
7
4> x
4>2
<f>3
a
X
P
Pi
Pi
P-i
Ljung-Box(12)
Shanghai
A-share
0.157*
(0.062)
0.008
(0.010)
0.072
(0.063)
-0.017
(0.014)
1.783*
(0.429)
0.217*
(0.094)
0.387*
(0.062)
-0.017
(0.025)
-0.014
(0.019)
-0.028
(0.023)
22.15
Shanghai
B-share
0.108*
(0.041)
0.195*
(0.082)
0.102
(0.085)
0.039
(0.027)
4.061*
(0.974)
0.183*
(0.055)
0.501*
(0.091)
-0.011
(0.019)
-0.019
(0.016)
-0.047
(0.039)
72.37*
Shenzhen
A-share
0.083+
(0.054)
0.012
(0.009)
0.095
(0.071)
-0.028
(0.024)
2.682*
(0.488)
0.406*
(0.116)
0.515*
(0.096)
-0.072
(0.069)
-0.016
(0.014)
-0.016
(0.020)
17.40
Shenzhen
B-share
0.184*
(0.066)
0.163*
(0.069)
-0.019
(0.011)
-0.005
(0.009)
1.933*
(0.643)
0.320*
(0.087)
0.559*
(0.078)
0.009
(0.008)
-0.010
(0.007)
-0.025
(0.028)
58.26*
in Shenzhen.
In contrast to the spill-over effect from Shanghai and Shenzhen to Hong
Kong, the spill-over of return from Hong Kong to Shanghai and Shenzhen
is statistically significant while the spill-over of volatility is not. The cur-
rent and past stock returns in Hong Kong have significant impact on the
Shanghai and Shenzhen securities exchanges, but for B-shares only. A pos-
sible explanation is that international traders who include B-share indices
as part of their global diversification portfolios place a significant portion
Summary 121
of their other types of China investment funds in Hong Kong market. As a
consequence, their investment adjustment decisions on B-share indices will
depend on their investment decisions on Hong Kong market. However, the
current and lagged stock-market volatility in Hong Kong has little or no
impact on the volatility in Shanghai and Shenzhen. An explanation is that
the overall liquidity is lower in Shanghai and Shenzhen than that of Hong
Kong.
3.9 Summary
In this chapter, we analyze the dynamic behavior of risks and returns in
Chinese stock markets. We find that the risk-adjusted mean stock returns
are low and the volatility of stock-market returns is high in China relative
to developed markets. Moreover, returns are positively autocorrelated to
greater extent in Chinese stock markets than in developed markets. The
famous Monday effect exists in both domestic and foreign share markets,
even if market trend is removed from daily return data. We find that
the random walk hypothesis is rejected for Chinese stock markets using
variance ratio tests while it can not be rejected for developed markets.
A residual-based cointegration test shows that the daily A share market
indices in Shanghai and Shenzhen tend to move together, but the daily B
share market indices do not.
We also establish an empirical model to capture the deterministic com-
ponents of the variation in the stock-market returns and characterize the
second order conditional moments using three error generation processes.
We find that the variance of stock-market returns is time-varying, mildly
persistent, leptokurtotic and is influenced by exogenous variables repre-
senting government market liberalization policies. We also find that the
Shanghai B-share market is more integrated with the world equity markets
than any other Chinese market, although even here the degree of integra-
tion appears to be weak. A share returns in Shanghai and Shenzhen tend
to respond similarly to common news and economic factors, leading me to
believe that there is potential for them to become a fully integrated national
stock market. Encouraging cross-listing and eliminating the distinction be-
tween different classes of shares will improve stock-market liquidity and
enhance capital mobilization.
In addition, we find that government's market liberalization policies
have contributed to the high stock-market volatility in China. We also
conjecture that market segmentation, both across different stock exchanges
and across different share types, and extraordinary large IPO underpricing
may have contributed to the high stock-market volatility as well.
Finally, we investigates the partial adjustment process with asymmetries
on stock indices in China. We find that stock-market returns and volatility
exhibit asymmetric adjustment behavior in all markets except the A-share
market in Shenzhen. The adjustment speed is faster for bad news than
for good news. We also find that current and past returns and volatility
in Hong Kong stock market have significant impact on B-share markets
in Shanghai and Shenzhen. Our empirical results are consistent with Chui
and Kwok (1998), who find significant cross-correlation between the B-share
markets in China and Hong Kong market.
The empirical model used in this chapter is a univariate model. There-
fore, some interesting questions are left unexplored. For example, what are
the sources of differences in expected returns and volatility in A- and B-
share markets? Has the degree of integration among Chinese stock markets
and world markets improve over time? Analyzing these questions requires
exploiting a multivariate factor asset pricing model with time-varying world
market integration.
Chapter 4
Ownership Restrictions and Foreign
Shares Discount
4.1 Introduction
It is widely documented that ownership restrictions and market segmenta-
tion result in price differentials among classes of shares in many emerging
stock markets. For example, Eun and Jankiramanan (1986) examine a two-
country model in which investors from each country attempt to maximize
the expected utility of wealth subject to a one-way investment barrier on
foreign investors and domestic investors can have no limits on security hold-
ing internationally. Errunza and Losq (1985) study a special case of Eun
and Jankiramanan model in which foreign ownership is completely banned
in one country and domestic investors are not subject to restrictions in the
other country. Hietala (1987) develops a single-period capital asset pric-
ing model (CAPM), incorporating specific investment restrictions imposed
by Finnish government (Finnish people must hold at least 80% of domes-
tic shares). Bailey and Jagtiani (1994) examine the effects of investment
barriers in Thailand and find that cross-sectional differences in local and
foreign share prices are correlated with proxies for foreign ownership limits,
liquidity and information availability. Stulz and Wasserfallen (1995) use
Hietala's model with added implication. They develop and test a price dis-
crimination model assuming domestic country is relatively small, relaxing
the holding restriction will not affect foreign investor's required return and
will therefore not affect the free share price. Johnson (1995) introduces a
multi-period model to explain the existence of a time-varying premium be-
tween free and restricted shares in the case of Sweden, assuming free shares
are available only for domestic investors and restricted shares are available
123
124 Ownership Restrictions and Foreign Shares Discount
for both domestic and foreign investors.
The price differential raises two questions: First, what economic factors
determine domestic and foreign investors' portfolio choices under ownership
restrictions? Second, why do firms issue both types of shares, if one type
sells at a higher price than the other?
In this chapter, we extend Stulz and Wasserfallen's one-period price
discrimination model to multiple domestic firms that issue A and B shares
at the same time, assuming A shares are only available to domestic investors
and B shares can only be owned by foreign investors, a case that mostly
resembles the restrictions in the Chinese stock markets. The model posits
that cross-sectional variations in the expected return between two classes
of sharesthose purchased and traded by domestic and foreign investors,
respectively, are determined by the shares' own market betas and betas
with respect to the international financial markets. We also formulate an
Intertemporal Capital Asset Pricing Model (ICAPM), assuming domestic
investors have a smaller investment opportunity set than foreign investors,
as is the case in China. The solution of the ICAPM also clearly shows
that cross-sectional variations in the expected return between domestic and
foreign shares are determined by the shares' own market betas and betas
with respect to the international financial markets. Given that we reach
the same results using two different theoretical models, we firmly believe
that we have solid ground for further empirical work. After testing the
predictions from both models using appropriate econometric techniques,
we find that:
Cross-sectional excess of domestic over foreign shares' average re-
turns are positively related to domestic shares' market betas and
negatively related to both foreign shares' market betas and betas
with the Hong Kong Hang Seng index.
The variance of shares' returns and the firm size do not appear to
influence cross-sectional return differences.
These results strongly support both price discrimination model and the
ICAPM formulation.
The rest of the chapter proceeds as follows: In Section 4.2, we describe
the effects ownership restrictions have on stock prices in China. In Sec-
tion 4.3, we develop a price discrimination model with effective market
segmentation. In Section 4.4, we analyze the ICAPM under ownership re-
strictions. In Section 4.5, we derive some testable implications from both
Ownership Restrictions in Chinese Stock Markets 125
models. In Section 4.6, we describe the methodology in empirical tests and
present the empirical results. We conclude this chapter in Section 4.7 with
a summary of findings.
4.2 Ownership Restrictions in Chinese Stock Markets
Chinese stock markets are segmented in two aspects. First, dual listing is
not allowed so that each company's issue is restricted to one of the two ex-
changes. In this way, the domestic and international share markets remain
distinct. In other words, a company officially listed in one exchange cannot
be listed in another exchange. Second, a listed company in either exchange
can issue to two types of shares: A-shares for domestic investors and B-
shares for foreign investors. Chinese citizens are only permitted to trade
A-shares, while international investors, including overseas Chinese residing
in Hong Kong, Macau, or Taiwan, can only participate in the trading of
B-shares in mainland China.
China's stock markets opened to international investors on February 21,
1992, when Shanghai Vacuum Electronics began issuing B-shares in Shang-
hai Securities Exchange. This was quickly followed by the listing of China
Southern Glass B-shares in Shenzhen Securities Exchange on February 28,
1992. For each company, A- and B-shares are entitled to the same rights
and dividends. Henceforth, international investors were allowed to partici-
pate in Shanghai and Shenzhen exchanges, but only in B-shares.
1
Column one in Tables 4.1-4.4 list the companies that have issued B
shares in either Shanghai or Shenzhen securities exchange by the end of
August 1995. Column four of the tables report the initial public offering
prices for A- and B-shares. A-share prices are converted to either U.S. dollar
prices or Hong Kong dollar prices for the ease of comparison. As shown in
the tables, A-shares are generally sold at substantially higher prices than
B shares.
1
In 1993, China created an additional class of stock to facilitate the direct listings of
Chinese companies on foreign stock exchanges, known as H-shares (for Hong Kong
Stock Exchange) and N-shares (for New York Stock Exchange). H-shares and N-shares
carry the same rights and obligations as A- and B-shares, but they can only be traded
on stock exchanges outside of China.
Table 4.1 Profiles for Companies Issuing Both A and B Shares in the Shanghai Secu-
rities Exchange (As of August 1995)
Company
Shanghai Vacuum Electro.
Shanghai Erfangji
Dazhong Taxi
Yongsheng Stationery
China First Pencil
China Textile Machinery
Shanghai Rubber Belt
Shanghai Chlor Alkai
Shanghai Tire & Rubber
Shanghai Refrigerator
Jinqiao Export & Import
Outer Gaoqiao
Lianhua Fiber
Shanghai Jin Jiang Tower
Forever Bicycle
Phoenix Bicycle
Shanghai Haixian Group
Yaohua Pilkington Glass
Dajiang Group
Shanghai Diesel Engine
Hero Pen
Sanmao Textile
Shanghai Friendship Store
Industrial Sewing Machine
Shang-Ling Refrigerators
Shanghai Steel Tube
Shanghai Merchandise Trading
Shanghai Automated Machine
Communication Equipment
Shuixian Electronics
Lujia Development
Huaxin Cement
Shanghai Xinya Group
Code
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
930
931
932
933
934
Number of
A-Shares*
12139
5882.8
1300
600
1320
2340
660
2300
2080
1300
7500
1950
900
2400
1300
2000
1100
2500
1950
1800
1300
1000
1204
1200
1800
1320
1100
2420
1300
1595
6300
1300
1200
Price
(in USD)
1.14
3.14
2.55
3.5
1.86
1.71
1.99
0.92
2.49
1.06
1.7
2.56
2.06
1.96
1.18
1.57
2.0
1.69
1.72
1.95
1.48
1.18
1.46
1.58
1.22
1.1
1.95
1.34
2.62
1.39
2.99
0.56
1.39
*: in 10,000 shar es
Table 4.2 Profiles for Companies Issuing Both A and B Shares in t he Shanghai Secu-
Company Number of I PO Da t e I PO Pr i ce
B-Shares* for B Shar e (in USD)
Shanghai Vacuum Electro.
Shanghai Erfangji
Dazhong Taxi
China First Pencil
Outer Gaoqiao
Lianhua Fiber
Forever Bicycle
Phoenix Bicycle
Dajiang Group
Hero Pen
Sanmao Textile
Shanghai Steel Tube
Shanghai Merchandise
Shanghai Automated Mac.
Communication Equipment
Shuixian Electronics
Lujia Development
Huaxin Cement
Shanghai Xinya Group
14250
19250
7800
3000
6600
10920
3300
33600
22100
6500
14300
16575
3600
9243
6000
10000
3850
10000
4500
11000
4550
3300
4800
7500
7900
8800
5500
7700
6000
10000
20000
4500
10000
Feb. 12, 1992
July 1, 1992
July 22, 1992
July 22, 1992
July 28, 1992
July 28, 1992
July 28, 1992
Aug. 20, 1992
Aug. 28, 1992
Nov. 16, 1992
May 21, 1993
July 29, 1993
Oct. 18, 1993
Oct. 18, 1993
Nov. 12, 1993
Nov. 12, 1993
Dec. 10, 1993
Dec. 10, 1993
Dec. 15, 1993
Dec. 28, 1993
Dec. 28, 1993
Dec. 31, 1993
Jan. 5, 1994
Jan. 18, 1994
Jan. 31, 1994
Mar. 15, 1994
Mar. 30, 1994
Apr. 29, 1994
Nov. 10, 1994
Nov. 10, 1994
Nov. 22, 1994
Dec. 9, 1994
Dec. 9, 1994
0.72
0.59
0.652
0.835
0.525
0.61
0.53
0.798
1.1
0.461
0.344
0.39
0.3792
0.3792
0.405
0.68
0.7
0.72
0.4023
0.523
0.439
0.3425
0.44
0.42
0.75
0.75
0.258
0.248
0.52
0.265
0.668
0.233
0.35
*: in 10,000 shar es
Table 4.3 Profiles for Companies Issuing Both A and B Shares in the Shenzhen Secu-
Company Code Number of
A-Shares*
Price
(in USD)
Shenzhen Vanke Co.
Gintian Industry
Property & Resource Devel.
China Southern Glass
Shenzhen Petroch. Co
Shenzhen Zhonghao Co.
Konka Electronics
Shenzhen China Bicycles
Victor Onward Textile
Shenbao Industry
Shenzhen Huafa Electronics
Chiwan Wharf Holdings
China Merchants Shekou Co.
Tellus Machinery
Fiyta Holdings
Yili Mineral Water
SEZ Real Estate
Shenzhen Lionda Holdings
Shenzhen Nanshan Power
China Marine Containers
Shenzhen Textile
Hainan Pearl River Ent.
Lizon Pharmaceutical Co.
Guangdong Electric Power
Foshan Electric Lighting
Jiangling Motors
2002
2003
2011
2012
2013
2015
2016
2017
2018
2019
2020
2022
2024
2025
2026
2028
2029
2030
2037
2039
2045
2505
2513
2539
2541
2550
5673
6877.6
4150
2030
2350
5972.2
3015
3280
5972.2
1940
2963
4600
9975
2598
3350
3000
11200
4350
4350
1776
8200
11340.5
4978.3
9855
55566
35324
23.77
19.65
23.3
19.85
18.15
9.74
20.01
11.6
9.74
8.85
13.33
13.15
11.24
10.54
12.67
12.56
9.65
10.67
11.3
10.08
5.93
3.89
3.47
4.44
8.22
3.88
T:
in 10,000 shares
All firms can issue A-shares, but as discussed further below, firms must
appl y for permission t o issue B-shares. To limit foreign ownership, t he
Chinese government allows no more t han 25% of an enterprise' s convertible
shares t o be B-shares. Unlike Indonesia and Thai l and, where individuals
are permi t t ed t o purchase local shares and have t hem registered as foreign
shares t hr ough a foreign sponsor, Chinese investors are allowed t o purchase
and t r ade only A-shares while foreign investors can purchase and t r ade only
B-shares. Fi r ms t hat issue B-shares are entitled t o favorable t ax t r eat ment
Table 4.4 Profiles for Companies Issuing Both A and B Shares in the Shenzhen Secu-
Company
Shenzhen Vanke Co.
Gintian Industry
Konka Electronics
Shenbao Industry
Tellus Machinery
Fiyta Holdings
Yili Mineral Water
SEZ Real Estate
Shenzhen Textile
Hainan Pearl River Ent.
Lizon Pharmaceutical Co.
Guangdong Electric Power
Foshan Electric Lighting
Jiangling Motors
Number of
B-Shares*
4500
3800
3000
1600
1500
2000
1000
2550
2000
1800
2350
4000
5000
2000
1500
2000
10000
3000
3000
1300
2500
5750
2132.5
9000
50000
17400
IPO Date
for B Share
May 28, 1993
June 29, 1993
Mar. 30, 1992
Feb. 28, 1992
May 6, 1992
June 16, 1992
Mar. 27, 1992
Mar. 31, 1992
June 16, 1992
Oct. 12, 1992
Apr. 28, 1992
May 5, 1993
June 7, 1993
June 21, 1993
June 3, 1993
Aug. 9, 1993
Jan. 10, 1994
Sept. 29, 1993
Sept. 29, 1993
Mar. 23, 1994
Aug. 15, 1994
June 29, 1995
June 28, 1995
June 28, 1995
Aug. 8, 1995
Sept. 29, 1995
IPO Price
(in USD)
10.53
10.71
5.3
5.3
4.28
2.50
5.5
5.3
2.5
3.6
2.69
2.83
3.23
4.03
3.48
3.17
3.5
3.5
3.5
7.54
2.39
2.56
4.29
4.29
5.61
1.1919
: in 10,000 shar es
and more freedom to import and export goods.
Although A- and B-shares represent claims to the same cash flow rights,
B-shares are generally sold and traded at substantially lower prices than A-
shares. Define A-share premium as (PA PB)/PB, where PA is the price for
A-share and PB is the price for the corresponding B-share. Tables 4.5 and
4.6 present the time-series average weekly A-share premia for 27 Shanghai
firms and 20 Shenzhen firms that issued both A- and B-shares before April
1994. The sample period is between April 1994 and October 1998. Define
1
3
0

O
w
n
e
r
s
h
i
p

R
e
s
t
r
i
c
t
i
o
n
s

a
n
d

F
o
r
e
i
g
n

S
h
a
r
e
s

D
i
s
c
o
u
n
t

(
0

n
>

^
_

(
0

.
c

C
O

m

-

c

<
c

<
t

T
O

a
>

n s s i <
D

(
"

t
-
j

(
0

F

k
_

i
s
:

r
^
.

t
:
r

994
*

i
_

i
.
J
_

<

n
>

*
_

n

C
D

-
Q

Table 4.5 Sample Statistics for A-share Premia for Shanghai Companies That Issued
A- and B-shares Before April 1994 (Sample Period: April 1, 1994-October 31, 1998)
Company
Shanghai Vacuum Electronics
Shanghai Erfangji
Dazhong Taxi
China First Pencil
Outer Gaoqiao
Lianhua Fiber
Forever Bicycle
Phoenix Bicycle
Dajiang Group
Hero Pen
Sanmao Textile
Shanghai Steel Tube
Shanghai Merchandise Trading
Mean
0.95
0.56
1.26
0.61
0.75
0.87
1.31
0.63
0.56
0.98
0.56
0.62
0.92
0.79
0.61
0.68
1.51
0.88
0.61
0.55
0.73
1.38
0.85
1.49
0.62
1.25
0.75
Std. Dev.
0.51
0.31
0.53
0.39
0.23
0.47
0.61
0.37
0.34
0.43
0.34
0.38
0.39
0.38
0.36
0.29
0.63
0.34
0.38
0.28
0.36
0.49
0.48
0.54
0.36
0.46
0.32
Minimum
0.13
-0.11
0.15
-0.26
0.33
-0.35
0.25
0.20
0.11
0.17
-0.09
-0.19
-0.05
0.24
0.08
0.30
0.42
0.21
-0.04
0.09
-0.13
0.26
-0.25
0.16
0.12
0.28
-0.06
Maximum
1.64
1.25
1.93
1.16
1.30
1.66
2.38
1.26
1.09
1.76
1.23
1.46
1.83
1.62
0.97
1.14
2.75
1.92
1.27
0.96
1.37
2.12
1.63
2.24
1.27
2.10
1.27
B-share discount as (PB PA)/PA- Figure 4.1 plots the unweighted average
weekly B-share discount for these firms during the sample period. As shown
in the figure, the magnitude of the discount is persistently large but varies
over time. The average B-share discount over the entire sample period is
59.18%, with the spreads in Shenzhen about half of those in Shanghai.
A complete analysis of this price difference would explain both why dif-
ferent investor groups pay different prices and why firms sell both shares
when their price differ. Any attempt to model firm's behavior would en-
Table 4.6 Sample Statistics for A-share Premia for Shenzhen Companies That Issued
A- and B-shares Before April 1994 (Sample Period: April 1, 1994-October 31, 1998)
Company
Shenzhen Vanke Co.
Gintian Industry
Konka Electronics
Shenbao Industry
Tellus Machinery
Fiyta Holdings
Yili Mineral Water
SEZ Real Estate
Mean
0.43
0.62
0.32
0.36
0.58
0.49
0.26
0.28
0.69
0.74
0.24
0.28
0.14
0.08
0.04
0.85
0.87
0.33
0.45
0.13
Std. Dev.
0.21
0.27
0.18
0.20
0.34
0.31
0.15
0.19
0.37
0.39
0.16
0.16
0.17
0.16
0.32
0.27
0.28
0.20
0.24
0.22
Minimum
0.08
0.09
-0.05
0.04
0.11
-0.06
-0.18
-0.23
0.09
0.14
-0.17
-0.14
-0.58
-0.24
-0.57
0.04
-0.09
-0.14
-0.28
-0.24
Maximum
0.87
1.08
0.68
0.83
1.07
1.12
0.62
0.75
1.41
1.57
0.70
0.69
1.06
0.46
0.80
1.27
1.56
0.93
0.76
0.53
counter a number of difficult issues. Stulz and Wasserfallen (1995) use a
one-period CAPM to analyze the price differentials between restricted and
unrestricted shares in the Swiss case, in which firms choose ownership re-
strictions voluntarily. They argue that price differentials can arise when a
representative domestic firm maximizes its value by discriminating domes-
tic and foreign investors.
To apply their framework to the Chinese case, we note that there is
enough evidence from the initial public offering process and daily trading
activities that domestic demand for A-shares is enormous while the foreign
demand for B-shares is weak and A- and B-share markets are effectively
segmented. The amount of new shares issued in each year is determined by
a national quota set by the State Planning Committee and the CSRC. The
quota is distributed to individual provinces and then to individual firms.
When a firm is approved for a listing, it usually sets an IPO price so that
the price-earning ratio is around 15 to 25. To issue foreign shares, firms
must obtain special approval from the CSRC and local security regulatory
authorities. An issuer of B-shares must, besides satisfying requirements
stated in the securities regulations, meet the following conditions:
(1) It must have obtained approval from the relevant authorities for its
use of foreign investment or for its conversion into a foreign-funded
enterprise.
(2) It must have a stable source of adequate foreign exchange income
and the total amount of its annual foreign exchange income must
be sufficient to pay the annual dividend.
(3) The proportion of B-shares to the total number of shares, in a
company which has been restructured from a state enterprise into
joint-venture or private enterprise, must not exceed the ceiling de-
termined by the relevant authority.
The aggregate amount of shares is fixed in each year and the total number
of firms allowed to issue foreign shares is also limited.
There are a few limitations on price discrimination approach: First, as-
sumptions are needed on whether firms are engaged in Cournot competition
or in cartel competition in allocating A- and B-shares, or whether firms is-
sue foreign shares simply to enhance their reputation. Second, It is unclear
whether the allocation of shares may result from the government's planning
behavior. Third, as Stulz and Wasserfallen (1995, p.1054) note, the price
discrimination strategy they describe is likely to be time-inconsistent in a
dynamic framework.
In light of these peculiarities, we analyze the price and return differen-
tials using another approach (besides price discrimination approach). We
do not model the firms' behavior. Instead, we use asset pricing theory to
explain why different investors pay different prices for exogenously deter-
mined quantity of shares. Since a stock's price equals the present value of
expected future dividends discounted at investors' required rates of return,
different shares with the same expected future cash flows should have the
same prices unless different investors have different required rates of returns
due to different valuation of risks. B-share investors are primarily foreign in-
vestors who are likely to have different risk exposure or altitude toward risks
in Chinese financial markets than domestic investors do. For example, in a
simple dividend growth model, P
t
= D
0
(l +g)/(ri g) where P$ is the price
1
3
4

O
w
n
e
r
s
h
i
p

R
e
s
t
r
i
c
t
i
o
n
s

a
n
d

F
o
r
e
i
g
n

S
h
a
r
e
s

D
i
s
c
o
u
n
t

A Price Discrimination Model 135
for a specific type of share, g is the dividend growth rate (assumed as con-
stant) and r is the investor's required rate of return for that type of share.
From simple manipulation, {P
B
-P
A
)/P
A
= (r^' "BJ/ C^B Q)- If dividend
growth rate is assumed to be zero, then (PB PA)/PA = {T~ATB)/TB- The
cum dividend average returns for A- and B-shares are 8.27% and 18.79%
in the sample period, respectively. These return figures are consistent with
an average B-share discount of 59.18%. Figure 4.2 plots the time-series
differences in average weekly returns between B- and A-shares during the
sample period, As shown in the figure, B-share returns are on average
57.44% higher than their A-share counterparts.
4.3 A Price Discrimination Model
In this section, we describe a one-period partial equilibrium model for stock
pricing and portfolio behavior in a two-country world economy with invest-
ment restrictions across countries. This one-period model is a variation of
the price discrimination model of Stulz and Wasserfallen (1995) in the sense
that there are multiple domestic firms who issue different types of shares
and domestic investors are not allowed to hold foreign securities. It is still
a static partial equilibrium model because the total supply of domestic A-
and B-shares are assumed to be a constant S and the rates of return for
domestic and foreign risk-free assets are assumed to be fixed and do not
depend upon the investor's maximization problem. The purpose of this
model is to demonstrate the difference in the prices for B-shares and the
corresponding A-share prices. The model also explains the difference in the
holding for premium-priced A-shares by domestic investors and discount-
priced B-shares by foreign investors in the presence of domestic ownership
restrictions.
4.3.1 Modeling Ownership Restrictions
Throughout this section, we assume that there are only two countries in
the world, the domestic country D (D = D U D
2
) and the foreign country
F (F = Fi U F
2
). An investor i is a domestic investor (foreign investor) if
i G Di (i G Fi). Di and JF\ are the sets of investors in domestic and foreign
countries, respectively, and Di = (1, ,N), i<\ = (1, , M). A
firm j is a domestic firm (foreign firm) if j e D
2
(j F
2
). D
2
and F
2
are
the sets of firms in domestic and foreign countries, respectively. I assume
there are Q domestic firms who issue two different types of shares but
there is only one foreign firm who issue a single type of shares. Therefore
D
2
= (l, , Q) a n d F
2
= (l ).
A representative firm in the domestic country issues a single stock but
in two types of shares: A- and B-shares. A-shares are sold to domestic
citizens only while B-shares are sold solely to foreign investors. A domestic
firm has to decide whether it is optimal to charge different prices for A-
and B-shares given the fact that markets for A- and B-shares are entirely
segmented. It will be shown later that a necessary and sufficient condition
for the representative domestic firm to price discriminate is that domestic
and foreign investors have different demand functions for the firm's stock.
A representative firm in the foreign country issues a single type of stock: C-
shares. C-shares are only available to investors residing in foreign countries.
A risk-free debt instrument (Lr},Lp) exists in each country. The following
ownership restrictions exist in the world economy:
(1) Domestic investors (i D{) are prohibited from buying B- and
C-shares;
(2) Foreign investors (i e Fi) are prohibited from buying A-shares;
(3) No short selling of A- and B-shares are allowed in domestic country.
Under the above situation, domestic investors face a smaller investment
opportunity set than foreign investors. Therefore it is likely that domestic
investor's demand for A-shares is less elastic than the foreign investor's
demand for comparable B-shares. A domestic firm can choose to price
discriminate between these two groups of investors in order to maximize
the proceed from selling its stocks.
Denote Pj
t
k as the price for type-fc stock j at time 0, Pj
:
k as the payoff
of type-fc stock j at time 1, rp, as the rate of return on the domestic risk-
free debt instrument Lp> at time 1, and rp as the rate of return on the
foreign risk-free debt instrument Lp at time 1. Pj
t
k includes capital gain
and dividend. It is unknown to investors at time 0 and is therefore itself
random variables.
Also let Dij
t
k = Di{Pj
t
k) as the demand function for type-fc stock j for
investor i and is differentiable with respect to Pj^. Also denote Sj
t
k as the
supply for type-fc stock j . Market equilibrium implies that Sj
t
k J2i î,j,k-
4.3.2 Investors' Maximization Problems
A domestic investor i chooses the optimal amount of A-share stocks to
maximize the indirect utility from her end-of-period wealth:
max E[U(Wi(l))] = E[Wi(l)] - \
D
Var{Wi(l)) (4.1)
{i,A>
s.t. Wi{0) = D'
iA
P
A
+ L
D
(4.2)
Wi(l) = D'
iA
P
A
+ (1 + r
D
)L
D
(4.3)
where A, A is a (1 x Q) demand vector for A shares for Q domestic firms
and D
iA
= ( A, I , A, A,2,A, A, Q, A) - A, J , A > 0, i.e., the holding of a
domestic firm's share cannot be negative. PA is a (Q x 1) price vector for
all A-shares at time 0 and P
A
= (Pi,A, PI,A, , PQ,A)- PA is a (Q X 1)
vector for A share payoffs at time 1 and PA = (A, A, PI,A, , PQ,A)-
XD is the coefficient for absolute risk aversion for a domestic investor i.
Wj(0) is her initial wealth and Wj(l) is her end-of-period wealth.
EWm - IXoVariW^l)) = D[
A
E(P
A
) + (1 + r
D
)[Wi(0) - D[
A
P
A
]
+ \*DD'
itA
V
AA
Di,
A
(4.4)
where V
AA
is the variance-covariance matrix of the A-share payoffs for all
Q firms at time 1. The solution to the above wealth maximization problem
yields the following demand vector for A-shares for domestic investor i:
A,A = r^-VTi
E(P
A
) - (1 + r
D
)P
A (4.5)
or,
where
A , A = 4>D- ^DPA (4.6)
I>D = T-VX\E(P
A
), and (4.7)
An
^D = ^-^VX}. (4.8)
Ar
Therefore, the demand for A-shares depends on domestic investor's ab-
solute risk aversion coefficient, future A-share payoff, the variance-covariance
matrix of A-share payoffs and the risk-free rate in the domestic country.
In equilibrium, the aggregate demand for A-shares for all N domestic
investors must equal to the aggregate supply of A-shares by Q domestic
firms. Hence,
S
A
= ND
i<A
= N{ct>
D
- ip
D
P
A
) (4.9)
l+r
D
{l + r
D
)N
A foreign investor chooses the optimal amount of B- and C-shares to
maximize the indirect utility from her end-of-period wealth:
max E[U(Wi(l))] = [Wi(l)] - \\
F
Var{W,{\)) (4.11)
s.t. Wi(0) = D'
iB
P
B
+ D
itC
Pc + L
F
(4.12)
Wi(l) - D'
iB
P
B
+ D
iiC
Pc + (1 + r
F
)L
F
(4.13)
where Di
t
s is a (Q x 1) demand vector for B-shares for Q domestic firms,
Di,j,B > 0, andD-
) B
= ( A, i , s , A, 2, s, , A. Q. B) - PB is a ( Qxl ) price
vector for all B-shares at time 0, P
B
= (PI,B, PI,B, , PQ,B) and PB is
a (Q x 1) vector for B-share payoffs at time 1. A, c = A, i , c, Pc = Pj,c
and Pc = Pj,c because it is assumed that there is only one foreign firm
that issues C-share. And X
F
is the coefficient for foreign investor's absolute
risk aversion.
E[U(Wi(l))] = E[Wi(l)]-\
F
Var(Wi(l))
= D'
itB
E(P
B
) + D
itC
E(P
c
)
+(1 + r
F
)(Wi(0) - D'
i>B
P
B
- D
iiC
Pc)
-
l
-\
F
(D'
UB
V
BB
D^
B
+Dl
c
a
2
c
+2D'
iiB
po
2
c
D
it
c) (4.14)
where V
B
B is the (Q x Q) variance-covariance matrix of B-share payoffs for
all domestic firms at time 1. j3 is the (Q x 1) vector of /3-coefficients of the B-
share payoffs with respect to the C-share payoffs, i.e., J3 (/3i, /?2, , A?)
and 0, =Cov(P
jtB
,Pc)/oc-
The solution to the above representative foreign investor's maximization
problem yields the following demand vector for B-shares:
D.
i,B
where
V
R R
-
VBCVBC
\ I 'BB 9
A
F
{E(P
B
)-(l+r
F
)P
B
-P [E{P
C
) - (1 + r
F
)Pc\ }
J _ 1 v '
V
BCVBC
(4.15)
(4.16)
[E{P
B
) - p [E{P
C
) - (1 + r
F
)P
c
] } (4.17)
ipf
l+r
F
\F
V,
V
BC
V
}
BB
BC
(4.18)
and VBC is a (Q x 1) vector of covariances of B-shares payoffs with C-shares
payoffs at time 1.
Therefore, the demand for B-shares depends upon the foreign investor's
absolute risk aversion coefficient, future B-shares and foreign C-share pay-
offs, variance-covariance matrix of B-shares payoffs, covariances between
B-share payoffs and C-share payoffs, /3-coefficients of the B-share payoffs
with respect to C-share payoffs, and risk-free rate in foreign country.
In equilibrium, the aggregate demand for B-shares by all M foreign
investors must equal to the aggregate supply of B-shares by Q domestic
firms. Hence,
S
B
= MD
itB
= M{4>
F
- ip
F
P
B
) (4.19)
E(P
B
)-{3 E(P
c
-(l+r
F
)P
c
A
F
(V
i
BB
l+r
F
V
BC
V'BC)
(4.20)
(l+r
F
)M
To find out whether the demand functions for A- and B-shares are the
same or not, we calculate the ratio of the slope coefficients and the ratio of
the constant terms from the two demand functions:
D<PF VXXE{P
A
)
V,
BB
VBCV
B
C
{E{P
B
-(3[E(P
c
)-(l+r
F
)Pc\}
1
(4.21)
V'nV'f
AF l+r
D
l+r
F
v xl
V,
BB
VBCV
BC
(4.22)
The implication of equations (4.21) and (4.22) can be summarized in
the following proposition:
Theorem 4.1 A necessary and sufficient condition for all domestic firms
to price discriminate between domestic and foreign investors is (J)D4
>
~
F
1
7^ 1
and ipDipp 7^ 1) where 1 is a (Q x 1) vector of 1.
Since A- and B-shares are identical except there are restrictions on own-
ership, i.e., domestic investors can only own A-shares and foreign investors
can only hold B-shares, the expected payoffs on A- and B-shares should
be the same. It is also reasonable to assume that the present value of the
expected payoff on C-shares is greater than the time-0 price and foreign
investors are more risk-averse than domestic investors in investing in Chi-
nese shares (XD < X
F
). Bailey (1994) provides empirical evidence that the
/3-coefficients of the B-share payoffs with respect to the foreign share (i.e.,
C-share) payoffs in Chinese stock markets are very close to 0. If the rates
of return on domestic and foreign risk-free debt instruments are the same
(r
D
= r
F
), and if variance-covariance matrices of the payoffs of A- and B-
shares are the same (VAA = VBB), then from equations (4.21) and (4.22),
it is not difficult to derive that ^D^
1
>
e a n
d ipDip]?
1
> e.
Therefore, it is optimal for domestic firms to charge higher prices for A-
shares than for B-shares, despite the fact that A- and B-shares are entitled
to the same type of ownership and liability of the firms' assets. Of course,
in order for a firm to price discriminate between two investor groups, the
firm has to be able to prevent resale (Carlton and Perloff, 1994).
4.3.3 Firms' Maximization Problems
A representative domestic firm faces identical domestic investors in their
demand for A-shares and identical foreign investors in their demand for
B-shares. It has to decide the optimal amount of A- and B-shares issued
in order to maximize its proceed from selling the stocks taken as given the
amount of A- and B-shares other firms issue. The equilibrium prices for A-
and B-shares will then be determined by domestic and foreign investors'
JV M
i = l t = l
ND
itA
P'
A
+ MD
iiB
P
B
a ,
h
-i
N
^p-S
A x
M(j)'
F
-S
B
>AV
D
^ h >BV
F
^ (4.23)
SA + SB = S
S
A
>0
S
B
>0
where Z is a (Ax 1) vector of firms' proceeds from selling the stocks, S
A
and
S
B
are (Q x 1) vectors of firms' supply for A- and B-shares, respectively. 0
is a (Q x 1) vector of zeros.
To get equation (4.23), we simply use equations (4.6), (4.16) and the
market equilibrium condition that aggregate demand for a firm's shares
equals aggregate supply, and make necessary transformations. The solution
to the firms' maximization problems yields the following vector on optimal
amount of A- and B-shares issued:
S*
A
= \NM{Mip
F
+ N^
D
)-
1
L
F
<t>
D
- ^
D
cf>
F
+ ^DS'\ (4.24)
demand functions.
max Z =
{S
A
,S
B
}
S*
B
= ^NM(Mip
F
+ N^
D
)~
l
U
D
<f>'
F
- y ,
F
4 + ^
F
S'\ (4.25)
The above solution to the firm's maximization problem is an interior
solution. In the next section, we prove that this interior solution is almost
guaranteed given reasonable assumptions on risk-free rates and covariances
of the B-shares payoffs with respect to the C-share payoff.
Substituting the inverse demand vector equations (4.8) and (4.18) into
equations (4.24) and (4.25) yields the following equilibrium price vectors
for A- and B-shares:
W 5
1
- 7;M(Mi,
F
+ N^D)-
1
^
D
^
D
l
+y
(4.26)
P
B
= < ^ / - \M{M^
F
+ N^
D
)~
l
U
D
4>'
F
^ -4>
D
+ | S ) (4.27)
After manipulation,
P*
2 i&DlpD
1
^F
1
)
or,
E(P
A
) _ E(P
B
) PlE(P
c
-(l+r
F
)Pc)
1 + r
D
1 + r
F
I +r
F
(4.28)
(4.29)
The price discounts on B-shares is determined by the expected payoffs
for A-, B- and C-shares, the /3-coemcients of the B-share payoffs with re-
spect to C-share payoff, and the rates of return of domestic and foreign
risk-free debt instruments. The following proposition summarizes the re-
sult:
Theorem 4.2 A necessary and sufficient condition for all domestic firms
to charge higher prices for A shares is
{<t>Di>D
l
~ <t>Fip
F
l
) >Qor
E(P
A
) E(P
B
)
|
B[E{P
C
1 +r
D
1 + r
F
(1 + r
F
)P
c
\
l+r
F
> 0
Taking the first derivative of (P
A
- P
B
) with respect to various right-hand
variables generates:
9{P\-PB) ^
Q
d(P
A
-P
B
)
<Q
d(E(P
A
)) - ' d{E{P
B
)) "'
d(P*
A
-P*
B
)
=
1 z d(PX-P
B
)_ -
d(E(P
c
)) l+r
F
P
' d(P
c
)
d(PX-P*
B
)
d(r
D
)
<0 ,
d(P*
A
~ P*
B
)
=
E(PB) ~ PE{Pc)
d{r
F
) ( l + r
F
)
2
Therefore, the higher the expected payoff on A-shares, the higher the
A-share price premium or B-share discount; the higher the expected payoff
on B-shares, the smaller the B-share discount. If the /3-coefficients of the
B-share payoffs with respect to the C share payoff are nonnegative, i.e.,
P > 0 , then d(P*
A
- P*
B
)/d{E(P
c
)) > 0 and d{P*
A
- P
B
)/d{P
c
) < 0. The
higher the expected payoffs on C-shares, the smaller (if not unchanged)
the B-share discount; the higher the price for C-share, the smaller (if not
unchanged) the B-share discount. And finally, the higher the domestic risk-
free rate, the smaller the B-share discount. The effect of a change in foreign
risk-free rate on B-share discount is ambiguous.
Again, since A- and B-shares are identical except restrictions on own-
ership, the expected payoffs on A- and B-shares should be the same for the
same firm. Employing the same assumption as before, i.e., E\Pc/'(l+rF)} >
Pc and r, = r
F
, then it is quite obvious that PjJ > P
B
. Therefore, given
proper assumption on domestic and foreign risk-free rates, as well as the as-
sumption that the present value of the expected payoff on foreign C-shares
exceeds its initial issuing price, there will be a discount on B-share price.
4.3.4 Corner Solutions for Domestic Firms' Maximization
Problems
It is of concerns on whether there will be corner solutions to the price
discrimination model under the one-way domestic ownership restrictions.
We show below that given the original conditions of the model (equations
(4.20)-(4.23)) and some reasonable assumptions, interior solution (4.24)
and (4.25) are guaranteed.
Consider the representative firm's maximization problem, to simplify
the expressions, we get rid of M and N in the equations by assuming a
composite domestic investor and a composite foreign investor with risk-
aversion coefficients AD and A^, respectively,
E
N l \-^M
=1 A
4
,
D
1 2_,i=l
Ai , .
X
D
N ' A
F
M
Further assume that the variance-covariance matrices of A- and B-share
payoffs are the same, i.e., VAA = VBB = V, and the expected payoffs on
A- and B-shares are the same, i.e., E(PA) = E(PB = \i. A representative
domestic firm chooses the optimal amount of A- and B-shares to maximize
its proceeds from selling the stocks taken as given the amount of A- and
B-shares other firms issued:
maxZ D
iiA
P
A
+ D^BPB
{SA}
= SA^
1
( 4 - S'
A
) + (S- SAW? [</> 'F ~ (S' - S
A
)
Note that
4>D = T ^ - V - V ,
AD
l+r
D
!
AD
1 / / VnaV,
- l
' n K - T {<-/> [*(fc)-(l
+
r,)ft]}
and
X
F
\ a
2
c
Therefore,
Z = S
A
V
n-P
E(P
c
)-(l+r
F
)P
c
l+r
D
l+r
F
+ -
2X
F
V-
VBC
YBC
l+r
F
S }-S
A
+S
n-P
E{Pc - (1 + r
F
)P
c
l + r
F
l+r
D
+ S
+
A
F
| V - ^ B C
l+r
F
S
A
\
F
[V- Yêpic
l+r
F
S
The first order condition to the above maximization problem yields:
8Z
dS
A
H-p [E{P
C
) - (1 + r
F
)P
c
]
1 + 7\D 1+rp
2\
F
(v - yss%sa\
l + r
P
XpV
l + r
D
+
X F [ V - ^ ^ L
l+r
F
SA
= 0
This gives the optimal amount of A-shares issued:
X
D
V
l + r
D
+
X
F
[V- YBcptc
l + r
F
- i
V
l+r
D
V-P
E{P
C
) - (1 + r
F
)P
c
l + r
F
+
2X
F
( V -
VBC
X
BC
l+r
F
Further assume that the risk-free rates are the same across countries,
i.e., TD =r
F
r, the above expression can be reduced to:
s;
X
D
V + Xp[V-
V
BC
V,
BC
+2X
F
V
VBCV
B
C
[p [E(P
C
- (1 + r)P
c
]
If B-share payoffs are uncorrelated with C-share payoff, i.e., VBC = Q,
and the /3-coemcients of the B-share payoffs with respect to C-share payoffs
are approximately 0, then
XD + X
F
Therefore, interior solution is guaranteed as long as the risk-averse co-
efficients for domestic and foreign composite investors are not too different.
However, if foreign composite investor is very risk-averse (Xp is a very large
number) while domestic composite investor is risk-neutral, then domestic
firms will only issue A shares for domestic investors (S^ = S) and pay them
risk-free rate for owning the firms' equities. On the other hand, if domestic
composite investor is very risk-averse (XD is a very large number) while
foreign composite investor is risk-neutral, then domestic firms will only is-
sue B-shares (S^ = 0). These two cases are too extreme and contradict to
reality.
4.3.5 Relaxing Ownership Restrictions
An interesting question arises concerning the effect on price differentials for
A- and B-shares when ownership restrictions are entirely or partially re-
moved. If domestic investors are allowed to purchase foreign C shares, then
domestic investors' investment opportunity sets expand and their wealth
functions will also include C-shares. Equation (4.5) becomes:
A,.
V.
V
AC
V
J
AC
\D
-P*
AA
'C
[E{P
A
)-{l+r
D
)P
A
E{P
C
) - (1 + r
D
)P
c
] }
(4.30)
where /?* is the ( Qx l ) vector of /^-coefficients of the A-share payoffs with
respect to the C-share payoffs, V
A
A is the (Q x Q) variance-covariance
matrix of A-share payoffs of all domestic firms at time 1 and VAC is the
(Q x 1) covariance vector of A-shares payoffs with C-shares payoffs at time
1. Equations (4.21) and (4.22) become:
?D<PF
- / ? *
v
AC
v
j AC
{E{PA
E{P
C
) - (1 + r
D
[E(P
B
- 0 [E(P
C
) - (1 + r
F
)P
c
] } ' (4.31)
TpDTp'f
AF fl+r
D
X
D
\l+r
F
V
AC
V'
AC
V
BB
-
VBCV
1
BC
(4.32)
Therefore, given the same assumptions as those with complete owner-
ship restrictions, domestic and foreign investors will have the same demand
functions for domestic Chinese firm's stocks if Aî = \p, 0 = J3* and
VAC = VBC- However, if the risk-averse coefficients are not the same across
domestic and foreign investors, or if the covariances of A-shares payoffs
with C-share payoff are not equal to the covariances of B-shares payoffs
with C-share payoff, it is still optimal for domestic firm to price discrimi-
nate between domestic and foreign investors. Otherwise, price discrimina-
tion will no longer exist. The results can be summarized in the following
proposition:
Theorem 4.3 The sufficient conditions for the elimination of B-share
discount are: (1) removal of ownership restrictions; (2) domestic and for-
eign investors have the same absolute risk-averse coefficients; (3) the co-
variance of B-share payoffs with respect to C-share payoff are identical to
the covariance of A-share payoffs with respect to C-share payoff.
In the next section, we show that the difference in the expected return
for domestic and foreign investors can be predicted from the equilibrium
asset pricing theory. A general equilibrium model accounting for the supply
side economy would certainly be an interesting extension to this chapter.
4.4 An Intertemporal Capital Asset Pricing Model
In this section, we formulate an ICAPM to model the optimal portfolio
selection for domestic and foreign investors under ownership restrictions
and effective market segmentation. Our objectives are: (1) to explain the
cross-section spread in expected returns for A- and B- shares; (2) to draw
testable implications from the model; and (3) to compare the empirical
predictions of ICAPM with those of price discrimination model.
4.4.1 Stock Price and Return Dynamics
Denote the cum dividend price of a type h stock i at time t as Ph,i(t)
(h = A,B,C), and assume that Ph,i(t) follows a diffusion process with
constant proportional mean and variance (geometric Brownian motion).
dP
hti
{t) = n
h
,iPh,i(-t)dt + <T
h
,iPh,i(
t
)
dz
hA
t
) (
4
-
33
)
where fXh,i and a\
i
are the mean and variance of the return on type h
stock i and dzhît) is a standard diffusion (Wiener) process.
Therefore, the distribution of stock price is log-normal and the rate of
return is
r
hM =
d
p(!} = W.,idt + (Th,idz
h
,i(t) (4.34)
The assumption of lognormal returns is a simplification of the actual
process, which has some serial correlation as discussed by Su and Fleisher
An Intertemporal Capital Asset Pricing Model 149
(1998).
4.4.2 Intertemporal Maximization by Domestic Investors
We assume that a composite domestic consumer-investor is infinitely lived
and has state-independent isoelastic utility of consumption,
2
U(C,t)
C(t)
1- 7D
1 -ID
(4.35)
where C(t) is the composite domestic investor's consumption at time t and
7r> is the measure of her constant relative risk aversion. This investor
maximizes
V(W, t) = max E
0
{C(t),
Waji
(t)} L
e-
pt
U(C,t)dt
Ut=o
(4.36)
where p is the rate of pure time preference, w
a>
i(t) = N
at
i(t)P
a
,i(t)/W(t)
is the fraction of wealth invested in the zth A-share stock at time t, i
1,2, ,N. In particular, w
at
N+i(t) is the fraction invested in safe asset.
N
a
,i(t) is the number of ith A-share stock investor k owns, and W{t) is his
total wealth.
The investor's wealth accumulation equation is
JV+l
dW{t) = ^2 w
a
,i(t)W(t)dr
aii
(t) - C(t)dt
(4.37)
i = l
Since there are N different A-share stocks and one riskless domestic
asset with time-invariant risk-free interest rate T\D, substituting r
a
^{t) from
(4.34) and using Yli=i
w
a,i(t)= 1> equation (4.37) can be rewritten as,
dW{t)
N
^2 w
a
,i(t)(fi
a!i
- r
D
) + r
D
W(t)dt
2
A composite investor's risk-aversion coefficient is the harmonic mean for all K individual
investors' risk-aversion coefficients, =
f c =
3 ,
7
'
=
'
0
.
ID
K
N
+ Y, w
a
,i(t)
w
(t)a,idz
aj
i{t) - C(t)dt (4.38)
Equations (4.36) and (4.38) constitute a continuous-time portfolio opti-
mization problem of the form solved by Merton (1971, 1973). The Bellman
equation yields the following optimality condition,
max
{C(t),u,
0|i
(t)}
dV* dV*
dt dW
( J2
w
*At)(Va,i ~ r
D
) + r
D
(tij W{t) - C(t)
N
wit)
2
d
2
v* ^ v-^
+ ~ ^ ^ T l_j 1^
W
aA
t
)
w
a,J W
a
a,i,j
DW
2
I I
(4.39)
where a
a
,i,j
a r e
the instantaneous covariance between the returns on the
zth and j t h A-share stock. Now, following Merton (1971, 1973), conjecture
that the value function V* is of the form
W/ + U- 7 D
V*(W{t)) = Ye-
pt
^ ,
1 ~1D
where Y is a constant. The choice variables are consumption and the
portfolio weights and the corresponding first-order conditions are:
017 _ dV^
DC ~ dW
(4.40)
f)V* B
2
V*
N
rW(t)(^ - T
D
) + ^ ^ ( t )
2
J ^ , j ( t ) < 7
B
, y = 0 (4.41)
dW
Equation (4.40) is the familiar envelope condition equating the marginal
utility of consumption to the shadow value of wealth and (4.41) is the
standard set of linear equations in the mean-variance portfolio problem.
Substituting in the conjectured value function yields the following optimal
portfolio selection for domestic composite investor,
3
{^-+K^)
2
^]}
3
We assume p > max { 0,7> ru + \ (
l
*^' .'
D
) (
1
_!
vr
^ ) }, which is necessary for
a well-defined solution. This restriction ensures t hat an investor cannot achieve infi-
N
(J>a,i ~r
D
= -YD^2wa,j(t)v
a
,iJ, (4-42)
i=i
The optimal portfolio choice is time-invariant because the investor's
risk-aversion and the means, variances and covariance of asset returns are
constant over time. Moreover, the optimal portfolio choice is independent
of the domestic investor's wealth, consumption, and time preference. The
greater the risk aversion, the lower the expected return, the higher the
risk-free interest rate, or the higher the risk, the less will be held in A-share
stocks.
The equilibrium (domestic market) portfolio is mean-variance efficient.
Therefore the equilibrium return for stock i satisfies the classical capital
asset pricing model (CAPM). To show this, note that Y^j=i
w
a,j(t)
<J
a,i,j =
&a,i,M and J2i=i
w
a,i(
t
)
<T
a,i,M <?l,M' where a
a
,i,M is the covariance of
the ith A-share return with the return on A-share market portfolio and
a
2
a M
is the variance of A-share market portfolio return. Then (4.42) can
be rewritten as fi
a<i
- r
D
= ~{DOa,i,Mi which implies \x
a
^
M
-TD= ID^M-
Therefore,
&a,i,M 9
< M
= AM.AKMQ.M -TD), (4.43)
which is the familiar single-beta CAPM.
4.4.3 Intertemporal Maximization by Foreign Investors
I assume a composite foreign consumer-investor who is also infinitely lived
and has state-independent isoelastic utility of consumption,
U(C,t) = ^> (4.44)
1 -IF
This foreign composite investor maximizes
nite expected utility by any plan. For example, if p < 0, which does not satisfy this
restriction, U(C, oo) = oo for any consumption plan with C > 0.
V(W, t) = max E
0
{C(t),w
bl
i(t),w
c
,j(t)}
/oo
/ e-
pt
U{C,
Jt=o
t)dt (4.45)
subject to
N+l
dW(t) = Y, v>b,i(t)W(t)dr
bti
(t)
i=l
M
+ Yl
w
c,j(t)W(t)dr
CJ
(t) - C{t)dt (4.46)
j = i
where w
bti
(t) = N
b}i
(t)P
bii
(t)/W(t) and w
cJ
(t) = M
CtJ
{t)P
cJ
(t)/W(t) are
the fractions of wealth invested in the ith B-share stock and j t h C-share
stock at time t, respectively, i = 1,2, ,N, j = 1,2, , M, N
bi
i(t)
and M
c
j(t) are the number of ith B-share stock and the number of j t h
C-share stock investor k owns.
There are N different B-shares, M different C-shares and one risk-free
foreign debt instrument with interest rate rp, which is assumed to be time-
invariant. Without loss of generality, number the safe asset as N + l. Since
Si =i
w
b,i{t) + X)j=i
w
c,j(t) = 1, (4.46) can be rewritten as:
dW(t) =
N M
Y2w
bi
i(t)(fj.
bti
- r
F
) + ^2w
Cij
(t)(iJ,
c
j -r
F
)+r
F
. 1 l
N
W(t)dt + ^w
bi
i(t)W(t)<T
bli
dz
bii
(t)
I
M
+ Y,w
c
,j(t)W(t)a
cJ
dz
Ctj
(t) - C(t)dt (4.47)
The optimality conditions for the foreign investor's maximization prob-
lem are
f dV* dV*
{C(t),w
bii
(t),w
oJ
(t)} { dt dW
/ JV 1V1 x
( ^w&.tMCMM - 7 " F ) + ^w
C J
- ( t ) ( / i
C J
-r
F
) + r
F
(t)jW{t)
^ 1 1 '
JV N
-C(t)
w{ty d
2
v*
2 dW
2
L
I I
M M AT M
+ E E ^ (
f
)
tW
c,j W"c,i,j + E E
Wb
'
i
(
f
)
W
<:,j M^b.c.i.j
1 1
{AX
where <Jb,i,j i
a
c,i,j)
a r e
the instantaneous covariance between the returns on
the zth and j t h A-share (B-share) stock, and (?b,c,i,j
a r e
the instantaneous
covariance between the returns on the ith A-share and the j t h B-share. The
corresponding first-order conditions with respect to the portfolio weights are
given by (4.49) and (4.50):
dV* d
2
V*
0 = ^
7
W{t){^
i
-r
F
) + W{t)
2
dW
N
dW
2
N M
\^2wb,j(t)(Xb,i,j +Y^
w
cj(
t
)
(J
b,c,i,j (4.49)
r)V* f)
2
V*
M JV
i i
ptWjt)
1
-^
The value function takes the form V*(W(t)) = Ye
can be rewritten as
(4.50)
i -
7
, > so (4.49)
fJ-b,i - TF = IF
The optimal holdings of B-shares are
N M
22
w
b,j(t)crb,i,j + >,Wc,j{t)crb,c,i,j
Wb,i{t)
1
IF
' M N
EE
(Mi>,j ~
r
F)<?c,i,j
<Tb,i,j<Tc,i,j - O
b,c,i,j
(4.51)
M N , ,
2^2.^ (j.. .
a
2 _ 2
1
! b,i,j
a
c
,i,j "b,c,i,j
(4.52)
Therefore, the foreign composite investor's optimal portfolio choice is
also time-invariant and is independent of his wealth, consumption and time
preference. The greater the risk aversion, the lower the expected return on
B-shares, the higher the expected return on C shares, or the higher the risk
on B-shares, the less will be held in B-share stocks. Moreover, less wealth
will be allocated to B-shares if the correlation between B- and C-share
stocks is higher.
To show that the CAPM holds under continuous time portfolio opti-
mization problem for the foreign investors, denote ab,%,M as the covariance
of returns between an ith B-share stock and the B-share market portfolio,
a\
M
as the variance of returns on B-share market portfolio, (Jb,i,c,M as the
covariance of returns between an ith B-share stock with a C-share market
portfolio and o
2
cM
as the variance of returns on C-share market portfolio,
since
N
2_^Wb,j{t)(Jb,i,i = &b,i,M>
.7 = 1
N
2__
l
w
b,i{
t
)
a
b,i,M &b,M>
t = l
M
2 _,
w
c,j(.t)vb,c,i,j = &b,i,c,M>
J = l
M
and 2jty
c
,i(0
o
'6,t,c,M = &c,M>
(4.51) becomes,
W>,t ~rp = Pb,i,M7F<Tb,M + Pb,i,c,MlF
a
c,M- (4.53)
4.5 Testable Implications From the Models
To derive testable implications from the price discrimination model, we ex-
press the share price equations (4.10) and (4.20) in expected return format
Testable Implications From the Models
155
for individual stocks as follows:
E(Rf) -r
D
=a
D+
(
XDU
^<A ^
A
(4.54)
E(R?) -r
F
= a
F
+ ( ^ " f ^ "
1
' * ) &,
B
+ [E(R?) ~ r
F
] *
2
c
fr,c (4.55)
where ao and a
F
are constants, E(Rf),E(Rf) and E(R
C
) are expected
returns on individual A-, B- and C-shares, respectively, Ilf = PÂSJ,A
and 11^ = Pj
t
BSj
t
B are market values for the j t h stock. < 7 ^ and a^g
are variances of the A- and B-share market returns. Note that in equations
(4.54) and (4.55), the coefficients that come before 0j
t
A and (3J,B contain U^
and 11^, respectively, hence they are not constant over j . The reason for this
lies in the choice of the constant absolute risk aversion utility for domestic
and foreign investors. With the constant relative risk averse utility, it
follows that X
D
U^/N X
D
Wi(0) = X*
D
, where X
D
is investor's coefficient
of constant relative risk aversion. It is clear that X*
D
is independent of j .
After combining equations (4.54) and (4.55), we obtain
\E{Rf) - r
F
] - [E{Rf) - r
D
] =a
0
+ a
x
p
jA
+ a
2
p
jiB
+ a
3
^
c
+ ij (4.56)
Therefore, according to the price discrimination model, the explanation
for the differences in A- and B-share prices can be framed in terms of a
model that explains the differences in expected returns for these two types
of shares.
To derive the testable implications from the ICAPM formulation, we
combine (4.43) and (4.53) and obtain the following equation for the varia-
tion of expected return between A- and B-shares:
{Va,i-'r
D
)-(n
bt
i-r
F
) = PaJMiVaM -TD) ~ 0b,i,M7FVb,M
-Pb,i,c,MlF
a
l,M (
4
-
5 7
)
Regressions (4.56) and (4.57) decompose the B-share return premium
into two components: The first component, aiPj
t
A +
a
2pj,B +OizPj,c, repre-
sents the part of B-share return premium that is related to cross-sectional
structure of risks, as measured by A-share market beta, B-share market
beta and betas with respect to the international market returns. The sec-
ond component, a\ + j, is the sum of residuals and intercept which is
uncorrelated with measures of risk. The lower the A-share market beta,
the higher the B-share market beta, or the higher the beta of an individual
B-share with respect to international financial market returns, the higher
the foreign investor's required B-share premium. Therefore, both (4.56)
and (4.57) provide a basis for testing the linear relationship between the
difference in the expected return for A- and B- shares and the market risk
factors in the Chinese stock markets.
Note that no other variables, such as the variance of returns or firm size,
appears in (4.56) or (4.57). Therefore, another testable implication of the
model is that the betas are complete measure of the stock risks.
4.6 Empirical Resul ts
4.6.1 Estimating Betas
The Chinese data consist of time-series of weekly A- and B-share market
indexes and weekly stock prices for all 47 firms that issued both domestic
A-shares and foreign B-shares before April 1994. 27 firms are listed in
Shanghai and 20 firms are listed in Shenzhen. The sample period is from
April 1994 to October 1998 with a total of 235 observations for each firm.
In calculating the rates of return on individual shares, we make the fol-
lowing adjustments to account for dividend payoffs, stock splits and rights
issued:
(1) The amount of cash dividend per share is added to the stock price
on the date it is paid.
(2) When there is a stock split at time t, P
t
is divided by the split ratio
so that
where SR is the stock split ratio. The rate of return between t and
t + 1 is calculated as usual.
(3) When shares begin trading ex-rights, the theoretical stock price
falls by the intrinsic value of rights. To consider the effect of rights
issued on the rate of return of a share, the following adjustment is
Empirical Results 157
made to the market price at the time trading ex-rights begins: Take
the number of rights required to buy one new share, multiply by
the market price per share before the ex-date, plus the subscription
price and then divide the result by 1 plus the number of rights
required to buy one new share.
Because A-shares are listed in Chinese renminbi yuan while Shenzhen
B-shares are listed in Hong Kong dollar and Shanghai B-shares in U.S.
dollar, we use the weekly exchange rates between the Chinese yuan, U.S.
dollar and Hong Kong dollar to convert all share prices to U.S. dollars.
Returns are calculated using U.S. dollar prices. The rate of return for the
six-month Chinese Treasury Bond is used as the risk-free rate for domestic
investors and the yield on the one-month U.S. T-bill is used as foreign risk-
free rate. Stock indexes for Hong Kong Hang Seng market, MSCI world
equity market, NASDAQ and NYSE are also obtained from April 1994 to
October 1998.
4
To test equations (4.56) and (4.57), I estimate betas for each company
from the following time-series regressions,
ri,t - rf,
t
= Ui + Pi,i(
r
M,i,t - 17,t) + Vi,t, (4.58)
where r^t is the rate of return for stock % in week t, rf,
t
is the corresponding
risk-free interest rate, rM,i,t is the rate of return for Zth market index at
time t, pi
t
i is the market beta for stock i with respect to type I market, and
Ui
t
is the error term. Six beta estimates, namely, Pi,A, P%,B, Pi,NASDAQ,
Pi,NYSE, Pi,MSCI and Pi,HS, are obtained for all companies. Tables (4.7)
and (4.8 present the beta estimates.
Many of the Chinese firms whose stocks have significant betas with
international financial markets are well-established world-class companies
with strong international connections. For example, Sanmao Textile (code
900922), which has a significant beta with respect to the Hong Kong market,
produces world-famous "SB" brand wool and its products are exported to
the U.S., Europe, Middle East, Japan and Hong Kong. Shanghai Phoenix
Bicycle (code 900916), which has significant betas with respect to the MSCI
and Hong Kong markets, owns state-of-the-art production lines and also is-
4
The dat a on Chinese stock prices are obtained from Shenyin International Securities, Xi-
amen Branch. The dividend, stock split and rights issued information is obtained from
"Chinese Stocks and Futures Encyclopedia" on a compact disc produced by Shanghai
Xian Zi Information Co., Ltd. and from China Securities Weekly magazine.
Table 4.7 Beta Estimates for 28 Firms t hat Issue A- and B-shares in Shanghai
Code
900901
900902
900903
900904
900905
900906
900907
900908
900909
900910
900911
900912
900913
900914
900915
900916
900917
900918
900919
900920
900921
900922
900923
900924
900925
900926
900927
900928
Pi,A
1.1935*
1.3070*
1.2090*
0.8340*
1.2325*
1.1456*
1.1780*
1.2925*
1.0574*
0.8165*
0.9733*
0.7780*
1.1365*
1.2409*
0.8813*
1.5244*
0.8859*
0.9004*
0.9685*
1.0776*
1.0321*
1.0816*
1.1830*
1.0953*
0.9475*
1.2173*
0.9910*
1.1303*
Pj,B
0.8495*
1.2242*
0.8525*
0.7177*
0.6826*
0.8163*
0.8565*
1.0348*
1.1499*
0.6009*
1.1079*
1.0583*
0.5538*
1.1381*
0.8732*
1.2782*
0.4777*
0.8159*
0.6332*
0.9866*
0.6260*
0.2031
0.6218*
0.5589*
1.0164*
0.9253*
0.6291*
0.9782*
Pj,NASDAQ
0.4455
0.0613
0.2201
0.9719*
0.0752
0.2541
0.2890
0.1819
0.3294
0.0754
0.2117
0.4058*
0.1840
0.2599
0.0190
0.3094
0.2818
0.3719
-0.1381
-0.0136
0.0403
0.0485
0.1293
0.0263
0.0857
0.1744
0.1691
0.1244
/3j,NYSE
-0.2472
-1.1188
-1.2434
-1.1257
-0.8016
-1.1800
-1.273
-0.9815*
-1.0379
-1.0806
0.2336
-0.1731
-0.0315
-0.3709
0.0718
-0.4234
0.5562
0.2326
0.5799
-0.6957*
0.1610
-0.0897
0.1071
-0.3797
0.0286
-0.2255
0.1905
-0.8847*
Pj,MSCI
0.0913
-0.1853
-0.1089
0.9776
-0.5958
-0.3151
-0.3996
-0.9015*
0.0992
0.0871
0.7002
0.4511
0.0432
0.7236
0.0857
1.0074*
0.583
0.4647
-0.0365
0.4105
0.1765
0.235
0.6271*
0.2491
0.26
0.3117
0.0049
1.3602*
Pj,HS
2.3292*
0.5589*
0.1833
0.4780
0.1167
0.2789
0.1499
0.2772
0.3632
0.3759
0.3666
0.2105
-0.1964
0.4688*
0.1323
0.7258*
0.0749
0.3544*
0.1674
0.2879*
-0.0932
-0.1963*
0.3454*
-0.0084
0.3782*
0.4345*
0.1643
0.3062*
St at i st i cal l y significant at 5% and 10% level, respectively.
sues Euro-dollar denominated B-share stock convertible bonds. Shanghai
Chlor Alkali (code 900908), which has significant beta with respect to the
NYSE and MSCI world market, is the biggest chemical manufacturer in
China and East Asia. It also issues H-shares which are listed on the Hong
Kong Stock Exchange and some of its B-shares have been traded on the
New York OTC market in the form of level ADR. Shenzhen Property & Re-
source Development (code 2011), which has a significant beta with respect
Table 4.8 Beta Estimates for 19 Firms t hat Issue A- and B-shares in Shenzhen
Code Pj
z
A Pj,B Pj,NASDAQ Pj.NYSE Pj,MSCI P},HS
2002
2003
2011
2012
2013
2015
2016
2017
2018
2019
2022
2024
2025
2026
2028
2029
2030
2045
2513
1.0536*
0.9882*
1.0503*
1.1240*
1.1258*
1.1905*
0.9872*
1.0827*
1.1707*
1.1656*
1.1087*
1.1068*
1.1616*
1.0567*
1.0342*
1.0305*
1.1953*
1.1132*
1.0958*
0.8246*
0.9494*
0.8515*
0.7203*
0.7451*
1.0442*
0.6120*
1.1580*
1.0497*
0.8618*
1.0550*
1.1574*
0.3598
0.8809*
1.0536*
0.7722*
1.2391*
1.1991*
0.4934*
0.1940
-0.0581
0.7717*
0.3178
-0.3533
0.6263
0.0557
0.0594
0.5214
0.2938
0.0936
0.0533
-0.6407
0.4690
0.2440
0.1702
0.3190
0.2255
-0.0436
-0.2188
-0.5975
-0.1537
-0.4439
-1.0418*
-0.9906
0.5724
0.1389
-0.9496
f
-1.1016
f
-0.3743
-0.6759
-0.6910
-0.1622
-0.7823
f
-0.3739
-0.3210
-1.0039
-0.4910
f
0.3041
-0.3401
0.2275
-0.1169
0.1725
0.7404
0.0292
-0.3561
0.9303
0.7119
0.0614
0.3936
-1.1334
0.2148
0.3909
0.6258
0.7026
0.027
0.0738
-0.0989
0.0065
0.2232
-0.0574
0.0533
0.0521
-0.0837
-0.0314
-0.1157
0.0757
0.0784
0.1863
-0.4887*
-0.2265
-0.0058
0.0002
-0.0140
0.0416
0.2036*
St at i st i cal l y significant at 5% and 10% level, respectively.
to the NASDAQ market, always has the largest A- and B-share market
capitalization on the Shenzhen Securities Exchange and is one of the most
actively traded stocks by international investors. Lizon Pharmatheutical
(code 2513), which has significant betas with respect to the NYSE and
Hong Kong Stock Exchange, has also issued level ADR on the New York
OTC market.
As described above, the price discrimination model and ICAPM formu-
lation predict that these firms will have higher B-share returns. Now, we
proceed with cross-section studies to test whether this prediction is valid.
4.6.2 Errors-in- Variables and Tests for B-share Premiums
To test hypothesis on B-share return premiums, we proceed in three steps.
First, we test the validity of equations (4.43) and (4.54) by estimating the
following cross-section regression:
Table 4.9 Cross-section Regression For Average A-Share Excess Returns
5o
-0.0705
(-1.3591)
Si
0.0737*
(2.1428)
{ 1.9890}
h
0.0348
(0.8441)
h
0.0425*
(1.9878)
R
2
0.2962
F
3.2265
[0.0406]
"Figures in parentheses are t-statistics for the GLS estimates. Figures in braces are
ElV-adjusted t-statistics. Figures in brackets are p-values. * and t denote statistically
significant at the 5% and 10% level, respectively.
Eîfry r
Dtt
)
=
^
+ s
j
f
w
+ 52{
psw
? + SsDi + & ( 4 - 5 9 )
where J2t=i(
r
a,i,t~
r
D,t)/T is the average excess return for the i-th A-share,
Di is a location dummy variable
D.
{ > :
if firm i is listed in Shanghai
if firm i is listed in Shenzhen.
a n d
< 5 i = 7 B < M-
In estimating regression (4.59), it is necessary to recognize that betas
estimated from (4.58) are measured with error. To correct for the errors-in-
variables (EIV) problem (specifically non-synchronous trading) and obtain
a consistent estimator for Pi,A, we follow Scholes and Williams (1977) and
use the lag, current and lead market returns to construct J3f^. In partic-
ular,
aSW _ A.A.-l + Pi,A + Pi,A, + l
Pi
'
A
~ l + 2p
A
where PA is the first-order autocorrelation coefficient of the A-share market
return. The variable (Pf)
2
i
s
included in (4.59) to test for linearity.
Table 4.9 presents the parameter estimates using Generalized Least
Squares (GLS). As shown in the table, 5\ is significantly positive at the
5% level, suggesting that A-share market beta is a good measure of risk for
an individual firm's share.
5
Both (5o and 82 are statistically insignificant,
5
Using principle component analysis (Campbell, Lo and MacKinlay, 1997), we find only
implying that non-beta risk does not affect the average returns across firms
and the relationship between individual firm's excess return and its A-share
market beta is linear. Furthermore, the representative domestic investor's
absolute risk aversion coefficient, 715, is S\ja
2
a M
= 0.0737/0.0328 = 2.25.
However, the EIV problem treated by Scholes and Williams remains in
the cross-section regression standard errors, i.e., 5i is not ./V-consistent.
6
To
adjust for the small sample bias in the cross-section regression estimates due
to measurement errors in the betas, we follow Shanken (1992) and compute
an EIV adjustment term from the sample covariance matrix of market risk
factors, c, which is positively related to the price of market risk factor and
inversely related to the variance of market risk factor.
7
In cross-sectional
regression (4.59), c 0.21. Then we compute the ElV-adjusted standard
error for <5i as:
\
1.21
0. 0737\
2
0.0328
2.1428/ ~ 235
0.0328 _
+
^ 3 5 - = -
0 3 7 2
where 0.0737 is S\, 2.1428 is the unadjusted t-statistic for 5i, 0.0328 is the
variance of A-share market excess return, a\
M
, and T = 235 is the num-
ber of weekly observations. The ElV-adjusted t-statistic is 0.0737/0.0372
1.9890 with p-value of 0.4902, as reported in the braces in Table 4.9. There-
fore, Si is still statistically significant at the 5% level after adjusting for the
EIV problem.
Second, we test the validity of equations (4.53) and (4.55) by estimating
the following cross-sectional regression:
oSW\2
+64(J3??y + 6
6
D
i
+
i
. (4.60)
one common factor affecting A-share excess returns among firms.
6
I t is not consistent when t he time-series length is fixed and t he number of assets, TV, is
allowed to vary (Shanken, 1992).
7
Given a single market risk factor, c is just t he squared value of t he famous Sharpe ratio
(Shanken, 1992).
Table 4.10 Cross-section Regression for Average B-Share Excess Returns
0o
a
0i
#2, NYSE
02,NASDAQ
02,MSCI
@2,HS
03
8A,NYSE
04,NASDAQ
@4,MSCI
04,HS
05
R
2
F
-0.0726
(-1.3920)
0.0551*
(2.3806)
{ 2.2642}
0.0239
(1.1528)
{ 1.070}
0.0396
(1.0261)
0.0093
(0.3028)
-0.0012
(-0.1866)
0.1688
2.6759
[0.0843]
-0.0527
(-1.0869)
0.0636*
(2.6301)
{ 2.4714}
0.0215
(0.9430)
{ 0.8941}
0.0406
(0.8990)
0.0068
(0.9233)
-0.0015
(-0.7862)
0.1803
2.8747
[0.07960]
-0.0651
(-1.1738)
0.0573*
(2.1433)
{ 1.9338}
0.0355
(1.5826)
{1.3254}
0.0462
(0.9860)
0.0318
(0.6109)
0.0008
(0.4133)
0.1526
2.4982
[0.0931]
-0.0367
(-1.0244)
0.0755*
(2.2062)
{ 1.9290}
0.1583*
(2.7408)
{ 2.3991}
0.0308
(0.8101)
0.0269
(0.9547)
0.0011
(0.2550)
0.3257
3.7909
[0.0186]
"Figures in parentheses are r-statistics for the GLS estimates. Figures in braces are
significant at t he 5% and 10% level, respectively.
where X^t=i(
r
M,* ~
r
F,t)/T is the average excess return for the i-th B-
share, J3f^ and J3fY
a r e
the Scholes-Williams consistent estimators for
B-share market beta and B-share beta with respect to c-th foreign market
(c G {NYSE, NASDAQ, MSCI,HS}), respectively. According to equa-
tion (4.53), 0i = JFÎM
a n d
6*2 = lF
a
\,M-
Table 4.10 presents the GLS parameter estimates. Using the usual t-
statistics, B\ is significantly positive at the 5% level in all four regressions.
Using the ElV-adjusted ^-statistics, it is still significant at the 5% level
(with p-value of 0.031, 0.018, 0.057, and 0.039, respectively), suggesting
that B-share own market beta is a good measure of risk for B-shares. In
addition, none of the foreign-market betas chosen in this study is statis-
tically significant except the B-share beta with respect to the Hong Kong
Hang Seng index, which is significant at the 5% level even under the EIV-
adjusted t-statistic. This result may be explained the geographical proxim-
ity of Hong Kong to mainland China (and hence more rapid transmission of
news) and the high proportion of B-share investors residing in Hong Kong.
The estimated parameters 8o, #3 and 84^s are all statistically insignifi-
cant, implying that non-beta risk does not affect the cross-section variation
in the average B-share excess returns and that the relationship between
B-share excess returns and B-share market beta and B-share beta with
respect to the Hong Kong market is linear. The representative foreign in-
vestor's absolute risk aversion coefficient, jp>
c a n
be computed in two ways:
V 6 , M = 0.0755/0.0128 = 5.90 or hlo\
SM
= 0.1583/0.0271 = 5.84, im-
plying that international investors are two-and-a-half times more risk-averse
towards Chinese shares than domestic Chinese investors.
Finally, we estimate the following cross-section regression:
(Y^=iiTa,i,t-r
D
^
t
)\ (YZ=i(n,i,t-r
F
,
t
)\ ~sw
I f I ~ I f I
= a
+
a
iPi,A
+a
2
0fj^ + a
3
P?
t
%
s
+ a
4
A + i (4.61)
Regression (4.61) decomposes the differences in average returns for A-
and B-share into two components: The first component, a\$f^[ + 012/3^ +
a
z@i^HS represents the part of the average return difference between A- and
B-shares that is related to the cross-section structure of risks, as measured
by consistent estimates of A-share market beta, B-share market beta, and
B-share beta with respect to the Hong Kong Hang Seng index. The sec-
Table 4.11 Cross-section Test for B-share Return Premiums
do"
0.0841
(1.2508)
Ql
0.1238
f
(1.8044)
{1.5174 }
&2
-0.1733*
(-2.3108)
{ -1.9453}
Ot
3
-0.2095*
(-2.9668)
{-2.5579 }
Q4
0.1118
(0.7925)
R
2
0.3662
F
4.2869
[0.0053]
"Figures in parentheses are t-statistics for t he GLS estimates. Figures in braces are
ond component, ao + e, is the sum of residuals and an intercept which is
uncorrelated with measures of risk.
Table 4.11 reports the GLS parameter estimates of equation (4.61).
The coefficient estimates for the three market-risk factors are not only of
right sign but also statistically significant using the conventional ^-statistics.
In addition, the constant terms are not significantly different from zero,
suggesting that the smaller the A-share market beta, the higher the B-share
market beta, or the higher the B-share beta with respect to the Hong Kong
Hang Seng market, the smaller the difference in average returns between
A- and B-shares. Using the ElV-adjusted ^-statistics, A-share market beta
becomes marginally significant (with p-value of 0.1503), B-share market
beta becomes significant only at the 10% level, but the B-share beta against
the Hong Kong Hang Seng market is still statistically significant at the 5%
level. Therefore, the cross-section results support the price discrimination
model and ICAPM formulation, although the empirical evidence is not
overwhelming strong after adjusting for the EIV problem.
4.6.3 Idiosyncratic Variance Effect and Firm Size Effect
To test the hypothesis that the betas are complete measures of risk and
that there are no systematic effects of non-beta risks on the difference in
expected excess returns for A- and B-shares, we estimate the following two
cross-section regressions:
(ZUraM-r
D
,
t
)\ (ZUr
b
M-r
F
,
t
)\
=
^
+
^ w
Summary 165
+a
2
0?g + a
3
P?
t
%s + a
4
Di + a
5
(<7^ - *<) + e
u
(4.62)
+a
2
0fj + a
3
p?%
s
+ a
4
Di + a
6
(ln MV5) + e
i ;
(4.63)
where a\
i
and of
4
are the sample variances of A- and B-share returns,
respectively, and In MV
t
is the logarithm of the sum of the average daily
outstanding market capitalization of A- and B-shares for firm i.
If the price discrimination approach or ICAPM formulation is valid in
explaining the cross-section differences in average returns, the variance of
share returns should not have any explanatory power and &s should not be
statistically significant. Both approaches also imply that firm size should
not have any power in explaining the cross-section variation of asset re-
turns, after controlling for risks. Therefore, &e should not be statistically
significant either.
The regression results in Tables 4.12 and 4.13 show that neither a$ nor
&Q is statistically significantly different from zero. Therefore, after the betas
are controlled for, the variance of returns and firm size do not systematically
affect the difference in expected returns for A- and B-shares in Chinese stock
markets. It is interesting to note that the ElV-adjusted t-statistics for $i, S
2
and $3 are smaller than the conventional ^-statistics, which weaken evidence
for the price discrimination and ICAPM formulations. Nevertheless, smaller
^-statistics for 5$ and 5Q after adjusting for the EIV problem indicate that
there are no idiosyncratic variance effect and firm size effect, thus render
further support to both approaches.
4.7 Summar y
In this chapter, we formulate a one-period price discrimination model under
ownership restrictions to explain the discounts on foreign-owned Chinese B-
shares relative to the prices of domestic A-shares. Under the one-period
price discrimination model, the differences in prices for A- and B-shares for
the same firm can be expressed in terms of the differences in their expected
excess returns. The model posits that cross-section variations in expected
Table 4.12 Idiosyncratic Variance Effect
d
0
a
0.0301
(0.4419)
d
4
0.4708
(1.2336)
di
0.1055
t
(1.6886)
{ 1.4089}
&5
0.2899
(0.1755)
{0.1438}
d
2
-0.1863*
(-2.4511)
{ -2.0426}
R
2
0.3347
&3
-0.1927*
(-2.7230)
{ -2.3271}
F
3.9715
[0.0107]
Table 4.13 Firm Size Effect
OtQ
-0.0275
(-0.2488)
d
4
0.3955
(1.0460)
di
0.1806*
(2.1169)
{ 1.8052}
&e
-0.0836
(-0.8735)
{-0.8168}
ai
-0.1604*
(-2.1033)
{ -1.8001}
R
2
0.3751
&3
-0.2205*
(-3.2008)
{ -2.8202}
F
4.5058
[0.0018]
excess returns between A- and B-shares depend on shares' own market
betas and betas with respect to the international equity markets.
Then we combine the ICAPM with circumstances peculiar to China to
explain the discounts on B-shares again. Under the assumption of infi-
nite investment horizon and representative consumer-investor, we formu-
Summary 167
late an ICAPM with restrictions on share ownership. With the one-instant
ICAPM, the differences in prices for A- and B-shares for the same firm are
again expressed in terms of the differences in their expected excess returns.
As the prediction of the one-period price discrimination model, the ICAPM
posits that cross-section variations in expected excess returns between A-
and B-shares are related to shares' own market betas and betas with respect
to the global equity markets.
Empirical tests of the models using the conventional GLS ^-statistics
indicate that cross-section excess A-share over B-share average return is
positively related to A-share market betas and is negatively correlated to
B-share own market betas and betas with respect to Hong Kong Hang Seng
market. After correcting for the EIV problem using Shanken's method, we
find that these relationships are somewhat weak. Nevertheless, non-beta
risk variables, such as the variance of returns and firm size do not systemat-
ically affect returns. Therefore, the empirical results by and large support
the hypothesis of effective market segmentation and price discrimination
as well as the ICAPM formulation in explaining the price differentials and
return spread between domestic A-shares and foreign B-shares.
We have only investigated the investors' side of the problem in this
chapter. A natural extension would be to account for the supply side as
well as the demand side in a general equilibrium multi-period asset pricing
framework.
Chapter 5
Excess Volatility in Domestic Share
Markets
5.1 Introduction
In this chapter, we attempt to resolve an intriguing problem in Chinese
stock markets: Why is return volatility for domestic A-shares so much
higher than for foreign B-shares, even though the two share categories are
entitled to identical rights and dividends within the same company?
Chinese stock markets are segmented in the sense that A-shares are
available only to Chinese citizens and B-shares can only be purchased by
foreign investors. Both types of shares are traded in the Shanghai and
Shenzhen exchanges, but companies are not cross-listed. Bailey (1994)
provides evidence that B-shares are sold and traded at various discounts
relative to their A-share counterparts. Su and Fleisher (1998a) find that
the volatility of stock-market return is much higher for A-shares than for
B-shares. In the sample used in the current paper, the unweighted means
of daily returns were 0.0303% and 0.0516% while the standard deviations
were 6.0146% and 3.8916%, for A- and B-shares, respectively.
Su (1999a, 1999b) analyzes the B-share discounts and finds that foreign
investors require a larger risk premium for holding B-shares than domestic
investors do for holding A-shares. Moreover, the differences in the average
excess returns for A- and B-shares are related to the shares' market risks
in a manner predicted by the price discrimination model and Intertempo-
ral Capital Asset Pricing Model. However, the sources of the volatility
differences between A- and B-shares remain a puzzle.
To explore the sources of volatility differences in A- and B-share mar-
kets, we specify a model under the Mixture of Distribution Hypothesis
169
170 Excess Volatility in Domestic Share Markets
(MDH) as modified by Andersen (1996) in which stock returns and trading
volumes are assumed to be contemporaneously dependent on an underlying
mixing variable representing the non-uniform intensity over time (relative
to an unobserved benchmark level) of the information flow. Since we do
not observe the actual level of the latent information flow, we assume (re-
alistically we believe) that the unobserved benchmark flow of information
to (domestic) A-share investors is no less than that to (foreign) investors
in B-shares.
The empirical model posits that the volatility of stock returns is identi-
cal to the latent information flow, and trading volume fluctuates in response
to news. To model positive autocorrelation in news arrival, we specify that
the latent information flow follows a stochastic autoregressive process. We
estimate the dynamic system relating stock returns and trading volume
using Generalized Method of Moments (GMM) and obtain parameter esti-
mates characterizing the distribution of the unobservable daily information
arrival for 24 companies that issue both A- and B-shares. We find that news
arrival (1) is on average 49.45% larger for A-shares than for B-shares; (2)
is more highly correlated with trading for A-shares than for B-shares; and
(3) displays larger persistence in A-shares than in B-shares. We conclude
that trading volume is a good approximation for the news flowing into the
stock markets and that the differences in news levels and fluctuations as
reflected in trading volume, as well as the way markets process news, can
explain the difference in the return volatility between A- and B-shares.
The rest of this chapter is divided into four sections. In Section 5.2, we
discuss the mixture of distribution theory and specify our empirical model.
In Section 5.4, we describe the data and detail the adjustment procedure
to remove long-run trend and other systematic calendar effects for stock
prices and volume data. In Section 5.5, we present our empirical results.
We summarize our findings in Section 5.6.
5.2 A Modified Mixture of Distribution Approach
We hypothesize that the main source of stock price movements is changes
in the flow of information perceived by market participants. Market mi-
crostructure theory suggests that trading volume, number of transactions,
and bid-ask spreads are related to the stock-return volatility. It is observed
that market volatility is linked to phenomena associated with information
A Modified Mixture of Distribution Approach 171
arrival such as trading volume, quote spreads, market closures and fore-
castable events such as dividend announcements and macroeconomic data
releases. The MDH posits that the joint distribution of daily returns and
volume can be modeled as a mixture of bivariate normal distributions,
specifically that they are contemporaneously dependent on an underlying
mixing variable representing the flow of information,
RtlKt-N^KuafKt) (5.1)
V
t
\K
t
~ N(^
2
K
t
,a
2
2
K
t
),and (5.2)
Cov(R
t
,V
t
\K
t
) = 0 (5.3)
where R
t
is the stock return on day t, V
t
is the trading volume on day
t, and Kt is a mixing variable, usually interpreted as the unobserved flow
of underlying information pertaining to the future dividends and/or the
liquidation value of a particular stock.
The MDH was first developed by Clark (1973), and has been extended
and tested by Epps and Epps (1976), Tauchen and Pitts (1983), Harris
(1986, 1987), Lamoureux and Lastrapes (1990). These authors find evi-
dence to support the hypothesis that trading volume is a good proxy for
the latent information arrival process and is significantly correlated with
the volatility of stock returns. However, recent and more comprehensive
work by Richardson and Smith (1994), Lamoureux and Lastrapes (1994)
and Gallant, Rossi and Tauchen (1992) provide more mixed evidence on
the validity of the MDH. In particular, these studies find that the nor-
mality assumption for the trading volume is unrealistic because trading
volume cannot be negative. In the simulation exercise of Lamoureux and
Lastrapes (1994), the distribution of the error term is truncated in order
to accommodate strictly nonnegative values of K
t
-
Anderson (1996) proposed an improved characterization of the return
volatility by assuming a time-invariant Poisson distribution for the trading
volume, ruling out negative values a priori. Anderson's modified MDH
(MMDH) can be modeled as
R
t
=
l
i + Z
t
y/K
t
(5. 4)
V
t
\K
t
~ c Po{m
0
+ m
x
K
t
) (5.5)
where Z
t
is a standard normal random variable, V
t
is the detrended, sta-
tionary trading volume series, m
0
is the daily arrival intensity of noise
trading, mi is the proportion of trading volume that fluctuates in response
to the underlying information flow (or news), c is a scaling parameter for
the level of the trading volume, and o\ (equation 5.1) is normalized to be
1. In equations (5.4) and (5.5), the volatility of daily returns is the latent
information flow itself.
We adapt Anderson's MMDH to model China's A- and B-share markets.
Suppose that a company has higher A-share volatility than B-share volatil-
ity. If the expected value of K^ and/or the estimate of cmf are larger than
for Kf and cmf, respectively, we can take this as evidence that a greater
magnitude of news arrival, operating through trading volume, explains the
company's "excess" A-share volatility.
In the spirit of Andersen (1996), we specify a dynamic process gov-
erning information arrivals, Kt, assuming that news arrivals are positively
autocorrelated and follow a lognormal stationary stochastic autoregressive
process in the following form:
In yfKt = a + /?ln y/W^l + -yu
t
, and K
t
> 0 (5.6)
where 0 < /? < 1, a > 0, 7 > 0, u
t
> 0, and u
t
~ i.i.d.N(0,1). Note that,
in (5.6), the information arrival process generating the dynamics of return
volatility displays positive conditional dependency similar to GARCH-type
models, which have been shown to be a reasonable representation of stock-
market return volatility in China by Su and Fleisher (1998a).
It follows from equation (5.6) that
E{\n^K
t
)=a/{l-(i),
Var{\n^K
t
) =
1
l(l-p
2
),
and
Cov(ln y/Kt, In ^ ~ ) = j3
j
Var(ln ^Kt) = {(3
3
i)/(l - f3
2
).
The unconditional moments for K
t
are:
Estimation: Generalized Method of Moments 173
E(y/K?) = exp nEQn y/K
t
) + Var(\n y/K
t
)
exp
na n
2
7
+
[ 1 - / 3 2(1-/ 32)
(5.7)
E(^K?-JK^~) = E(^K?)-E(JK^)
mnCov (In yK
t
, In y/K
t
-j)
e xp
exp
(n + m) a (n
2
+ m
2
+ 2mn(3
3
)^
1- / 3 2(1 - / 3
2
)
(5.8)
where n can be any positive integer.
5.3 Estimation: Generalized Method of Moments
Our dynamic system consists of equations (5.4)-(5.6). The parameterization
of the system involves seven unknown coefficients, 6 = (//, a, /3,7, c, m
0
, mi ),
which can be estimated by the GMM developed by Hansen (1982).
Let v
t
be an (p x 1) vector of variables that are observed at date t, 9 be
an unknown ( gxl ) vector of coefficients, h( #, v
t
) be an ( r x l ) vector-valued
function where h : ( R
p
x R
q
) R
r
, and 60 be the true parameter vector.
Then the r orthogonality conditions can be written as .E[h(0o,
v
t)] = 0.
Let y = (v'
1;
, v'
T
) be a (Tp x 1) vector containing all the observations
in a sample of size T, then the vector of sample moments can be denoted
as g(0, y) = ( Ef =i
h
( 0>
v
t ) ) / r . The GMM estimator 6
T
is obtained by
choosing 6 to minimize the scalar
Q(0,y)='[g(0,y)]'w
T
[g(0,y)] (5.9)
so that the sample moments are as close as possible to the population
moment. Wj - is a (r x r) positive definite weighting matrix that is a
function of the data y. Hansen (1982) shows that g(0, y) A E[h(0
Oj
v
t
)] if
vt is strictly stationary and h(-) is continuous. Thus the value of 0T that
minimizes (5.9) offers a consistent estimate of 6Q .
The following unconditional moments can be exploited to estimate the
unknown coefficients:
E(R
t
) = /i
E(V
t
) = cm
0
+ cmiE(K
t
) = V
Var(Rt) = E(K
t
)
Var(V
t
) = c
2
m
2
Var{K
t
)
E(R
t
, V
t
) = fi(cm
0
+ c
mi
E(K
t
)) = ixV
E(\R
t
-
(i
\) = Ê(y/K'
t
)
E[(R
t
- mfVt] = cm
0
E(K
t
) +
cmi
E(K
2
)
E(\R
t
- p\V
t
) = cmoÊiy/Kl) + cm
iy
jÊ{^lQ)
E[(R
t
- tfiRt-j - M)
2
] = -E(K
t
,K
t
-j)
77
E[(V
t
- V)(V
t
-j - V)] = c\m\Cov{K
u
K
t
^)
E(\R
t
- nWRt-j - /i|) = -E(y/K~
t
, y/i^~)
7T
where j = 1, , 6. Our dynamic system has a total of 26 moment condi-
tions and 7 unknown parameters and therefore 19 overidentifying restric-
tions.
Denote I I as the asymptotic variance-covariance matrix of the sample
meanof h( 0
o
, v
t
) . Then II = l i m r ^ T E{[g(0
0
,y)][g(0o,y)]'}- Assum-
ing that h(#o,Vt) is serially uncorrelated,
T
tl
T
= (1/T) 5 > ( 0
T
, vt)][h($
T
, v
t
)]' 4 n
t=l
Data Adjustments 175
for any consistent estimate for 0
Q
. Then our optimal weighting matrix
is given by W
T
= n^
1
- To ensure that Wj is positive semidefinite, we
follow Newey and West (1987) to estimate II:
(IV + f ' J (5.21)
where
1
T
f
* = T [M0>vt)][h(0, vt-, )]' (5-22)
t=v+l
and I is a parameter representing the maximal order of autocorrelation for
v
t
.
5.4 Da t a Adj ust ment s
The raw data consists of daily closing prices and trading volume for all 24
companies represented by both A- and B- share issues for the period August
6, 1993 through September 25, 1997 for a total of 1021 observations. Cum-
dividend returns are computed using ln(P
t
+ D
t
) l nP
t
_i . Tables 5.1 and
5.2 present summary statistics for the percentage returns. As shown in
the tables, average daily returns are uniformly higher for B-shares than
for A-shares, but the volatility of returns, as measured by the standard
deviation of daily stock returns, is higher for A-shares than for B-shares.
Most of the return series are skewed to the right, although 5 of 24 A-shares
and 10 of 24 B-shares exhibit negative skewness. All return series display
excess kurtosis, meaning that large one-day returns are frequently observed.
Tables 5.3-5.4 reports the autocorrelation of daily returns up to lag 8. The
Ljung-Box portmanteau statistics for 12-th order autocorrelation in returns
and squared returns are all statistically significant at the 1% level.
Tin
To + E
(i + 1).
Table 5.1 Summary Statistics for
Company Mean
(%)
Vacuum A 0.0189
Electronics B 0.0426
Shanghai A 0.0235
Erfangji B 0.0330
Dazhong Taxi A 0.0214
B 0.0715
Yongsheng A 0.0523
Stationery B 0.0728
China A 0.0282
First Pencil B 0.0470
China Textile A 0.0577
Machinery B 0.1006
Shanghai A 0.0397
Rubber Belt B 0.0400
Shanghai A 0.0306
Chlor Alkali B 0.0493
Shanghai Tire A 0.0120
fe Rubber B 0.0228
Shanghai A 0.0024
Refrigerator B 0.0129
Jinqiao Export A 0.0181
& Import B 0.0495
Outer Gaoqiao A 0.0170
B 0.0943
Returns for A- and B-Shares in Shanghai
Std. Dev. Skewness Kurtosis
1%)
5.5104 1.1417* 9.3754*
4.3738 0.0646 1.8447*
6.8080 0.8821* 7.2607*
4.1857 0.6738* 3.9148*
4.8206 -5.6027* 106.15*
4.0776 -3.1307* 56.639*
5.1929 1.2784* 42.506*
3.6338 -0.2321* 5.2527*
4.4095 -0.4829* 22.052*
3.7804 -0.6932* 11.684*
4.5440 0.6768* 16.318*
3.7172 0.6592* 9.6953*
4.3691 0.9262* 10.082*
4.1637 -0.6671* 9.8001*
4.3436 1.2151* 10.936*
3.2005 -0.1058 3.0754*
4.1982 1.1762* 9.0307*
3.4573 0.1220 3.3694*
4.2352 0.3674* 11.293*
3.6603 -0.855* 8.276*
4.4087 -2.3892* 56.288*
3.1875 0.4043* 4.1316*
3.8895 -1.2698* 25.579*
3.2147 -0.4841* 16.294*
*: Statistically significant at the 5% level.
Table 5.2 Summary Statistics for
Company Mean
(%)
Shenzhen A 0.0144
Vanke Co. B 0.0268
Property & A 0.0624
Resource Devel. B 0.0717
China A 0.0326
Southern Glass B 0.0393
Shenzhen A 0.0469
Petroch. Co. B 0.0493
Shenzhen A 0.0547
Zhonghao Co. B 0.0631
Konka A 0.0549
Electronics B 0.0674
Shenzhen A 0.0573
China Bicycles B 0.1431
Victor Onward A 0.0323
Textile B 0.0330
Shenbao A 0.0021
Industry B 0.0049
Chiwan Wharf A 0.0168
Holdings B 0.0352
China Merchants A 0.0079
Shekou Co. B 0.0191
Tellus A 0.0237
Machinery B 0.0480
Returns for A- and B-Shares in Shenzhen
Std. Dev. Skewness Kurtosis
!%)
3.9284 -0.15223* 10.8596*
3.368 0.9935* 28.684*
8.8933 -0.0514 326.05*
3.9288 0.3238* 6.1907*
23.6045 0.0700 93.6294*
4.1852 0.4169* 16.698*
4.3663 0.5319* 11.479*
3.7285 -1.0372* 15.375*
9.0626 0.0505 245.95*
4.5195 0.8339* 8.8355*
4.3610 -0.8690* 164.91*
3.7465 -1.1687* 22.921*
4.3966 0.1085 9.5413*
4.2169 1.4624* 33.141*
4.5932 1.0872* 9.0665*
4.3152 0.6208* 24.753*
4.9079 0.5246* 7.5238*
4.5281 0.3851* 10.766*
4.0620 0.9916* 8.4955*
3.6050 -0.2657* 9.8367*
9.7604 0.6713* 384.20*
4.0811 1.5517* 18.538*
5.6844 0.8061* 71.997*
4.5225 0.7864* 7.161*
Table 5.3 Autocorrelation for Daily Returns for A- and B-Shares in Shanghai
Company
Vacuum
Electronics
Shanghai
Erfangji
Dazhong Taxi
Yongsheng
Stationery
China
First Pencil
China Textile
Machinery
Shanghai
Rubber Belt
Shanghai
Chlor Alkali
Shanghai Tire
& Rubber
Shanghai
Refrigerator
Jinqiao Export
& Import
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
Lag 1
-0.2168
-0.2064
-0.2285
-0.2016
-0.2084
-0.2071
-0.2071
-0.2215
-0.2058
-0.2079
-0.2072
-0.1866
-0.2068
-0.2285
-0.2043
-0.2084
-0.2074
-0.2264
-0.2207
-0.2071
-0.2077
-0.2075
Lag 2
-0.1493
-0.1538
-0.1430
-0.1491
-0.1537
-0.1570
-0.1554
-0.1887
-0.1557
-0.1566
-0.1557
-0.1289
-0.1553
-0.1445
-0.1514
-0.1534
-0.1555
-0.1444
-0.1883
-0.1545
-0.1581
-0.1552
Lag 3
0.0262
0.0265
0.0261
0.0234
0.0220
0.0245
0.0221
0.0099
0.0209
0.0219
0.0219
0.0216
0.0262
0.0257
0.0221
0.0217
0.0264
0.0270
0.0100
0.0218
0.0214
0.0216
Lag 4
0.1560
0.1626
0.1492
0.1588
0.1658
0.1632
0.1674
0.2003
0.1626
0.1647
0.1666
0.1327
0.1628
0.1487
0.1581
0.1656
0.1639
0.1511
0.1999
0.1618
0.1653
0.1665
Lag 5
0.0118
0.0036
0.0044
0.0055
0.0052
0.0043
0.0065
0.0039
0.0043
0.0061
0.0052
0.0112
0.0035
0.0048
0.0045
0.0056
0.0041
0.0051
0.0037
0.0039
0.0063
0.0060
Lag 6
-0.1718
-0.1698
-0.1583
-0.1723
-0.1772
-0.1731
-0.1763
-0.2066
-0.1780
-0.1776
-0.1768
-0.1457
-0.1702
-0.1599
-0.1732
-0.1771
-0.1696
-0.1589
-0.2061
-0.1770
-0.1766
-0.1769
Outer Gaoqiao A -0.2274 -0.1868 0.0121 0.1939 0.0074 -0.2049
B -0.2065 -0.1541 0.0258 0.1635 0.0042 -0.1727
Table 5.4 Autocorrelation for Daily Returns for A- and B-Shares in Shenzhen
Company
Shenzhen
Vanke Co.
Property
Development
China
S. Glass
Shenzhen
Petroch. Co.
Shenzhen
Zhonghao Co.
Konka
Electronics
China
Bicycles
Victor
Textile
Shenbao
Industry
Chiwan Wharf
Holdings
China
Merchants
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
Lag 1
-0.2284
-0.2036
-0.2162
-0.2074
-0.2268
-0.2186
-0.2055
-0.2084
-0.2060
-0.2264
-0.2054
-0.2273
-0.2036
-0.2182
-0.2070
-0.2282
-0.2021
-0.2042
-0.2079
-0.2067
-0.2217
-0.2060
Lag 2
-0.1445
-0.1504
-0.1494
-0.1544
-0.1450
-0.1865
-0.1532
-0.1571
-0.1556
-0.1849
-0.1546
-0.1440
-0.1506
-0.1488
-0.1530
-0.1443
-0.1480
-0.1532
-0.1556
-0.1550
-0.1875
-0.1560
Lag 3
0.0253
0.0224
0.0258
0.0265
0.0255
0.0099
0.0224
0.0207
0.0216
0.0122
0.0214
0.0189
0.0219
0.0253
0.0268
0.0255
0.0222
0.0221
0.0214
0.0212
0.0100
0.0208
Lag 4
0.1481
0.1579
0.1564
0.1636
0.1496
0.1984
0.1630
0.1654
0.1664
0.1936
0.1620
0.1534
0.1582
0.1562
0.1627
0.1491
0.1573
0.1637
0.1654
0.1670
0.1998
0.1631
Lag 5
0.0052
0.0040
0.0107
0.0041
0.0045
0.0040
0.0042
0.0063
0.0068
0.0066
0.0046
0.0067
0.0044
0.0116
0.0041
0.0051
0.0047
0.0048
0.0058
0.0070
0.0040
0.0044
Lag 6
-0.1593
-0.1735
-0.1712
-0.1699
-0.1584
-0.2039
-0.1778
-0.1765
-0.1765
-0.2035
-0.1773
-0.1653
-0.1733
-0.1722
-0.1704
-0.1584
-0.1701
-0.1767
-0.1767
-0.1758
-0.2068
-0.1774
Tellus A -0.2087 -0.1543 0.0214 0.1644 0.0073 -0.1780
Machinery B -0.2030 -0.1510 0.0217 0.1585 0.0040 -0.1745
Tables 5.5 and 5.6 present summary statistics for the raw daily trading
volumes corrected for stock splits. As shown in the tables, the mean and
standard deviation of trading volume are much larger for A-shares than for
B-shares. Tables 5.7 and 5.8 report the autocorrelation in the logarithm
of trading volume up to lag 6. The autocorrelation coefficients are large
and decay only slowly.
1
This behavior is very typical for non-stationary
moving average process. Since the MMDH is built on the assumption that
trading volume is stationary, we test the null hypothesis of a unit root in
the time-series of logarithm of trading volume against the trend-stationary
alternative using Phillips-Perron unit root tests. The test results show that
unit root hypothesis can be strongly rejected for all the trading volume se-
ries. (The Phillips-Perron unit root test results are available upon request.)
Moreover, the time-series plots of the logarithm of trading volume clearly
indicate a constant upward trend for all A- and B-shares. Therefore, we
believe that the Chinese trading volume data are trend stationary.
To detrend the trading volume time-series, we follow Gallant, Rossi
and Tauchen (1992) and use the following set of dummy and time-trend
variables in the adjustment regression:
Day-of-the-week dummies (Tuesday through Friday).
Dummy variables for the number of non-trading days preceding the
current trading day (dummies for each of 1, 2, 3 and 4 and more
days preceding the current trading day). These dummy variables
capture the systematic effects of weekends and holidays.
Dummy variables for each month (January, and March through
December).
Dummy variables for each year (1994 to 1997).
A time trend variable = ! , - , 1021.
1
Obviously, the presence of zero-values for some observations on B-share volume presents
a problem here. In performing all estimation procedures involving log volume we have
set log 0 = 0.
Table 5.5 Summary Statistics for the Raw Trading Volume for A- and B-Shares in
Shanghai
Company
Vacuum
Electronics
Shanghai
Erfangji
Dazhong Taxi
Yongsheng
Stationery
China
First Pencil
China Textile
Machinery
Shanghai
Rubber Belt
Shanghai
Chlor Alkali
Shanghai Tire
& Rubber
Shanghai
Refrigerator
Jinqiao Export
& Import
Outer Gaoqiao
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
Mean
(xlO
3
)
32.114
2.5204
11.787
4.7763
4.6959
2.3143
2.2968
0.4547
3.4404
1.6512
7.3257
2.2739
2.2419
0.563
5.1942
5.9343
3.7693
4.5577
3.3995
3.8023
22.0144
4.6155
4.2763
5.0448
Std. Dev.
(xlO
3
)
46.575
4.042
17.586
6.6654
7.1137
3.3539
4.4025
1.5289
5.1296
3.2055
12.7792
5.9229
4.5986
1.1667
9.7279
7.4490
6.7162
6.2202
5.4373
5.3126
44.8383
5.6830
8.9029
5.6617
Skewness
2.9166*
3.3309*
3.2307*
3.0039*
3.1867*
3.2553*
6.5944*
11.1901*
2.8465*
7.5157*
4.2259*
10.5660*
6.9131*
4.2031*
4.5083*
3.0992*
4.3760*
3.9702*
3.8467*
2.8662*
5.4912*
3.3453*
8.8738*
2.7272*
Kurtosis
10.2976*
14.4421*
13.1728*
11.9879*
13.3447*
16.3812*
68.6105*
184.148*
9.5381*
104.426*
26.9474*
171.536*
72.2882*
22.6308*
28.623*
15.2082*
27.0091*
25.6410*
21.5183*
10.5796*
40.0738*
19.8269*
142.068*
10.3127*
Table 5.6 Summary Statistics for the Raw Trading Volume for A- and B-Shares in
Shenzhen
Company
Shenzhen
Vanke Co.
Property
Development
China
S. Glass
Shenzhen
Petroch. Co.
Shenzhen
Zhonghao Co.
Konka
Electronics
China
Bicycles
Victor
Textile
Shenbao
Industry
Chiwan Wharf
Holdings
China
Merchant
Tellus
Machinery
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
Mean
(xlO
3
)
29.4483
1.1971
15.1792
0.9688
17.088
3.2343
11.5309
0.4597
7.7190
0.6052
15.4057
0.6391
18.8657
4.6383
5.0691
0.6896
7.2486
0.9655
10.9633
1.6686
8.7451
2.3985
10.1318
1.1188
Std. Dev.
(xlO
3
)
45.5784
2.8022
24.2258
3.5524
29.8682
8.4756
19.4081
1.4796
11.2185
1.4470
22.1406
2.2242
30.3878
10.8935
8.4574
2.3477
12.3383
3.2586
15.7688
4.4655
11.4356
5.3088
16.1771
2.9569
Skewness
2.7029*
5.5819*
3.4490*
10.7562*
3.6385*
7.1944*
3.1606*
7.8157*
2.9985*
4.8237*
2.4938*
9.4728*
2.8935*
4.8382*
3.6083*
7.8718*
3.5413*
9.7383*
2.8565*
7.7874*
2.4407*
6.6242*
3.2258*
7.5704*
Kurtosis
9.0481*
42.3339*
16.8830*
145.704*
17.9250*
75.4146*
12.1630*
78.2487*
11.5989*
27.7306*
7.8229*
118.868*
10.0227*
27.1395*
18.0816*
90.5425*
15.7309*
139.379*
9.9275*
87.0522*
6.9676*
72.4576*
13.0613*
90.6199*
Table 5.7 Autocorrelation for the Log of
Shanghai
Company Lag 1 Lag 2
Vacuum A 0.7042 0.5968
Electronics B 0.6973 0.5996
Shanghai A 0.7019 0.5963
Erfangji B 0.6888 0.5925
Dazhong Taxi A 0.7092 0.6048
B 0.6928 0.5989
Yongsheng A 0.7038 0.5995
Stationery B 0.7029 0.6043
China A 0.7094 0.5983
First Pencil B 0.7141 0.6083
China Textile A 0.7132 0.6091
Machinery B 0.6958 0.5917
Shanghai A 0.7056 0.6039
Rubber Belt B 0.7077 0.6006
Shanghai A 0.6903 0.5950
Chlor Alkali B 0.7061 0.5968
Shanghai Tire A 0.7022 0.6109
& Rubber B 0.7099 0.6039
Shanghai A 0.6982 0.5982
Refrigerator B 0.7090 0.6006
Jinqiao Export A 0.7138 0.6161
fc Import B 0.7188 0.6121
Outer Gaoqiao A 0.7023 0.5970
B 0.7100 0.6054
Trading Volume for A- and B-Shares in
L ag 3 L ag 4 L ag 5 L ag 6
0.5080 0.4086 0.3343 0.2799
0.5062 0.4138 0.3360 0.2876
0.4924 0.3912 0.3115 0.2566
0.4998 0.3984 0.3214 0.2759
0.5156 0.4231 0.3372 0.2865
0.5044 0.4109 0.3286 0.2850
0.5089 0.4104 0.3360 0.2817
0.5132 0.4149 0.3433 0.2857
0.5021 0.4043 0.3368 0.2833
0.5110 0.4136 0.3301 0.2733
0.5151 0.4236 0.3490 0.2990
0.4990 0.4042 0.3281 0.2728
0.5087 0.4116 0.3422 0.2896
0.4941 0.3909 0.3092 0.2500
0.5036 0.4021 0.3251 0.2802
0.5091 0.4208 0.3351 0.2806
0.5152 0.4198 0.3384 0.2940
0.5099 0.4080 0.3291 0.2696
0.5102 0.4169 0.3384 0.2835
0.5008 0.4000 0.3285 0.2771
0.5253 0.4278 0.3477 0.2938
0.5150 0.4200 0.3408 0.2855
0.5053 0.4075 0.3360 0.2783
0.5090 0.4103 0.3352 0.2786
Table 5.8 Autocorrelation for the Log of
Shenzhen
Company Lag 1 Lag 2
Shenzhen A 0.7243 0.6188
Vanke Co. B 0.6929 0.5982
Property A 0.7069 0.5979
Development B 0.7047 0.6147
China A 0.7087 0.6062
S. Glass B 0.7014 0.6010
Shenzhen A 0.7117 0.6021
Petroch. Co. B 0.7103 0.6144
Shenzhen A 0.7162 0.6107
Zhonghao Co. B 0.7039 0.6000
Konka A 0.7059 0.5947
China A 0.7005 0.6012
Bicycles B 0.7087 0.6003
Victor A 0.7006 0.6032
Textile B 0.7083 0.6055
Shenbao A 0.6895 0.5939
Industry B 0.7094 0.5990
Chiwan Wharf A 0.6957 0.6018
Holdings B 0.7092 0.6008
China A 0.7065 0.6059
Merchants B 0.7022 0.5910
Tellus A 0.7136 0.6057
Trading Volume for A- and B-Shares in
0.5147 0.4135 0.3289 0.2706
0.5078 0.4075 0.3308 0.2866
0.5067 0.4143 0.3312 0.2793
0.5197 0.4256 0.3452 0.3017
0.5088 0.4069 0.3299 0.2750
0.5092 0.4119 0.3293 0.2755
0.5098 0.4174 0.3353 0.2857
0.5222 0.4238 0.3446 0.2929
0.5193 0.4238 0.3472 0.2958
0.5090 0.4119 0.3411 0.2837
0.4981 0.4017 0.3343 0.2798
0.5132 0.4164 0.3339 0.2787
0.5024 0.4042 0.3315 0.2863
0.5088 0.4148 0.3321 0.2778
0.5049 0.4142 0.3326 0.2880
0.5086 0.4052 0.3263 0.2723
0.5015 0.4002 0.3235 0.2787
0.5107 0.4205 0.3393 0.2911
0.5070 0.4131 0.3298 0.2869"
0.5138 0.4189 0.3395 0.2878
0.5157 0.4176 0.3438 0.2868
0.4940 0.3972 0.3295 0.2750
0.5087 0.4122 0.3276 0.2723^
0.5042 0.4076 0.3368 0.2926
We first regress the logarithm of trading volume on the above set of
adjustment variables,
lnV
t
= Xp + e
t
.
Each regression produces In V
t
, the trading volume that is assumed to be
due to factors not systematically related to news or information arrival.
Then we divide each volume observation, In V
t:
by the corresponding "non-
constant noise" component, \nV
t
, for that day to obtain the detrended
volume series.
Table 5.9 presents estimated coefficients of the detrending equation for
Shanghai Vacuum Electronics and China Textile Machinery. Monday has
a higher volume, which can be seen by noting that the coefficients of all
other day-of-the-week dummies are negative. There are seasonal patterns
by months of the year, with higher volume in January, March and April.
The estimated coefficients for the time trend variable and the year dummies
are all significantly positive, indicating that there is a pronounced upward
trend in trading volume from 1993 to 1997.
To adjust for infrequent trading, we use a nonparametric kernel estima-
tion procedure based on Gaussian kernel to further smooth the detrended
trading volume data. This approach captures a wide variety of nonlinear-
ities in the data without assuming any particular type of functional form.
The nonparametric kernel estimator replaces 0 trading volume observation
with weighted average of those V
t
in a small neighborhood around 0, where
the weights decline as the V
t
's get farther away from 0. Denote the non-
parametric smoothing estimate for the trading volume as V
t
,
V
t
TE^
T
t = i
V-Vt
h
(5.23)
where /C^(-) refers to the kernel, V* is the reference index (the choice of
location of V
t
), and h is the bandwidth. We use the most popular kernel,
the Gaussian kernel, therefore
Kh
V*-V
t
1
hV27T
exp
(5.24)
Selecting the optimal bandwidth h is crucial in kernel estimation, since
it determines the amount of smoothing undertaken. We use the method of
cross-validation with fast Fourier transformation, in which the bandwidth
is chosen to minimize a weighted-average squared error of the kernel esti-
mator. (We use np^dens.src, a Gauss code written by Mico Loretan (1996),
to compute the kernel estimates of the trading volume series. By plotting
the bandwidths against the cross validation scores, which validate the suc-
cess of the kernel estimator in fitting V
t
series across T 1 subsamples,
each with one observation omitted, the bandwidth is chosen to minimize
the cross validation score function.)
Tables 5.10 and 5.11 report summary statistics for the detrended volume
series V
t
for A and B shares for 24 companies. The mean is near unity
and the standard deviation is stabilized, which is to be expected given the
normalization procedure. Skewness is mostly negative for A shares but
positive for B shares. However, compared with the skewness and kurtosis
parameters for the raw trading volume series, the number of significant
parameters is reduced by more than one third in the detrended sample.
Table 5.12 and 5.13 report the autocorrelation of the detrend volume. As
shown in the tables, the autocorrelation coefficients are much smaller than
those in the raw trading volume series and display a smooth decline from
lag 1 to lag 4, becoming randomly positive or negative and statistically
insignificant. In sum, the detrended volume series appear to be stationary
and exhibit the type of characteristics that are required by the empirical
model.
Table 5.9 Adjustment Regressions for the Log of Volume for A and B Shares
Trend
Intercept
t
Day of the week
Tuesday
Wednesday
Thursday
Friday
Number of non-trading days
One day
Two days
Three days
Four and more days
Shanghai Vacuum
Electronics
a
A
10.1559*
(0.7140)
0.017*
(0.0054)
-0.5697+
(0.3337)
-0.5517+
(0.3332)
-0.5528
f
(0.3328)
-0.6011+
(0.3302)
-0.2049
(0.6158)
0.5790
f
(0.3315)
0.6971*
(0.3484)
0.7268+
(0.4206)
B
7.7634*
(0.6440)
0.0121*
(0.0048)
-0.1559
(0.3009)
-0.0568
(0.3005)
-0.1561
(0.3001)
-0.2358
(0.2978)
0.1517
(0.5554)
0.1570
(0.2990)
0.1075
(0.3142)
0.1857
(0.3793)
China Textile
Machinery
A
5.6967*
(0.7711)
0.0105*
(0.0058)
0.2678
(0.3603)
0.2701
(0.3598)
0.2807
(0.3594)
0.3645
(0.3566)
0.6395
(0.6650)
0.4107
(0.3580)
0.0442
(0.3762)
0.044
(0.4541)
B
6.0316*
(0.7947)
0.0111*
(0.0059)
0.2250
0.3714)
0.2633
(0.3708)
0.3373
(0.3704)
0.8472
(0.3675)
0.8472
(0.6854)
0.3956
(0.3689)
0.3647
(0.3877)
0.3582
(0.4681)
(continue on the next page)
"Standard errors in parentheses and * and t denote significance at the 5% and 10% level,
respectively.
{continued)
Month
January
March
April
May
June
July
August
September
October
November
December
Year
1994
1995
1996
1997
Shanghai Vacuum
Electronics
A
0.6113*
(0.2057)
0.4986*
(0.2067)
0.1810
(0.2796)
-0.0931
(0.3713)
-0.6748
(0.4685)
-1.1874*
(0.5744)
-0.3049
(0.6846)
-0.9524
(0.7916)
-1.0678
(0.9110)
-1.2164
(1.0184)
-2.2157+
(1.1329)
4.0286*
(1.3399)
7.9731*
(2.6768)
11.2462*
(3.9957)
4.9623*
(1.2940)
B
0.8181*
(0.1855)
0.7392*
(0.1864)
0.8062*
(0.2522)
0.3971
(0.3349)
-0.0651
(0.4225)
-0.3856
(0.5181)
0.4315
(0.6174)
-0.1924
(0.7140)
-0.1780
(0.8217)
-0.0251
(0.9185)
-0.6425
(1.0218)
2.2612+
(1.2085)
5.3672*
(2.4143)
7.6414*
(3.6039)
9.5176*
(4.7748)
China Textile
Machinery
A
0.2675
(0.1603)
0.7640*
(0.2231)
0.7061*
(0.3019)
0.7550+
(0.4001)
-0.2702
(0.5059)
-0.6146
(0.6203)
0.1391
(0.7392)
-0.3383
(0.8548)
-0.4755
(0.9838)
-0.8339
(1.0997)
-1.1126
(1.2234)
0.6063
(1.4470)
4.2099*
(1.8906)
6.3185*
(3.0148)
7.8986*
(3.7167)
B
0.7069*
(0.2289)
0.3956
(0.3649)
0.9862+
(0.3111)
0.7434+
(0.4132)
-0.0211
(0.5214)
-0.4240
(0.6392)
0.3920
(0.7619)
0.0710
(0.8811)
0.1578
(1.0140)
-0.1852
(1.1335)
-0.9613
(1.2609)
1.6064
(1.4913)
4.850*
(1.9792)
7.1678*
(3.4471)
9.4826*
(4.8920)
Table 5.10 Summary Statistics for the Detrended Trading Volume for A- and B-Shares
in Shanghai
Company
Vacuum
Electronics
Shanghai
Erfangji
Dazhong Taxi
Yongsheng
Stationery
China
First Pencil
China Textile
Machinery
Shanghai
Rubber Belt
Shanghai
Chlor Alkali
Shanghai Tire
& Rubber
Shanghai
Refrigerator
Jinqiao Export
& Import
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
Mean
0.9989
1.0014
0.9999
1.0001
1.1097
0.9778
1.7229
1.0029
1.1083
0.9083
1.3808
0.9930
1.164
1.009
1.0203
0.9097
1.0317
0.9845
1.0208
0.8841
1.1058
1.0011
Std. Dev.
0.3250
0.2795
0.2026
0.2931
0.3474
0.2772
0.7028
0.4719
0.3367
0.3156
0.520
0.3259
0.3571
0.3146
0.2151
0.1938
0.2589
0.2288
0.2978
0.2602
0.1701
0.1579
Skewness
0.2782*
-0.0869
0.2813*
0.3691*
-1.0437*
0.1209
-1.2679*
0.1704*
-0.7865*
-0.039
-0.2591*
-0.0717
-0.1665*
0.590*
-0.1924*
0.2928*
-0.1306
0.1639*
-0.2250*
0.419
0.1984*
0.7633*
Kurtosis
0.1451
0.0419
0.2407*
2.7334*
0.4297*
0.7557*
2.1036*
0.2073*
0.4380*
0.1382
0.1198
0.4752*
0.1487
0.2374*
0.3395*
0.490*
0.1198
0.1778*
0.3276*
0.1131
0.2371*
1.5953*
Outer Gaoqiao A 1.0029 0.1550 0.1145 0.1367
B 0.9960 0.1211 0.4793* 1.1669*
Table 5.11 Summary Statistics for the Detrended Trading Volume for A- and B-Shares
in Shenzhen
Company
Shenzhen
Vanke Co.
Property
Development
China
S. Glass
Shenzhen
Petroch. Co.
Shenzhen
Zhonghao Co.
Konka
Electronics
China
Bicycles
Victor
Textile
Shenbao
Industry
Chiwan Wharf
Holdings
China
Merchants
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
Mean
1.2427
0.8766
1.3605
0.9723
1.2274
1.0002
1.0957
0.9566
1.1796
1.0279
1.0616
0.8915
1.370
0.9588
1.1822
1.0103
1.3070
0.9484
1.124
0.9974
1.1528
0.9742
Std. Dev.
0.4526
0.2995
0.5123
0.3443
0.4501
0.3619
0.2139
0.1663
0.3223
0.2957
0.1619
0.1220
0.2235
0.1884
0.2486
0.1899
0.2497
0.1969
0.2898
0.2275
0.2003
0.1814
Skewness
-0.9507*
0.0615
-1.1003*
0.1844*
0.-2286*
0.6686*
-0.2276*
0.2763*
-0.1129
0.3502*
-0.1176
0.2653*
0.1093
0.1992*
-0.3710*
0.8209*
-0.1022
-0.0735
-0.1181
0.2418*
0.1255
0.2418*
Kurtosis
2.1414*
0.1559
2.7906*
0.0793
0.3429*
0.8692*
0.5272*
0.1255
0.0759
0.2475*
0.0809
0.3944*
0.1058
0.2953*
0.4172*
0.6215*
0.0857
0.0391
0.0740
0.2472*
0.1103
0.4355*
Tellus A 1.158 0.2209 0.1021 0.0810
Machinery B 0.9551 0.1989 0.540* 1.1313*
Table 5.12 Autocorrelation for the Detrended
in Shanghai
Company Lag 1 Lag 2
Vacuum A 0.2866 0.1509
Shanghai A 0.2286 0.1611
Erfangji B 0.2208 0.1673
Dazhong Taxi A 0.2377 0.1552
B 0.2253 0.1933
Yongsheng A 0.2172 0.1644
Stationery B 0.2941 0.1457
China A 0.3091 0.1745"
First Pencil B 0.3229 0.2453
China Textile A 0.3204 0.1907
Shanghai A 0.3341 0.2277
Rubber Belt B 0.3026 0.2324
Shanghai A 0.3278 0.1622
Chlor Alkali B 0.3427 0.2313
Shanghai Tire A 0.3615 0.2431
& Rubber B 0.3451 0.1615
Shanghai A 0.3208 0.1865
Refrigerator B 0.3316 0.2351
Jinqiao Export A 0.2885 0.1593
& Import B 0.3395 0.2273
Outer Gaoqiao A 0.3002 0.2372
B 0.2864 0.1881
Trading Volume for A and B Shares
0.0507 -0.0792 -0.0269 0.0080
0.0409 -0.0885 0.0317 0.0963
0.0159 0.0480 -0.0008 -0.1031
0.1043 0.0157 -0.0069 -0.1829
0.1121 0.0169 -0.0042 -0.0686
0.1038 0.0043 -0.0017 -0.1108
0.1551 0.0130 -0.0052 -0.1889
0.1049 0.0142 -0.0699 -0.0977
0.1011 0.0441 -0.0723 0.0788
0.1020 0.0435 -0.0882 -0.0229
0.0660 -0.0568 -0.0493 0.0396
0.1607 0.0665 -0.0240 -0.1009
0.1394 0.0404 -0.0940 -0.0466
0.1145 0.0695 -0.0285 -0.0950
0.1125 0.0839 -0.0731 0.0506
0.1138 0.0733 -0.0337 -0.1095
0.1216 0.0838 -0.0492 0.0373
0.0735 0.0070 -0.0693 -0.1364
0.0705 -0.0241 -0.0420 0.0653
0.1787 0.0541 -0.0436 -0.1570
0.0800 -0.0284 -0.1108 0.0173
0.1231 0.0371 -0.0417 -0.1307
0.1242 0.0673 -0.0752 -0.1206
0.0520 -0.0531 0.0703 -0.0229
Table 5.13 Autocorrelation for the Detrended Daily Trading Volume for A- and B-
Shares in Shenzhen
Company
Shenzhen
Vanke Co.
Property
Development
China
S. Glass
Shenzhen
Petroch. Co.
Shenzhen
Zhonghao Co.
Konka
Electronics
China
Bicycles
Victor
Textile
Shenbao
Industry
Chiwan Wharf
Holdings
China
Merchants
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
Lag 1
0.3209
0.3062
0.3492
0.3318
0.2682
0.2271
0.2537
0.2480
0.2983
0.2722
0.2938
0.3054
0.3248
0.3142
0.3299
0.2693
0.3078
0.2618
0.2732
0.2525
0.2687
0.2630
Lag 2
0.2348
0.2315
0.2405
0.2388
0.1348
0.1764
0.1338
0.1302
0.1949
0.1977
0.1817
0.1865
0.2393
0.2117
0.2398
0.1529
0.1902
0.1783
0.1893
0.1624
0.1452
0.2156
Lag 3
0.1470
0.1438
0.1084
0.1244
0.0716
0.0774
0.0422
0.0481
0.1239
0.1158
0.1243
0.1170
0.1269
0.1577
0.1223
0.1079
0.1366
0.1027
0.0994
0.0745
0.0471
0.1274
Lag 4
0.0399
0.0663
0.0291
0.0666
-0.0114
0.0265
-0.0488
-0.0449
0.0762
0.0561
0.0488
0.0501
0.0569
0.0607
0.0421
0.0478
0.0823
0.0231
0.0490
-0.0083
-0.0400
0.0545
Lag 5
-0.0611
-0.0767
-0.0681
-0.0240
0.0647
-0.0532
-0.0101
0.0549
-0.0394
-0.0125
-0.0419
-0.0368
-0.0341
-0.0384
-0.0518
-0.0027
-0.0301
-0.0589
-0.0469
0.0451
0.0451
-0.0208
Lag 6
-0.1243
0.0584
-0.0804
-0.1201
-0.470
-0.1145
0.0628
-0.0200
0.0028
-0.0655
0.0073
-0.0891
0.0843
-0.0615
0.0492
-0.0974
-0.0964
0.0180
0.0604
0.0992
-0.0187
-0.0834
Tellus A 0.2730 0.1724 0.1021 0.0241 -0.0584 0.0295
Machinery B 0.2585 0.1662 0.0743 -0.0210 0.0528 -0.0467
5.5 Empirical Resul ts
In estimating our dynamic return-volume system (5.4)-(5.6) by GMM pro-
cedure, the weighting matrix used in the objective function (5.9) is calcu-
lated according to Newey and West (1987a).
2
A modification of Durbin's
method is used in the procedure to determine the optimal maximal or-
der of autocorrelation in the sample (I) and to obtain the autoregression
coefficients from Yule-Walker equations. The weighting matrix and the pa-
rameters are iterated until convergence, so the final parameter estimates
reflects the optimal value of the weighting matrix. Since we have 26 mo-
ment conditions and 7 unknowns, we use a J-statistic suggested by Hansen
(1982) to test whether the 19 overidentifying restrictions are needed. In
particular,
J = ^[ g( 0, y) ] ' w
T
[ g( 0, y) ] ~ x
2
19)
. (5.25)
Table 5.14 presents the parameter estimates as well as Hansen's J-
statistics. As shown in the table, the overidentifying restrictions are rejected
for Jinqiao Export & Import, Shenzhen Petrochemical Co. and Shenbao
Industry, meaning that our empirical model does not fare well for these
three companies. For the remaining 21 companies, the J-statistics are not
statistically significant even at the 10% level, suggesting that the MMDH
cannot be rejected and our dynamic system is quite informative in study-
ing the stock return and trading volume for these companies.
3
Therefore,
we only report the distributional characteristics for return volatility, or the
latent information variable, for 21 companies in Tables 5.14 and 5.15.
To see how well the MMDH helps us to understand the greater volatil-
ity of returns among Chinese A-shares, we first recall from equation (5.12)
that the volatility of returns is the expected value of K
t
. Therefore, an
estimate of E(K
t
) larger across A- than across B-shares may be viewed
as supporting our hypothesis that a greater volume of news arrival is a
fundamental explanation of the excess of volatility in the A-share markets
over the B-share markets. As shown in Tables 5.16 and 5.17, E(K
t
) are
2
We use a GAUSS program written and made publicly available by Hansen, Heaton and
Ogaki for the GMM procedure.
3
We should point out t hat we have estimated our dynamic system of stock return and
volume using different number of moment conditions. Our results are quite robust to
the specification of the moment conditions.
O
l

N
H

X

f
-
C
Q
^

a

S

o

u

a
.

>
-
>

a

c
S

Q
.

Com]
E
x
c
e

I
V

O

i
v

C
O

T
-
H

C
M

L
O

T
1

T
-
H

d

C
O

O
l

i
v

o
o

o

T
-
H

C
O

L
O

o

d

C
O

o
>

o
>

0
0

C
M

L
O

C
O

C
M

T
-
H

T
t

T
-
H

C
M

O
l

r
H

d

L
O

i
v

^

o

o

o

d

<
Vacuum
c
S

J
3

b
o

a

C
O

,

,

o

C
O

o

o
o

p
.

p
"

*

L
O

o

d

C
O

0
0

o
o

T
-
H

o

C
M

O
l

T
-
H

p

p
^

P

C
O

C
M

O
S

o

c
T

^

o

"
*

d

i
o
"

t
~
-
t
v

p

d

^

L
O

O
l

o

o

o

p
^
s
s

i
n

C
O

o

C
O

C
M

C
M

C
O

i
n

T
-
H

d

o

C
M

*

c
o

o

o

t
v

C
M

o

d

c
i

i
v

L
O

C
O

o

C
M

c
o

L
O

I
-
;

d

^
c

C
O

c
o

^
h

d

C
M

c
o

o
i

o

o

o

d

C
O

'
3

o

t
-
l

o

C
D

V
o
f
o

,

,

T
-
H

*

C
M

P
.

P
"

C
M

b
-
O

d

0
0

i
v

C
M

C
M

O

T
-
H

o

T
-
H

p

p
^

p

O
l

<
M

C
M

o

o
o
"

C
O

O
l

T
-
H

d

"
5
"

T
-
H

L
O

T
-
H

d

S
"

I
V

o

C
M

O

o

p
^
t
i
l
i
t
y

i
i

C
O

O
l

L
O

^
t
f

T
-
H

o

C
O

C
O

T
-
H

d

o
o

c
o

O
l

t
v

o

o

I
V

c
o

p

d

o
o

T
-
H

C
M

T
-
H

C
M

L
O

I
V

o
o

O
l

d

C
O

^

T
l
<

C
M

d

C
M

c
o

r
-
H

o

o

p

d

<
Erfangji
n
j

X
I

b
O

c
S

T
H

O

C
O

t
~
08614) [0
d

T
-
H

C
O

L
O

C
M

p
_

^

o

T
(

O

P
^

o
T

O
l

C
O

o
o

o

c
o
"

L
O

C
M

r
H

d

^

C
O

o
o

o

d

c
o
"

t
-
o

o

o

o

^

%

D

o
o

C
O

p

O
S

T
-
H

C
O

C
M

L
O

o

d

^

C
M

C
O

L
O

o

C
O

o

o

p

d

^
f

o

L
O

o
o

o

o
o

o

C
M

o
o

d

O
l

c
o

*

^

d

*

C
N

r
-
H

o

o

o

d

o
m
t

L
O

C
O

C
O

*

o
^

o
T

C
M

C
M

o

p
^

C
O

C
O

o
o

<
M

o

C
M

o

o

o

d

o
T

T
(

C
M

C
O

o

c
o
"

I
V

I
V

C
M

d

T
-
H

L
O

t
"
~

T
-
H

d

o
o
"

L
O

o

o

o

o

p
^

s
t
i
c

S
h

C
O

T
-
H

I
v

L
O

T
-
H

0
0

0
0

I
V

T
-
H

d

O
l

T
-
H

-
*

I
V

o

C
M

O
l

L
O

O

d

*

C
O

C
M

1
-
-
C
M

L
O

T
-
H

T
-
H

C
O

T
-
H

I
V

C
M

I
V

T
-
H

d

c
o

c
o

L
O

o

o

o

d

<

'
3

b
O

t
i

o

x
s

N

Q

(

(

C
M

C
M

O
l

p

P
,

C
O

*

C
O

o

d

"
^

i
v

C
M

t
^

T
-
H

o

c
o
"

C
O

T
-
H

o

p
^

c
o

O
l

o

C
M

T
-
H

O
l
"

L
O

T
-
H

L
O

d

i
o
"

*

L
O

O

O
^

L
O
"

L
O

0
0

T
-
H

o

o

p
^
a
r
e

*

C
M

C
O

d

C
M

O
l

T
-
H

o

c
o

d

o
o

O
l

c
o

L
O

o

o

I
V

T
-
H

o

d

o

0
0

T
-
H

C
O

T
-
H

C
M

0
0

o

p

T
-
H

C
O

t
-
T
-
H

C
O

d

o
o

T
-
H

t
-
o

o

p

d

P
Q

M
o

t
^

T
-
H

o

^

p
.

T
-
H

C
M

L
O

T
-
H

d

O
l

O
l

c
o

T
-
H

O
^

i
P

L
O

o

p

p
^

C
O

c
o

o

o
o

o

O
l
"

0
0

o

c
o

d

C
M
"

C
M

-
t
f

T
-
H

p
^

p
"

o

O
l

T
-
H

o

p

p
_

r
k
e

C
M

o
o

I
V

^

T
-
H

O
l

O
l

0
0

T
-
H

d

C
M

I
V

I
V

)
V

o

C
O

L
O

-
t
f

o

d

O
l

C
O

C
O

f
~

T
(

C
M

0
0

C
O

0
0

d

I
V

I
V

O
l

C
M

d

0
0

o

C
O

T
-
H

o

p

d

<

M

a

C
O

J
H

C
O

b
D

a

t
s

T
v
"

C
O

L
O

l
v

p
.

L
O
"

0
0

0
0

O

O
^

O
l

o
o

O
l

C
M

o

o
o
"

C
O

T
-
H

p

p
^

C
O
"

C
M

o

L
O

o

C
O
"

o

o
o

C
O

d

c
o
"

^
t
<

O
l

p

p
^

C
M

O
l

"
*

T
t
<

o

p

p
^

C
O

C
O

C
O

o
i

T
-
H

C
O

O
l

L
O

T
-
H

d

T
-
H

o

C
O

C
O

C
O

L
O

T
-
H

o

d

I
V

O
)

C
M

O
l

o

o
o

C
O

O
l

C
O

d

C
M

C
O

C
M

L
O

d

L
O

C
O

C
O

T
-
H

o

p

d

m

!
>
,

t
~
i

C
D

a

o

C
O

I
V

C
M

C
O

-
*

p
.

^
5
"

o
o

C
O

o

d

o

C
M

O
l

C
M

O

C
M
"

L
O

o

o

d

c
o
"

L
O

L
O

C
O

o

P
"

o

o
o

C
M

d

L
O
"

^
F

0
0

T
-
H

p
.

C
M

c
o

c
o

0
0

o

p

p
^

a
,

-
t
o

H

C
O

e

e

o

e

s
>

o

E
m
p
i
r
i
c
a
l

R
e
s
u
l
t
s

1
9
5

X

e
-
<
=
a

-
a

s

S

a

S

o

O

22.3862
[0.2710]
>
*

C
M

O

l
O

T
(

C
O

C
M

O

o

o

^

v

C
O

f
-
C
O

-
^

0
0

O

-
#

T
-
l

d

d

v

'

r

v

0
0

-
*

c
o

t
-
-
C
O

O
S

I
1

i
1

d

d

v
'

c
o

t
-
5
*

O

O

0
0

-
*

o

d

d

T
1

C
O

C
M

O
S

O

t
-
-
l
O

T
-
H

o

o

^

v

c
o

c
o

0
0

O

O
S

C
M

c
o

i
i

o

o

0
0

O

C
O

C
O

^
f

T
-
H

o

o

o

o

o

o

d

d

v

'

<

O
S

T
-
H

T
1

c
o

C
M

"
*

r
-
o

(
M

O

c
o

c
o

N
-
(
M

d

T
-
H

o

C
O

T
(

d

C
O

C
O

T
-
H

-
*

o

o

T
-
H

T
-
H

i
n

o

C
O

o
o

o

^
f

o

o
o

T
-
H

t
-
o

o

p

d

P
Q

ina
st Penci
-
a

.
s

O

f
e

O
S

^

C
M

T
-
H

d

m

T
-
H

o

T
-
H

o
^

^
_
^

^
f

C
O

O
S

O

d

*

"

,

v

C
M

i
n

c
o

p

d

"
*

'

^

C
O

C
O

0
0

T
-
H

o

^

o

i
n

o

C
M

o
_
_

^

^

O
S

C
O

t
^

T
-
H

O

C
O

m

c
o

o

o

o

d

"
*

'

(
M

O

p

^

T
-
H

T
-
H

0
0

a
s

T
-
H

o

0
0

C
M

l
r
~

!
>
;

d

r
-
H

0
0

(
M

O

d

C
M

*

O

r
-
H

T
-
H

o

a
s

a
s

5
J
1

o

t
^

T
-
H

O
S

C
O

O

a
s

^
f

i
n

o

O

o

d

<
<

J
D

I
S

O

C
M

o
o

i
n

r
-
d

-
*

o

O
S

o

o

^

v

a
s

c
o

o

C
M

d

s

'

^

v

i
n

O
S

o

p

d

"
^

^

C
M

0
0

C
O

o

t
^

m

c
o

T
-
H

o

v

C
O

T
-
H

T
-
H

T
-
H

o

C
O

O

C
O

O

O

O

d

N
'

C
O

o

o
q

a
s

T
l
T
-
H

C
O

0
0

C
M

o

c
o

T
-
H

T
-
H

p

d

a
s

a
s

C
O

p

d

0
0

o

c
o

C
O

o

C
O

O
S

C
O

^
f

o

^

i
n

^

m

o

C
O

o

T
-
H

O

O

d

P
Q

>
>

*
H

C
D

a

2

o

o
3

s

C
M

0
0

^

d

o

>
*

C
M

t
-
H

o
^

^

v

C
O

C
O

c
o

C
M

d

*

'
_
,

s

^

o

T
-
H

p

d

*

'
^

C
O

O
S

a
s

T
-
H

o

^

0
0

^
t
<

i
n

T
-
H

o
^

_
,

s

o
o

o

i
^

T
-
H

o

m

a
s

^

o

o

o

d

"

'
i
n

c
o

o

t
^

T
-
H

o

T
-
H

C
M

C
M

O

C
O

a
s

0
0

C
O

d

C
M

C
M

^
f

O

d

C
D

T
-
H

T
-
H

^
f

T
-
H

t
~

0
0

O

i
n

o

a
s

C
O

C
O

C
O

o

T
-
H

o
o

t
~

o

o

p

d

<

'
3

f
c
s
O

a

a
3

X
!

i
n

0
0

^
f

^
p

i
n

d

c
o

o

t
>

o

o

v

o

a
s

a
s

T
-
H

d

s

^

v

^
p

C
O

T
-
H

p

d

"
"

C
M

t
-
c
o

i
n

o

c
o
"

C
O

r
-
H

C
M

O

^

C
M

T
-
H

^

T
-
H

o

m

0
0

T
-
H

o

o

p

d

N
'

T
-
H

C
M

o
q

0
0

T
-
H

C
M

a
s

o

C
M

O

^
f

o
o

o
o

i
n

d

o

i
n

p

d

a
s

o

a
s

i
n

o

T
-
H

^
t
<

O
S

C
O

o

o

o

T
-
H

i
n

o

m

i
n

O
S

o

o

p

d

P
Q

-
M

'
a
s

P
Q

C
D

p

P
4

<
M

O

O
S

^

d

T
-
H

T
-
H

t
-
o

o

^

^

C
O

o

b
-
T
i

d

^

'

T
-
H

p

d

b
-
C
O

T
-
H

T
-
H

o

^

i
n

a
s

T
-
H

C
M

O
^

^

^

C
M

C
O

C
M

T
-
H

O

C
O

C
O

^
f

o

o

o

d

*
~
~
^

C
M

o
o

a
s

^

T
-
H

C
O

c
o

T
-
H

r
-
H

o

^

C
O

O
S

t
-
d

C
M

C
M

-
*

O

d

o

r
-
o

^
f

C
M

C
M

T
-
H

i
n

o

T
i

t
>

m

'
t
f

C
M

O

^

C
D

i
n

o

O

p

d

<
:

'
3

-
a

b
O

a

J
3

<
X
l

T
-
H

o

0
0

1
>
-
d

T
t

^
t
<

C
O

o

o

^

^

a
s

o

c
o

T
-
H

d

^

'

x

^

m

0
0

T
-
H

p

d

^
'

m

r
-
C
O

r
-
o

^

^
f

T
-
H

C
O

O

^

C
O

C
O

o

T
-
H

o

o
o

o

C
M

o

o

p

d

"

0
0

o
o

a
s

^

C
M

t
-
-
T
-
H

i
n

C
M

O

T
-
H

m

a
s

C
O

d

a
s

b
-
r
-
p

d

i
n

I
-
-
i
>

c
o

o

o
o

C
D

O

t
-
o

a
s

^

0
0

i
n

o

o

o
o

o
o

o

o

o

P
Q

1

<

!
H

o

3

O

m

C
D

o

C
M

d

T
-
H

^
f

0
0

o

o

^

^

O
S

a
s

-
^

T
-
H

d

^

'

^

^

m

C
O

C
M

O

d

^

'

o
o

T
-
H

T
-
H

C
M

o

^

C
M

T
-
H

C
O

C
M

O
^

^
_
^

C
O

o

a
s

i
i

o

T
-
H

O

C
O

o

p

d

a
s

M

0
3

-
^

X

a
s

a

a

o

-
a

C
D

a

'
s

o

O

1
9
6

E
x
c
e
s
s

V
o
l
a
t
i
l
i
t
y

i
n

D
o
m
e
s
t
i
c

S
h
a
r
e

M
a
r
k
e
t
s

I
N
-
*

X

t
~

<
Q
.

T
3

3

e

c
3

a

a

a
,

S

o

O

o
o

o
>

C
O

0
>

C
M

0
0

d

"
?

C
N
I

O

^

^

L
O

C
O

C
O

O

0
0

C
M

o

o

d

d

^

0
0

T
-
H

L
O

O
)

T
-
H

o
o

C
O

T
-
H

o

o

v

,
s

I
V

C
M

C
O

C
O

I
V

T
-
H

o

o

d

3

o

o
o

C
N
I

C
O

^

*

T
-
H

o
o

C
M

d

T
-
H

t
-
0
0

O

T
-
H

-
r
f

O
)

C
M

d

d

^

v

i
o

i
~
-
o
o

o
o

C
O

C
O

C
M

O

d

d

"
"

,

v

0
5

C
M

C
O

0
2

T
f

T
-
H

o

2

o

o

o

o

d

3

<
i

T
-
H

"
*

0
0

-
t
f

C
M

C
O

o

C
M

T
-
H

d

i
v

i
t

t
v

c
o

o

o

0
0

O
l

o

d

O
)

C
O

o
o

p

i
<

L
O

*
I
V

T
-
H

T
(

C
O

c
o

T
*

c
o

d

C
O

o

I
V

o

o

o

d

P
Q

J
H
ghai Ti
ibber
0
3

P
S

G
O

<
$

^
C
M

0
5

T
-
H

d
.

^

C
O

C
O

L
O

o

d

0
0

o

c
o

T
t

o

^
'

_
,

v

*
o

C
M

O

d

N

'

i
v

T
-
H

0
5

C
M

d

c
o

C
O

L
O

C
O

d

^

-
*

o

i
t

r
1

d

*
~
^

_
.

^

C
M

T
*

C
M

o

o

o

d

*
'
L
O

o

t
-
;

0
0

T
-
H

C
O

C
O

T
-
H

C
M

d

L
O

c
o

o
o

L
O

o

^

0
0

c
o

p

d

o
>

C
O

T
-
H

"
*

T
-
H

C
M

o

T
-
H

0
0

d

C
M

o
o

o
>

c
o

d

o

C
M

C
O

o

o

p

d

<

'
3

4
3

b
O

o
3

4
3

C
O

L
O

0
0

T
-
H

L
O

d
.

^

*

I
V

I
V

o

d

*

^

c
o

0
0

o

C
M

o

^
'

^

^

0
0

0
0

o

p

d

a
>

C
M

C
O

C
O

d

L
O

o

t
v

C
O

d

*

i
v

T
-
H

T
-
H

d

*

'

y

v

o
o

o
o

C
O

o

o

o

d

*
'
o

L
O

0
0

C
M

C
M

C
M

O

o

T
-
H

d

T
-
H

I
V

I
V

T
*

o

L
O

C
M

o
o

p

d

0
0

o

o
>

t
-
d

o
o

o

o

L
O

d

C
O

o

o

C
O

d

r
t
1

^
*

T
-
H

o

o

p

d

P
H

u

o

0
3

J
H

0
>

M

c
u

O
l

C
O

-
#

C
M

P

^

-
^

T
f

*

O

d

'

'

o
>

o
>

(
M

T
-
H

o

'

'

^

^

c
o

I
V

o

o

d

s

'

L
O

O
l

O
l

C
M

d

t
v

C
M

c
o

T
-
H

d

_
^

O
l

0
0

o
>

T
(

d

'

'

^

^

L
O

I
V

T
-
H

o

o

p

d

^
'

74.2219
C
M

T
-
H

0
0

O

d

o

T
-
H

o

C
O

o

L
O

0
0

C
M

T
-
H

d

C
O

o

0
0

d

C
O

o
o

C
M

>
*

T
t

C
O

C
O

*

p

d

C
O

t
-
-
C
M

o

o

p

d

<

-
M

S
H

o

P
H

X

W

o

o
3

'
E
r

T
-
H

o

o

o

p
.

^
^

0
0

O
)

r
}


L
O

d

o
o

c
o

c
o

L
O

d

^

t
>

T
-
H

o

T
-
H

d

^
'

^

^

a
i

c
o

T
-
H

o

o

p

o
^

T
T

o

C
M

c
o

d

o

T
t

^
f

C
O

T
-
H

C
M

d

i
v

C
M

O
l

T
-
H

o

L
O

C
O

L
O

T
-
H

d

C
O

p

T
l
i

0
5

C
M

c
o

b
-
d

^

o
o

L
O

o

d

a
i

L
O

t
-
o

o

o

d

P
Q

O

a

B

t
-
H

o

o

p

p
.

^
^

C
M

o
>

I
>

p

d

^

y

C
O

0
0

T
-
H

T
-
H

o

*
~
-
^

^
,

^

C
M

^
f

o

p

d

^

o
o

L
O

T
-
H

o
o

c
o

o

^
t
<

C
5

C
O

d

^

0
2

^
f

c
o

o

d

^
'

^

^

C
O

L
O

C
M

O

o

o

p
_
^
C
M

T
-
H

C
O

C
M

C
O

T
-
H

d

o

0
)

O
)

i
v

o

o

0
5

T
f

p

d

C
O

C
O

C
M

^

T
-
H

L
O

T
-
H

p

d

T
-
H

a
>

o

^

d

^

^

t
v

O

o

p

d

<

o

0
3

O

o
3

o

l
-
i

o

T
-
H

C
O

0
0

I
V

p
.

C
M
~

O
)

L
O

p

d

^

^

T
-
H

0
0

0
0

T
-
H

g

o
o

C
O

o

o

d

^

'

T
-
H

O
l

0
0

^

d

I
V

I
V

C
O

C
O

d

S
*

o
o

c
o

T
-
H

p
^

_
,

t
-
-
C
O

C
M

O

o

p

p
^

*

C
O

"
*

o
o

T
-
H

C
O

L
O

C
M

T
-
H

d

C
M

o

I
V

L
O

o

o

L
O

T
-
H

o

d

o

o
o

o
o

"
*

d

O
)

T
-
H

o

I
V

d

t
-
-
C
O

C
O

L
O

d

C
O

o

T
-
H

T
-
H

o

p

d

P
H

C
O

0
5

0
5

^

p
.

o
T

o

T
t
<

o

d
^

C
T
)

T
-
H

I
V

T
-
H

o
^

r
f

^
f

o

p

d

^

'

C
M

c
o

C
O

T
-
H

d

T
-
H

C
O

o

C
M

d

o
o
~

L
O

o
o

T
-
H

p
^

^

C
M

L
O

T
-
H

o

o

o

c
i
.

E
m
p
i
r
i
c
a
l

R
e
s
u
l
t
s

1
9
7

(
U

S

S

C
D

w
na
0
0

P
O
-
T
S

C
S

c
S

<

o

E

>
>

e
n

e

3
Vol
c

+
J

K

a
>

3

0
3

>

&
.

-
a

a

c
c
j

C
O

C
D

C
O

C
D

+
5

a

0
)

*
H

a
3

a

a

C
O

O

S
H

(
H

C
D

T
3

a
3

-
a

f
l

0
3

o

C
N
^

X

c
-
<
Q
.

e

S

o

s

o

3
.

0
3

a

a

o

o

I
1

o

C
M

o
i

T
-
H

O
l

C
M

o

1
1

d

0
0

C
M

C
M

C
O

o

o

i
n

o

o

^
f

o

T
-
H

i
n

T
1

C
O

c
o

C
O

o

T
-
H

O
l

0
0

O
l

C
M

o

C
O

o

C
M

o

o

o

o

<

a

C
D

J
=
!

N

a

C
D

,
d

t
/
3

C
O

0
0

C
M

^

O

C
O

C
M

m

O

o

T
-
H

<
*

O
l

T
-
H

o

o
o

o

o

o

o

i
p

C
M

t
-
l
O

o

O
l

T
-
H

c
o

C
O

o

C
D

<
*

O

T
-
H

o
^

C
M

O

C
M

O

O

o

o

C
5

0
0

O

C
O

C
M

O

c
o

C
O

T
-
H

o

C
M

O
l

^

"
*

o

l
~
-
T
1

T
-
H

o

o

^
f

O
l

o

C
O

o

L
O

C
M

0
0

o
o

o

c
o

T
-
H

e
n

^

o

c
o

C
M

^
*

o

o

p

o

P
Q

6

O

^

>

i
n

m

C
O

C
M

O

^

^

o

*

p

o

C
O

O
l

o

T
-
H

o
.

0
0

C
O

o

p

o

t
^
~

o

C
O

C
M

o

c
o

T
-
H

T
-
H

C
M

o

T
-
H

0
0

^

H

o

O
l

o

C
M

o

o

o

o

^

o

T
-
H

C
O

T
-
H

O
l

O
l

t
-
T
-
H

o

^

c
o

o

l
>
-
o

T
-
H

^

T
-
H

o

o

C
M

O
l

T
-
H

c
q

T
-
H

c
o

0
0

m

o

T
-
H

T
-
H

0
0

o

c
o

o

T
-
H

i
n

o
o

o

o

o

o

<
^

<
3

4
^

S
H

C
D

O
H

o

U

P
H

t
~
-
C
M

T
l
"

C
O

O

C
M

T
-
H

C
O

O

O
^

p

o
o

C
O

T
-
H

o

c
o

c
o

o

o

o

o
T

T
-
H

O
l

*

C
D

C
M

^
f

^
f

c
o

o

O
l

c
o

T
-
H

T
-
H

o

T
-
H

O
l

C
O

o

o

o

o

C
M

i
n

i
n

t
^

T
-
H

o

0
0

i
n

p

d

i
n

C
M

t
~

C
O

o

0
0

o
o

C
J
)

p

o

>
*

T
-
H

T
-
H

T
-
H

o
o

c
o

O
l

o

O
l

o
o

c
o

-
^

o

t
-
O
l

C
O

o

o

p

o

P
Q

>

C
D

o

C
D

C
J

F
H

0

O

C
O

"
2

P
H

0
0

o

C
M

i
n

o

^
p

o
o

o

p

d

c
o

t
-
O
S

T
-
H

o
^

T
-
H

^

C
M

p

O

o
"

i
n

C
M

C
M

O

c
o

C
O

o

C
M

o

t
-
T
-
H

T
-
H

T
-
H

o

C
D

O
l

T
-
H

o

o

p

o

0
0

C
M

?

c
o

T
-
H

O
l

C
O

i
n

T
-
H

o

o

C
M

C
M

0
0

O

T
-
H

T
-
H

m

p

o

O
)

O
l

-
C
O

T
-
H

I
>

T
-
H

^

0
0

O

C
M

O
l

C
M

c
o

o

-
s
f

C
O

0
0

o

o

p

o

<

0
3

a

-
C
H

o

C
O

o
o

0
0

t
~

o

^
_
^

C
M

o
o

^
f

p

o
^

i
n
"

o

e
n

T
-
H

o

o
o

O
l

o

p

o

S
"

o
s

O
l

-
^
t
;

o

C
M

c
o

b
-
C
N

O

O
l

o

^

T
-
H

o

i
n

c
o

^

o

o

o

o

O
l

c
o

T
-
H

c
o

C
M

<
M

T
-
H

T
-
H

C
M

C
D

r
~
-
C
O

C
O

<
*

o

T
-
H

o
o

^

o

d

C
D

^
f

o

o
o

d

O
l

T
-
H

C
M

C
M

T
-
H

i
n

C
M

0
0

c
o

o

O
l

C
O

o
o

o

o

p

d

P
Q

C
O

C
O

O

H

C
D

,
n

-
^

p

o

C
/
2

0
0

C
O

C
O

<
M

d

<
M

i
1

0
0

o

d

o
T

C
O

T
-
H

T
-
H

o
^

t
^

0
0

O

o

d

m

i
n

o
o

<
M

d

o

t
>

b
-
C
O

o

^

1
^

<
M

T
-
H

o
^

T
-
H

C
O

^
P

o

o

p

d
^

42.9838
C
D

O

0
0

T
-
H

d

o

^

O
l

C
M

O

c
o

C
O

C
O

p

d

^

0
0

c
o

i
n

d

e
n

o

T
-
H

C
N

O

m

o
o

o

c
o

o

m

c
o

-
*
*

o

o

o

d

<
!

C
D

^
3

N

1
=
1

C
D

J
3

m

T
-
H

o

o

o

d

^

O
l

c
o

T
-
H

C
D

d

i
n

c
o

o

T
-
H

o

c
o

o
o

o

p

d

^
F
~

t
-
-
o
o

C
M

d

T
-
H

C
O

m

T
-
H

O

r
-
O
i

O
l

T
-
H

o

t
-
O
l

T
-
H

o

o

o

d

49.4439
[0.0001]
_
_
^

C
M

C
O

b
-

c
o

C
O

O

T
-
H

i
1

d

d
^

C
O

S
T

T
-
H

l
O

C
O

T
-
H

C
O

T
-
H

o

o

C
M

T
-
H

i
n

o
i

o

o

T
-
H

O

d

d

C
N

c
T

r
-
H

C
O

O
l

C
N

*

C
M

d

d

t
^

o
o

O

C
M

T
-
H

C
M

t
~
-

^

o

o

C
O

^
P

O

T
j
i

O

C
O

T
-
H

T
-
H

o

o

C
D

T
-
H

O
l

0
0

C
N

T
-
H

O

O

o

o

o

o

d

d

P
Q

d

V

o

o

F
-
!

+
J

C
D

P
H

O
l

C
D

s

E
x
c
e
s
s

V
o
l
a
t
i
l
i
t
y

i
n

D
o
m
e
s
t
i
c

S
h
a
r
e

M
a
r
k
e
t
s

-
?
-
h
o
-
e

a

a

S

\
6

C
O

C
O

C
O

o

o

*

i
n

^

T
-
H

O

o

*

0
0

C
M

c
o

^
f

"
"
!

P

o

o

N

'

^
^

c
o

c
o

C
O

O

C
O

O
S

t
-

H

o

o

c
o

a
?
"

i
1

0
0

C
O

O

q

o

o

o

s

o

t
-
L
O

C
N

^

O
S

C
N

c
o

T
-
H

O

^

O
S

L
O

T
-
H

t
-
o
o

*

"
^

T
-
l

o

o

C
M

i
-
H

C
O

C
O

C
O

C
M

c
o

i

3

_
,

s

L
O

b
-
O
S

T
-
i

-
r
t
>

C
N

O

O

O

O

O

O

C
'

S
-
<

t
o

L
O

t
-
d

C
M

C
M

o
o

c
o

T
1

o

C
M

o
o

*

L
O

o

L
O

o

C
O

q

o

o

c
o

l
-
H

L
O

o

c
o

C
O

a
s

t
-
o

o
o

C
O

o
o

"
*

o

T
-
H

f
r
-
O
S

o

o

q

o

P
Q

6

O

henzhen
honghao
C
O

I
S
!

C
O

I
)

C
O

"
?

o

f
r
-
l
>
-
L
O

q

o

'

'

_
,

^

L
O

*

t
-
I
1

o

^
5
"

o

T
-
H

q

o

N

'

o
s

o

c
o

T
1

C
D

^

L
O

^

,
1

C
M

O

o
c
T

t
-
o

T
t

o

o
o

o

c
o

o

o

q

C
D
.

o

-
*

O
S

C
D

C
M

(
-
c
o

c
o

T
-
H

o

^

l
>
-
o

L
O

o

C
O

C
O

t
-
q

o

T
t

o

C
O

C
O

T
t

C
O

c
o

i
-
H

C
M

T
-
H

o

T
-
H

T
-
H

C
O

o

O
S

C
M

o

T
-
H

o

q

o

<

a

o

W

C
M

0
0

o

^

o

0
0

O
S

*
r

q

o

"
"
'

^

^

C
M

C
O

0
0

T
-
H

o

o
"

T
-
H

T
-
H

q

o

s

L
O

c
o

o
s

L
O

C
D

^

o
o

c
o

c
o

L
O

o

p

T
-
H

o

T
-
H

o

v

o
o

C
O

c
o

o

q

p
^

T
t

T
~
^

t
~
-
o
d

T
-
H

C
O

o
s

O
S

T
-
H

C
D

t
-
c
o

0
0

L
O

o

o

C
M

o

q

C
D

T
~
A

0
0

O

C
O

C
D

L
O

L
O

t
-
o

T
t

C
O

T
P

^
f

L
O

o

L
O

C
M

t
-
o

o

o

C
D

P
Q

C
O

C
J

lectron
W

C
O

C
O

T
-
H

L
O

d

C
O

c
o

o
o

q

C
D

^
-
^

v

L
O

^
f

T
-
^

C
M

O

P
"

o

o

q

C
D

v

'

c
o

L
O

o
o

C
M

C
D
^

o
o

c
o

^

o

^

C
O

o

C
M

o

,

^

o

c
o

C
M

o

o

q

C
D
,

L
O

C
M

C
M

"
*

T
-
t

L
O

C
O

o

T
-
H

C
D

-
*

t
-
o

0
0

o

C
O

o

O
S

q

C
D

t
-
C
O

o

C
N

T
-
H

C
N

i
i

L
O

C
O

o

M
<

L
O

0
0

C
O

o

c
o

"
t
f

0
0

o

o

q

o

<

a
henzhe
m

^
S
f

T
-
H

t
-
t
^

C
D

O
S

T
-
^

c
o

q

c
s

C
O

C
O

C
O

r
-
^

o
^

0
?

L
O

C
N

q

C
D

^

L
O

f
r
-
e
e

C
O

C
D

O
S

C
O

c
o

C
M

O

L
O

t
-
<
M

T
-
H

o
^

^

^

L
O

T
-
H

C
O

o

o

q

C
D
,

T
t

T
-
H

c
o

r
j
<

C
M

T
t

C
N

f
r
-
C
N

s

C
O

C
O

o

*

o

O
S

o
o

o
o

q

C
D

L
O

l
>
-
0
0

C
O

C
D

O

T
-
H

C
O

m

o

t
-
^

T
t

o
o

o

T
-
^

"
*

o

T
t

O

q

d
>

m

C
O

J
D

o
hina B
U

L
O

C
O

C
M

C
M

C
D

*

O
S

f
r
-
O

d
>

v

^

t
-
o

0
0

T
-
H

o

e
?

c
o

^
f

q

C
D

L
O

0
0

L
O

T
-
H

C
D

^

T
-
H

1
I

L
O

C
N

o

^

C
O

C
O

T
-
H

o

L00455)
o

C
M

C
O

o
q

O
S

T
t

o

L
O

t
-
q

o

^
<

o
s

o

C
O

o

T
-
H

T

1

o
o

q

C
D

o

^
f
l

C
N

q

C
M

O
S

T
t

T
t

0
0

o

c
o

o

^
t
<

C
M

o

o

o
o

L
O

o

o

q

C
D



o
s

o

o

*

C
D

C
O

o
s

T
-
H

o

d
>

s
^

^

o
o

O
B

t
-
T
-
H

O

L
O

c
o

T
-
H

q

C
D

^

y

o
o

>
*

o
o

q

C
D

T
l
<

t
-
o

<
M

O

P

C
O

o
o

o

o
^

^

^

L
O

t
-
C
N

o

o

o

C
D

T
t

o

^

L
O

C
N

O
S

T
-
^

o

r
H

o

C
M

C
M

C
O

C
O

o

o

L
O

f
-
q

C
D

C
O

^
f

O
S

q

T
-
^

t
-
f
r
-
^
f

T
-
H

T
t

O
S

o
o

c
o

c
o

o

T
t

T
t

C
O

o

o

q

C
D

m

_
c
u

+
J

X

H

!
nward
O

L
O

o
o

o
o

T
-
H

C
D

T
t

C
N

T
l
<

o

C
D

^
_
_

C
O

o

o
o

T
-
H

o
.

^

L
O

T
t

q

o
.

O
S

C
M

o
o

C
N

C
D

L
O

C
O

c
o

C
O

o

t
^

^

o

T
t

o

0
0

O
S

C
N

o

o

o

o
.

c
d

o
_

X

c
u

a

a

o

T
3

c
u

a

-
M

a

o

O

E
m
p
i
r
i
c
a
l

R
e
s
u
l
t
s

1
9
9

c
o

3

8

-
t
o

e

a

0
1

(
N
~

X

i
~

<
o
_

a

_
o

f
Z

o

S

o

;
i

0
3

a

a

o

o

53.6720
c
o

^

i
i

T
-
H

o

o

L
O

I
*
-
C
M

o

C
O

C
M

o

T
"
H

o

o
>

T
"
H

T
1

o
o

o

T
-
H

c
o

C
T
>

L
O

o

c
o

o
o

C
O

i
i

o

C
n

o

c
o

o

o

o

o

<

o

0
3

C
O

C
O

I
1

o

o

o

o

_
_
^

^
p

T
(

C
O

q

o

o
c
T

c
o

C
O

T
-
H

o

c
o
"

c
o

C
M

o

o

C
n

I
V

r
-
^

o

^

'

T
-
H

L
O

C
M

C
O

o

^

-
&

t
-
o

!
i

C
D

c
n
~

I
V

T
1

o

o

q

q
^

69.3371
0
0

o

C
O

1
1

o

C
O

I
-
-
T
-
H

C
O

o

T
-
H

0
0

C
M

o

o

I
*
-
c
o

C
M

C
O

L
O

C
n

o
o

c
n

C
O

o

^
p

C
M

L
O

q

o

I
V

L
O

T
1

o

o

q

C
Q

>
>

t
/
3

3

-
a

a

1
1

T
"
H

o

o

q

o

^

L
O

^
p

C
O

q

q
^

c
o
"

I
V

C
O

1
1

o

^

I
V

o

o

q
^

C
M

C
O

C
O

"
*

T
-
H

0
0

C
N

^
P

c
o

o

^

C
O

f
-
O
)

q

o

C
O

c
o

r
1

o

o

q

I
V

L
O

L
O

^
P

T
-
H

^
P

C
O

C
O

!
-
H

O

o
o

C
O

I
V

I
V

o

L
O

O

0
0

o

o

C
O

C
n

C
n

T
-
H

C
O

H

c
n

r
-
H

^
P

C
N

C
D

C
O

c
n

C
M

O

o

o

C
D

S
H

c
3

Chiwan W
0
0

0
0

C
M

I
V

O

^

T
1

L
O

c
o

o

o
^

L
O

L
O

C
O

i
-
H

o

o
"

C
M

T
1

o

o

I
V

C
n

T
=
*

o

o
o

C
M

C
M

c
o

o

^

L
O

T
1

I
V

o

q
^
_

L
O

C
M

T
-
H

O

o

q

C
D

L
O

O
O

C
O

^
P

C
M

^
P

O

C
n

r
-
i

C
D

O

0
0

C
O

c
j

C
M

C
M

o

T
<

o

H

C
O

C
T
)

"
*

C
D

C
n

C
O

C
O

C
M

T
-
H

C
O

0
0

c
o

^
p

C
D

o

I
V

^
p

o

O

O

C
D

C
Q

C
O

g

i
s

x
P

O

C
M

C
M

C
D

0
0

c
o

t
-
q

q
^

c
o
"

C
O

o

1
1

o

o
T

C
M

C
M

o

o

L
O

o

L
O

i
-
H

C
D

'

"
*

^
P

I
V

1
1

c
o

o

^
p
"

o

T
1

T
-
H

q
^

c
o

I
)

C
M

O

o

q

q
^

C
M

C
M

C
O

c
o

T
(

c
o

C
O

C
O

q

o

L
O

H

C
O

o
o

o

I
V

C
O

^
p

o

o

o
o

C
M

o

o
q

c
o

^
p

C
M

C
M

C
O

C
M

0
0

C
n

o

T
-
H

C
D

1

1

o
o

C
O

o

o

o

o

1

<

chants a Mer
L
O

C
M

r
H

0
0

C
D

^
P

^
P

C
M

q

^

P

i
i

^
p

i
i

o

c
o
"

a
>

o

o

o

L
O

L
O

C
O

o
q

C
D

^

'

T
I

C
O

C
M

I
V

o

^

^
p

T
-
H

^

q

q
^

b
^

C
M

C
M

O

o

q

q
^

o
o

T
-
H

q

^
*

C
M

O

L
O

c
o

T
-
H

C
D

0
0

C
M

T
-
H

-
t
f

o

(
S
i

<
s
>

C
O

o

o

t
~
~

C
M

T

1

T

1

T
-
H

^

L
O

O

S
f

T
-
H

c
o

t
-
t
^
-
C
M

C
D

C
M

L
O

c
o

o

o

q

C
Q

ou Co
C
O

^
p

I
V

c
o

C
M

O

^
_
^

I
V

o
o

C
O

q

o

^

L
O

o

T
-
H

q
^

c
o
"

^
p

T
-
H

o

q
^

C
M

o
o

C
O

C
M

q
^

L
O

C
M

T
-
H

C
O

O

S
"

T
-
H

^

C
O

T
-
H

o

o

q

q
^

C
O

o
o

q

^

T
-
H

L
O

c
o

o
o

q

o

<
s
s

C
M

C
O

I
V

o

o

^
p

C
M

o

o

c
o

C
O

c
o

"
*

T
-
H

C
M

L
O

T
-
H

0
0

O

^
P

o

o

c
o

C
D

<
S
i

c
n

i
i

o

o

q

C
D

<

C
O

J
3

L
O

T
-
H

o

I
V

o

^
^

I
V

T
-
H

C
M

q

o

^

T
-
H

C
O

T
-
H

o

"
"

C
O

o

o

q
^

c
n

I
V

O
i

C
M

C
D

^

^

L
O

o
o

C
M

C
M

O

C
M

I
V

I
V

q

o

c
o

o

T
-
H

o

o

q

q
^
_

^
p

C
M

C
O

^
p

C
M

L
O

c
n

C
M

i
1

C
D

o
o

c
o

o

x
p

o

o

C
M

o

T
-
H

o

c
n

C
O

C
M

t
~
~

C
D

C
O

T
-
H

X
P

o

T
-
H

0
0

o

C
O

c
o

C
D

c
o

o
o

c
n

o

o

q

o

C
Q

>
>

C
D

.
3

-
C
H

o

L
O

o
o

C
M

C
M

C
D

^
_
^

C
M

^
P

C
O

q

o

o
o
"

C
O

c
n

T
-
H

o

L
O

C
M

C
M

o

q
^

L
O

C
O

o

C
M

C
D

^

'

C
O

T

1

C
O

C
M

O

^

t
~
-
T

1

T
-
H

q
^

L
O

T
-
H

C
O

o

o

q

d

uniformly higher for A-shares than for B-shares, ranging from 1.9187 to
4.7095 with an average of 3.0810 for A-shares, and from 1.1685 to 2.5890
with an average of 2.0615 for B-shares. These estimates imply that daily
news arrivals have been on average 49.45% larger across all 21 A-shares
than across B-shares. Moreover, the point estimates for /x, c,TOO, mi, a, (3,7
also appear to be consistent with the MMDH. The parameter cmo, repre-
senting the average fraction of the daily trading volume unrelated to news
arrival (noise trading) ranges from 0.1699 to 0.3774 with an average of
0.2574 for A-shares and from 0.23978 to 0.5854 with an average of 0.3873
for B-shares. This implies, of course, that the proportional importance of
informed trading volume is greater for A-shares. Moreover, the parame-
ter CTOI, associated with the proportion of informed trading volume due
to news arrival, ranges from 0.161 to 0.5914 with an average of 0.4768 for
A-shares and from 0.1968 to 0.5034 with an average of 0.3331 for B-shares,
offering further support to the role of information flows in determining A-
and B-share relative volatility.
The estimates for persistence in return volatility (a + j3) are large and
statistically significantly positive across companies. This is consistent with
the findings in Su and Fleisher (1998a), who documented a GARCH effect
in the Chinese stock markets using A- and B-share market indices. The
persistence in return volatility ranges from 0.5807 to 0.9357 with an average
of 0.7749 for A shares and from 0.4034 to 0.7713 with an average of 0.5503
for B shares.
In summary, we find that: (1) News flows more intensively into A-share
markets than to B-share markets; (2) News is more highly correlated with
trading for A-shares than for B-shares; and (3) News is more persistent for
A-shares than for B-shares.
5.6 Cross-Sect i onal Anal ysi s
It is irresistible to ask why, if rights and dividends are identical across A-
and B-shares, information relevant to domestic and foreign investors is not
also identically distributed. The answer, we believe, lies in the proximity of
domestic investors to their information sources. Moreover, the larger num-
ber of and, presumably, higher geographical density, of domestic investors
intuitively leads to larger amount of news entering the A-share markets
and possibly a more rapid propagation of news, both of which have greater
Cross-Sectional Analysis
Table 5.16 Distributional Characteristics for the Mixing Variable in Shanghai
Company crop croi E{K
t
) Var(K
t
) Kurtosis
Vacuum A 0.2164 0.557 3.0286 32.0876 455.2750
Electronics B 0.3277 0.2851 1.9784 7.4158 16.3885
Shanghai A 0.2415 0.519 2.9869 29.9097 604.8968
Erfangji B 0.3644 0.3775 1.1685 1.1685 48.9391
Dazhong A 0.2783 0.4703 3.5048 48.0010 915.0005
Taxi B 0.3199 0.5134 2.5243 28.6541 136.4124
Yongsheng A 0.2495 0.526 3.9190 29.2303 795.4628
Stationery B 0.3646 0.4864 1.8443 6.4042 161.9072
"Chhia" A (X2 0O61 2.9404 17.2789 181.9652
First Pencil B 0.2086 0.1689 2.2397 7.2794 102.9925
China Textile A 0.1955 0.4325 3.0384 27.2149 353.5309
Machinery B 0.2398 0.1968 2.4247 6.8356 79.0401
Shanghai A 0.1699 0.4713 3.0486 19.8525 127.9680
Rubber Belt B 0.4169 0.3549 2.4179 5.1797 32.9673
Shanghai A 0.3226 0.5642 2.3192 24.6291 358.129
Refrigerator B 0.3008 0.475 1.7771 2.1483 48.6289
Shanghai A 0.2583 0.5914 2.2416 38.2823 100.3267
Chlor Alkali B 0.4134 0.3963 1.9065 7.5179 46.2556
Outer A 0.3744 0.5824 3.3905 38.3418 396.878
Gaoqiao B 0.3978 0.2765 1.1365 2.5558 71.2696
Shenzhen A 0.3179 0.4515 2.4670 3.9808 61.4547
Vanke Co. B 0.4336 0.2994 2.4372 2.1024 22.5456
impact on A-share trading volume. Higher trading volume that is propor-
tional to news leads to higher A-share volatility, consistent with our results.
We should note, though, that the MDH's definition of news is value-free, in
the sense that "truth" versus "gossip" and public versus private informa-
tion are indistinguishable in the MMDH model as specified in Section 5.2.
"Information is what information does," and anything leading to non-noise
("non-liquidity") trading is defined to be "information." In an attempt to
gain some insight into the nature of the differences in the intensity of infor-
mation flows to the A- and B-share markets, we examine the cross-sectional
variation across firms and across share types for the following variables: (1)
Table 5.17 Distributional Characteristics for the Mixing Variable in Shenzhen
Company
Property
Development
China
S. Glass
Shanghai Tire
& Rubber
Konka
Electronics
Shenzhen
China Bicycles
Victor
Textile
Shenzhen
Zhonghao Co.
Chiwan Wharf
Holdings
China
Merchants
Tellus
Machinery
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
cmo
0.3261
0.4117
0.2771
0.4674
0.1699
0.354
0.3774
0.5854
0.251
0.4326
0.1953
0.389
0.1765
0.3852
0.2743
0.5557
0.2879
0.3897
0.2449
0.3758
cm i
0.4989
0.5034
0.553
0.3078
0.4713
0.3014
0.4137
0.331
0.4651
0.3157
0.487
0.371
0.4559
0.2482
0.4828
0.2163
0.4175
0.3086
0.4405
0.2612
E(K
t
)
2.2595
2.2417
4.5870
1.9940
4.0638
1.9229
4.7095
2.5890
2.9304
1.8482
1.9230
1.5996
2.2607
2.0701
4.0927
2.170
3.0713
1.7569
1.9187
1.8257
Var(K
t
)
15.8543
2.6924
19.3756
7.2787
4.8339
3.8033
8.9309
2.0293
13.2296
7.9808
2.2616
1.5317
2.5196
1.8481
29.5482
6.7597
15.7529
2.8314
4.0782
3.1546
Kurtosis
216.9141
35.6494
744.2686
153.5039
75.3546
29.1589
156.1073
36.2989
268.1887
146.5652
46.8391
16.5769
66.9716
27.4972
439.4244
101.2753
129.9115
29.0362
80.7247
45.8312
SE(K
t
), which is the return volatility-related expected intensity of infor-
mation flow; and (2) cm\E(Kt), which is the amount of trading volume
that fluctuates with the intensity of information arrival, or the amount of
informed trading volume. In particular, we test the following hypotheses:
(i) The cross-section variation in the volatility-related expected inten-
sity of information flow is positively related to the standard de-
viation of profit-per-share, firm size, and the average number of
investors over the sample period.
This hypothesis is tested jointly with the supporting hypothesis
that news observable by both A- and B-share investors (or publicly
available information) is positively correlated with the standard
deviation of profit-per-share, firm size and the average number of
Cross-Sectional Analysis 203
investors. The first variable, a firm's profit variation over time, is
of obvious interest to investors. It is entirely plausible that the
greater variation in current and expected future profits, the more
intensely will investors search for and analyze news items about a
firm. Given fixed costs of processing and publicizing information
about a firm, larger firms are likely to find these expenditures less
burdensome than small firms, cet. par. Thus, large firms are likely
to provide more news than small firms. The larger the number of
investors in a stock, the greater is the probability that a given news
item will be observed and passed on, or propagated, through the
investor pool. Therefore, the intensity with which news is perceived
by investors should be positively related to the number of investors
in a particular stock,
(ii) The relationship between the volatility-related expected intensity
of information flows and the standard deviation of profit-per-share,
firm size, and the average number of investors is more intense
(greater) for A-shares than for B-shares.
Again, this is a joint test. The supporting hypothesis is that the
gap between news observed by informed A-share investors but un-
observed by B-share investors is positively correlated with informa-
tion observable to all investors as proxied by the above variables.
The rationale for this hypothesis is that if a firm's time-series of
profit-per-share is flat and/or there is little other types of published
data about the firm, nobody will find it worthwhile thinking, talk-
ing and gossiping about it. But if there is considerably published
information and/or actual and potential changes in a firm's fu-
ture earnings, then it may be worthwhile for investors to acquire
information and gossip about the firm. Such communication, it
seems to us, is more likely among domestic Chinese than among
disparate foreign investors. Furthermore, the larger the size of a
firm, the more likely are domestic investors to gossip about it, and
the larger is the expected volatility-related intensity of information
flow.
To investigate the cross-section pattern of estimated news arrival in
greater depth, following the market microstructure theory described in An-
dersen (1996), we assume that the probability of a single informed investor
trading on the arrival of a given amount of information is constant across
firms. As a result, cross-section variation in mi , the expected number
of daily trades given a certain amount of benchmark information arrival,
should be explainable in terms of the number of a firm's stockholders, which
is assumed to be positively related to the number of informed traders. In
addition, cross-section variation in cmiE(Kt), the amount of trading vol-
ume that fluctuates with news, should be explainable in terms of variables
affecting the proportion of trading volume that fluctuates with news arrival
and by proxies for information arrivals as perceived by investors. Therefore,
we hypothesize:
(iii) The cross-section variation in the amount of trading volume fluctu-
ating with the intensity of news arrival is positively related to the
standard deviation of profit-per-share, firm size, and the average
number of investors.
4
(iv) The relationship between amount of trading volume fluctuating
with the intensity of news arrival and the standard deviation of
profit-per-share, firm size, and the average number of investors is
more intense (greater) for A shares than for B shares.
Tests of the above four hypotheses should also be viewed as joint tests
with the following two hypotheses:
(v) Noise trading, or liquidity trading, is uncorrelated with any proxy
for information flow,
(vi) The proportion of noise traders is uncorrelated with the number of
investors.
To test hypotheses (i) and (ii) empirically, we regress the log of volatility-
related expected intensity of information flow (ln[5E(K
t
)]) on the following
variables:
In STDPROF? = the logarithm of the standard deviation of
calendar year's profit-per-share during 1993-97
(i = 1,2, , 24, j e {A-share, B-share}).
SD\ = a share-type dummy that takes the value of 1
for A shares and 0 for B shares.
4
Andersen (1996) assumes t hat the benchmark information flow is time-invariant for an
individual firm, but not necessarily invariant across firms.
Cross-Sectional Analysis
205
= the logarithm of the average daily number of
investors during the sample period.
= the logarithm of the average daily stock-market
capitalization during the sample period.
= a stock-exchange dummy, ED\ = 0 if firm i
is listed in Shanghai, and ED\ = 1 if it is
listed in Shenzhen.
fa + fa In STDPROF? + faSDj
+fa(SD \nSTDPROF){ + fa XnlNVESTORl
+fa In FIRM SIZE? + fa{SD \nFIRMSIZE)\
+faED{+4 (5.26)
The regressions are estimated using Generalized Least Squares. Table
5.18 presents parameter estimates. As shown in the table, the estimated
elasticity of the volatility-related expected intensity of news arrival variable
with respect to the standard deviation of profit-per-share is about 0.11,
statistically significant at the 5% level. The estimated elasticity of the
volatility-related expected intensity of news arrival variable with respect
to the number of investors is about 0.06, and is also statistically signif-
icant at the 5% level. The variable representing firm size (and assumed
negatively related to information asymmetry) is positively related to the
volatility-related expected intensity of information flow. The estimated co-
efficient for this variable is statistically significant at the 10% level without
the presence of stock-exchange dummy, but is marginally significant at best
when the stock-exchange dummy is included. The stock-exchange dummy
is positively related to SE(K
t
), although it is only marginally significant in
the specification in which both the number of investors and firm size are
included. Perhaps companies listed in Shenzhen generate a higher degree
of intensity of information flow (both publicly available information and
private information, or gossip) and volatility, in part because of their prox-
imity to foreign investors. The econometric results reported in table 5.18
imply that hypothesis (i) is not rejected by the data. Cross-section vari-
ation in volatility-related expected intensity of information arrival among
firms can be explained by cross-section differences in information-related
\n INVESTOR{
\nFIRMSIZEf
ED\
H5E(K
t
)}{ =
Table 5.18 Cross-section Analysis on Volatility-Related Information Flow
Constant
In STDPROF
SD
SD- In STDPROF
In INVESTOR
In FIRMSIZE
SD-lnFIRMSIZE
ED
R
2
lu.[5E(K
t
)] as dependent
0.32
(1.41)
0.12*
(2.42)
0.05
(0.26)
0.06
(0.87)
0.09*
(3.59)
0.78
0.42+
(1.97)
0.12*
(2.63)
0.15
(0.84)
0.07
(1.07)
0.07*
(2.81)
0.08*
(2.57)
0.81
-0.26
(-0.55)
0.09*
(1.85)
0.02
(0.03)
0.06
(0.90)
0.06*
(2.15)
0.03*
(1.93)
0.01
(0.30)
0.81
variable"
0.02
(0.04)
0.11*
(2.05)
0.02
(0.03)
0.06
(0.91)
0.06*
(2.06)
0.02
(1.11)
0.01
(0.33)
0.05
(1.23)
0.81
"We pool both A and B shares for 24 firms in our sample. The regressions are estimated
using Generalized Least Squares. Figures in parentheses are t-statistics. * and t denote
5% and 10% levels of significance, respectively.
variables.
The estimated coefficient for InlNVESTOR suggests that the differ-
ence in the number of A- and B-share investors helps account for the dif-
ference in the volatility-related expected intensity of news between A- and
B-shares. However, the coefficients of the A-share dummy variable (SD)
and its interaction terms with In STDPROF and In FIRM SIZE are quite
insignificant. This implies that there are no excluded variables, correlated
with stock type, that help explain variation in the dependent variable,
5E(K
t
), and that the effects of In STDPROF and hxFIRMSIZE are no
more intense in A-share markets than in B-share markets. Therefore, hy-
Cross-Sectional Analysis 207
pothesis (ii) must be rejected.
To test hypotheses (hi) and (iv), we specify the following regression,
relating the amount of trading volume that fluctuates with the expected
intensity of news arrival to the same right-hand variables specified for equa-
tion (5.26):
Inicm.EiKt)}
3
, = p
0
+ p^nSTDPROF? + p
2
SD\
+p
3
(SD \nSTDPROF)\ + p
4
\nINVESTOR\
+p
5
InFIRMSIZEf + p
6
(SD In FIRMSIZE){
+
P7
ED1 + 4 (5.27)
Table 5.19 presents the econometric results. In comparing the coefficient
estimates for equation (5.27) with those for equation (5.26), we note that
the R
2
, or adjusted R
2
, is higher and that the estimated coefficients for
InSTDPROF and In INVESTOR are quite robust, both in magnitude
and statistical significance. The coefficient for the variable In FIRM SIZE
remains positive, and is statistically more significant in equation (5.27) than
that in equation (5.26). Therefore, hypothesis (hi) cannot be rejected.
In Table 5.19, the coefficient of the A-share dummy variable is sig-
nificant in the estimation of equation (5.27) when firm size and stock-
exchange are both included, suggesting that proxies for public informa-
tion do not fully explain the differences in cmiE{K
t
) between A- and B-
shares. The coefficients of the interaction terms (SD InSTDPROF and
SD hi FIRM SIZE) are insignificant, implying that firm size and profit
variation do not differentially affect informed trading volume of A-shares
compared to B-shares. Therefore, hypothesis (iv) must be rejected.
We emphasize however, that the estimated coefficient for the A-share
dummy in equation (5.27) indicates that something is yet to be learned
about the nature of the events that leads A-share investors to engage in
trading on news not perceived by owners of B-shares. We conjecture that
our ignorance partially lies in the structure and type of information flows
perceived by A- and B-share investors. This interesting question awaits
further investigation.
In summary, hypotheses (i) and (iii) are not rejected by the data, while
hypotheses (ii) and (iv) are rejected. The evidence suggests that differences
in volatility-related expected intensity of information flow and differences
in how information impacts trading volume across firms can be explained
Table 5.19 Cross-section Analysis on the Proportional of Trading Volume That Fluc-
tuates with News
Constant
In STDPROF
SD
SD- In STDPROF
In INVESTOR
In FIRMSIZE
SD-lnFIRMSIZE
ED
R
2
In [cm
:
-1.35*
(-2.81)
0.09+
(1.97)
0.29
(0.80)
0.05
(0.42)
0.13*
(2.76)
0.80
iE(K
t
)} as dependent variable
0
-1.03*
(-2.44)
0.09+
(1.91)
0.56+
(1.72)
0.08
(0.69)
o.os+
(1.82)
0.20*
(3.69)
0.84
-3.84*
(-4.79)
0.09+
(1.86)
1.84+
(1.78)
0.15
(1.27)
0.08+
(1.85)
0.13*
(4.27)
-0.05
(-1.15)
0.87
-3.14*
(-3.68)
0.08+
(1.79)
1.83*
(1.80)
0.15
(1.31)
0.07+
(1.80)
0.10*
(3.05) [3pt]
-0.05
(-1.12) [3pt]
0.10+
(1.65) [3pt]
0.87
"We pool both A and B shares for 24 firms in our sample. The regressions are estimated
using Generalized Least Squares. Figures in parentheses are ^-statistics. * and t denote
5% and 10% levels of significance, respectively.
by differences in information-related variables and the intensity with news
is propagated among investors in particular stocks. The news-related vari-
ables are the standard deviation of profit-per-share, firm size and the num-
ber of investors. Since the number of A-share investors is always greater
than their B-share counterparts, equations (5.26) and (5.27) do help explain
the gap between A-share and B-share volatility. However, the forces prox-
ied by firm size and standard deviation of firm profits do not differentially
impact the trading of A- and B-shares.
Time-Series Analysis 209
5.7 Time-Series Analysis
In an attempt to explore further the implications of our hypothesized
information-volume-volatility relationship as it pertains to A- and B-share
price behavior, we examine whether there exists any time-series relationship
between the differences in A- and B-share expected intensity of informa-
tion flows and the average B-share discounts. The idea is that international
investors have less information about Chinese shares than the domestic in-
vestors do. The information asymmetry increases international investors'
required risk premium for B-shares and reduces their incentives to trade.
As a consequence, information-induced B-share trading volume is less than
that of A-share, and so is the B-share volatility. To investigate this question,
we estimate the relationship between 5
A
E(K
A
) - S
B
E(K^),t = 1, ,T,
and ) L i (
PtB
p/
tA
), where S
A
E(K
A
) - S
B
E(K
t
B
) is the difference in
the estimated volatility-related expected intensity of information arrivals
between value-weighted A- and B-share portfolios during time period T,
and j , 2_j
t=1
( '
P
A * j is the average B-share discount during time period
T. (P
B
is the price of the B-share portfolio at time t while P
A
is the
price of the A-share portfolio at time t.) After considering the number
of observations T that is required to estimated the MMDH model and
the number of moment conditions specified, we choose 10 moment condi-
tions (equations (5.10)-(5.17), and (5.18) and (5.19) without lags) and set
T = 60. We estimate the MMDH model for the return and trading volume
for A- and B-share portfolios to obtain 17 pairs of estimated 5
A
E{K
A
) and
5
B
E{K
B
). Then we compute the average B-share discounts in the corre-
sponding 17 time periods. Figure 5.1 plots the average B-share discount
against the differences in the volatility-related information flow for A- and
B-share portfolios (6
A
E(K
A
) - S
B
E(K
B
)). As shown in the figure, the
two time series seem to be positively correlated. In fact, the correlation
coefficient between the time-series of the differences in volatility-related
expected intensity of information flows and the time-series of the average
B-share discounts is 0.59, which means an R
2
of approximately 0.36%. The
result suggests that one of the reasons B-shares are priced lower than their
corresponding A-shares, even though both share types are entitled to the
same rights and dividends, is that the intensity of information arriving at B-
share markets is smaller than for A-share markets. The result lends support
to the hypothesis that information asymmetry is important in explaining
B-share discounts, which is consistent with the findings of Chakravarty, et.
al (1998).
In order to test for the persistence of volatility-related expected intensity
of information flow differences between A- and B-share portfolios over time,
we estimate the MMDH model (equations (5.5)-(5.6)) using the same 26
moment conditions described in Section 2 (equations (5.10)-(5.20)) for the
A- and B-share portfolios over four 255-trading day periods (one calendar
year). The econometric results are reported in Table 5.20. They show that
the differences in the return volatility persists over the entire sample period,
and the MMDH model describes the portfolio return and aggregate trading
volume data quite well.
5.8 Summar y
The main focus of this paper is to identify the sources of the substantial
volatility differences across the Chinese stock markets segmented between
domestic (A-share) and foreign (B-share) investors. In the spirit of Ander-
son's MMDH, we hypothesize that daily stock return and trading volume
are contemporaneously dependent on an underlying mixing variable repre-
senting daily flow of information, or news. We assume that news follows
a lognormal stochastic autoregressive process and is non-uniform through
time. Using GMM, we estimate a dynamic system of return and volume
equations and find that the MMDH cannot be rejected for 21 out of 24
companies in our sample.
After comparing the parameter estimates that characterize the joint dis-
tribution of the news and return variables, we conclude that the differences
in return volatility between A and B shares are directly related to differ-
ences in the flow of news as reflected in trading volume and the differences
in the way news is incorporated into trading decisions. News arrives more
intensively for holders of an individual company's A shares than for holders
of its B shares, so the volatility of returns is higher for A shares, and a larger
fraction of news is incorporated into trading volume for A shares. One may
be tempted to ask why, if rights and dividends are identical across A- and
B-shares, information relevant to domestic and foreign investors is not also
identically distributed. The answer, we believe, lies in the proximity of do-
mestic investors to their information sources. Moreover, the larger number
of and, presumably, higher geographical density, of domestic investors intu-
S
u
m
m
a
r
y

2
1
1

C
O

f
f
l

1

H

f
t
!

S

P
Q

b
o

Table 5.20 Time-series Analysis for the Value-weighted Portfolio
Company"
2
August 15,
August 14,
August 15,
August 14,
August 15,
August 14,
August 15,
August 14,
August 15,
August 14,
August 15,
August 14,
August 15,
August 14,
August 15,
August 14,
1993-
1994
1994-
1995
1995-
1996
1996-
1997
1993-
1994
1994-
1995
1995-
1996
1996-
1997
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
Sample
Mean
of-Rt
0.017
0.023
-0.026
-0.022
0.019
0.024
0.022
0.026
6E(K
t
)
3.09
2.46
2.69
1.95
2.47
2.05
3.51
2.62
Sample
Variance
of R
t
3.74
3.11
3.68
3.06
3.52
3.25
4.20
3.46
Var{K
t
)
18.72
7.85
14.18
6.27
20.26
11.90
23.11
12.40
crriQ
0.42
0.59
0.27
0.44
0.38
0.52
0.37
0.58
OTi
0.51
0.30
0.61
0.40
0.55
0.40
0.58
0.29
Kurtosis
294.60
102.83
207.24
89.71
336.88
192.64
445.65
184.77
a
We construct value-weighted portfolio for A and B shares, using 24 firms in t he sample.
We estimate the MMDH model during a one-year interval and examine t he persistence
in t he volatility differences.
itively leads to a more rapid transmission of news and, likely, a greater affect
on trading volume, consistent with our results. We should note, though,
that the MMDH's definition of news is value-free, in the sense that "truth"
and "gossip" are homogeneous within the definition of the unobserved K
t
.
"Information is what information does," and anything leading to non-noise
("non-liquidity") trading is defined to be "information."
Our empirical model so far has formally treated A-share and B-shares
Summary 213
as independent, but as though they can be compared via estimates of the
expected values Kf and Kf, assuming that the benchmark flow of infor-
mation the A-share markets is not less than that to B-share markets. In
subsequent work, we hope to develop a joint model of the A-share and
B-share markets, allowing for the possibility of both common and idiosyn-
cratic information flows. Moreover, our estimation results so far have been
obtained under the assumption that the innovation in news process, u
t
,
is lognormally distributed. It would be worthwhile to consider alternative
specifications such as Gamma, Student t and Stable distributions in further
studies.
Chapter 6
The Underpricing of Initial Public
Offerings
6.1 Introduction
Initial public offering (IPO) underpricing, or high IPO initial return, is
a phenomenon common to most stock marketsboth in developed and
emerging economies (Loughran, Ritter, and Rydquist, 1994). A common
perception is that underpricing of IPOs is a contradiction to market effi-
ciency and may hurt emerging firms trying to raise capital for expansion.
Therefore it has spawned an extensive literature attempting to explain this
apparent financial anomaly. A number of theories of IPO underpricing have
been put forth and tested against the data of various stock markets.
This chapter attempts to explain the cross-sectional differences in under-
pricing of Chinese IPOs using data compiled for 308 firm-commitment new
issues from 1987 to 1995. The Chinese case is of particular interest because
of the extreme magnitudes observed after market trading of stocks began
in late 1990. Defining IPO initial return as IPORETN = {P
l
- P
0
)/P
0
,
the mean IPORETN in our sample is 948.6%! In other words, the first-
day market closing price was on average almost eleven times as high as the
initial price offered to domestic Chinese investors.
A frequent and understandable reaction to the striking statistics for
initial returns of Chinese IPOs is that the extraordinarily low IPO prices
relative to first-day market prices must represent either irrational behavior
or, more likely, give-aways (presumably bribes) to officials who allocate the
rights to go public. For instance, Basu and Li (1998) argue that offering the
shares of newly privatized enterprises to the bureaucrats that most closely
controlled these State-Owned Enterprises was a necessary condition for
215
216 The Underpricing of Initial Public Offerings
efficient privatization, because: (1) these bureaucrats were the possessors
of inside information on which enterprises would most likely be successful
in a world of hard budget constraints; and (2) they (the bureaucrats) would
not reveal this information and risk their control over the enterprises under
their supervision without compensation (bribes), which took the form of
offering them shares at par value. We believe, however, that the bribery,
while it may well occur, does not explain underpricing for three reasons.
First, underpricing an entire public offering in order to allocate some of
the underpriced shares to officials in return for favors is very costly relative
to other means of paying illicit bribes. Second, underpricing of shares can
be explained as rational behavior under asymmetric information. Third,
the bureaucrat-insiders had strong incentives to reveal information about
which firms were good prospects; "bribery" and asymmetric information
may not be mutually exclusive alternative hypotheses. We believe that
the conditions for equilibrium underpricing exist in China, and that high
underpricing is caused by basic problems derived from the microeconomic
uncertainty and information asymmetry.
Allen and Faulhaber (1989), Grinblatt and Hwang (1989), and Welch
(1989) have proposed a class of signaling models in which IPO underpricing
is an equilibrium outcome where issuers possess superior information than
investors do. As long as the revelation probability for the issuers' quality is
neither too large nor too small, there exists a separating equilibrium where
high-value issuers signal their quality by retaining a portion of shares and
underpricing initial offerings and low-value issuers sell all of their shares
and do not underprice. In a pooling equilibrium, no signaling occurs, al-
though high-quality firms will be observed, after the fact, to underprice.
Underpricing, averaged over all firms in the pool, will be zero.
Allen and Faulhaber (1989) showed that underpricing in a separating
equilibrium implies a shortage of IPO shares, which must be allocated by
the issuer or underwriter in some fashion. Granting favored public officials
the right to purchase IPO shares may be one of the allocation methods
chosen, just as an underwriter may favor long-term customers with purchase
rights. In this sense, bribery is a byproduct of underpricing, not a cause.
We believe that our econometric results support this view.
The rest of the chapter proceeds as follows. In Section 6.2, we present
institutional details of the new-issue offering process in China and describe
variables used in our empirical work. In Section 6.3 we test various hy-
The New-issue and Offering Process 217
potheses derived from the signaling models against data of Chinese IPOs
and subsequent market behavior. In Section 6.4, we consider the hypothe-
sis that IPO underpricing in China is primarily explainable as a means of
bribing public officials, and we examine an assertion of the World Bank that
allocating IPO shares by lottery has been a cause of extreme underpricing.
In Section 6.5, we extend the above approach to examine the difference in
initial returns between A shares (available only to Chinese investors) and
B shares (available only to foreign investors). We summarize the findings
and propose future research in this area in Section 6.6.
6.2 The New-issue and Offering Process
There are several interesting characteristics of the new-issue and offering
process in China:
First, the aggregate amount of new shares issued each year is deter-
mined by a quota set by the security regulatory authorities, namely, the
State Planning Commission, the People's Bank of China (the central bank),
and the China Securities Regulatory Committee (CSRC). The quota is then
distributed to individual provinces. The stated criteria used for allocation
of new issues among provinces reflect the central security regulatory au-
thorities' perceived regional development needs and provincial differences
in production structure and industrial base. Within each regional quota,
the local security regulatory authorities invite enterprises to request a list-
ing and make a selection based upon enterprise performance and sectoral
development objectives. Infrastructure enterprises, especially those special-
izing in electricity and water supply, are given priority.
Second, the Chinese government has introduced a variety of share cate-
gories to allow ownership of state-owned enterprises to be dispersed among
the government itself, other state-owned enterprises, firms' own employ-
ees, domestic public and foreign investors. There are currently five types
of shares: (1) government shares, which are retained in the state institu-
tions and government departments and are non-tradable; (2) legal entity
shares, or C shares, which can only be held by other state-owned enter-
prises. C shares can not be listed in the two official exchanges (Shanghai
and Shenzhen Security Exchange), but a very small number are traded on
the Security Trading and Automatic Quote System (STAQS) and National
Electronic Trading System (NETS); (3) employee shares, which are non-
tradable until the firm allows their convertibility; (4) ordinary domestic
individual shares, or A shares, which can only be purchased and traded by
private Chinese citizens in the two official exchanges in China; (4) foreign
individual shares, which can only be purchased and traded by the foreign in-
vestors in security exchanges in China (B shares), in Hong Kong (H shares)
or in NYSE (N shares).
1
Sample statistics of the proportion of each kind of share in new-share
offerings are shown in Table 6.1. Note that most stock sales are partial
sales. The government maintains control in varying degree over many firms.
The size of government ownership at the time of IPO ranged from none
to 74% with an average of 10%. Only 89 out of 308 issuers going public
between December 1986 and January 1996 reported no retained government
shares, and even these enterprises reported IPO sizes that did not exceed
50% of firm's intrinsic value, indicating that a large portion of shares were
controlled by other state-owned enterprises. The concept of an enterprise's
intrinsic value (formally defined to be the discounted value of all future net
cash flows) is essential to analysis of the causes of IPO underpricing and to
other important aspects of corporate finance. Yet, in world of imperfect,
not to mention asymmetric, information, intrinsic value is unobservable to
outsiders and, hence to researchers. Our operational definition of intrinsic
value is a measure of total market capitalization, measured by the value of
the sum of all types of shares from the IPO through all and all subsequent
equity offerings.
Third, most stock sales are partial sales. The government still main-
tains control in varying degree over many firms. The size of government
ownership ranges from 10% to 88%. Only 89 out of 308 issuers going pub-
lic between December 1986 and January 1996 do not report government
ownership of shares. However, none of these 89 issuers has reported IPO
size that is above 50% of its total market capitalization, which indicates
x
An issuer of B shares must, besides satisfying requirements stated in the securities
regulations, meet the following conditions: (1) It must have obtained approval from the
relevant authorities for its use of foreign investment or for its conversion into a foreign-
funded enterprise. (2) It must have a stable source of adequate foreign exchange income
and the total amount of its annual foreign exchange income must be sufficient to pay
the annual dividend. (3) The proportion of B shares to the total number of shares must
not exceed the ceiling determined by the relevant authority. The aggregate amount of
shares is fixed in each year and the total number of firms allowed to issue foreign shares
is also limited. An issuer of H or N shares is not subject to the quota restriction, but
is subject to case-by-case approval.
The New-issue and Offering Process 219
Table 6.1 Sample Statistics for Share Offerings
Variable Mean Median Min. Max. Number of
Observations
Full sample of 308 firm-commitment IPOs
between December 1986 and January 1996
(in million U.S. dollar)
A Share IPOs
Government shares
Other State Enterprise
Employee Shares
Seasoned Equity
B Share Offerings
16.83
15.91
7.10
0.80
114.42
4.06
11.99
3.66
2.71
0.36
33.60
2.54
0.29
0
0
0
0
0.82
249.40
486.09
289.13
22.89
8513.8
23.07
308
219
264
294
268
57
(as a fraction of firm's intrinsic value, %)
A Share IPOs
Government shares
Other State Enterprise
Employee Shares
Seasoned Equity
15.02
10.05
5.81
0.87
49.93
12:6
4.41
2.67
0.39
41.58
0.07
0
0
0
0
81.25
74.27
33.85
11.09
99.87
308
219
264
294
268
t hat a larger port i on of its shares are still controlled by ot her state-owned
enterprises.
Fourt h, it is also not ewort hy in t er ms of t he signaling hypot heses t est ed
in t hi s paper t hat initial offering size relative t o firm' s intrinsic value is on
average 15%, which is small relative t o subsequent seasoned equity offerings
(50% on average). Seasoned equity offerings (SEOs) are very frequently
observed. About 91% of t he Chinese firms t hat went public before June 30,
1994 have issued subsequent equities. By comparison, in t he Uni t ed St at es,
between 1980 and 1986, only about 20% of firms going public subsequently
issued seasoned offerings. (See Jegadeesh, Weinstein and Welch, henceforth
J WW, 1993.)
Fifth, average t i me elapsed between t he announcement of an I PO and
t he first day of market t radi ng is considerable greater in Chi na t han in
other countries260 calendar days for A shares and 72 calendar days for B
shares. Ther e are a number of steps a firm must t ake after it is selected for
initial public offering and before t he market t radi ng begins. Some typical
steps include: (1) publication of a prospectus in newspapers and selection
of underwriters; (2) purchase of application forms by prospective investors;
(3) a lottery to determine which individual and institutional investors will
be allowed to purchase new issues at the IPO price; (4) delivery of shares
to the lottery winners after payments are made.
Sixth, the offering mechanism adopted by most Chinese firms is quite
different from those observed in mature stock markets and has undergone
several substantial changes over time. The offer price is chosen months be-
fore the market trading starts, and in the great majority of offerings there is
no feedback mechanism through market demand that allows adjustment in
the offer price. The lottery mechanism, which remains the primary method
of share allocation, has undergone several substantial changes. Before 1992,
the security regulatory authorities designed a lottery system based on a
pre-announced fixed number of application forms. Each retail investor was
allowed to purchase a limited number of lottery forms from the central bank
and its subsidiaries. Lottery winners were entitled to a certain number of
shares per winning form. With the number of lottery forms pre-determined,
the odds of winning the lottery was known to investors. In 1993, the se-
curity regulatory authorities introduced two new lottery mechanisms: One
mechanism was based on an unlimited number of application forms. The
central bank sold as many lottery forms as investors were willing to buy.
Therefore, the odds of winning the lottery were unknown to investors at the
time of lottery. The other lottery mechanism was based on savings deposit
certificates. Investors were required to deposit a certain quantity of funds
into a special saving account when submitting an application for shares,
which could not be withdrawn until the lottery was completed. These spe-
cial saving accounts were given relatively low interest.
Under the lottery mechanisms, the IPO prices were fixed for all in-
vestors. In the early stage of development of Chinese stock markets, some
initial issues were even offered at the shares' face values, which were Ren-
minbi 1 yuan. Companies that went public before January 1991, such
as Shanghai Vacuum Electronics, Jinbei Automotive, Phoenix Chemical,
China Textile Machinery, Shenzhen Vanke Co., Gintian Industry, Shen-
zhen Zhenye Co., and Shenyang Materials Development, all offered shares
at Renminbi 1 yuan. Initial IPO returns were over 1000% for these shares.
In 1994, two kinds of auction mechanisms were introduced. Under the
first auction mechanism, an issuer set an initial price and investors were
The Role of Financial Variables in the Pricing of IPOs 221
required to bid for the price and quantity. The final offer price was set at
the level where the accumulative quantities demanded by investors equaled
the total number of new shares available. Under the second auction mech-
anism, the IPO price was fixed and investors were invited to bid for the
quantity of shares. In case of oversubscription, all investors were guaran-
teed a certain amount of shares and the remaining shares were distributed
on a pro-rata basis. Lottery mechanisms as described above remained as
options, however, so firms could choose either a lottery method of distribut-
ing IPO shares or an auction mechanism. The proceeds from the lottery
become the property of the state banks that sell shares for the firms.
Seventh, Chinese A-share IPOs are sold to relatively unsophisticated
retail investors while B shares are usually sold to international institutional
investors such as foreign mutual funds.
6.3 The Role of Financial Variables in the Pricing of IPOs
6.3.1 The Chinese Data
We use data of all the firm-commitment IPOs of A-share common stocks
occurring between 1987 and 1995. In order to study the effects of subse-
quent equity offerings on IPO underpricing, we also extract a sub-sample of
firms that went public between 1987 and June 1994. The June 1994 cutoff
allows 548 days for a firm to issue seasoned equities.
Variables used in our empirical work include:
IPOSZ = the number of shares issued at the time of an IPO,
including A- and B-shares.
IPOVALUE = IPOSZ x offer price, or the proceeds from an IPO
(in U.S. dollars).
LNIPOSZ = logarithm of IPOVALUE.
SEOSZ
q
= the number of shares issued at the time of an SEO,
q = 1,2,3. Y, SEOSZ
q
= SEOSZ
g
SEOVALUE = ] T SEOSZ
q
x offer price
g
,or the proceeds from
all SEOs.
TOTALSHA
LNTOSIZE
LNGOVNT
LNEMPLOY
INTRINSC
PROFIT =
INFORMED
AGE
TIMEIPO
AFTRETNl
AFTRETN2
EARNINGS
LEVERAGE
IPOSZ + SEOSZ + government shares
+ employee shares +state enterprises shares,
or the total number of shares outstanding.
logarithm of the sum of IPO and SEO values.
logarithm of the value of government shares.
logarithm of the value of employee shares.
IPOVALUE + SEOVALUE + stated, or face,
value of government shares, employee shares
and other state enterprises shares.
past calendar year's profit at the time of IPO.
(in U.S. dollars).
the sum of government shares, manager and
employee shares, and legal entity shares
divided by TOTALSHA, or the percentage
of shares held by insiders and institutional
investors.
age of the firm at the date of its IPO.
number of days elapsed between announcement
of an IPO and first-day of market trading.
after-market return based on closing prices from
the closing of the first day of trading to the
end of the second week of trading.
after-market return between the closing prices
from the beginning of the third week of
trading to the end of the fourth week.
earnings forecast dummy, EARNINGS = 1 if
the issuer provides earnings forecast in
the prospectus, EARNINGS = 0 otherwise.
the book value of pre-IPO debt (short term
and long term) divided by the book value
of all assets.
MKTRUNUP = cumulative daily returns on the Shanghai
or Shenzhen security exchange 30 trading
days before an IPO, an indicator for the market
conditions surrounding a new issue.
PREMSTD = standard deviation of daily returns on the
Shanghai or Shenzhen security exchange 30
trading days before an IPO, an indicator for the
market uncertainty surrounding a new issue.
STD = standard deviation of daily after-market returns
estimated over a 100-trading day period after
inception of market trading.
EXD = stock exchange dummy, EXD = 1 for a company
that is listed on the Shanghai Securities
Exchange, EXD = 0 for a company that is
listed on the Shenzhen Securities Exchange.
SIC(k) = six industry dummies: durable goods (SIC1),
non-durable goods (SIC2), transportation
and public utilities (SIC3), finance,
insurance and real estate (SIC4), services
including restaurants, department stores and
hotels (SIC5) and domestic and foreign
trade (SIC6).
YEAR(t) = IPO year dummies, t = 1 if a firm went public
before January 1, 1991; t = 2 5 for
going public in the years 1992 through 1995.
Descriptive statistics for some of the above variables are presented in
Tables 6.2 and 6.3. The correlation matrix for some of the variables is
presented in Table .
224 The Under-pricing of Initial Public Offerings
Table 6.2 Descriptive Statistics for Variables to Explain IPO Initial Returns
Variable Mean Median Std. dev. Min. Max.
A-share full sample with 308 firm-commitment IPOs
between December 1986 and January 1996
-18.58%
-0.9268
-0.9173
0
5.4806
7.1701
0.6931
3
0
0
0.29
2.77
0.65%
2.61
0.07%
38300%
1.7039
0.9444
2.5938
12.241
15.772
4.4773
1868
12.908
9.8522
249.40
3788.3
51.27%
1856.4
81.25%
6. 3. 2 A Benchmark Regression
As an empirical exercise t o analyze t he cross-sectional differences in I PO
underpricing among Chinese firms, we est i mat e t he following baseline cross-
sectional regression using ordi nary least squares (OLS) for t he full sampl e
and sub-sampl e of firms.
IPORETN, = a
0
+ a
x
RCPIPOi + a
2
LNIPOSZi
+a
3
PROFSHA
l
+ a
4
LNTOSIZEi
+a
5
LNAGEi + a
6
TIMEIPOi
+a
k
SIC(k) + a
t
YEAR(t) + e* (6.1)
The OLS regression est i mat es for A-share I POs presented in Table 6.4
show t hat :
(1) The smaller t he I PO price, t he larger is t he I PO initial ret urn.
(2) The smaller t he size of initial offering, t he higher is t he I PO initial
IPORETN
AFTRETNI
AFTRETN2
InPo
LNIPOSZ
LNTOSIZE
LNAGE
TIMEIPO
LNGOVNT
LNEMPLOY
IPOVALUE
PROFIT
STD
INTRINSC
IPOSZ
INTRINSC
948.59%
-0.0264
-0.0275
1.1966
9.0781
10.681
2.4849
260.12
6.0294
5.6087
16.83
466.58
5.89%
161.23
15.02%
231.25%
-0.054
-0.045
1.335
9.2053
10.628
2.7701
135
8.0194
5.7038
11.99
221.99
4.74%
86.01
12.6%
2967.7%
0.2428
0.1841
0.6352
0.9955
1.0975
0.8929
341.24
4.0055
1.5882
22.56
98.13
5.11%
237.43
11.22%
Table 6.3 Descriptive Statistics for Variables to Explain IPO Initial Returns
Vari abl e Mean Medi an St d. dev. Mi n. Max.
A- shar e sub- sampl e wi t h 268 f i r m- commi t ment I POs
between December 1986 and June 1994
-10%
-0.9268
-0.9173
0
5.4806
8.2295
0.6931
3
0
0
0.29
2.77
0.65%
2.61
3
0
0
38300%
1.704
0.693
2.3253
12.241
15.772
4.477
1831
12.908
9.852
249.40
3788.3
51.27%
1856.4
1831
99.87%
1217
return; the larger the size of total offerings, which is the sum of IPO
and all SEOs, the higher is the IPO initial return. These findings
imply that firms with small IPOs relative to their total offerings
have a relatively high degree of IPO underpricing.
(3) Time elapsed between the announcement of an IPO and the first-
day market trading does not affect the IPO underpricing. This
contradicts Chowdhry and Sherman (1996), who show that IPO
underpricing is positively related to the time period between IPO
date and first trading date.
(4) The age of the firm and the profit per share variables do not seem
to be related to the IPO initial return, indicating that available
information about a firm at the time of the IPO is not related to
IPO underpricing. This is in contrast to that of Ritter (1991), who
found significant negative relationship between IPO underpricing
and past information for the U.S. firms.
IPORETN
AFTRETN1
AFTRETN2
InPo
LNIPOSZ
LNTOSIZE
LNAGE
TIMEIPO
LNGOVNT
LNEMPLOY
IPOVALUE
PROFIT
STD
INTRINSC
TIMEIPO
SEOSZ
INTRINSC
TIMESEO
1043.1%
-0.2171
-0.0712
1.1805
9.0635
10.77
2.7743
251.04
6.0898
5.6935
16.89
430.64
6.01%
160.03
251
49.93%
284
271.24%
-0.0215
-0.0375
1.315
9.177
10.678
2.6391
142
8.0283
5.7038
11.65
211.13
4.74%
88.59
142
43.85%
220
3166.3%
0.2519
0.1684
0.6271
1.0099
1.0583
0.9016
305.98
4.0101
1.3461
23.61
83.64
5.42%
236.72
306
28.95%
217
Table 6.4 OLS Regression Estimates for the Baseline Empirical Model
Full"
Sub
Full
Sub
Full
Sub
Full
Sub
Constant
-69.6048*
(-3.0612)
-93.6159*
(-3.7446)
LNTOSIZE
10.0695*
(6.1995)
11.1672*
(5.9527)
SIC4
2.8004
(0.4782)
3.8525
(0.5565)
YEAR4
-3.9135
(-0.5362)
-3.6153
(-0.6443)
RCPIPO
22.9758*
(3.4019)
25.1627*
(3.4513)
TIMEIPO
-0.0172
(-1.0161)
-0.0141
(-1.2724)
SIC5
5.0636
(0.9657)
8.193
(1.384)
YEAR5
0.7963
(0.1014)
LNIPOSZ
-5.0583*
(-2.4754)
-4.1532
f
(-1.8391)
SIC1
7.8147
f
(1.7232)
9.9033*
(1.9049)
YEAR1
43.5243*
(3.3936)
44.4975*
(3.3733)
R
2
0.3581
0.3757
PROFSHA
-4.1638
(-1.4312)
-4.4099
(-1.4059)
SIC2
10.7099*
(2.1654)
14.0239*
(2.4707)
YEAR2
-7.6043
(-0.8122)
-8.5893
(-1.0246)
LNAGE
2.3892
(1.4162)
2.5793
(1.3741)
SIC3
1.6218
(0.2802)
3.2262
(0.4773)
YEAR3
0.6983
(0.0868)
-0.9173
(-0.1385)
"The dependent variable is the IPO initial return. The independent variables are the
reciprocal of IPO price (RCPIPO), logarithm of IPO size (LNIPOSZ), profit per
share (PROFSHA), logarithm of the age of t he firm (LNAGE), logarithm of t he size of
total offerings (LNTOSIZE), time elapsed between the offer date and the first trading
date (TIMEIPO), industry dummies (SIC(k)) and IPO year dummies (YEAR(t)).
Figures in parentheses are t-statistics. *, t denote 5% and 10% level of significance,
respectively.
(5) Durable and non-durable goods industries have larger IPO under-
pricing than other industries, suggesting that the proportion of
high-value firms going public is larger in durable and non-durable
goods industries than in other industries.
(6) IPO initial returns were significantly higher at the early stage of
The Adverse-Selection Models 227
development of Chinese stock markets.
Since the LNIPOSZ, TIMEIPO and In P
0
variables are endogenous,
OLS estimation gives inconsistent estimates for the coefficients for these
variables. Therefore, I use the instrumental variables (IV) approach in
estimating the structural equation (6.1).
To instrument the LNIPOSZ, LNTIMEIPO and l nP
0
variables,
we regress them against the exogenous variables LNAGE, LNPROFIT,
SIC(k)jmd YEAR(t)jismg OLS, and obtain the fitted values LNIPOSZ,
LNTIMEIPO and lnP
0
- As will be shown in subsequent sections, IPO
prices are determined from firms' value-maximization problem and first-
day market prices reveal whether a firm is of high-value or not. There is no
compelling reason for the IPO price and first-day market price to depend
upon industry and IPO year dummies. Therefore, SIC(k) and YEAR{t)
variables are excluded from the structural equation (6.1).
The results from the 2SLS regression presented in Table 6.5 show that:
(1) The smaller the IPO price, the larger the IPO initial return.
(2) The smaller the size of initial offering, the higher the IPO initial
return. This implies that firms with small IPOs have higher degree
of IPO underpricing.
(3) Time elapsed between the announcement of an IPO and the first-
day market trading is positively related to the degree of IPO un-
derpricing. This is consistent with Chowdhry and Sherman (1996).
(4) The age of the firm and the past profit variables are positively
related to the IPO initial return. This implies that firms with
better performance and low-priced new issues have higher-priced
shares after market trading begins.
The benchmark model successfully relates the cross-sectional differences
in IPO underpricing to some economic variables. In the next section, I test
signaling hypotheses that may help explain the extraordinarily high A-share
IPO underpricing that characterizes the Chinese markets.
6.4 The Adverse-Selection Models
Rock (1986) and Beatty and Ritter (1986) propose that there is information
asymmetry between informed and uninformed investors. Because quantity
rationing instead of price adjustment occurs when there is excess demand
Table 6.5 2SLS Regression Estimates for the Benchmark Model
Variable
Constant
lrfPo
LNIPOSZ
LNPROFIT
LNAGE
LNTIMEIPO
R
2
Full Sample"
210.4286*
(3.1832)
-120.1951 +
(-1.753)
-43.7689*
(-4.2017)
20.9588*
(4.4772)
5.8186*
(3.1548)
13.9308+
(1.9101)
0.2415
Sub-Sample
271.942*
(3.2998)
-146.3093+
(-1.901)
-53.3934*
(-4.2471)
25.223*
(4.5008)
6.8196*
(3.2696)
14.3106+
(1.7707)
0.2561
"The dependent variable is the IPO initial return. The independent variables are the
instrument for the logarithm of IPO price variable (In PQ), instrument for the logarithm
of IPO size variable measured in dollars (LNIPOSZ), logarithm of past year' s profit at
the time of IPO (LNPROFIT), logarithm of the age of the firm (LNAGE) and instru-
ment for the logarithm of time elapsed between the offer date and the first trading date
(LNTIMEIPO). InPo, LNTPOSZ and LNTIMEIPO are obtained by regressing
l nP
0
, LNIPOSZ, and LNTIMEIPO on LNAGE, LNPROFIT, industry dummies
(SIC(k)) and IPO year dummies (YEAR(t)). Figures in parentheses are t-statistics.
*, + denote 5% and 10% level of significance, respectively.
for shares, the informed investors crowd out the uninformed investors for
allocations of profitable issues. The allocations received by uninformed
investors are biased toward less-profitable issues. Therefore, in order to
attract uninformed investors to participate in the IPO market, firms must
underprice their IPOs sufficiently to allow uninformed investors a reason-
able return for their ex ante uncertainty, and to enable them to recover
the losses resulting from their purchase of overpriced securities. There are
two important empirical implications: (1) Issues in which the public have
a priori knowledge that insiders and institutional investors have little or
no participation will be less underpriced. The reason is that uninformed
investors will not face winner's curse and allocation disadvantage, and un-
derpricing will not be required to induce them to participate in the IPO
markets. (2) Expected underpricing increases in the ex ante uncertainty
The Adverse-Selection Models 229
surrounding an issue.
To examine cross-section variations in the underpricing of Chinese IPOs,
we regress the degree of IPO underpricing (IPORETN) against a number
of proxies for ex ante uncertainty. That is, we test the joint hypotheses that
there is a positive relationship between risks and the degree of IPO under-
pricing and proxies used in the following regression are adequate measures
of risks.
IPORETN, = a
0
+ a
x
(
Ip0
y
ALUE
* l.OOO.OOo) (6.2)
+a
2
EARNINGSi + a
3
USAGE
l
+aJNFORMEDi + a
5
PROFSHAi
+a
6
LEVERAGEi + a
7
In AGEi
+a
8
MKTRUNUP
l
+ a
9
PREMSTDi
+a
w
STDi + a n InTIMEIPOi
+a
12
LDi + aisEXDi + a
14
TIMEi
+a
k
SIC(k)i + a
1/IPOVALUE is the inverse of proceeds raised in an A-share IPO
expressed in terms of 1993 purchasing power. It reflects the maintained hy-
pothesis that smaller offerings are more speculative, on average, than larger
offerings (Beatty and Ritter, 1986). This implies that a.\ is positive. In gen-
eral, firms are unwilling to give out detailed information on what they will
do with the net proceeds from IPOs because of increasing legal liability and
exposure to competitors. However, the China Security Regulatory Com-
mittee (CSRC) requires highly speculative issues to provide future earnings
forecasts and disclose IPO funds usage in their prospectus. As a result of
this regulation, issues that list earnings forecasts and usage of IPO funds
are associated with greater ex ante uncertainty.
2
a
2
and a
3
, the coefficients
for EARNINGS and USAGE, should both be positive. The percentage
of shares held by insiders and institutional investors (INFORMED), a
proxy for the severity of the winner's curse problem, is hypothesized to be
positively related to underpricing.
In addition, information on the past performance of a firm, such as
2
For example, it is well known t hat Tsing Tao Brewery stated one use of its proceeds in
the prospectus and then use the money for other purposes, which turned out to yield
very disappointing returns to investors.
calendar year's profit prior to an IPO divided by the number of IPO shares
(PROFSHA), debt-to-total assets ratio (LEVERAGE) and the age of
the issuing firm (In AGE), is related to the ex ante information asymmetry
(Ritter, 1991, and Su, 2000d). In particular, the higher the past profit-per-
share, the better is a firm's growth potential. 0:5 should be negative. Su
(2000c) argues that a high pre-IPO leverage ratio raises ex ante uncertainty
about the financial strength of a firm, because debt financing for investment
projects is not a viable choice for imposing a hard budget constraint on
management, while a small pre-IPO leverage conveys good news to the
market. This suggests that a@ is negative. Ritter (1991) finds that there is a
strong negative relationship between the age of the firm and the IPO initial
return, which is consistent with the notion that risky issues require higher
average returns and that age is an useful proxy for this risk. Therefore, 07
should be negative.
Prevailing market conditions (MKTRUNUP and PREMSTD) influ-
ence the assessment of firm risk. If market returns are high and the variance
of returns is low at the time a firm goes public, the IPO initial return will
naturally be high. Moreover, risky firms have an incentive to go public
when market conditions are favorable. Hence, Qs should be positive while
ag should be negative. Furthermore, Ritter (1984) uses the variability of
stock returns of the issuing firm in the after-market period as one of the
proxies for ex ante uncertainty. He finds significant relationship between
the variability of after-market returns and the degree of IPO underpric-
ing for a sample of natural resource companies. He interprets his findings
as giving support to the claim that the greater the uncertainty about the
true price of new shares, the larger is the discount that an issuer must
offer in selling IPOs, and that there is no reason to restrict risk proxies
to ex ante observable characteristics. Therefore, we hypothesize that the
degree of IPO underpricing is positively related to AFTMSTD2, i.e., aio
is positive.
Chowdhry and Sherman (1996) propose that IPO initial returns will be
higher if the underwriter gets some direct compensation based on the size
of the demand for IPO shares. In addition, when the offer price of an IPO
is set many days before the actual issue day, there is a possibility that the
first-day price at which the issue would trade in the secondary market may
leak to the public. It is well known that most Chinese A-share IPOs are sold
through a lottery mechanism and the underwriter (typically a state bank)
sells lottery application forms. The revenue from the sale of lottery forms
The Adverse-Selection Models
231
belongs to the underwriter. The IPO price is also set long before shares
are sold-49 days on average. The empirical prediction is that the risk of an
issue and hence, the degree of underpricing, will be higher if In TIMEIPO
is larger and/or a lottery mechanism is used in share allocation. This
suggests that both a n and a.\2 are positive.
Since most firms listed in Shanghai are formerly large-scale state enter-
prises while most firms listed in Shenzhen are relatively small joint-venture
companies, we add a variable EXD to capture the differential risks of
these two types of firms. 0:13 is hypothesized to be positive. Finally, we
use SIC(k) and TIME to control for industry-specific risks and risks as-
sociated with the year of IPOs.
3
We note that to raise a given amount of funds from IPO (IPOVALUE),
a firm can sell a smaller fraction of ownership and thus, retain a larger frac-
tion of shares held by government, managers, employees and other institu-
tions (INFORMED). To detect possible simultaneous equation bias, we
estimate regression (1) with and without
IPO
VALUE ^
or t n e e n
t
l r e
sample
of 371 firms and a sub-sample of 239 firms. All regressions are corrected for
heteroskedasticity using the generalized least squares (GLS) procedure. We
present regression results of initial-day underpricing on issue characteristics
and proxies of informational uncertainty in Table 6.6.
We find that there are only minor differences among the regressions.
The explanatory power of the model, as measured by the adjusted-i?
2
, is
between 0.62 and 0.70. The coefficient estimates for INFORMED are
significantly positive at the 5% level, indicating that IPO underpricing is
smaller for issues in which the public knows a priori that government and
other institutions have little or no participation. To further investigate
the relation between IPO underpricing and the participation of insider and
institutional investors, we rank firms in ascending order of INFORMED
and group firms into quintiles. Table 6.7 compares IPO initial returns and
some fundamental characteristics of firms that are grouped into the lowest
and highest quintiles of INFORMED. As shown in the table, the mean
IPO initial return is significantly larger for the high INFORMED group
than for the low INFORMED group. This suggests that the "lemon's
problem" is less pronounced in a market in which uninformed investors have
3
Because firms going public in China typically use large state banks, which in turn use
their branch networks, to distribute shares, there is little or no variation in underwriters'
reputation. We do not use variables such as investment banker prestige, venture capital
backing and the use of warrants to underwriters as proxies for ex ante uncertainty.
Table 6.6 GLS Regression Estimates for the Relationship Between IPO Underpricing
and Proxies of Uncertainty
Pul l -sampl e Sub- sampl e
Constant
1,000,000
IPOVALUE
EARNINGS
USAGE
INFORMED
LEVERAGE
In AGE
MKTRUNUP
PREMSTD
STD
LD
EXD
TIME
SIC(l)
SIC(2)
adjusted- R
-3.1461**
(-5.1636)
0.8664*
(2.1883)
1.6282*
(2.3530)
2.5051**
(3.1630)
1.0343*
(2.1180)
-2.5209
(-1.0216)
4.7535**
(3.6186)
-1.4661+
(-1.7620)
-1.1442
(-1.0620)
3.7522**
(4.1737)
2.5730*
(2.3595)
-6.3106**
(-7.1883)
2.0184+
(1.8695)
2.9155*
(2.3472)
0.6240
-4.2073**
(-6.5922)
10.8424**
(3.2911)
1.4015**
(4.0682)
1.3358*
(2.1004)
1.7472*
(2.4705)
0.8551+
(1.8495)
-2.6864
(-1.2883)
3.9609**
(2.9301)
-1.2054
(-1.5163)
-1.3059
(-1.2191)
4.1086**
(4.8452)
2.0288+
(1.8803)
-7.7296**
(-8.6960)
2.3353*
(2.0597)
2.5930*
(2.1306)
0.6749
-3.9426**
(-6.1407)
0.7155*
(2.0377)
1.4420*
(2.1646)
2.3118**
(3.0069)
1.2177*
(2.5099)
-2.7038
(-1.3170)
5.2459**
(4.1784)
-1.6295+
(-1.8942)
-1.2684
(-1.1742)
3.9901**
(4.3174)
2.1916*
(2.1066)
-6.9350**
(-7.8054)
1.7728+
(1.7820)
3.2475*
(2.5185)
0.6508
-4.9050**
(-7.2825)
12.0629**
(4.8463)
1.3407**
(3.6299)
1.2512*
(2.0319)
1.4550*
(2.1504)
0.8835+
(1.9033)
-2.7599
(-1.3656)
4.4620**
(3.3861)
-1.3305
(-1.5955)
-1.4117
(-1.2838)
4.5360**
(5
:
2044)
1.8633+
(1.7110)
-8.3260**
(-9.5414)
1.9660+
(1.8204)
2.7305*
(2.0608)
0.7033
prior knowledge of a lack of informed investors. This result is consistent
with the winner's curse hypothesis but is contradictory to the prediction
of the signaling model of Perotti (1995), who argues that underpricing of
state-enterprises' IPOs is an equilibrium outcome for a government issuer
to signal its future commitment to pro-market privatization policies.
T
h
e

A
d
v
e
r
s
e
-
S
e
l
e
c
t
i
o
n

M
o
d
e
l
s

n

c
o

0
0

C
O

T
f

0
5

t
^

l
O

C
O

C
O

i
n

0
0

o
o

C
N

C
N

T
(

l
O

l
O

0
0

C
N

s
o

<
>

6
?

o

o

C
O

T
i
l

*

Qui
P

C
O

C
D

o

H
-
l

g

'
3

O
"

+
J

C
O

<
D

s

0

g

l
e
g

C
J

i
-
J

S
i
t
e

C
D

O

d
"

B

C
D

t
o

*
*

"
S
.

g
i
g

^

-
3

S

v
O

*

a

"
S

2

"
i

t
o

C
D

s
^

C

,
_

3

.
2

c
r

o

c

c
S

c

c
d

w

l
a

C
D

T
3

.

-

(
3

J
3

O

3

3

_

4
f
f
l
H

+
J

*
t
?

^

W

O

o

J

l
i

"
j

"
^

C
D

C
D

O

-
g

-
3

+
j

+
J

+
j

.
S

c

.

3

8
f
c

^

2

O

&

C
O

I
i

a
>

e
n

Q
.
O

I
D

3

^

h
O

c
3

C
D

<

f
c

K
l

X

^

B
E
;

^

g

5

^

-

3

t
t
]

a
3

C
D

C

-
t
e

a
)

(
a

*

5

~
3

C
D

C
D

'
-
I

J
3

X
I

+
J

^

^

C
D

c
c
j

X

c
j

=
3

_
,

3

'
-
^

s

a

r
a

>

>

o

s
,

"
f
5

J
f
e

s

c
a

J
3

1
-
S

c

-
^

l
a
.

S
3

H
I

C
O

,
-
S

3

c
e

Q

C
D

C
D

C
D

>
>

S

O

C
O

-
3

C
D

|

^

C
t
3

c
f
i

3

O

*
H

3

u

C
D

C
D

>

3
*

g
o

"
g

.
"

X
I

X

>
>

C
D

S

i
s

x

g

-
S

-
g

C
D

g

*
^

o

3

.

>
>

C
D

'
~

^
3

S

_

C
D

C
D

~

^

C
D

*
j

t
n
;

f
c
e

s

2

t
o

T
3

C
D

C
D

O

J
S

c
a

^

f
T

M
-
,

c
a

f
t

"
<

o

>

In Table 6.6, the coefficient estimates for EARNINGS and USAGE
are both significantly positive at the 5% level, suggesting that firms required
to provide earnings forecasts and detailed usage of IPO proceeds experience
higher IPO underpricing. The coefficient estimates for the reciprocal of the
A-share proceeds from an IPO are significantly positive at the 1% level,
indicating that smaller offerings have substantially higher average initial
returns. The coefficient estimates for LEVERAGE are significantly pos-
itive at least at the 10% level, which implies that investors perceive the
debt-to-total assets ratio as a correlate with the ex ante uncertainty of the
firm, and that the higher the pre-IPO leverage ratio, the larger is the ex-
pected initial return. In addition, there is evidence that prevailing market
conditions influence the assessment of firm risk because the coefficient esti-
mates for MKTRUNUP are significantly positive at the 1% level and the
coefficient estimates for PREMSTD are significantly negative at the 10%
level when JPQVALUE
l<& n o t mc m
de d in the regression. Investors seem
to believe that risky firms go public when market conditions are favorable
(i.e., the pre-IPO market return is high while the standard deviation of
returns is low) and require higher initial returns for risky firms. However,
in contrast to empirical findings of the U.S. data by Ritter (1984), the vari-
able AFTMSTD2 is not only statistically insignificant, but also of wrong
sign. Since we are testing joint hypotheses here, it is possible that the
after-market standard deviation of returns is not a very good proxy for the
ex ante uncertainty of the firm.
Furthermore, we find conflicting evidence from the Chinese data against
the theories of Chowdhry and Sherman (1996). The coefficient estimates
for LD are significantly positive at the 1% level, indicating that under-
pricing is higher, on average, for firms that use a lottery mechanism in
allocating shares. However, the coefficient estimates for \nTIMEIPO are
insignificant at the conventional level, although it is of right sign. There is
lack of evidence that the degree of IPO underpricing is related to the time
elapsed between when the IPO price is set and when the issue is actually
sold. The coefficient estimates for EXD are significantly positive, suggest-
ing that firms listed in the Shenzhen Securities Exchange are, on average,
more risky and have higher expected IPO underpricing. TIME, the vari-
able representing the year of an IPO, is significantly negative. This implies
that firms going public in the early stages of development of Chinese stock
markets were generally perceived as more risky firms, probably because the
degree of information asymmetry among investors was larger at that time.
The Signaling Models 235
The coefficient estimates for SIC(1) and SIC(2) are significantly positive
at the 10% and 5% level, respectively, indicating that durable and non-
durable goods industries are underpriced more than the other industries,
whose coefficient estimates are insignificant and hence, omitted from Ta-
ble 6.6 for brevity. Finally, the coefficient estimates for PROFSHA and
In AGE are of right signs, but are insignificant at the conventional level.
In summary, we find evidence that investors interpret the percentage of
insider and institutional investors' ownership, the size of offerings, whether
the issuer provides earnings forecast and IPO fund usage in the prospectus,
leverage, market conditions prior to IPOs, allocation mechanism, listing
exchange, listing time, and the belonging industry as correlates with ex
ante uncertainty, and that there is a positive relation between ex ante
uncertainty among investors and expected underpricing. Our empirical
results strongly support the winner's curse hypothesis of IPO pricing.
6.5 The Signaling Models
6.5.1 Empirical Implications from the Signaling Models
Allen and Faulhaber (1989), Grinblatt and Hwang (1989), Welch (1989)
and Chemmanur (1993) have proposed a class of signaling models of IPO
underpricing in which issuers have superior information than investors do
about the intrinsic value of the firms. The discussion in the remainder
of this section draws heavily upon these papers. In the cited models, an
issuer maximizes the value of the firm through the IPO and subsequent
SEOs. Because they do not have complete information, in the absence
of a signal, investors cannot distinguish between "high value" and "low
value" firms. In a separating equilibrium, a "high value" issuer signals
its quality by underpricing. It can afford to underprice its IPO because it
expects to capture larger revenues through SEOs. In contrast, a "low value"
issuer does not signal because it does not expect to recoup its investment
in underpricing through after-market SEOs. The best a low-value issuer
can do is to "take the money and run" when its stock is initially offered.
When a separating equilibrium occurs, the average risk-adjusted IPO return
over all new issues will be positive, the quantity of shares demanded for
underpriced issues will exceed quantity supplied, and shares will be rationed
by a mechanism other than the offer price.
The papers mentioned above have somewhat different setups that lead
to similar, but not identical, testable hypotheses. In Allen and Faulhaber's
framework, as long as the probability that a high-quality firm will remain
good after it implements its production technique, product, or whatever it
does, is not "very large", the high-value firm will find it advantageous to sig-
nal by underpricing its IPO while a low-value firm will not find it worthwhile
to underprice. Therefore, underpricing occurs in a separating equilibrium.
In Grinblatt and Hwang (1989), if the expected value and the variance of
future cash flow for a risk-averse firm is unknown to the investors, then,
corresponding to the revelation probability of the firm's intrinsic value (be-
tween 0 and 1), there exists an equilibrium signaling schedule relating two
signals, fractional holding and IPO underpricing. The firm signals its value
to the market by choosing the optimal combination of underpricing and
IPO size. In Welch (1989), low-value firms must invest in imitation ex-
penses to appear to be high-value firms, and with non-zero probability this
imitation will be discovered between the dates of IPO and SEO. Under-
pricing by high-value firms at the time of IPO can add sufficient signaling
costs to these imitation expenses to induce low-value firm to separate and
thus reveal their value voluntarily.
In a recent paper, Chemmanur (1993) models the long held intuition
that underpricing motivates the market to produce information about the
issuing firm. High value issuers underprice to motivate outsiders to produce
information and make informed bids for the IPO. Low value issuers mimic
this behavior, but their type will be revealed after the IPO bids. IPO un-
derpricing results from insiders inducing information production to obtain
a more precise valuation of their firms in seasoned equity offerings.
The testable implications from the signaling models with separating
equilibrium include:
(1) The correlation between the degree of IPO underpricing and IPO
proceeds is negative. In the signaling model of Grinblatt and Hwang
(1989), there exists a monotonic correspondence between offering
price at which the issue is being sold and the degree of IPO un-
derpricing in a separating equilibrium, holding constant the firm's
fractional shareholding (or, equivalently, holding constant the frac-
tional size of the IPO). That is, if investors observe an issuer's
fractional ownership and can conjecture the value of a benchmark
lowest-variance issuer by looking the relative size of its IPO, they
can also infer the degree of IPO underpricing for any arbitrary is-
The Signaling Models
237
suer given the issuer's initial offering price and relative IPO size.
(2) There is an optimal signaling schedule relating the firm's intrinsic
value, the degree of underpricing, and relative IPO size. In partic-
ular, the degree of underpricing and the issuer's intrinsic value are
positively correlated, holding constant the issuer's fractional owner-
ship and the variance of returns; the degree of underpricing and the
project variance are positively correlated, holding constant the is-
suer's fractional ownership and the variance of returns; the degree
of underpricing and the fractional holding are positively related,
holding constant the variance of returns and the firm's intrinsic
value.
(3) Issuers with a larger degree of IPO underpricing are more likely to
return to the secondary market and offer larger amount of SEOs,
more quickly than issuers with lower IPO underpricing.
There are various "stories" rationalizing this signaling hypothesis.
One linkage between the underpricing signal and SEO behavior is
that an issuer gives out "free samples" to the public by underpric-
ing and induces the public to learn more about the issuer. The
learning process leads to a higher price on the first day of mar-
ket trading than would otherwise occurbut for high-value issuers
only. This effect on the market price allows a high-value issuer to
quickly return to the market with SEOs and thereby reap the re-
turn from underpricing its IPO. In another version, investors are
more passive and "learning" occurs through exogenous revelation
of information after implementation of the firm's innovation. The
IPO is underpriced to induce investors to furnish sufficient startup
funds to enable implementation of the innovation. Firms are will-
ing to underprice because they expect a higher-than-normal return
on their investment.
The alternative to a separating equilibrium is a pooling equilibrium
where IPO prices are a weighted average of the present value of high-value
and low-value issuers. When information is revealed after market trading
starts, price differentiation occurs, but the first-day market price cannot be
predicted from the IPO price and there is no risk-adjusted "excess" IPO
initial return between the offer date and the first-day market trading for a
large sample of firms.
In a pooling equilibrium, high-value firms do issue SEOs, but only in
response to market-provided information about their value. That is, the
market possesses superior information than the issuers. If the after-market
returns are high, then firms will issue seasoned equities. If they are low,
then there are no seasoned issues. No equilibrium signaling schedule exists.
This is the so called market feedback hypothesis.
6.5.2 Correlation Between IPO Returns and IPO Proceeds
The first set of empirical implications in Section 6.4.1 indicates a mono-
tonic negative correspondence between the price of the initial offering and
IPORETN, holding constant the relative size of the IPO in a separating
equilibrium under the signaling hypothesis. In a pooling equilibrium, since
the IPO price is a weighted average of the intrinsic value of low- and high-
quality firms, there is no implied correspondence between the offering price
and realized IPO initial return.
To test this hypothesis empirically, we need to consider that measuring
offering price consistently across firms is complicated by the fact that the
number of shares a firm declares when it corporatizes is arbitrary. There
is no reason, a priori why two firms with the same intrinsic value would
declare the same number of shares. Ideally, we would measure offering
price as the observed offering price per share divided by intrinsic value per
share. However, intrinsic value is unobservable. Since we do not observe
the firm's intrinsic value, offering prices are measured with error. This
measurement error does not affect the variable IPORETN, because the
unobserved value of intrinsic value per share would appear in both the
numerator and denominator, and thus cancels out. Moreover, there could
be a spurious negative correlation between the observed offering price and
the observed IPORETN, even under a pooling equilibrium.
In light of the above complications, we use the following two proxies
for the offering price: (1) In IPOV ALUE, which is the logarithm of the
product of the number of share issued at the time of an IPO and the
observed offering price. (2) OPTOBOOK, which is the observed offering
price per share divided by the book value per share. To correct for the
errors-in-variable (ElV)problem, we adopt a two-stage approach to derive
unbiased estimates of the relationship between the logarithm of IPO return,
IPO price, and the relative size of IPO (the number of shares offered at the
time of the IPO divided by the total number of shares outstanding, the
complement of the firm's fractional shareholding). In the first stage, we
Table 6.8 2SLS Regression Estimates for the Correlation Between IPO Underpricing
and IPO Price
Variable"
1
Constant
In IPOSZ
IPOSZ
MKTCAP
R
2
Pull Sample
15.1907*
(23.2712)
-0.8612*
(-11.3981)
-11.2312*
(-10.4522)
0.5396
Sub-Sample
13.9522*
(20.1465)
-0.7021*
(-8.5542)
-12.3773*
(-10.7187)
0.5469
a
The dependent variable is t he logarithm of IPO initial return (In IPORETN). The in-
dependent variables are the instruments for the logarithm of IPO proceeds (In IPOSZ)
and the relative size of IPO ( MKTCAP ) ' ^'
l e
instruments are obtained by regress-
ing In IPOSZ and MKTCAP
o n e x
S
e
nous variables including In AGE, In PROFIT,
InTIMEIPO, stock-exchange dummy (EXD), industry dummies (SIC(k)) and IPO
year dummies (YEAR(t)). Figures in parentheses are t-statistics. *, t denote 5% and
10% level of significance, respectively.
regress In IPOVALUE or OPTOBOOK against the exogenous variables
In AGE, In PROFIT, InTIMEIPO, EXD, SlCJk) and YEAR{t), and
obtain the instruments In IPOV ALUE or OPTOBOOK. In the second
stage, we estimate the following regression:
JPQSZ
In IPORETN, = fa + ftln I POV ALUE + (3
2 TQTALSHA
+ <* (6-3)
We interpret Grinblatt and Huang (1989) to imply that equation (6.3)
holds for issuers of given initial wealth. We assume that the issuer is the
Chinese government, and therefore the initial wealth in the cross section
of issuers is roughly constant and can thus safely be omitted from the
regression.
The estimation results reported in Table 6.8 show that: (1) holding
constant the relative size of initial offerings, a one percent decline in the
price of initial offering is associated with approximately a 1.98 percent
increase in IPO initial return and that (2) a one percentage-point increase
in relative IPO size is associated with a 6 percent decline in the IPO return.
Both coefficients are highly significant.
6.5.3 Correlations Among IPO Underpricing and Issuer's
Intrinsic Value, Fractional Ownership and Project
Variance
The second set of empirical implications in Section 6.4.1 suggests a posi-
tive relationship between an issuing firm's IPO return, relative IPO size,
variance of future returns, and intrinsic value (under a separating equilib-
rium). Testing these implications is complicated by the necessity of mea-
suring the variance of future returns and lack of an error-free measure of
the issuing firm's intrinsic value. As a proxy for intrinsic value, we use
the I NT RI NSC variable which is the sum of the proceeds from IPO, all
SEOs, employee shares, government shares and state enterprises shares. We
proxy the variance of future returns with variance of daily A-share returns
over the first 100-day period of market trading. Moreover, the complement
of the fraction of the firm's value retained by the issuer, the relative size
of IPO is determined endogenously. Again, to correct for these EIV and
endogeneity problems we estimate the following regression by 2SLS using
the same instruments as in equation (6.3) to obtain the fitted values of the
regressors:
In IPO RET Ni = f3
0
+ j3
x
In INTRINSC
+/3
2
ln IPOV'ALUE + foSTD +
i
(6.4)
The estimation results reported in Table 6.9 show that: (1) holding
constant the firm's intrinsic value and the project variance, a one percent
decrease in the size of the initial offering is associated with approximately a
two percent increase in the degree of IPO underpricing; (2) holding constant
the size of the initial offering and the project variance, a one percent increase
in the firm's intrinsic value is associated with approximately a one percent
increase in the degree of IPO underpricing; and (3) holding constant the
firm's intrinsic value and the size of the initial offering, an increase of one
standard deviation in the variation of future returns is associated with
almost an eleven percent increase in the degree of IPO underpricing, which
implies an elasticity of approximately 0.6 at the mean value of the regressor.
All of the coefficients are statistically significant, even at the 1% level.
It should be noted that we have calculated another measure of firm's
intrinsic value, which is the outstanding market capitalization as of De-
cember 29, 1995. We find that both measures of firm's intrinsic value are
Table 6.9 2SLS Regression Estimates for the Correlation Among IPO Underpricing and
Issuer's Intrinsic Value, Fractional Ownership and Project Variance
Variable
11
Constant
\n IPOSZ
In MKTCAP
STD
R
2
Full Sample
-5.1423*
(-2.9377)
-1.8383*
(-19.9304)
1.1406*
(12.0549)
10.9575*
(4.0374)
0.6216
Sub-Sample
-6.2608*
(-3.1859)
-1.7676*
(-17.411)
1.1646*
(10.9617)
10.2704*
(3.7736)
0.5967
a
The dependent variable is the logarithm of IPO initial return (In IPORETN). The
independent variables are the instrument for the logarithm of the size of initial offer-
ings (In IPOSZ), instrument for t he logarithm of firm's intrinsic value (In MKTCAP),
and instrument for the project variance (STD). The instruments are obtained
by regressing In IPOSZ, In MKTCAP, and STD on exogenous variables including
In AGE, In PROFIT, InTIMEIPO, stock-exchange dummy (EXD), industry dum-
mies (SIC(k)) and IPO year dummies (YEAR(t)). Figures in parentheses are t-
statistics. *, t denote 5% and 10% level of significance, respectively.
highly correlated and yield almost the same results in all of our estima-
tions. Therefore, we only report the estimation results using the first mea-
sure. Our empirical results in this section support a separating equilibrium
under the signaling hypothesis.
6.5.4 IPO Underpricing and SEOs
All the signaling models discussed in Section 6.4.1 imply that high-quality
issuers underprice their IPOs so that they can subsequently issue seasoned
equities at more favorable prices than they would otherwise have received.
Issuers that do not plan to sell seasoned equities subsequent to the IPOs
do not discount their initial sales. In reality, not all IPOs that appear to
be underpriced are followed by SEOs. About 91% of Chinese firms going
public before June 30, 1994 issued subsequent equities before January 1,
1996. As JWW (1993) point out, some of the firms that underprice their
IPOs with the intention of issuing subsequent equities may fail to do so
because of unexpected economic shocks. However, such shocks are less
likely to prevent firms of very high quality to stick to their original plans. In
addition, some underpricing may be unintentional or unexpected. Therefore
we view the relationship between IPO underpricing and subsequent offerings
probabilistically, that (1) issuers with larger IPO underpricing are more
likely to issue subsequent equities than issuers with lower expected IPO
underpricing. Additional hypotheses related to IPO-SEO behavior are:
(2) Issuers with larger IPO underpricing will issue larger amounts of SEOs
relative to their intrinsic value, and (3) Issuers with larger IPO underpricing
will issue SEOs more quickly after the initial sales.
4
Empirical tests need to distinguish between the implications of a sepa-
rating and a pooling equilibrium, where firms to do not signal their value to
the investors through underpricing. A firm's specific value is revealed only
after market trading begins. This suggests the "market-feedback hypoth-
esis," in which high post-IPO returns lead firms to issue seasoned equities
and low post-IPO returns will discourage issues of additional shares.
To test the hypothesis that issuers with larger expected IPO underpric-
ing are more likely to issue SEOs, we follow the spirit of JWW (1993) and
estimate the following logit model,
/ pSEO \
In f
i
J
pSEO
j = 4>o + <PiIPORETN
i
+ 4>
2
AFTRETNl
i
+<p
3
AFTRETN2i (6.5)
where p?
EO
is the probability that the ith issuer will issue subsequent
equity offerings after the initial sale. The independent variables are the
observed degree of IPO underpricing (IPORETN), the after-market return
for the first two weeks (AFTRETNl) and the after-market return for the
third and fourth weeks (AFTRETN2).
JWW (1993) include a variable representing IPO size in their equation.
Their rationale for including IPO size is "Since firms that raise relatively
small amounts of capital at the IPO may be more likely to return with a
seasoned equity offering, we include the natural logarithm of IPO size as
an additional explanatory variable." We believe that inclusion of IPO size
4
J WW (1993) test t he signaling models using U.S. dat a and find weak evidence t hat firms
t hat underprice their IPOs are more likely to issue seasoned equities and on average
have larger subsequent offerings.
Table 6.10 Logit Regression Estimates for t he Relationship Among I PO Underpricing,
After-Market Returns and Probability of SEOs
Vari abl e" Coefficient i -st at i st i c
Cons t ant 2.0242* (6.5531)
IPORETN 0.001326* (2.1301)
AFTRETN1 0.031597* (2.1831)
AFTRETN2 0. 014543 (0.8885)
"The dependent variable is the probability for the i t h firm to issue SEOs {Pf
EO
). The
dependent variable takes value 1 if SEOs are observed and 0 otherwise. The independent
variables are the observed IPO initial return (IPORETN), after-market first two-week
return (AFTRETN1) and after-market next two-week return (AFTRETN2). We
allow 548 days for a firm to issue SEOs, therefore our sample only consists of firms
who went public between December 1986 and June 1994. There are 268 firms in the
sample. Figures in parentheses are asymptotic t-statistics. *, t denote 5% and 10%
level of significance, respectively.
as a regressor is inappropriate when the hypothesis being tested is that
underpricing reduces the size of IPOs relative to SEOs. IPO size becomes
endogenous and may well "pick up" the effect of underpricing we are trying
to estimate. In addition, JWW (1993) do not take into consideration the
EIV problem when using observed IPO initial returns instead of a proper
instrument for the expected IPO initial returns. Unfortunately, neither are
we able to find any observable exogenous variables that are unknown to the
investors under the asymmetric information assumption. Thus, following
JWW (1993), we do not use an instrumental-variable or 2SLS approach in
estimating equation (6.5).
The logit regression estimates are presented in Table 6.10. The slope
coefficient for IPORETN (^-statistic) is 0.001326 (2.1301), which indicates
a positive relationship between the degree of IPO underpricing and the
probability for a firm to issue SEOs.
To compare the explanatory power of the signaling versus market feed-
back hypotheses, we calculate (dP
SEO
/dIPORETN) dIPORETN and
(dP
SEO
/dAFTRETNl) dAFTRETNl and evaluate these two expres-
sions using the standard deviation of the respective regressors, obtaining
4.2 and 0.80, respectively, implying that a one standard-deviation increase
in IPO return is associated with an increase in the magnitude of the prob-
ability of an SEO that is more than five times larger than a one standard-
244 The Underwriting of Initial Public Offerings
deviation increase in after-market return over the first two weeks of market
trading after the IPO date. Since the estimated coefficient for AFTRETN2
is much smaller and not significantly different from zero, we conclude that
the empirical power of the market signaling hypothesis is greater than that
of the market feedback hypothesis.
To test the implication that firms with larger IPO underpricing tend to
issue a larger amount of seasoned equities, we estimate a Tobit regression
that relates IPO underpricing and after-market returns to the ratio of the
market value of all SEOs to the firm's intrinsic value which is proxied by
the sum of the proceeds from IPO and all SEOs,
5
/SEOVALUE\
V INTRINSC )
The coefficient estimates of equation (6.6) are contained in Table 6.11.
The coefficient estimate for IPO RET N (t-statistic) is 0.00003 (6.13), which
indicates that the higher the IPO underpricing, the larger is the size of
seasoned equity issues.
To compare the relative explanatory power of the signaling and market
feedback hypotheses, we follow the same procedure as with the estimates of
equation (6.5) and calculate [d (^T^MNSC) /dIPORETN) -dlPORETN
a n d
(
d
(im^rMi) /dAFTRETNl) dAFTRETNl using the standard
deviation of the respective regressors. We obtain 0.09 and 0.05, indicating
that a one standard-deviation increase in IPO return and first two-week
after market return are associated, respectively, with an increase in relative
SEO size of 10% and 5%, respectively. This comparison suggests that the
signaling hypothesis has almost twice as much "strength" in explaining SEO
relative magnitude than does the market-feedback hypothesis. As with the
probability of an SEO, the coefficient estimate for AFT RET N2 is positive,
but is statistically insignificant. We conclude that while the Tobit regression
estimates support the signaling models relating A-share IPO underpricing
to the relative size of SEOs, they do not allow us to reject the alternative
5
We allow until January 1, 1996 for a firm to issue its first SEO. About 91% of the
firms t hat went public between December 1986 and June 1994 issued subsequent equity
offerings before this date. Because our sample is censored as of the cutoff date of
January 1, 1996, a Tobit specification is desirable.
d
0
+ dilPUKtill\i
+S
2
AFTRETNli , ,
+5
3
AFTRETN2
l
+e
i
if RHS> 0
[
' '
0 otherwise
Table 6.11 Tobit Regression Estimates for the Relationship Among IPO Underpricing,
After-Market Returns and Size of SEOs
Var i abl e
0
Coefficient i -st at i st i c
Cons t ant 0.4312* (25.8477)
IPORETN 0.0000309* (6.127)
AFTRETN1 0.002015* (2.6024)
AFTRETN2 0. 000483 (0. 5501)
a
The dependent variable is t he ratio of total seasoned equity offerings as a fraction
of the intrinsic value for the i t h firm ( MKTCAP I ^ ^
e
independent variables
are the observed IPO initial return (IPORETN), after-market first two-week return
(AFTRETN1) and after-market next two-week return (AFTRETN2). We allow 548
days for a firm to issue SEOs, therefore our sample only consists of firms who went pub-
lic between December 1986 and June 1994. There are 268 firms in the sample. Figures
in parentheses are asymptotic t-statistics. *, t denote 5% and 10% level of significance,
respectively.
hypothesis that the size of SEOs is also partially related to after-market
returns.
Finally, we examine the relationship between IPO underpricing and the
time elapsed between the IPO and the first SEO using the following Tobit
model:
TIMESEOi = I
7o + 71 IPO RET
]
Ni
+
l2
AFTRETNU
+~j
Z
AFTRETN2i + e< if RHS> 0
[
' '
0 otherwise
where TIMESEO is the number of days elapsed between the IPO date
and the first SEO date.
The Tobit regression estimates are presented in Table 6.12. The slope
coefficient (^-statistic) for IPORETN is -0.002 (-0.52), indicating that
firms with a higher degree of IPO underpricing tend to return to the sec-
ondary market and issue seasoned equity offerings more quickly than firms
with a lower degree of IPO underpricing and larger IPO sizes, although the
estimated coefficient of IPO return is not significant at conventional levels.
In comparing the relative explanatory power of the market-signaling
and market-feedback hypotheses, we note that the median time elapsed
Table 6.12 Tobit Regression Estimates for the Relationship Between I PO Underpricing,
After-Market Returns and Time Elapsed Between IPO and First SEO
Vari abl e" Coefficient t - st at i st i c
Const ant 291. 0871* (22.3196)
IPORETN -0.00222 (-0.5245)
AFTRETN1 1.0864* (2.2203)
AFTRETN2 0.7852 (0.9593)
"The dependent variable is t he number of days between t he I PO dat e and t he first
SEO date for the i t h firm TIMESEO. The independent variables are the observed
IPO initial return (IPORETN), after-market first two-week return (AFTRETNl) and
after-market next two-week return (AFTRETN2). We allow 548 days for a firm to issue
SEOs, therefore our sample only consists of firms who went public between December
1986 and June 1994. There are 268 firms in the sample. Figures in parentheses are
asymptotic t-statistics. *, t denote 5% and 10% level of significance, respectively.
between IPO and SEO dates is almost twenty weeks, far longer than the
period over which we measure after-market returns. Thus, we predict a
negative coefficient on the two after-market return variables as evidence
supporting the market-feedback hypothesis. However, the estimated co-
efficients (t-statistic) for AFTRETNl and AFT RET N2 are 1.09 (2.22)
and 0.79 (0.96), respectively, which we interpret as contradictory to the
market-feedback hypothesis.
To summarize, we find strong evidence from the Chinese A-share data
that supports the signaling models that link IPO underpricing to SEOs. We
find that Chinese issuers who underprice their A-share IPOs more heavily
are more likely to return to the secondary market and issue larger amounts
of after-market equities. The alternative, namely, the market feedback
hypothesis, does not fare as well in explaining the high underpricing in the
data.
6.6 Bribery and Lottery Hypotheses of IPO Underpricing
6.6.1 Bribery Hypothesis: Underpricing of Chinese IPOs
are Gifts to Public Officials or Fav ored Purchasers
Allen and Faulhaber (1989) show that underpricing implies an excess de-
mand for IPO shares. It follows that either the issuing firm or its agent
Bribery and Lottery Hypotheses of IPO Underpricing 247
(e.g. the underwriter) has power to allocate valuable assets to favored as-
sociates in return for their loyalty, favors, or whatever. In this context it
seems natural to consider that issuers may allocate underpriced shares to
politicians or bureaucrats who have substantial control over some of their
planned resources. For example, under the quota system in China, a firm is
not unlikely to allocate underpriced shares to bureaucrats in order to gain
permission to issue shares
6
We reject a priori the likelihood that bribery by underpinning under
the alternative frameworks of full information or a pooling equilibrium with
asymmetric information between issuers and investors accounts for observed
IPO returns in China. Why should a firm underprice its entire IPO in order
to create an opportunity to bribe a few officials? This seems very costly
indeed, given that other means are available. It would be much cheaper for
the firms to transfer funds directly to bureaucrats in order to gain favors.
Admittedly, there may be a few cases involving very small IPOs where the
extra cost of using underpricing as a bribe relative to alternative means is
minimal, but we believe that such occurrences, if they exist, do not shape
the broad pattern of underpricing observed in the data.
As a test of the importance of bribery in the IPO process, we con-
jecture, that with bribery, many bribe recipients would want to sell their
underpriced shares as quickly as they can, or as soon as the stock prices go
up after market trading begins, leading to significant depression of returns
in the secondary market. Moreover, high-quality firms that allocate under-
priced shares to gain permission to go public will be more likely to delay
their SEOs to avoid depressing stock prices further and give the politicians
or bureaucrats a chance to cash in.
Evidence pertaining to the price-depression effect noted above can be
deduced from calculating alternative measures of IPO underpricing, us-
ing the after-market price from the first day, first week, second week, and
subsequent weeks' closing prices. We find that compared to the first-day
IPO return, which has a mean of 948.59% (with a minimum of -18.58%
and maximum of 38300%), the average degree of IPO underpricing using
first-week, second-week, third-week and fourth-week market closing prices
remains approximately the same, 950.98%, 955.9%, 950.49%, and 898.39%,
6
Note t hat bribery itself can be construed as a signal of firm value in t hat only a firm
t hat expects to recoup its investment in bribery will bribe. It is questionable, however,
whether a bribe can be a signal to the average investor, who presumably is not privy
to the illicit transaction.
respectively. (The minimums are -17.64%, -27.95%, -38.72%, -42.05% and
the maximums 48920%, 50410%, 52210%, and 52470%, respectively.) We
take this as evidence that bribes (if they occurred) were not "cashed in"
during the first four weeks of market trading. Since IPO underpricing was
the largest at the early stage of development of stock markets in China
when regulations governing new issues were not yet well developed, it is
natural to suspect that this was a fertile period for bribery. Therefore we
calculate different measurement of the average degree of IPO underpricing
for 29 firms that went public before the emergence of formal stock markets.
However, we do not find any evidence of a large after-market price drop for
these issues, either
7
.
We also note that the estimation results from the Tobit regression 6.7,
offers no evidence that a higher IPO return is associated with a delay in
offering SEOs. In addition, the average time elapsed between the first-day
market trading and SEO dates for those 29 firms going public before the
establishment of formal stock markets is 254 days, which is less than the
overall average of 284 days for the entire sub-sample of firms going public
before July 1994. We take this as additional evidence against the bribery
hypothesis.
To sum up, while bribery may in principal be a byproduct of under-
pricing in a separating equilibrium under asymmetric information, direct
evidence is by the nature of the process difficult to find. We find little if
any indirect evidence that bribery occurred in a large enough proportion of
cases to cause to observable effects on aftermarket returns and/or the tim-
ing of SEOs. We conclude that the signaling hypothesis characterized by a
separating equilibrium seems to be capable of explaining the a substantial
portion of underpricing behavior of the Chinese firms.
7
The average degree of underpricing for 29 new issues t hat went public before the emer-
gence of formal stock markets using t he first-day market price is 5664.29%. (The
minimum is 748% and t he maximum is 38300%.) The average degree of IPO under-
pricing using first-week, second-week, third-week and fourth-week market closing prices
for these 29 new issues is 6053.4%, 6181.39%, 6230.83%, and 5964.11%, respectively.
(The minimum is 782%, 728%, 679%, 610%, and the maximum is 48920%, 50410%,
52210%, and 52470%, respectively.)
Bribery and Lottery Hypotheses of IPO Underpricing
249
6.6.2 Lottery Hypothesis: Lottery Mechanism in Share Al-
location Contributes to High IPO Underpricing
Five different offering mechanisms have been used in allocating A - shares
in China. They are: a lottery mechanism with a fixed number of appli-
cation forms (OD(l)), a lottery mechanism with an unlimited number of
application forms (0.0(2)), a lottery mechanism based on certificate of de-
posit receipts (OD(3)), an auction mechanism with only quantity bids and
fixed IPO prices (0.0(4)), and an auction mechanism with quantity and
price bids (0.0(5)). A team of World Bank specialists argued that offering
mechanisms affected the degree of underpricing ("China: The Emerging
Capital Markets" Vol. II, p. 96).
. . . the allocation mechanism adopted for the new share
issue affects the degree of underpricing. Non-discretionary
allocation of shares, by mechanisms such as a lottery, ex-
acerbate the tendency to underprice.
We disagree with their assertion. It was hypothesized and empirically
supported that a lower offer price and smaller IPO size leads to higher IPO
underpricing. As mentioned in previous sections, IPO underpricing implies
an excess demand for shares initially and necessitates some form of non-
price rationing. All types of lottery mechanisms and auction mechanism
with only quantity bids are best viewed as simply means to allocate over-
subscribed shares, and do not determine the after-market demand for shares
and/or the initial supply of shares. Therefore, these mechanism themselves
do not cause IPO underpricing. An alternative to the lottery, an auction
in which both price and quantity bids are submitted, does create an offer
price that is not predetermined by the issuing firm. Investors are invited
to submit a single bid for the price they are willing to pay and the amount
of shares they are willing to purchase. The final offer price is set at a
level where the accumulative quantity demanded equal the amount of new
shares available. In case of oversubscription, shares are allocated on a pro
rata basis. Eighteen firms used this mechanism in issuing and allocating
new shares prior to January, 1996. The average degree of IPO underpricing
is only 1.04% for these 18 issues, which is negligible. It is critically im-
portant to note that these firms chose the auction method when a lottery
mechanisms was available. With the choice of share-allocation mechanism
endogenous, it would be difficult to conclude that the auction led to virtu-
ally zero underpricing. Rather, the opposite may be inferred, namely, that
these firms chose not to signal their quality by underpricing and used the
auction mechanism as a means to assure they received the largest possible
revenue at the IPO, because they had little nor no intention of returning
to the secondary market with SEOs.
Even though we reject a priori the notion that the IPO allocation mech-
anisms in China have caused the large magnitudes of underpricing observed,
we nevertheless find it interesting to investigate the relationship between of-
fering mechanisms and IPO initial return cross-sectionally, holding constant
the variables associated with IPO return under the hypothesis of asymmet-
ric information with a separating equilibrium. To do this, we reformulate
equation (6.3) as follows, which we estimate using 2SLS:
In IPO RET Ni = r/
0
+ ?ftln IPOSZi +
%
l n MKTCAP,
+r
l2
,Sfb
l
+ r]4,LDi +
Vi
(6.8)
where
LD dummy variable that takes value 1 if a firm uses a lottery
mechanism or auction mechanism with only quantity bids in
allocating shares and 0 if it uses auction mechanism with
both quantity and price bids.
The estimation results presented in Table 6.13 show that the coefficient
for LD variable is highly positively significant, indicating that underpric-
ing is present under lottery mechanisms and auction mechanism with only
quantity bids. Under the auction mechanism with both price and quantity
bids, which is designed to eliminate excess demand for shares at the first
place, underpricing is on average negligible.
We also test the hypothesis that the mean IPO underpricing differs
under different offering mechanisms by estimating the following expansion
of equation (6.1) using 2SLS:
In IPORETN, = 0
O
+ 9{ln IPOSZi + 0
2
ln MKTCAPi
+e
3
sfD
t
+ diOD(l)i + Vi (6.9)
Underpricing of Foreign-Share IPOs
251
The estimation results in Table 6.13 show that underpricing is on av-
erage larger under the various lottery mechanisms than under the auction
mechanisms. In particular, IPO underpricing is the largest under the lot-
tery with fixed number of application forms, is the second largest under
the lottery with unlimited number of application forms, and is the small-
est under auction mechanisms. To interpret these results, we emphasize
that when more than one mechanism was available, as they were in 1993
through 1995, choice of mechanism was endogenous. We reiterate that un-
der a asymmetric information with a separating equilibrium, underpricing
is a signal for firm's true value, there will be excess demand for shares and
some type of offering mechanism has to be used to eliminate the excess
demand. Offering mechanisms themselves do not cause underpricing.
6.7 Underpricing of Foreign-Share IPOs
We now examine the underpricing of B-share IPOs using data for 57 firms
that issued both A and B shares between February 1992 and January 1996.
Table 6.14 presents the sample statistics for firms issuing both A and B
shares. The average degree of IPO initial returns is 37.13% for the 57
B shares but is 838.91% for the corresponding 57 A shares. A further
comparison of the sample statistics in Tables 6.2 and 6.14 shows that, on
average, firms issuing both A- and B-shares have experienced smaller degree
of IPO underpricing than firms who raise equity capital only in China. The
difference in average IPO underpricing raises two questions: First, why
will a firm issue foreign-owned B shares? Second, what determines the
differences in underpricing of these two classes of shares?
We believe that Chinese firms issue foreign-owned shares to: (1) obtain
foreign capital that is otherwise difficult to get under government's foreign
currency control regime; (2) expand markets for raising funds and enhance
the reputation of the firm. Therefore, firms with larger stock-market capi-
talization and better performance in terms of higher profit before the IPOs
are more likely to issue B shares than firms with smaller stock-market cap-
italization and smaller profit per share. A comparison of the sample statis-
tics in Table 6.14 with those for all firms issuing shares in Table 6.2 shows
that on average, Chinese firms that offer B shares are considerably larger
in terms of total market capitalization; have higher profit before the IPOs;
have larger total offerings of A shares; and have experienced smaller degree
Table 6.13 2SLS Regression Estimates for t he Relationship Between I PO Underpricing
and Offering Mechanisms
CONSTANT
In IPOSZ
In MKTCAP
STD
LD
ODl
OD2
OD3
ODA
R
2
Full sample"
-3.6962*
(-2.624)
-1.4253*
(-17.6848)
0.8819*
(11.2355)
8.8957*
(4.075)
1.3404*
(12.9301)
0.7565
-2.4563*
(-1.9631)
-0.9214*
(-11.0507)
0.6002*
(8.2209)
4.2494*
(2.2415)
2.942*
(16.1233)
1.9509*
(11.6627)
1.2655*
(6.7202)
0.8824*
(4.6508)
0.8319
Sub-sample
-4.5103*
(-2.8844)
-1.3736*
(-15.8996)
0.8947*
(10.2935)
8.7891*
(4.0686)
1.4428*
(12.5193)
0.7467
-3.1875*
(-2.2574)
-0.927*
(-10.6195)
0.6327*
(7.8407)
4.3763*
(2.2835)
2.9669*
(14.6956)
1.9991*
(10.5858)
1.2849*
(6.0654)
0.8319*
(3.7219)
0.8198
"The dependent variable is the logarithm of IPO initial return (In IPORETN). The
independent variables are the instrument for the logarithm of the size of initial offer-
ings (In IPOSZ), instrument for the logarithm of firm's intrinsic value (In MKTCAP),
instrument for the project variance (STD) and lottery dummy (LD) or offering mecha-
nism dummies (OD(l)). Figures in parentheses are t-statistics and *, t denote 5% and
10% level of significance, respectively.
of IPO underpricing and larger after-market returns on their A shares than
the firms that have not offered B shares.
A convenient way to sort out the characteristics distinguishing Chinese
firms that offer B shares from those of firms that offer only A shares is
to estimate the following logit model, which I apply to all firms issuing A
shares.
In ( i - g ] = 0o + falPORETNi + fo In IPOVALUEi
Under-pricing of Foreign-Share IPOs 253
Table 6.14 Descriptive Statistics for Variables to Explain B-share IPO Initial Returns
Variable Mean Median Std. dev.
57 B-share IPOs
Min. Max.
IPORETN
AFTRETN1
AFTRETN2
InPo
LNIPOSZ
LNTOSIZE
LNEMPLOY
LNGOVNT
LNPROFIT
LNAGE
TIMEIPO
IPOVALUE
INTRINSC
37.13%
-0.0502
0.0393
-0.7009
7.7492
10.553
5.7674
6.058
8.5055
2.6791
71.807
33.71
297.91
21.43%
-0.0455
0
-0.7351
7.6535
10.3187
5.8295
8.4683
8.3707
2.3979
44
21.08
192.54
47.55%
0.1425
0.2156
0.4333
0.7936
1.0204
0.6972
4.5147
0.8856
0.9309
81.98
38.85
318.70
-21.14%
-0.542
-0.275
-1.812
6.528
8.5172
3.3322
0
7.001
1.0986
6
6.84
60.44
236.45%
0.24
1.0933
0.3834
9.8601
13.517
7.1493
11.113
11.175
4.3307
348
191.52
1856.43
57 Corresponding A-share IPOs
IPORETN
AFTRETN1
AFTRETN2
l nP
0
LNIPOSZ
LNTOSIZE
TIMEIPO
IPOVALUE
TIMEIPO
8.3891
0.0172
-0.018
-0.5691
9.1039
11.14
208.54
14.15
208.54
2.7059
-0.0221
-0.0631
-0.5147
9.1844
11.2241
142
11.74
142
12.27
0.3061
0.2244
0.433
0.7826
1.053
225.13
11.32
225.13
0.0457
-0.9257
-0.8535
-1.7492
6.1137
8.2941
3
0.54
3
45.143
1.335
0.9444
0.2249
10.751
13.081
980
56.23
980
+<j>
3
In INTRINSCi + 4>
A
In PROFITi
+</>
5
XuTIMEIPOi + foAFTRETNli
+4>
7
AFTRETN2i + <j>
k
SIC{k)i (6.10)
where P? is t he probability t hat t he i t h firm will issue B shares after it
has offered A shares. P? = 1 if firm % issues B shares and 0 otherwise.
The logit regression est i mat es presented in Table 6.15 indicate t hat :
(1) Holding const ant t he firm's stock-market capitalization, t he smaller
t he size of A-share I PO, t he more likely t he firm will issue B shares.
Table 6.15 Logit Regression Estimates for the Probability A Firm Will Issue B Shares
Variable"' Coefficient t - st at i st i c
CONSTANT
IPORETN
AFTRETNl
AFTRETN2
In IPOVALUE
In PROFIT
InlNTRINSC
InTIMEIPO
SIC1
SIC2
SIC3
SICA
SIC5
-30.99*
-0.00012
0.011+
0.009
-1.86*
0.48
f
1.73*
-0.41*
1.18
1.95*
0.36
-0.74
0.93
(-5.42)
(-1.04)
(1.94)
(1.06)
(-4.98)
(1.86)
(5.73)
(-2.05)
(1.56)
(2.32)
(0.41)
(-0.56)
(0.88)
"The dependent variable is the probability for the i t h firm to issue B shares after it
completes offering A shares (P^)- The independent variables are t he IPO initial return
(IPORETN), logarithm of IPO proceeds excluding B-share offerings (In IPOVALUE),
logarithm of firm's intrinsic value excluding B-share offerings (InlNTRINSC), after-
market first two-week return (AFTRETNl), after-market next two-week return
(AFTRETN2), logarithm of past year' s profit prior to IPO (In PROFIT), logarithm of
time elapsed between the offer date and t he first trading date (InTIMEIPO). We also
allow the probability of issuing B shares to vary across industries (SIC(k)). Figures in
parentheses are asymptotic t-statistics. *, t denote 5% and 10% level of significance,
respectively.
(2) Holding constant the IPO size, the larger the firm's stock-market
capitalization, the more likely the firm will issue B shares.
(3) The larger the profit prior to the IPO, the more likely it will issue
B shares.
(4) The smaller the degree of IPO underpricing, or the higher the after-
market return on its A shares, the more likely it will offer B shares.
(5) The shorter the time elapsed between the announcement of A-share
IPO and first-day trading, or the more efficient the firm's A-share
IPO process, the more likely it will issue B shares.
One of the most interesting features distinguishing the IPO process for B
shares from that of A shares is that lottery mechanism has never been used
in allocating B shares. Moreover, foreign securities firms such as Sassoon
Underpricing of Foreign-Share IPOs 255
and J. P. Morgan are allowed to participate in the B-share underwriting
process. The IPO prices for B shares are announced approximately one
month prior to the target market trading date and foreign investors are
invited to bid for the quantity of shares they wish to purchase. The ab-
sence of lottery mechanisms for B shares indicates that in contrast to A
shares, oversubscription is not a big problem in the new issue and offering
process for B shares. There is no persistent excess demand for B shares
by foreign investors at a "normal" price-earning ratio. Therefore, it is very
likely that the difference in the average degree of underpricing between A
and B shares can be explained by the differences in domestic and foreign
investors' available investment opportunities as well as the differences in
their investment sentiments.
When we apply the above empirical tests for the signaling models to the
B-share data, we find that neither signaling hypothesis nor the market feed-
back hypothesis discussed in Section 4 appears capable of explaining the
underpricing of B-share IPOs. However, when we estimate the benchmark
model (6.1) for the cross-sectional differences in underpricing for B-share
IPOs, we find that available information about the issuers, as reflected in
past profit prior to the IPOs, is positively and significantly related to the
IPO initial return. These two contrasts are consistent in the sense that
both suggest foreign investors rely more on information gathered prior to
purchasing shares than do domestic investors. A complementary explana-
tion is that Chinese firms participating in both domestic and foreign share
markets are those that can signal their quality through available informa-
tion about their histories and performance and do not need to "signal" as
strongly as firms participating only in domestic equity markets through the
mechanism of underpricing their IPOs. A comparison of the A-share sam-
ple statistics in Table 6.14 with those in Table 6.2 show that the difference
between the logarithms of total offering size and IPO size is 2.04 for the
subsample of firms issuing B shares and only 1.70 for the entire sample
of firms issuing both A-share IPOs and SEOs, while the mean IPO initial
return for the subsample is about 20% less than for the entire sample of
firms that make both IPOs and SEOs. These comparisons suggest that
the subsample participating in international equity markets do not need to
underprice their IPOs to the same extent as firms issuing only A shares to
stimulate demand for their SEOs.
6.8 Long-run Performance of IPOs
Recent work by Ritter (1991) and Loughran and Ritter (1995) has docu-
mented that in the long-run, the U.S. IPOs are overpriced. The long-run un-
derperformance of IPOs has been confirmed in many countries (Loughran,
Ritter, and Rydqvist, 1994). Ritter (1991) and Loughran and Ritter (1995)
interpret the long-run underperformance of IPOs as indicating that the high
initial returns at the offering dates occur due to over-reactions by investors.
They also argue that managers have some ability to time their offerings for
periods when investors are over-optimistic.
Shiller (1984) and De Bondt and Thaler (1985, 1987) argue that a stock
price may temporarily deviate from its fundamental value due to over-
optimism or over-pessimism on the part of investors. Fads or over-reactions
by investors are more likely to occur for small, less certain stocks or stocks
held by "noise traders" such as individual investors (Camerer, 1989, Lee,
Shleifer and Thaler, 1991, and Chopra, Lakonishok and Ritter, 1992). IPO
firms are, by nature, small and less certain than listed firms and tend to be
held by individual investors. Thus, IPOs are a good candidate for investi-
gating fads or investors' over-reactions. One of the empirical implications
which follows from the over-reaction hypothesis is: the greater the initial
return at the IPO date is, the greater the degree of subsequent correction
of overpricing will tend to be (Shiller, 1990).
As a long-run performance measure, we use (1) cumulative benchmark-
adjusted return and (2) the wealth relative. Both measures do not include
the first partial-month return. The average benchmark-adjusted return on
a portfolio of n stocks for event month t, AR
t:
is defined as
1
n
AR
t
= -
s
j
ar
it
(6.11)
t = i
The cumulative benchmark-adjusted return on event month t, CAR
t
, is
defined as
t
CAR
t
= Y^
AR
j (
6
-
12
)
3 = 1
The wealth relative based on the holding period return is used as an al-
ternative of cumulative benchmark-adjusted returns. The holding period
return on firm i from the closing price for the first partial-month to the
Long-run Performance of IPOs 257
closing price for event month t, R
it
, is denned as
R
it
= P~P' . i
( 6
.
1 3 )
Pi,l
where p
i>t
is the closing price for the tth calendar month if t > 2, and is
the closing price for the first partial month if t = 1. The holding period
return on the benchmark during the corresponding period for firm i, R
l
m t
,
is similarly calculated. The wealth relative of n IPOs up to event month t,
WR
t
, is obtained by
WRt = " i r r
1
(6.i4)
1
+ n l~,i=\
n
m,t
A wealth relative of less than 1 can be interpreted as IPOs underperforming
a benchmark. For the cross-sectional analysis of long-run performance, we
calculate the wealth relative up to month t for firm i, WR\, as
W
^ = TTW
L ( 6
'
15)
Ritter (1991) suggests that the long-run performance of IPOs is sensitive
to the selection of benchmarks. We use the Shanghai and Shenzhen A-
share indices as benchmark. Since the returns on IPO firms do not include
dividends, the returns on Shanghai and Shenzhen A-share indices also do
not include dividends.
Tables 6.16-6.17 shows the average benchmark-adjusted returns (AR
t
)
and the cumulative benchmark-adjusted returns (CAR
t
) excluding the first
partial-month returns over the 36 months of seasoning after the IPO month,
^-statistics for AR
t
is calculated as (AR
t
^/n^/sd
t
) where n
t
is the number
of observations in event month t and sd
t
is the cross-section standard de-
viation of benchmark-adjusted returns in month t. t-statistics for CAR
t
is calculated as (CAR
t
^/nt)/^/t var + 2(t l)cov where var is the aver-
age of the cross-section variance over 36 months and cov is the first-order
autocovariance of the AR
t
series. We use Shanghai and Shenzhen A-share
indices as benchmark for Shanghai and Shenzhen IPOs, respectively. Figure
6.1 plots the CAR series for Shanghai and Shenzhen IPOs.
Table 6.16 Average Benchmark-adjusted Returns and Cumulative Benchmark-adjusted
Returns for Shanghai IPOs in December 1990-December 1995
Mont h of Seasoni ng ARt i -st at i st i c CAR
t
i - st at i st i c
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
-0.008
0.012
0.009
-0.008
-0.028
-0.006
-0.012
-0.017
-0.005
0.005
-0.004
-0.009
-0.003
-0.004
0.016
0.006
0.004
-0.007
0.007
-0.008
0.008
-0.002
-0.007
-0.004
-0.012
0.012
-0.002
0.015
0.011
0.018
0.012
-0.002
-0.003
0.001
-0.004
-0.014
-1.13
1.71
1.13
-1.16
-4.46
-0.92
-2.16
-2.48
-0.71
0.81
-0.66
-1.53
-0.49
-0.53
1.93
0.89
0.66
-1.13
1.08
-1.00
1.00
-0.23
-1.18
-0.57
-2.01
1.80
-0.36
2.06
1.25
2.37
1.69
-0.33
-0.54
0.17
-0.53
-2.51
-0.008
0.04
0.013
0.005
-0.023
-0.029
-0.041
-0.058
-0.063
-0.058
-0.062
-0.071
-0.074
-0.078
-0.062
-0.056
-0.052
-0.059
-0.052
-0.060
-0.052
-0.054
-0.061
-0.065
-0.077
-0.065
-0.067
-0.052
-0.041
-0.023
-0.011
-0.013
-0.016
-0.015
-0.019
-0.033
-0.81
0.36
0.95
0.34
-1.37
-1.61
-2.13
-2.84
-2.92
-2.57
-2.63
-2.92
-2.93
-2.96
-2.29
-1.99
-1.78
-1.97
-1.68
-1.91
-1.60
-1.61
-1.80
-1.86
-2.17
-1.80
-1.83
-1.38
-1.07
-0.58
-0.26
-0.32
-0.40
-0.37
-0.45
-0.79
Long-run Performance of IPOs 259
Table 6.17 Average Benchmark-adjusted Returns and Cumulative Benchmark-adjusted
Returns for Shenzhen IPOs in December 1990-December 1995
Mont h of Seasoni ng ARt t - st at i st i c CARt ^-st at i st i c
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
-0.001
0.013
0.014
0.002
-0.027
-0.009
-0.013
-0.018
-0.010
-0.004
-0.004
-0.011
-0.008
-0.008
0.005
0.004
0.005
-0.013
0.003
-0.011
0.007
-0.002
-0.006
-0.002
-0.008
0.010
-0.010
0.014
0.011
0.011
0.012
-0.005
0.000
0.006
0.001
-0.011
-0.19
2.10
1.92
0.29
-5.09
-1.50
-2.57
-3.03
-1.51
-0.82
-0.75
-1.97
-1.32
-1.38
0.69
0.70
0.87
-2.46
0.58
-1.56
0.86
-0.37
-1.00
-0.30
-1.49
1.77
-1.72
2.24
1.30
1.67
2.07
-0.90
-0.09
1.08
0.20
-2.18
-0.001
0.012
0.026
0.028
0.001
-0.008
-0.021
-0.039
-0.049
-0.053
-0.057
-0.068
-0.076
-0.084
-0.079
-0.075
-0.070
-0.083
-0.080
-0.091
-0.084
-0.086
-0.092
-0.094
-0.102
-0.092
-0.102
-0.088
-0.077
-0.066
-0.054
-0.059
-0.059
-0.053
-0.052
-0.063
-0.14
1.16
2.14
2.04
0.02
-0.54
-1.28
-2.21
-2.60
-2.70
-2.78
-3.17
-3.39
-3.62
-3.29
-3.01
-2.73
-3.15
-2.96
-3.29
-2.98
-2.99
-3.12
-3.11
-3.29
-2.91
-3.14
-2.65
-2.29
-1.92
-1.55
-1.68
-1.66
-1.48
-1.42
-1.71
For benchmark-adjusted returns in Shanghai, the CAR exhibits a de-
clining pattern for the first 14 months and continues to be significantly
negative from month 8 to 17. However, the CAR fluctuates over months
16 to 28 with only 4 months having a statistically significant negative value
at the 0.05% level. From months 28 to 37, the CAR recovers and is not
significantly different from zero. For benchmark-adjusted returns in Shen-
zhen, the CAR exhibits a similar pattern. In months 4 and 5, the CAR is
significantly positive and then decreases with the month of seasoning up to
month 15. From months 16 to 26, the CAR fluctuates, showing a slightly
declining pattern. The CAR is significantly negative from months 9 to
30. However, the CAR increases and is only insignificantly negative from
months 31 to 37.
The results from the CAR indicate that while the after-market returns
on the Chinese IPOs severely underperform the market in the first and
second year of seasoning, they on average exhibit less abnormal performance
for the longer-term period. The underperformance of IPOs appears to
indicate that corrections to overpricing at the offerings are taking place.
However, the amount of underperformance (the lowest CARs of -0.077 and
-0.102 for Shanghai and Shenzhen A-shares, respectively) is considerably
small compared to the high initial return observed in the Chinese data.
However, our results could arise simply because we ignore the returns on
the first partial-month, which are considerably negative in value.
We now turn to the cross-section analysis of long-run performance of
Chinese IPOs, using wealth relatives exclusive of the first partial-month
returns. Table 6.18 shows wealth relatives by initial returns.
The results for the wealth relatives by initial return indicate that only
the IPOs in the fourth quartile of initial returns underperform the benchmark-
adjusted returns for the three-year holding period. This seems to be con-
sistent with the hypothesis that only extremely high initial returns at the
IPO dates may be associated with fads or overreactions by investors.
6.9 Summar y
In this chapter, we have empirically identified some causes of the cross-
sectional differences in underpricing of Chinese IPOs using data compiled
for 308 firm-commitment A-share IPOs and 57 B-share IPOs. We first for-
mulate and estimate a benchmark empirical model that relates IPO initial
S
u
m
m
a
r
y

2
6
1

C
O

C
O

O

Q
_

<
D

C
O

o
>

U

C
O

c
:

4
1

X
3

0
)

4
<

C
O

Z
3

T
?

C
O

J
*

i
_

C
O

E

_
s
z

C
J

Z

<
D

D
O

>

'4
J

=
3

E

=
3

U

C
O

C
M

r
-
-
C
N

i
n

C
N

c
n

C
N

v

C
M

C
D

O
)

1
-
c

C
J

l
/
J

m

m

(
/
)

M

C
J

r

4
-
J

O

u
n

C
O

0
3

Q

a

o

c

a

<

1

c

3

O

&

O

a

<

3

O

C
O

a
)

L
i
.

C
O

C
D

o

o

o

o

o

o

5
1
"

C
N

O

C
N

^
T

C
O

G
O

0
0
0
0
0
0
a

o

o

o

o

C
N

b
o

s

y
v
o

T
h
e

U
n
d
e
r
p
r
i
c
i
n
g

o
f

I
n
i
t
i
a
l

P
u
b
l
i
c

O
f
f
e
r
i
n
g
s

s

C
D

C
f
l

.
>
,

o

*

f
e

0
)

-
C
D

J
=
S

-
a

-
W

N

Weal
(Shen
_
,

s

C
O

c
o

j
z

i
i

c
u

.
r
t

u

n
3

1
3

b
O

Weal
(Shan
e

t
-
t

3

c
u

s

3

-
^
>

.
5

i
>
-
c
o

i

C
M

i
n

C
M

C
M

c
o

T
-
H

C
M

t
>

c
o

i

C
M

m

C
M

C
M

C
O

T
1

C
M

T
3

_
o

'
C

v
w

C
D

o

a

m

b
O

-
^

-
^
>

.
3

T
3

1
1

.937
o

T
-
H

C
O

C
D
,942
o

,934
o

,945
o

,945
,070
i
.

,124
i
I

o

p

1
1

236
T
t
.337
i
l

.091
,902
o

,894
o

,952
o

.953
o

.915
o

928
.902
o

,009
i
l
.983
o

.053
i
1

.976
o

.978
C
M

T
t

o

.708
o

.826
o

,619
o

.657
o

.788
T
1

t
-
o

>
#

r
-
H

O

C
M

7
i

T
-
H

C
O

C
M

C
M

C
D

a
>

^

c
o

C
M

0
0

c
o

a
s

o

o

o

o

:
i

^

a

C
O

o

C
M

t
-
H

<

_
g

T
3

C
D

e
n

3

^

<

u

a
3

3

-
^

C
O

T
-
H

c
d

3

a
*

X
)

c

C
M

a
3

3

c
r

T
3

S
H

C
O

3

o
1

J
3

-
*
J

"
*

Summary 263
returns to variables widely used in studies of IPO underpricing. Applying
this model to A-share IPO data, we find that:
(1) IPO underpricing was largest at the earliest stage of development
of stock markets in China.
(2) Available information about a firm at the time of IPO is not related
to IPO underpricing.
(3) Time elapsed between the announcement of an IPO and the first
market trading day is not related to IPO underpricing.
(4) Durable and non-durable manufacturing industries exhibit higher
IPO initial returns than other industries.
We also find that a framework in which IPO underpricing is an equi-
librium outcome in a world of uncertainty and information asymmetry is
a useful paradigm for understanding the Chinese equity offering process.
Underpricing arises to compensate uninformed investors for the "lemon's
problem" they face. The extent of insider ownership, IPO size, the content
of prospectus, pre-IPO leverage, pre-IPO market conditions, share alloca-
tion mechanism, listing exchange, listing time, and the belonging industry
are important variables that correlate with ex ante uncertainty among in-
vestors, and that there is a positive relation between ex ante uncertainty
and expected underpricing. Consistent with the adverse-selection models,
underpricing is significantly smaller for issues in which individual investors
know a priori that they face little or no winner's curse problem.
Then we contrast implications of the hypothesis of a separating equilib-
rium where firms signal their quality by retaining shares and underpricing
against the alternative hypothesis of a pooling equilibrium where no sig-
naling schedule exists and "underpricing" is unrelated to IPO returns or
the subsequent pattern of SEO behavior across firms. We find that the
signaling hypothesis explains the pattern of underpricing behavior among
Chinese issuers rather well, although the market-feedback hypothesis has
some explanatory power and cannot be entirely rejected. On theoretical
grounds, we reject the use of alternative IPO-share allocation mechanisms
as causes of the observed underpricing behavior. In addition, we believe
that while bribery may well help explain the pervasiveness and high level
of IPO underpricing in China, its presence is consistent with a signaling
explanation of the pattern of underpricing.
Our major findings in support of the signaling hypothesis with a sepa-
rating equilibrium are:
(1) The correlation between the degree of IPO underpricing and initial
offer price for the proportion of the firm offered to the public is
negative, given the issuer's retained ownership.
(2) The degree of IPO underpricing is positively related to proxies for
the issuer's intrinsic value, the variance of future returns, and the
issuer's fractional ownership.
(3) Issuers with larger IPO underpricing are more likely to raise larger
amounts of capital through SEOs and to do so more quickly than
issuers with a smaller degree of IPO underpricing, although the
latter relationship is weak.
When we apply our approach to the B-share data, we find that:
(1) The underpricing of B-share IPOs is on average much smaller than
that of A-shares. The difference in IPO underpricing among A- and
B-shares can evidently be explained by the differences in domestic
and foreign investors' investment opportunities.
(2) Traditional signaling equilibrium models for IPO underpricing do
not explain differences in B-share IPO initial returns. Instead, past
information about the issuers have better explanatory power. We
take this as evidence that international investors in Chinese equi-
ties rely more on prior acquisition of information than do Chinese
domestic investors. Perhaps this is because the current sample of
international investors in Chinese equities is heavily weighted with
experienced, professional investors, while the vast majority of Chi-
nese investors in A-shares are inexperienced in equity markets.
We also examine the over-reaction hypothesis by analyzing the long-run
performance of IPOs. For the average IPOs, our empirical results indicate
that long-run underperformance of IPOs is consistent with cross-section
implications of overreaction hypothesis.
Despite our faith in the hypothesis that underpricing of Chinese IPOs
can be explained in terms of a separating equilibrium under asymmetric
equilibrium, we still face the question, why is the degree of underpricing in
China so very large? We note that the mean IPO return was over 5700% for
offerings made before 1991 (when formal stock markets commenced), about
133% in 1991, and it declined steadily to "only" 69% by 1995. We believe
that three factors help explain the extraordinarily high (and declining) level
of underpricing: (1) the initial dearth of financial instruments others than
Summary
265
savings accounts for Chinese investors, compared to other countries with
lower IPO returns; (2) the large degree of uncertainty and information
asymmetry inherent in China's transition to a market economy with modern
financial markets and the initial lack of sophistication of domestic Chinese
investors; and (3) the very long time between IPO dates and the first day of
market trading for IPOs issued before the inception of formal stock markets.
Obviously, our "casual" empiricism on this fascinating and important topic
must eventually be bolstered by rigorous empirical analysis that is beyond
the scope of the present paper. A rigorous test of the hypothesis that the
differences in underpricing between two classes of shares can be explained by
differences in investors' investment opportunity sets and differences in their
methods of formulating expected returns on investments requires explicit
modeling of firms' share supply decisions and investors' portfolio choices.
Moreover, we have not considered possible rent-seeking behavior by the
government or individuals who have access to the limited quota during the
new-issue and offering process. These problems await further research.
Chapter 7
Corporat e Governance and Post -IPO
Financing
The Chinese government, wishing to avoid the political and economic tur-
moil that accompanied the mass privatizations of the former Soviet Union
and other Eastern European governments, has chosen as a cornerstone of its
political survival the commercialization and partial privatization of claims
over profits and assets of its state-owned enterprises (SOEs). In general,
government is reluctant to initiate deep institutional reform and is loath to
surrender its control over enterprises, because doing so increases the cost
of maintaining political support in the form of attaining worker satisfac-
tion and a "quiet" population. At the same time, the inefficiencies and
waste inherent in socialist economies in practice limit government's ability
to maintain its base without resorting to totalitarian controls that have
proven impossible to sustain in a few societies. Therefore, privatization
of SOEs has been adopted only because it is the only viable means to as-
sure the survival of the existing political order, which in the case of China,
means the survival of the Chinese Communist Party. The government, by
reforming the agricultural industry, restructuring SOEs via partial privati-
zation, and encouraging rapid growth of Township and Village Enterprises,
seeks to obtain the benefits of a market economy for mobilization of pri-
vate savings, economic growth and an improved standard of living, while
retaining political control. In this chapter, we analyze post-IPO financing
choices of newly privatized SOEs as they reflect conflicts between govern-
ment's desire to retain political control, private investors' desire for return
on their investment, and management's desire to appropriate resources for
its own benefit. We thus hope to shed some light on the costs and benefits
of partial privatization.
267
268
Corporate Governance and Post-IPO Financing
7.1 Shareholders' Behavior and Corporate Governance
One of the main goals of China's enterprise reforms is to clarify property
rights and delineate the roles of the state and of the enterprises. In 1992,
a new regulation gave managers of SOEs "Fourteen Autonomous Manage-
ment Rights", including the right to set prices, the right to hire and fire
workers, and so on. Implementation was uneven, however, with few indus-
trial companies enjoying all 14 rights. The transformation of SOEs into
commercially oriented companies was proceeding slowly.
Under China's company law, an enterprise's investor must be identi-
fied in the corporate articles of association, a daunting task. It is not a
question simply of tracing funds but, rather, of assigning property rights.
The competing claimants various government departments and agencies of-
ten cannot reach consensus on who the investor is to be. This dilemma
is compounded inasmuch as most SOEs have large liabilities. Everybody
wants the assets, but nobody wants the liabilities. With the unclear as-
signment of property rights, there is an asymmetry in the allocation of
rights and obligations for good and bad assets. In effect, while assets are
privatized, liabilities are socialized. Property rights are fuzzy because gov-
ernment departments and agencies exercise the roles of both shareholder
and administrator. Indeed, multiple government departments exercise the
shareholder role. Across agencies both horizontally and vertically there is
fragmentation and partial exercise of the ownership function, with no single
entity responsible for the enterprise's bottom line. Consequently, managers
enjoy more autonomy than is officially sanctioned, and, without proper
checks and balances, are able to engage in opportunistic behavior.
In 1995, China initiated the "Modern Enterprise System" which aimed
to transform the SOEs into modern shareholding corporations. However,
the profitability and operational efficiency of SOEs have not improved by
the introduction of the system, partly because of the lack of effective mon-
itoring system of management. The state, as an owner of the SOEs, has
failed to adequately supervise management because of high costs associated
with checking the performance of management. Under such circumstance,
managers of the SOEs are able to operate their enterprises according to
their own interests without being disciplined by owners. The management
without owners' supervision inevitably gives rise to inefficiency. Therefore,
it is significant for the SOEs to construct effective corporate governance
structure for monitoring management and improving their efficiency.
Shareholders' Behavior and Corporate Governance 269
7.1.1 Organizational Structure and Board Composition
The primary actors of the modern corporation are shareholders, the board
of directors, and managers. These three entities can be found in compa-
nies in developed countries, although variant forms of organization exist in
some countries. Shareholders have all the rights related to property rights.
These rights include right to make decisions on production, right to appoint
managers, and right to designate investment beneficiaries. Shareholders
express their interests through annual meetings which have the supreme
control of a company. The board of directors is the top decision-making
organ in a company and supervises management on behalf of shareholders.
The main responsibility of the board is strategic planning in management
decision-making process, such as merger and acquisition, and appointment
and dismissal of executives. Managers conduct daily business of the com-
pany under the instruction and supervision of the board.
Organizational changes have been the most visible aspects of China's
enterprise reform process. In November 1993, the 14th Central Commu-
nist Party of China issued The Decision on Issues Concerning the Estab-
lishment of a Socialist Market Economic Structure, which emphasized the
importance of introducing the modern corporate system to the SOEs. In
the following year, the State Council announced a nation-wide enterprise
experiment based on carefully selected 100 large and medium industrial-
ized firms was conducted by the State Council. The primary objective
of the experiment is to transform the SOEs into modern corporations by
clearly defining ownership rights and establishing efficient corporate gover-
nance structure. The corporatized SOEs are required to set up the board
of directors representing owners' interests, to produce separate corporate
entities that supervise the management of each enterprise, and to create the
supervisory board. An important aspect of effective corporate governance
system is the separation of government administrative functions from the
commercial functions of the SOEs. Therefore, members of the board usu-
ally include outside specialists of business management. This pilot program
was formally implemented under the Company Law in early 1995.
With the establishment of a network of state asset management insti-
tutions and group companies, the State Council, China's cabinet, acts as
the ultimate owner of SOEs at the national level on behalf of the Chinese
people, with the National Administrative Bureau of State-Owned Property
acting as the agent. Similar upper-tier bodies exist in provinces and cities.
270 Corporate Governance and Post-IPO Financing
An intermediate tier is composed of provincial- and municipal-level holding
enterprises and state asset management companies. The SOEs are merely
the third tier.
In the newly formed organizational structure, however, the lines of au-
thority are unclear, and the boards of directors and the senior executives
are often the same people. In addition, board members are nominated not
by commercially oriented owners or their representatives, but by govern-
ment or party bodies. With few exceptions, sector bureaus and ministries,
as well as large enterprises, have been transformed into new entities, but
the corporate structure has not been modernized. Many managers see little
difference between the old line bureaus and the new entities other than a
name change. Not surprisingly, the multi-tiered network is burdened with
conflicts of interest that prevent the separation of business from govern-
ment.
In addition, the shareholders congress is required to obtain permission
for their decisions from its upper authorities. The general manager of a
shareholding company is usually the director of the board. The chairman
of the board is appointed by the government or Party officials rather than
by members of the board. Members of the board and the supervisory com-
mittee in corporatized SOEs are dominated by government officials or their
delegates, and the state shareholders are over-represented in proportion
to individual shareholders. Such corporate governance practices inevitably
lead to strong government intervention in business decision-making.
7.1.2 The Role of Gov ernment
In western countries, governments influence enterprise operation mainly
through monetary and fiscal policies and regulations. In Japan and other
newly industrialized countries, the governments' intervene into targeted
sectors using preferential policies such as such as interest subsidies, price
subsidies, tax breaks, informational assistance, and regulations concerning
technical and quality standards, to name a few. Although situations very
across countries, the general principle is that there is a limit of government
intervention so that firms can be held accountable for their own losses
and profits. Excessive government intervention may cause serious incentive
problems that reduce the firms' (including the board and the management)
efforts in enhancing performance.
The Chinese SOEs still face interventions from various levels of gov-
ernment, especially from local government agencies. Many expect that the
shareholding system can solve the problem of government interference in
China. However, evidence from the experimental stock companies suggests
that very few companies have achieved this goal. First, the ownership of
state property is, in many cases, represented by the industrial bureaus,
which, at the same time, oversee branches of industry and issue regula-
tions. Second, it is not always the case that the industrial bureaus initiate
the intervention. Some firms tend to rely on industrial bureaus for various
preferential treatments regarding credits, materials and subsidies.
The 1992 regulation regarding macro-management of shareholding com-
panies states that the government should gradually reduce its mandatory
production plan and wage control, and eliminate investment project ap-
proval requirement. This reform is an important breakthrough in the di-
rection of separating government administration from enterprise manage-
ment. The dilemma is that, however, while the shareholding companies'
irrational behaviors are not corrected by some incentive mechanism, many
government restrictions are not likely to be removed in the near future. For
example, if the companies tend to over- provide benefits to the employees,
the government will have to control wages and bonuses. But in the presence
of such restrictions, the firms are likely not to be responsible for their own
performances.
Table 7.1 shows changes in proportions of the central government's
projects and local projects in the total capital construction investment of
state-owned sector. Investment decisions of central projects are generally
made by the line ministries and the State Planning Commision, and that
of local projects are made by local governments and enterprises. It can
be seen that after 1992 the local project investments exceeded the central
government projects investments. Obviously investment decisions of SOEs
are now shared between governments at various level and enterprises.
Paralleling with the down-shifting of investment decisions, financial
decisions have been gradually separated from investment decisions, and
sources of investment funds have been dispersed. Roughly speaking, state
bank loans have taken over the state-budget funds as a dominant finan-
cial source of investment of SOEs, regardless of investment decisions being
made by governments or enterprises. The retained profits and non-bank
debts are also used by SOEs to finance their investment and operation.
The state budget funds now account only for the tiny part of the total
investment. Furthermore, even these state budget funds have been taken
Table 7.1 Changes of Proportions of Central Projects and Local Projects in Capital
Construction Investment of the State Sector (units: 100 million yuan)
Year Total Investment Central Projects Local Projects
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1074.37
1176.11
1343.10
1574.31
1551.74
1703.81
2115.80
3012.65
4615.50
6436.74
7403.62
8610.84
575.24
632.52
761.66
873.71
837.71
919.15
1060.40
1341.69
1834.90
2430.75
2970.67
3379.33
499.13
543.59
581.44
700.60
714.03
784.67
1055.37
1670.96
2780.60
4005.99
4432.95
5231.50
Source: China Statistical Yearbook 1997
as "loan-for-grant" channeled from t he government t o enterprises t hr ough
the state banks rather than directly appropriated to enterprises (that is,
in SOE's balance sheet , they are recorded as special debts rather than
equity).
Table 7.2 and Figure 7.1 present changes in financial sources of fixed
asset investment of the state sector. From the table, we see that the pro-
portion of state budget funds declined from 79% in 1979 to only 4.6% in
1995; domestic loans increased from almost zero in 1979 to 16.2% in 1982,
to 23.6% in 1996. We should point out that most of foreign funds were in-
jected as investment loans into enterprises rather than as the government-
equity investment. About half of the funds come from bank loans, and only
roughly equal part is from retained profits, retained depreciation funds,
extra-budget funds of local governments and other sources (such as bonds
and shares issued to employees and other institutions). It is typical that
enterprises finance their long term fixed investment from short-term bank
loans. If we assume 50% of fund-raising and others were bank loans, the
total bank loans would account for about 56% of financial sources of fixed
investments in 1996.
Based on the above discussion, we can conclude that corporate finance
S
h
a
r
e
h
o
l
d
e
r
s
'

B
e
h
a
v
i
o
r

a
n
d

C
o
r
p
o
r
a
t
e

G
o
v
e
r
n
a
n
c
e

o

i
H

^

-
J

r
i

9

p
,

O

W

4
1

V
I

L

.
,
4

<
&

e

a

u

+
J

o

p

m
+

r
>

(
1
0

o

t
-
o

t
o

o

i
n

o

o
"

i

o

C
O

o

C
M

o

c

*
1

1979
D

i
n

C
n

e
n

C
O

e
n

e
n

e
n

e
n

C
O

e
n

t
-
c
o

e
n

m

C
O

e
n

C
O

C
O

e
n

C
O

e
n

Table 7.2 Financial Sources of Fixed Asset Investment (%)
Year
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
State
Appropriation
79.0
52.0
48.5
36.6
40.6
39.0
24.0
22.2
20.5
14.7
13.4
13.2
10.2
6.3
6.0
5.0
5.0
4.6
Domestic
Loan
16.2
16.1
15.4
23.0
22.7
24.6
24.2
20.9
23.6
28.1
30.4
25.4
25.7
23.7
23.6
Foreign
Funds
1.9
1.7
2.2
5.3
6.5
9.0
9.0
10.2
9.1
8.3
8.0
6.1
7.2
7.9
6.7
Fund-raising
45.3
43.4
39.2
40.4
38.4
47.6
40.5
42.8
42.1
43.1
46.6
48.0
51.0
48.7
50.9
others
4.2
7.3
10.2
11.7
12.8
11.9
10.3
8.7
14.5
11.6
15.9
14.4
Source: China Statistical Yearbook 1983, 1991, 1994, 1996 and 1997
of Chinese SOEs has fundamentally shifted from the state-budget finance to
the state bank loan dominant finance. The role of government in corporate
financing decisions has diminished over the years.
7.1.3 Ownership Concentration and Firms' Performance
The privatization process in China operates under the Cooperative Share-
holding System (CSS) characterized by separation of ownership and control.
Ownership of an SOE's assets are initially distributed among the govern-
ment, other state enterprises, managers, employees, and outside domestic
and foreign investors. Effective control rights are assigned to management,
which generally has a very small, or nonexistent, ownership stake. Separa-
tion of ownership and control creates conflict of interest between manage-
ment and newly enfranchised private investors that is the root cause of the
Table 7.3 Industrial Out put of State-Owned and Non-State Owned Enterprises (%)
1992 1993 1994 1995 1996 1997 1998
State-owned
enterprises 51.5 43.1 35.3
Collectively-owned
Individually-owned
Other types of
ownership 7.6 6.0 4.0
Source:China Statistical Year Book(1999)
principal-agent problem in traditional corporate finance theory (Jensen and
Meckling, 1976). Moreover, because government desires to retain some con-
trol (in part through partial retained ownership) of commercialized SOEs,
further conflicts arise between politicians and firms (Shleifer and Vishny,
1994).
Although various reform programs have been carried out since 1979,
such as decentralization of decision-making on management and enhance-
ment of workers incentives to work, economic performance of the SOEs has
not improved as much as that of collective or individual owned enterprises.
As Table 7.3 shows, the share of SOEs in total industrial output has been
declining, while that of non SOEs have been increasing. As Table 7.4 shows,
the ratio of increase in total factor productivity is larger in the non-state
sector than in the state sector. Responding to this fact, various discus-
sions about the causes of the SOEs' inefficiency arose, and these arguments
can be roughly divided into two schools of thought: ownership focused and
market focused ones.
The school of thought that focuses on ownership blames an unclear def-
inition of ownership and the separation between ownership and control for
the SOEs' inefficiency. According to them, the state, as an owner of the
SOEs, has failed to monitor management effectively because of high super-
vision costs, and insufficient supervision has allowed managers to pursue
their private objectives at the expense of owners' interests such as maxi-
mizing the profit of enterprises. As policy makers try to solve the efficiency
31.1
55.9
12.9
29.4
60.3
10.3
26.4
62.6
11.0
25.7
63.0
11.3
Table 7.4 Estimated Rates of Annual Productivity Growth in Chinese Industry (%)
1980-1984 1984-1988 1988-1992
Total factor productivity
State sector 1.8 3.0 2.5
Urban and township 3.4 5.9 4.9
Village and private 7.3 6.6 6.9
Labor Productivity
State sector 3.8 6.2 4.7
Urban and township 8.6 7.0 13.8
Source: Jefferson and Rawski (1994)
probl em embedded in t he SOEs, t hi s school of t hought advocat es t he in-
t roduct i on of t he shareholding syst em t hat permi t s t he SOEs t o sell t hei r
stocks t o employees or t o individual investors. They assert t hat t he intro-
duct i on of t he shareholding syst em improves t he efficiency of SOEs. They
argue t hat selling shares t o t he public gives workers consciousness as owners
of their enterprises and make t hem care more about t he profit of SOEs.
On t he ot her hand, t he second school of t hought claims t hat creating a
well-functioned market economy is more i mpor t ant t han reforming owner-
ship st ruct ures in improving t he performance of SOEs. According t o t hem,
poor performance of SOEs is due t o t he lack of compet i t i ve market en-
vi ronment rat her t han an ambiguous definition of ownership. In cont rast
t o t he assertion t hat separat i on of ownership and management is t he root
cause for t he inefficiency of SOEs, t hey argue t hat conflicts t hat arise from
separat i on of ownership and management have existed ever since t he ap-
pearance of moder n corporat i ons and t hat t he ownership st r uct ur e is not
t he focal point of SOEs' efficiency problem. Thi s school of t hought asserts
t hat price di st ort i on, soft-budget constraint, and high costs for moni t ori ng
management i mpede t he performance of SOEs, and t hat such conditions
are generat ed by an i mmat ur e market system. According t o t hem, t he in-
t r oduct i on of t he shareholding syst em would not improve t he efficiency of
SOEs wi t hout existence of a fair market mechanism. Therefore, establishing
a well-functioned market syst em is a prerequisite for an effective sharehold-
ing syst em. Some i mport ant aspect s of SOE reform include establishing
well-functioned financial and manageri al labor market s, hard-budget con-
Political Costs and Agency Costs of Equity Financing 277
straints, and reducing government interventions in SOEs' decision-making
process.
7.2 Political Costs and Agency Costs of Equity Financing
In this section, we discuss how earlier attempts to commercialize SOEs
without privatizing them created adverse incentives that defeated the in-
tent to increase profitability and reduce the need for continual government
subsides. This failure led to the establishment of the CSS, with its polit-
ical costs and agency costs of equity financing. We show that given the
conditions of China's SOE reform, debt-financed buyouts by existing man-
agement is not a viable choice for imposing a "hard budget constraint"
on management. The inability for Chinese SOEs to use debt to finance
profitable investments also reduces management's ability to send favorable
signals to the market under the asymmetric information paradigm appli-
cable to mature market economies where debt conveys insiders' superior
knowledge of firm's investment opportunities.
In traditional corporate finance theory, private firms are viewed as a
nexus of contracts between various economic entities, particularly owners
(principals) and managers (agents). In the absence of complete and fully
enforceable contracts, principal-agent problems arise because owners want
to maximize firm value while managers maximize their own utility, which in
general reduces the value of the firm. When government is also an actor, the
traditional agency approach will not be the whole story since government
is interested in its survival, which may require the use of resources that
might otherwise contribute to management's utility or private investors'
wealth (firm value). These interrelationships are complicated by the fact
that government is by no means monolithic and constitutes a collection
of agents with conflicting goals: politicians, civil employees, and citizens
themselves.
In studying firms' post-IPO financing decisions, we look for evidence
pertaining to the rationale behind the Chinese government's reform strat-
egy. We hope to shed light on complex games played by the government,
management, and outside investors. We attempt to assess whether and to
what extent government's political concerns and management's self-dealing
behavior influence the choice of financing methods and on firm value (po-
litical costs of government intervention and agency costs of managerial dis-
cretion).
7.2.1 The Failure of Management Responsibility Contract
System
The reform of SOEs in China has been characterized by efforts to decentral-
ize, commercialize, and of necessity, to separate ownership and (effective)
control. It began with the Management Responsibility Contract System
(MRCS) in 1987 in which government transferred management authority
to the enterprises and allowed them to retain some of their profits.
1
Under
the MRCS, managers of SOEs were given specific control rights in produc-
tion, investment, sales, profits, personnel management, and distribution of
fringe benefits via contracts, therefore managers have partial incentives to
generate cash flows and expand wages in the form of bonuses and fringe
benefits. However, the contractual relationship between managers and the
government was asymmetric and incomplete. A critical defect was that
profit retention by management had no downside. The state, in fact, re-
mained responsible for final losses, mainly through the state-owned banks,
which were not contractual parties in the enterprise-commercialization pro-
cess. Employee bonuses and welfare expenditures, including housing, med-
ical and schooling expenses and other fringe benefits, were deductable from
enterprise revenue in calculating profits. Therefore, it is not difficult to
see why profit maximization was not management's primary objective and
why SOEs continued to function as government agencies or social security
institutions providing "cradle-to-grave" services to employees rather than
as even approximately profit-maximizing businesses.
Grossman and Hart (1986) argue that residual rights of control, not
specific rights via contracts, are critical determinants of managerial incen-
tives. Shleifer and Vishny (1994) also argue that commercialization without
allocation of residual rights to management is unlikely to guarantee more
productive or profitable uses of resources. Moreover, when government
can indirectly subsidize management to maintain bloated employment lev-
els, employee housing, and other political objectives through control over
transfers from the treasury, thus creating a soft budget constraint, privati-
zation alone is unlikely to enhance productivity. Under MRCS, managers
1
SOEs had to operate at state-controlled prices and remit all profits to the state before
MRCS.
had little residual rights of control and were subject to strong political in-
fluences to maintain SOE's social responsibilities. Consequently, they have
little interest in technology investment, profit maximization and efficient
resource allocation, all of which inevitably led to high political costs of
government control.
Groves, Hong, McMillan and Naughton (1994) find that MRCS strength-
ened workers' incentives, increased productivity and workers' compensa-
tion, but did not lower government subsidies nor increase overall profits.
With the adverse incentive structure as described above, SOEs were able
to rely on easy bank credit to pay production and other "operating" ex-
penses. Thus, as of 1994 over 40% of SOEs were unprofitable, and while
SOEs accounted for 34% of GDP, they absorbed three quarters of domestic
credit. Their ever increasing triangular debt (unpaid bills between state en-
terprises, state banks, and the government) had accumulated to over 4900
billion Yuan (or 95% of GDP), and bad debt was estimated to be as much
as 25% of state bank assets (China Business Review, January 1996). Many
troubled SOEs were unable to pay back these accrued loans, thus threat-
ening the collapse of China's banking system. Figure 7.2 illustrates the
distribution of sources of debt by the end of 1994. Figure 7.3 presents the
percentage of SOEs with independent accounting operating at losses and
the amount of losses during a period from 1978 to 1997. MRCS did not
achieve the goal of effectively reforming SOEs and was terminated nation-
wide in 1994. The process of partial privatization through the creation of
joint-stock companies (corporatization) began through the establishment
of Cooperative Shareholding System (CSS).
7.2.2 Cooperative Shareholding System
Under the CSS, there are five types of shares. Those shares not retained
by government, state enterprises, managers, and employees, are transferred
to outside investors through initial public offerings (IPOs) and seasoned
equity offerings (SEOs).
2
Each share type is entitled to the same voting
2
Among the five share types, state shares are retained by the State Assets Management
Bureau (SAMB) for central government or local government and are not tradable. Legal
person shares are held by other SOEs and non-bank financial institutions (including
investment companies, finance corporations, and mutual funds) and can not be traded
in the two official exchanges. A limited number of legal person shares are traded on
t he Security Trading and Automatic Quote System (STAQS) and National Electronic
Trading System (NETS). Employee shares are those sold to managers and workers
2
8
0

C
o
r
p
o
r
a
t
e

G
o
v
e
r
n
a
n
c
e

a
n
d

P
o
s
t
-
I
P
O

F
i
n
a
n
c
i
n
g

3

Q

3

O

C
O

I

P
o
l
i
t
i
c
a
l

C
o
s
t
s

a
n
d

A
g
e
n
c
y

C
o
s
t
s

o
f

E
q
u
i
t
y

F
i
n
a
n
c
i
n
g

2
8
1

i
n

^
*

O
l

r
o

C
O

C
O

t
>
-
C
M

i
1

<
N
J

I
T
)

-
H

w

o

i
n

<
H

o

(
D

t
J
)

<
f
l

H

C

I
D

U

>
J

0
1

P
-
.

i

W

0
}

W

W

O

i
-
4

<
H

O

+
i

o

4

<

y

1
1

*

1

1

*

:

4
i

\
.

4

j
'
4
.

.

4

'
"
'

.

'

>

*

*

"
%
\_

1

,
^

'

1

4

4

^
.

*
*
J
*

*
,

T
*
'
^

V
*

A

*

1

1
J

f

1

1

4

4

/

4

\

- _

_

.

>

{

\

)

{

V

i

t
o

o
n

a
*

t
M

C
J
1

O

t
X
J

C
O

C
O

O
l

C
M

C
O

C
f
l

o

o
o

C
O

O
l

o

o

t
>

O

O

*
3
"

O

o

C
-
J

o

o

o

o

o

C
O

o

o

I
D

o

o

*
r

o

o

<
M

(
i
r
e
u
j

T
i
o
i
^
I
i
i
y

0
0
1
)

s
s
s
s
o
q

j
o

i
i
i
n
o
u
i
Y

rights and dividend. After corporatization, all shareholders have residual
claims on the company's assets. Boards of directors are established to
represent the interests of all shareholders. A major characteristic of Chinese
joint-stock companies is that share ownership is dominated by the state or
legal person shareholders. The State Assets Management Bureau (SAMB)
retains majority ownership for about one-third of the listed firms and legal
person shareholders have controlling shares for the other two thirds.
To facilitate the privatization process and to wean doomed firms from
eternal subsidies, the government set up an annual reserve fund of 20 billion
Yuan in 1994 to support mergers and bankruptcies, to train surplus workers
for new jobs, to free enterprises from the responsibility to provide social
services, and thus to promote a system in which enterprise survival depends
only on market-oriented performance. By the end of 1996, 675 SOEs had
declared bankruptcy, 1801 had been merged or downsized, 9200 SOEs had
been corporatized, and of these, 485 of the most profitable had been listed
in the two stock exchanges with a total market capitalization of about 400
billion yuan (equivalent to approximately 50 billion U.S. dollars).
The extent to which SOE partial privatization leads to benefits in the
form of more rapid productivity and GDP growth is limited by the political-
and agency-cost problems. The games played by government, management,
and investors become more and more complex. An exacerbating character-
istic of the Chinese situation is that the vast majority of private shares
are held by small investors. Dispersed public ownership creates a classic
free-rider problem: Small shareholders do not have the incentive to monitor
management because the benefit is enjoyed by all while the cost is totally
borne by a few, active investors.
Two factors that can mitigate agency costs of managerial discretion are
debt (in conjunction with appropriate bankruptcy procedures) and takeover
threats. Jensen (1986) and Stulz (1990) show that debt (with the threat
of bankruptcy) imposes a hard budget constraint on managers and limits
management's control over firm's cash flows. When managers are faced with
choices of reducing empire-building and perquisites or going bankrupt, they
are likely to choose the first option. Therefore, leverage reduces agency costs
within a company and can not be traded within the first 3 years. Domestic individual
shares, or A shares, are those held and traded by private Chinese citizens in the two
official exchanges in China. Foreign shares are those held and traded by foreign investors
in security exchanges in China (B shares), in Hong Kong (H shares) or in NYSE (N
shares).
of managerial discretion. Shleifer and Vishny (1986) show that takeover
threats discipline management in the presence of many small sharehold-
ers. The reason is that if a company is badly managed, then there is an
incentive for someone to acquire a large stake of the company, improve its
performance and profit on the shares purchased. The threat of such action
can persuade management to act in the interest of shareholders.
Unfortunately, because of the political nature of the privatization pro-
cess itself, bank-debt financing of corporate investments is not a viable
means to assure a hard budget constraint in transitional economies such
as China (Shleifer and Vishny, 1994). Bank loans have traditionally been
viewed as grants from the state designed to bail out failing firms. State-
owned banks retain a monopoly in the Chinese banking sector and profit
is not their overriding objective. If political favor is deemed appropriate,
subsidized loans, rescheduling of overdue debt or even outright transfer of
funds can be arranged with SOEs (soft budget constraints). A banking sys-
tem that is so heavily controlled by the government's political objectives
can hardly exert any real influence on management. Moreover, a market for
private, nonbank debt, has yet to be established.
3
There is no active merger
or takeover activity in stock markets to discipline management. Informa-
tion available in the capital markets is insufficient to keep at arm's length of
the managerial decisions. Despite the existence of a bankruptcy law since
1986, the rate of bankruptcies has remained at a very low level.
4
In light of
the above peculiarities, direct control by (large) shareholders through the
board of directors appears to be the only means for shareholders to monitor
managers.
With good corporate governance structure, boards of directors would
consist of delegates of large institutional shareholders and minority in-
vestors, representing all shareholders' interests. Such boards would make
critical investment and re-organization decisions, select top managers, and
provide managers with incentives and compensation appropriate to in-
vestors' interests. However, board members in Chinese joint-stock com-
panies consist mainly of representatives or officials from the SAMB, local
government, and other state enterprises, whose interests are not the same
3
By the end of December 1996, only 12 companies have their corporate bonds listed on
Shanghai and Shenzhen exchanges.
4
For all forms of ownership in China, the rate of bankruptcy is 0.03% in 1996 compared
to 1.1% in the U.S..
as those of outside investors. On average, 90% of the board members are
government officials and delegates of other state enterprises. Moreover,
there is at best an inadequately functioning managerial labor market that
can discipline managers and help solve incentive problems caused by the
separation of ownership and control. All top managers must be approved
by the Organization Department of the local or central government. The
government maintains ultimate rights for changing managerial positions.
With state controlled stock companies, political considerations still domi-
nate economic performance in the selection of managers and adversely affect
managerial incentives. Managers no doubt care more about carrying out
wishes of the Party and the government, such as avoiding worker layoffs and
maintaining some level of worker social security than about the concerns
of shareholders. As a result, many management decisions still reflect the
old political concerns (political costs). In addition, managerial and board-
member compensation packages are determined by the government, and are
seldom directly linked to profits. Management is not severely constrained
from self-dealing behavior, such as diverting SOE assets outside the state
sector for private use, such as charging meals and travel expenses to com-
pany accounts, obtaining large apartments, and purchasing fancy cars and
furniture for personal consumption. Spending cash flow on acquisitions or
other unprofitable projects undertaken with the aid of unrealistically high
forecasts of future profitability gives management a bigger company to run,
thereby increasing their power and prestige in the community (agency cost).
Allocation of effective control rights to management without clearly de-
fined corporate laws and bankruptcy procedures gives managers a chance
to enrich themselves by such means as undervaluing state assets and spin-
ning off subsidiaries to be controlled by the manager's relatives. In a study
conducted by Kernen (1997), some managers of SOEs admitted to have cre-
ated a number of collective or private enterprises for their relatives using
SOEs' assets. Furthermore, China's embryonic regulatory regime and rudi-
mentary accounting standards provide very little protection for investors'
rights, because they limit boards' ability to monitor abuses of governmen-
tal influence and managerial discretion and thus limit political and agency
costs that diminish the productivity-enhancing effects of privatization and
reduce firm value.
In summary, partial privatization and separation of ownership and con-
trol have transferred residual rights of control from government to managers
Choices of Post-IPO Financing 285
but have not provided effective monitoring mechanisms. The traditional
government goals characterized by the "iron rice bowl" concept still exert
some control through the selection of managers and board members. Man-
agement has considerable leeway to seek its own self interest at the expense
of shareholders. In the next section, we present data bearing on the in-
fluence of agency and political costs revealed in the financing decisions of
Chinese joint-stock companies.
7.3 Choices of Post-IPO Financing
7.3.1 Soft Budget Constraints and The Role of Debt
Corporate governance is a institution governing contractual relationships
among equity-holders, management and creditors. It is structured for solv-
ing the two basic problems inherent within the firm. The first is the man-
agerial incentive problem; that is, how to motivate and discipline manage-
ment, given that managerial behaviour is hard to measure? The second is
the management selection problem; that is, what kind of mechanism can
ensure that only the most entrepreneurial people are employed to fill in the
management position, given that entrepreneurial ability is hard to observe?
For corporate governance to be effective and efficient, the following condi-
tions must hold: First, equity-holders as the owners of the firm must have
adequate incentives to select the highly competent manager and to disci-
pline the manager. Second, the manager must have adequate autonomy as
well as adequate incentives to make business decisions with the business
judgment rule, and is well-motivated to work hard. Third, creditors must
have adequate incentives as well as effective measures to enforce debt con-
tacts through bankrupt procedure. Non of these conditions hold in Chinese
SOEs. The debt finance has evolved into the debt crisis in China because
corporate governance has changed little.
The state bank as the major creditor has failed to enforce the debt con-
tract. In a market economy, debts play an important role in corporate gov-
ernance through bankruptcy threat. China enacted the P.R.C. Bankruptcy
Law in 1986, which became effective in late 1988. Theoretically, when enter-
prises become insolvent, creditors will take over the control, and the threat
of bankruptcy can discipline the management. Nevertheless, this is not a
case in China. Rather, bankruptcy has been widely used by enterprises
and local governments as a way to write off debts instead of disciplining
managers. After bankruptcy is declared, either through reorganization or
through liquidation, most incumbent managers still run the firms. The the
only discernible difference is that considerable debts have been canceled
(and in some cases the enterprises are renamed). Because of this, managers
are more than willing to file for bankruptcy. In contrast, state banks as
dominant creditors, have neither access nor incentive to execute their cred-
itor rights. They have been very passive in dealing with distressed firms.
Typically, when firms default on their debt, creditor banks take passive
actions such as extending the payment period for loans and capitalizing
unpaid interest rather than pursuing their claims through bankruptcy or
other active means. Indeed, very few bankruptcies have been filed by banks.
The government has subsidized the SOEs to help their poor economic
performances, and the large amount of the subsidy has become a major
cause of serious government budget deficit. We examine the behavior of 308
SOEs that went public before January 1996. Over the period from January
1994 through December 1996, 280 of these firms recorded a total of 459
announcements of new shares after their IPOs, including announcements
of 281 stock dividends, 126 rights issues, and 52 underwritten offers.
5
The
sample classification is presented in Table 7.5. In exploring evidence of
agency and political costs in financing the privatization process, we contrast
the agency- and political-cost paradigm discussed in the previous section
with the roles of asymmetric information under the well-known "pecking-
order" paradigm suggested by Myers and Majluf (1984).
Under asymmetric information, insiders (management) convey positive
news to outside investors by means of the financing method chosen, with in-
ternal financing (no new issues or possibly stock dividends) being the high-
est in the pecking order, riskless debt the next highest, followed by equity
issues. In choosing a flotation method, managers who know of profitable in-
vestment projects signal their true value in the choice of a subscription price
in an uninsured rights offer, while managers with lower-quality projects
make fully underwritten issues (Smith, 1977, Heinkel and Schwartz, 1986,
5
We note t hat a Chinese joint-stock company, subject to certain regulations and ap-
proval, can issue equities in the form of A shares (available only to domestic investors
and traded in Chinese exchanges), B shares (available only to foreign investors and
traded in Chinese exchanges), H shares (listed and traded in Hong Kong), N shares
(listed and traded in New York) or a combination of these share types. Because of
the small number of observations associated with the issuance of B-, H-, and N-shares
through alternative flotation methods, we focus only on A-share decisions in this paper.
Table 7.5 Sample Classification
Types of offering
Year Stock Dividend Uninsured Rights Underwritten All
1993 117 52 8 177
1994 94 29 14 137
1995 70 45 30 145
Total 281 126 52 459
and Eckbo and Masulis, 1992). Since new bank debt is likely to be a signal
of bad news in the Chinese situation (and in transitional economies with
soft budget constraints in general), and since markets for private nonbank
debt have not yet emerged, loan financing of new, profitable investment
projects does not appear in the pecking order for newly privatized SOE's.
Consistent with this argument, we observe no corporate debt issues over the
period covered in this study. This leaves internal financing (no new issues
or stock dividends) as most preferred under the pecking-order paradigm in
the current Chinese situation, followed by uninsured rights (being relatively
inexpensive to float but exposing existing owners to the risk of share-price
dilution), followed by underwritten offers.
We compare the financial characteristics of the 28 firms that announce
no post-IPO share issues with those that do issues new shares after their
IPOs. We calculate the average of the following ratios that represent firm's
liquidity, debt capacity, and investment opportunities: cash flow as a frac-
tion of total assets (CASHTA), long-term debt as a fraction of the book
value of total assets (LEVERAGE), market-to-book ratio (MKTBOOK),
and long-term investment expenditure to total asset ratio (LONGINV).
If a firm issues new shares in 1994, we include it in computing the financial
ratios of 1993, and so on and so forth.
Comparing firms with no post-IPO equity issues with all those that en-
gaged in some form of additional equity financing, we note that issuers ex-
hibit more liquidity and evidence of better investment opportunities. Cash
flow as a fraction of total assets is higher for issuer than for non-issuers in
every year from 1993 through 1995, with the gap growing from 1.5 percent-
age points in 1993 to over 4 percentage points in 1995. The market-to-book
ratio, a measure of investors' evaluation of a firm's investment opportunities
Table 7.6 Firm Characteristics by Post-IPO Stock-Issuing Behavior
No stock issue Stock dividend All issues
Mean Median Mean Median Mean Median
CASHTA 1993 0.0614 0.065 0.0773 0.0826 0.0741 0.0786
1994 0.0829 0.0843 0.0939 0.0984 0.1011 0.1139
1995 0.0882 0.0936 0.1004 0.1107 0.1302 0.1381
LEVERAGE 1993 0.0265 0.0194 0.0484 0.0307 0.0343 0.021
1994 0.0361 0.0252 0.0716 0.0593 0.0512 0.0406
1995 0.0453 0.0317 0.0655 0.0502 0.0618 0.0483
MKTBOOK 1993 1.2474 1.1511 1.0907 1.0226 1.3835 1.2642
1994 1.4012 1.3326 1.2806 1.2149 1.5257 1.4699
1995 1.4279 1.395 1.3413 1.2594 1.6576 1.5706
LONGINV 1993 0.0602 0.0571 0.0888 0.0713 0.1193 0.0968
1994 0.0828 0.0693 0.1427 0.1192 0.1545 0.1236
1995 0.0754 0.0648 0.1584 0.1260 0.1493 0.1108
and future profitability, is on average higher for issuers t han for non-issuers,
again exhibiting a wider gap in 1995 t han in 1993. Expendi t ur es on long-
t er m investment projects as a fraction of t ot al assets is nearl y twice as high
in firms t hat issue stock after t hei r I PO t han it is for firms t hat don' t .
Average l ong-t erm debt is very low for firms in all categories, for reasons
discussed in Section 2, al t hough it grows slightly, as a fraction of t ot al as-
sets, over t he period covered, and firms wi t h post - I PO stock issues have
slightly higher averages t han t he ot hers.
Fi r ms t hat issue stock dividends are represented in t he middle column of
Table 7.6. These firms have t he lowest market -t o-book rat i os in each year,
on average, but t hei r mean long-term investment relative t o t ot al assets is
higher t han for non-issuers, almost as high as t he means for all issuers t aken
t oget her. We conjecture t hat t hi s may reflect a dim view of t he market on
t he profitability of t he long-term investments undert aken by an average
stock dividend issuing firm compared t o firms t hat issue uni nsured rights
or underwri t t en offers. We interpret t he comparat i ve st at i st i cs of Table
7.6 as inconsistent with the pecking-order paradigm, because firms relying
predominantly on internal financing after their IPOs appear to have worse
financial characteristics than those that go to the market with new equity
issues.
7.3.2 Costs of Equity
The government as the majority equity-holder has malfunctioned with both
selecting and disciplining management because of the agency problem on
bureaucrats' side. In particular, the government still holds full authority of
appointing management. Because of the adverse selection problem, select-
ing good management is hard work. It requires that selectors must have
adequate incentives to find information about candidates' abilities and to
install high quality candidates. Adverse selection is most serious in China,
because, with no personal stake to signal ability, too many people pre-
tend that they are qualified for management. Worse is that bureaucrats,
unlike their capitalist counterparts, have the right to select, but do not
bear consequences of their selections. This implies that, not only would-be
managers, bureaucrats themselves also have the adverse selection problem.
They have no adequate incentive to search for good managers, and even
if they know some are capable, they still lack the adequate incentive to
install them. Observation suggests that bureaucrats too often base their
selections on personal connections rather than merits. Appointing friendly
managers is the most effective way for bureaucrats' rent seeking. Because
of this, management pool of SOEs is full of lemons.
Although reform has gradually shifted considerable autonomy to man-
agement through various ways, the government still exercises arbitrary ad-
ministrative intervention in the SOEs' businesses. Separation of the gov-
ernment from enterprises has been assumed as the basic purpose of the
reform since the reform was inaugurated. Nevertheless, this purpose has so
far never been materialized. This is true even after the state enterprise is
corporatized in which case the government is formally and legally defined
as the equity-holder and bureaucrats are appointed as board directors. The
fundamental reason for this is that when the state is the owner of the firm,
bureaucrats are delegated with ownership rights to control the firm but
bear no financial risk for their intervention. By definition, any owner has
to supervise management through control rights. The state as a (and in
most cases the only) equity-holder will naturally intervene. The key prob-
lem is how to determine the boundary of such intervention. In the absence
of a very well-defined division of rights between equity-holders, the board
of directors, and management, it is not clear to every party involved who
should do what. For a stockholder or board member who bears financial
risk, his or her decision of whether to intervene depends on how much
trust he or she places on the manager. If a stockholder or board member
trusts the manager, he/she will not intervene even if the manager is doing
something that fundamentally alters the enterprise. If the stockholder or
board member does not trust the manager, even if the manager is doing
something trivial, he/she may still intervene. However, this type of tacit
understanding between a stockholder and manager in dealing with the pub-
lic domain of control rights cannot be duplicated between bureaucrats and
managers of a corporatized SOE. Bureaucrats arbitrarily intervene simply
because they do not bear the consequences of intervention. With arbitrary
administrative intervention, management has little accountability for their
performance. For instance, when a debt-financed investment decision is
made by the government department, how could the manager be blamed
for poor performance or default?
SOEs suffer not only from arbitrary administrative intervention but
also from inadequate managerial incentive. Although the reform has im-
proved the management's incentive to make short profits through profit
sharing, the long-term incentive problem has yet to be solved. Managers
of SOEs prefer to distribute retained profits to employees or make invest-
ment in quick revenue-generating projects rather than to make investment
in long term productivity-enhancing projects and in research and develop-
ment (R&D). They have more incentives to expand their empire by bor-
rowing from banks and to keep insolvent enterprises as going concerns, but
little incentives to make efficient use of borrowed money and to repay. In
many cases, abnormal short-term profits are made at the large expense of
long-term productivity. The reason for management myopia is that, given
that there is little personal capital stake, the manager's enjoyment of ben-
efits from the firm cannot beyond his or her firm tenure, which is mainly
dependent upon bureaucratic preferences and have little relation with firm
performance. With bureaucrats making selection of management, good
performers are just as equally likely to be removed as bad performers, if
not more. Once a firm becomes highly profitable, bureaucrats have every
incentive to collect rents by replacing the incumbent with their favorite.
Thus, the best way for the incumbent to secure his position is to make
the firm not too good and not too bad. This also partly explain why SOE
managers typically underreport their profits.
To examine the extent of agency and political consideration as revealed
in firm's financing choices, we explore in greater depth the market's reac-
tion to various types of post-IPO stock issues. We first compute abnormal
returns over a two-day period that includes the day of the China Secu-
rity Daily announcement and the day preceeding the announcement for
each firm. The abnormal return is calculated based on the market model
(Fama, Fisher, Jensen and Roll, 1969):
ARit = Rit ~ (&i " PiRmt) (7-1)
where ARn is the abnormal return for firm i on date t, Ra is the return
for firm i on date t, R
m
t is the stock-market return on date t, (R/t is
the risk-free rate as measured by the rate of return on six-month Chinese
Treasury Bonds.) on and fa are the ordinary least squares estimates of firm
i's market model using return data from t = 120 to t = 10 and from
t = 10 to t 120 in relation to the announcement day, t = 0. Therefore,
the abnormal return for each firm's stock is calculated as the difference
between the actual and expected returns to its stock.
6
Average abnormal
returns for each day are calculated by
ARt^-YÂRu (7.2)
where iV is the number of stocks with abnormal returns during day t. There-
fore, the equity-issuing announcement two-day average abnormal return is
ADAR = AR-! + AR
0
. (7.3)
The i-statistics are calculated for ADAR by
6
In case of missing returns, we exclude both the day of missing return and the day after
t hat in estimating Qj and /3j.
ADAR
tADAR = I _ _
y ^ k ( E ! = : ( ^ - A%y + EtzllÂRt - A%y)
(7.4)
We find that the mean announcement day abnormal returns (t-statistics)
are-1.47% (-1.6138), 7.04% (4.6178), -9.08% (-4.0597), and-0.43% (-0.2198)
for stock dividends, uninsured rights, a combination of stock dividends and
uninsured rights, and underwritten offers, respectively. We view the prin-
cipal question raised by these statistics to be, what is the negative news
conveyed by stock dividend announcements, compared to announcements
of uninsured rights issues? This question is especially puzzling, because the
Chinese market's negative reaction to stock dividends is opposite to that
in many capital markets, where stock dividends are usually perceived as
conveying positive news. Grinblatt, Masulis, and Titman (1984) interpret
stock dividends as an implicit cash dividend increase in percentage terms,
because firms are loath to cut cash dividends. Thus, only firms confident
about their future cash flows will announce stock dividends, thus leading
to positive announcement day returns. On the face of it, then, the nega-
tive announcement day abnormal returns associated with stock dividends
appear to contradict the asymmetric information paradigm in the Chinese
situation.
7
To enhance our search for the underlying causes for the Chinese stock-
dividend announcement anomaly, we construct Table 7.7, which presents
summary statistics for the abnormal return data and other characteristics
of all firms issuing stocks after their IPOs, categorized by types of issue.
An "eyeball" analysis of Table 7.7 yields comparisons that, to us, are
anomalous in light of asymmetric information, because our abnormal an-
nouncement day return calculations and the statistics reported in Table
7.7 suggest that the most favorable characteristics belong to firms issu-
ing uninsured rights alone. Perhaps their most striking feature is that
CAR360, which is cumulative average return for the year following an an-
nouncement, which in the 120 days before an announcement, is distinctly
7
We conjectured the possibility t hat stock dividends in China represent an at t empt by
management to dilute the degree of government ownership, but since government and
"legal person" shareholders are entitled to and receive dividend t reat ment identical to
t hat of private investors, this explanation of the market' s negative reaction to stock
dividends must be rejected.
C
h
o
i
c
e
s

o
f

P
o
s
t
-
I
P
O

F
i
n
a
n
c
i
n
g

J
3

6
?

S

O

0
0

t
-

(
M

m

o

0
0

l
>

o

.

I
M

i
-
l

S
"
?
^
^
S
S
6
?
S
s
?

o
b

C
M

o

m

c
o

o

"

C
M

<
N

c
o

,
-
i

^
r

C
M

C
M

c
o

l
O

O

b
-
o

t
o

m

*

2

d

^

i
-
H

C
O

0
0

C
O

i
n

t
o

m

o

i
-
i

o
o

C
O

0
0

t
o

^

^

^
H

<

t
?

(
S

K

S
?

K

o

-
r
*

l

t
o

T
t

C
O

C
M

C
O

t
o

f
c
u
O

>

6
?

0
0

O

o

o
o

o

d

t
O

<
M

o

t
o

C
O

C
O

i
n

O

d

d

C
M

t
>

I
O

i
n

i
n

o

C
O

i
-
H

t
o

i
o

(
M

i
-
H

i
n

T
^

i
-
H

~
^

^

C
O

x

r
r
>

r
N
i

0
s
-
o

i
n

o

i
-
H

^
H

O

O

i
-
H

0
0

t
o

t
^

i
-
H

I
O

i
p

I
O

f
:

6

8
?

6
?

$
;

t
o

c
o

c
o

^
H

o

t
o

^

r
t

O
i

C
M

t
o

0
5

I
O

I
O

H

0
0

c
o

t
o

f
;

r
t

i
n

2

6
?

o

S
B

o

C
M

C
O

0
0

o

C
M

C
M

t
D

C
M

I
"
-

i
-
H

I
O

O

t
o

C
O

i
n

C
T
>

C
O

T
-
H

C
M

C
O

>
^
p

-
r
H

-
*
H

6
"
-
H

0
0

3

6
?

0
0

t
o

6
?

6
8

0
0

C
M

0
0

O

C
O

H
O

i
n

t
o

i
n

.
-
H

c
o

0
0

t
o

d

2

v
p

m

o

o

C
O

-
T
*

0
0

o

t
o

c
o

1
0

O

i
n

-
*
*

C
T
>

0
0

S
S
^
S
^
g
f
e
?
^
^
^

C
M

C
M

i
n

t
o

c
o

i
-
H

r
-
H

"
0

S

c
o

C
M

b
-
b
-

f
-
t
o

O

t
O

r
H

C
O

^

T
3

a

-
g

'
>

I

'
S
I

-
*
l

o

I
f

!
D

(
-

^
1

S

C
O

C
M

i
-
H

t
o

o

"
^

0
0

o

<
*

a
;

O

C
O

o
>

t
o

i
n

m

m

o

c
o

o

C
T
>

<
*

0
0

>
#

i
n

o

0
0

o

6
?

f
e
?

o

o

t
o

C
}

c
o

r
-
H

O

C
M

1
>

b
-
0
0

1
>

m

&
q

^

C
O

t
~

T
-
H

i
-
H

*

O

O

0
5

C
N

*

C
T
>

C
M

T
t
<

^
H

C
M

O

T
-
1

O

-
>

f
e
;

-
1

O

f
e
;

O

I
-
J

^

t
t
i

C
O

k
,

O

^

a
,

C
M

S
S

C
M

C
O

O

O

C
M

C
M

I
-
H

r
H

I

+

o

t
o

c
o

+

<
l

0
0

C
O

O

f
e
:

K
|

a
,

O

o
o
o
m

i
o

h

c
o
o
o
6
?
b
?
b
P
i
-
-
>
p
f
c
?
i
P
v
p

C
M

o

positive for the uninsured rights issuers, but not for firms in any of the
other financing categories. Profit per share is also by far the highest for
firms using uninsured rights, roughly 3- to 8- times higher than for firms
using other external financing methods, although this may simply be a
matter of the number of shares outstanding rather than intrinsic profitabil-
ity. However, given that MKTBOOK is also highest for firms issuing only
uninsured rights, it is plausible that these are genuinely more profitable
firms. Another distinguishing feature of firms that issue uninsured rights
alone is that the variable MVE, market value of equity one day for an
announcement, is far larger than for firms using other forms of financing.
This suggests firms issuing uninsured rights are bigger and/or more highly
valued than others. Perhaps sheer size has an informational advantage as
well, in that economies of scale increase the liklihood that investors will
know something about these firms enabling them to make better informed
investment decisions when opportunities to purchase new shares arise. De-
spite their relatively large size, firms issuing uninsured rights have only the
third-highest fraction of government ownership and largest concentration of
management ownership.
8
They have the lowest concentration of ownership
among institutional or private investors.
Firms issuing stock-dividends, which exhibit the lowest announcement-
day abnormal returns, have the lowest CAR in the 360 days following an
offer, and profits per share are much lower. MKTBOOK is only three-
fourths the magnitude as for firms issuing uninsured rights (and roughly
equal to firms issuing stock dividends alone), and market value of equity
just before announcement is by far the lowest for this group. The stock-
dividends and rights issuers have a somewhat higher level of government
ownership and the highest concentration of ownership among the top ten
institutional or private investors.
To summarize, then, firms engaging in some form of internal finance
appear to be "worse" than those seeking new funds from external sources
(uninsured rights only or underwritten offers), although within the group
seeking external funding, the "pecking order" appears to apply, with firms
issuing underwritten offers being apparently less profitable and with fewer
good investment opportunities than those issuing uninsured rights alone.
8
The proportion of shares t hat are government owned must be interpreted with caution,
given the interlocking ownership of government and government-owned firms among
shareholders.
The Information Content of Different Financing Choices 295
7.4 The Information Content of Different Financing Choices
We use the following variables as proxies for agency costs of equity, invest-
ment opportunities, risk of the firm, and subsequent stock return perfor-
mance:
Agency costs of equity: Because managers will be more likely to
strip assets and transfer resources from SOEs to themselves if the
managers' fractional investment is small, we use the percentage of
managerial holding of equities (MGOWN) as one of the proxies of
agency costs of equity. Because agency cost of equity is related
to the effort of monitoring by the large shareholders, we use the
percentage of the top ten shareholders (TOPTEN), including insti-
tutional and private investors but excluding state government, as
another proxy of agency costs of equity.
Political costs of equity: With large government ownership and in-
efficient governmental monitoring, managers are more likely to pur-
sue objectives other than maximizing the firm's value, e.g., man-
agers are more likely to use financing methods that dilute state
ownership and expand their empires. We use the percentage of
state government ownership (GOVNT) as a proxy for the political
costs of equity.
Investment opportunities: We use market-to-book (MKTBOOK)
defined as the ratio of market value of equity plus total assets mi-
nus book value of equity to total assets, cash flow to total asset
ratio (CASHTA), profit per share before the financing announce-
ment (PROFSHA) and long-term investment to total assets ratio
(LONGINV) as proxies.
Risk of the firm: We use a firm's stock market beta (BETA) and
standard deviation of stock returns (STD) calculated over a period
from 120 days before the announcement and 360 days after the
announcement as proxies.
Subsequent stock return performance: We measure a firm's subse-
quent stock performance using the cumulative abnormal returns
120 days after (CAR+120) and 360 days after (CAR+360) the fi-
nancing announcement.
We hypothesize that:
HI: With agency problems, stock dividends and combination of stock
dividends and rights are used to disperse ownership and further
reduce monitoring of management by existing shareholders. The
probability that a firm will use stock dividend is positively related
to the agency and political costs of equity and the risk of the firm,
and negatively related to the investment opportunities and subse-
quent stock price performance.
The alternative to this hypothesis is that issuing new shares in lieu of
dividends conveys information to the market that firms have ample slack
because retained earnings must be reduced by the dollar amount of the
stock dividends. Consequently, there is a positive correlation between the
probability of stock dividend announcement and subsequent stock price
performance (Grinblatt, Masulis and Titman, 1984). Since stock dividend
conveys good news, the probability of announcing stock dividend or a com-
bination of stock dividend and rights is positively related to the investment
opportunities and negatively related to the risk of the firm. There is no sta-
tistical relationship between ownership concentration and the probability
of stock dividends and combination of stock dividends and rights.
H2: The probability that a firm will use uninsured rights and under-
written offers is negatively related to the agency and political cost
of equity and the risk of the firm, and positively related to the
investment opportunities and subsequent stock price performance.
H3: Between the choices of stock dividends and a combination of stock
dividends and rights, the probability that a firm will choose stock
dividend and rights at the same time is more positively related to
the agency and political costs of equity and the risk of the firm, and
more negatively related to the profitability of investment projects
and subsequent stock return performance.
H4: Between the choices of uninsured rights and underwritten offers, the
probability that a firm will employ an underwriter is less negatively
related to the agency and political costs of equity and the risk of the
firm, and less positively related to the profitability of investment
projects and subsequent stock return performance.
To test the aforementioned hypotheses, we estimate the following multi-
nomial logit model (MNL). Let the value to firm i choosing financing
method j be y*', where j {stock dividend, rights, underwritten offers}.
We assume that y* depends upon some observed characteristics X^ and
The Information Content of Different Financing Choices 297
an error (j,^.
Vij =
x
i i / ^ + My
where X
y
includes MGCWJVy, TOPTENij, GOVNT
ij:
MKTBOOK
ih
BETA
i:i
, STD
i:j
, PROFSHAij, LONGINV
i:j
, CAR-12%, CAR+360y,
CASHTAij, the degree of IPO underpricing {IPORTN
i:j
), the logarithm of
IPO size (INIPOSZij), taxes as measured by tax payments divided by total
assets (TAXTAij), and the logarithm of firm's total asset (LOGTAij). Let
the vector of indicators y $ = (y
t
i, y
i2
, ya, yn) where j/y = 1 if firm i chooses
financing method j and j/y = 0 if firm i does not choose financing method
j . Then
Pij
= Pr[y
ij
= l\X
i
} = Pr[y*
j
>y*
k
,\/k^j\X
l
]
= Pr[Xij0 + fiij > X
ik
j3 + m
k
, Vfc ^ j]
= Pr[fj,ij - fj,
ik
< (Xij - X
ik
)(3, Vfc ^ j]
Let fi*
jt
= fiij-Hik and X*^ = (X^-X
ik
) for all k ^ j , and let F(-) be
the distribution function of (i*-
t
= {(Jt*ji, H*j2,lîa), the 3-dimension vector
of error differences, then
,x*.
2
/3 ^ p
= / / dF(^)
J oo Joo
- oo
coo /*oo
/
oo /*oo
/ l [^<X*
j 7
/ 3, Vfcî ]rfF(^. )
-oo /oo
The MNL results presented in Table 7.8 indicate that between the
choices of uninsured rights and stock dividends, firms with lower govern-
ment and institutional ownership, smaller debt to asset ratio and higher
cash to total assets and long-term investment to total assets ratios are
more likely to issue uninsured rights. These firms typically are bigger, sub-
ject to higher tax rates, and have better long-term share performance as
measured by CAR+360. However, the market run-up (CAR-120) and risk
surrounding the issue (STD) are not statistically significant. Between the
choices of underwritten offers and stock dividends, firms that have higher
managerial ownership, lower IPO size, and better liquidity are more likely
to use underwriters. Between the choices of issuing uninsured rights and
stock dividends together and stock dividends alone, firms that have lower
Table 7.8 MNL Regression on Financing Choices and Firm Characteristics
Uni nsur ed ri ght s St ock di v. & r i ght s Under wr i t t en offer
vs Stock div. vs Stock div. vs. Stock div.
CONSTANT
LOGTA
STD
TAXTA
CASHTA
LEVERAGE
IPORTN
CAR-120
LNIPOSZ
MKTBOOK
LONGINV
TOPTEN
GOVNT
MGOWN
BETA
CAR+360
-0.0007
(-0.0177)
0.0968*
(2.2926)
-0.0101
(-0.2415)
0.1532*
(2.7184)
0.0562+
(1.6650)
-0.3417+
(-1.9161)
-0.0125
(-1.1888)
-0.6637
(-1.2889)
0.0763
(1.5776)
0.1461
(0.3551)
0.0688+
(1.8372)
-0.5115*
(-2.1178)
-0.0788+
(-1.8970)
0.0031
(0.0079)
-0.0041
(-0.8431)
0.1457*
(2.1351)
0.0095
(0.2274)
-0.1472+
(-1.9411)
0.0459
(1.1104)
-0.1646*
(-2.4902)
-0.0113
(-0.2695)
0.0333
(1.1204)
0.0058
(1.5151)
0.0179
(0.4302)
-0.0722+
(-1.7498)
-0.1951*
(-3.0438)
-0.0470*
(-2.8819)
0.1451*
(2.8259)
0.0137
(0.3290)
-0.3138*
(-2.2175)
0.0030
(0.5797)
-0.0772
(-1.5325)
-0.0041
(-0.0982)
0.0206
(0.5315)
-0.0043
(-0.1031)
0.1265*
(2.4110)
0.1315+
(1.7879)
-0.0146
(-0.7604)
-0.0113
(-1.1998)
0.0024
(0.0574)
-0.1390*
(-2.4939)
0.0428
(0.9474)
0.0226
(1.0816)
0.0131
(0.3145)
-0.0493
(-1.0952)
0.1003*
(2.4082)
0.0012
(0.0230)
0.1414
(0.3405)
managerial ownership, higher institutional ownership and worse investment
opportunities as measured by market-to-book and long-term investments to
total assets ratios are more likely to issue both rights and stock dividends
at the same time. These firms that typically smaller and pay less taxes.
Institutional Transformation to Improve Corporate Governance 299
The econometric results by and large support the agency and political cost
approach in explaining the financing choices of newly privatized SOEs.
7.5 Institutional Transformation to Improve Corporate Gov-
ernance
In China, as in other transitional economies, a healthy and competitive
corporate sector is vital for sustained economic growth. Just as shared
governance is important in the public sector, corporate governance is as
important in the private sector. As there is an international trend towards
global competition, Chinese firms need to tap domestic and international
sources of funds in increased its competitiveness. The rules of the game
are changing: Banks, pension funds and individual investors will base their
investment decisions not only on a firm's commercial prospects, but also
on its reputation and its governance.
As the recent Asian financial crises demonstrated forcefully, macroeco-
nomic difficulties can expose and exacerbate systemic weaknesses in cor-
porate governance. A central element of the response to the recent crises
has been to strengthen corporate governance. This includes legal and regu-
latory systems, accounting and auditing standards, banking practices, the
regulation of capital markets, transparency, the work of corporate boards,
and the rights and responsibilities of different classes of shareholders. It is
important that China establishes sound corporate governance along lines
suggested below
7.5.1 Priv atization
Our previous analysis suggests that any successful state enterprise reform
strategy must address both the triangular debt problem among the gov-
ernment, SOEs and state banks and the corporate governance problem
simultaneously. Privatization of SOEs is not only inevitable but also desir-
able. However, there are at least three reasons that China's state enterprise
reform can not choose the same way of privatization by some western coun-
tries: One is the functional reason. The SOEs in China are firstly existent
as the important material foundation of socialism. It plays a dominant
role in the national economy. It does not only appear in situations where
the market does not work. The second is the number. The numbers of
China's SOEs, the fields they cover, the capital scale they occupy and their
proportion in the whole national economy are all much bigger than those
in any other countries in the world. Third, the healthy and efficient oper-
ation and development of China's national economy largely depend on the
vitality, dynamics and the future of development of the SOEs. So the aim
of China's state enterprise reform is not how to pull out from the compet-
itive market, but how to adjust and optimize the structure and operation
system of the SOEs and make them more adapted to the market economy
and develop more efficiently.
Given that most SOEs cannot continue their operation with the existing
debt burden, new equity funds have to be injected. However, the state as
the incumbent owner has no surplus funds available to bail out failing firms.
The triangular debt problem can only be solved by introducing new, non-
state shareholders to the SOEs. This can be easily done by transforming
household investors from debt-holders of state banks into equity-holders of
corporatized SOEs. Once individual investors replace the government as the
controlling equity-holders, corporate governance will change accordingly.
In other words, China does not need to follow Eastern European-style of
voucher privatization. China has got its own way to privatization. This kind
of privatization will not only resurrects the financial structure of SOEs but
also strengthens corporate governance.
Privatization can take various forms. For small-sized enterprises, man-
agement (or employee) buy-outs are a popular solution. For instance, 77.2%
of Zhucheng municipality's enterprises were bought out by managers and
employees from 1992 to 1994. For large and medium-sized enterprises,
forming joint ventures and going public are the two most attractive ways
to privatize. By forming joint ventures with domestic or foreign private
enterprises, local governments can partially privatize their SOEs with min-
imum disruption to operations. In order to attract foreign investors, local
governments compete to set up development zones and to offer substantial
tax concessions and other favorable treatments to foreign investors. As
for going public, the central government has recognized that two official
stock exchanges in Shanghai and Shenzhen can play an important role in
solving the debt crisis. All local governments are eager to get listing quo-
tas. Although the government often holds an absolute majority of shares
when they are issued, its stake should be further reduced. More recently,
privatization through Hong Kong has become fashionable. For example,
Shanghai and Beijing have packed their SOEs for listing on the Hong Kong
Stock Exchange. Other regions are following their footsteps. In many cases
the state-owned shares are taken over by non-state owners through the stock
market. Some Hong Kong companies may also directly take over SOEs.
7.5.2 Denationalization
Providing incentives for sound corporate governance is a challenge. In
China, the lack of clear identification of the owners of SOEs undermines
corporate governance because it leaves open the issue of who should be
monitoring the managers. The lack of delineation of the enterprises' com-
mercial and social functions and the fact that the state is both the ultimate
owner and the regulator of SOEs make for unclear governance objectives.
Moreover, effective corporate governance is difficult to exercise because
few of the institutions are responsible for managing state assets regularly
receive timely, accurate, and useful information about the financial perfor-
mance of the firms they control. Although the introduction of individual
shareholders through public listings has not resulted in a clear separation of
ownership and management interests, where ownership has been diversified
to include institutional investors, new governance structures have emerged.
Still, relatively few outside monitors, especially banks, exercise strong dis-
cipline on China's SOEs. The four specialized banks are mainly agents of
the state. Although they are attempting to transform themselves into com-
mercial entities, they have a long way to go to establish their independence.
Mergers (or threats of takeovers) can be effective disciplining devices against
poor management. But most mergers of large SOEs are engineered by the
state and are not market driven. Without reorienting governance incentives
toward a market system of checks and balances, many of China's SOEs will
continue to operate in a corporate governance vacuum, with managers and
other insiders exercising de facto control over the enterprises.
Hardening budget constraints and reducing subsidies provided to SOEs
through the fiscal system are important strategies in denationalization of
SOEs. State-owned Holding Corporation is a new form of the reform of
the management system of the state-owned asset that meets the request
of optimize the resources and improve the overall operation benefit of the
state-owned asset.
9
It is a meaningful exploration of the separation of the
9
Since 1987, Shen Zhen city has established a three-layer state-owned asset management
system, which includes the state-owned asset commission, asset management corpo-
ration and state-owned enterprises. This is a big step towards the socialist market
government's function from enterprise management and the separation of
the government from the asset. Although some reduction in state owner-
ship has taken place, the four main state-owned banks still often roll over
unpaid credits to SOEs. Interest rates have not been liberalized enough
for the banks to differentiate terms among enterprises according to credit
risks, which gives rise to cross-subsidies. The banks need to step up their
governance involvement with these enterprises, especially in calling in debt-
service obligations. But even where fiscal and financial system subsidy con-
straints have hardened, SOEs may be able to avoid financial restructuring
by resorting to inter-enterprise financing. Alternatively, they may arrange
for soft financing on non-market terms from public utilities or other enter-
prises.
More intense competition, whether from domestic sources or from abroad
would pressure SOEs to denationalize and restructure. Although the SOEs'
national market shares have already been significantly eroded by non-state
investors, because of the legacy of the planned economy, some SOEs in
heavy industry are protected by artificial regional autarky. Complex and
non-transparent business-licensing procedures keep new entrants out of key
sectors, such as electronics, petrochemicals, and machinery. Measures that
foster inter-regional trade and investment will boost enterprise denational-
ization and revitalization.
7.5.3 Diversification
The diversification of shareholders is conducive to the formation of standard
corporation corporate governance. Increasingly, enterprises are being par-
tially divested to non-state interests through minority shareholding, mainly
through Sino-foreign joint ventures and shares sold on stock exchanges in
China and elsewhere. There has been great progress in the reform in the di-
versification of shareholders. But in general, the degree of the diversification
economy. This system was further improved in 1996: the construction and real estate
enterprises t hat were originally under the Investment Management Corporation were
merged into Construction Investment Holding Corporation, and the t rade and circu-
lation enterprises t hat were originally under the Investment Management Corporation
were merged into Trade Investment Holding Corporation. This restructuring has given
the state-owned asset management corporation not only the right of operation and ben-
efit of the asset, but also the personnel power and the important investment decision
power, which optimize the relationship of responsibilities within the state ownership and
promote the improvement of the three-layer state-owned asset management system.
is still not high. This is reflected in the following two aspects: First, there is
still a great part of the reformed enterprises that have not implemented the
diversification of shareholders. For example, among the 2066 reformed en-
terprises in the pilot program of establishing the modern enterprise system
organized by various regions and departments, 31.7% have been reformed
into corporations fully owned by the state, the total asset of which is 49%
of that of all the reformed enterprises. Second, in the enterprises reformed
into corporation with multi-shareholders, most of them are absolutely con-
trolled by the state-ownership. For example, among the 1412 enterprises
reformed into limited liability companies with multi-shareholders or limited
liability stock companies in the pilot program of establishing the modern
enterprise system organized by various regions and departments, 65.7% are
companies absolutely controlled by the state-ownership. So, it is imperative
to take measures to actively deepen the diversification of shareholders.
Except the industries and fields where state-owned economy shall be
in control, the transfer of state-owned shares shall not be limited by state
control or state participation in the ownership. In enterprises that have
appropriate conditions, the transfer of state-owned shares shall give due
consideration to the bringing the strategic investment. Concerning the
shares owned by the capital management and operation companies that
were originally transferred from the debts of the SOEs, it would be more
effective if capital management and operation companies transfer the shares
to the non state-owned legal person or natural person at home and abroad
on the basis of sticking to the relevant state principles in adjusting the
structures of state-owned economy.
Diversifying ownership alone, however, is not enough, and, without the
proper institutional framework of checks and balances, may compound in-
sider control problems. The keys to success are to ensure that ownership
diversification is carried out through transparent and competitive proce-
dures; that it is cross-regional and cross-sectoral; that the process provides
for transferability of shares; and, most importantly, that it provides for
investments on behalf of the state to be managed by independent profes-
sionals, custodians or trustees whose remuneration is linked to their per-
formance. At the same time, organizational layers need to be eliminated
or simplified wherever feasible. The first step is to clarify the relationship
between the board of directors, managerial group and the supervision com-
mittee. The board of directors is a decision-making body, exercising the
power of personnel approval and decision making and some of the powers
of shareholders meeting. The managerial group is an executive body, exer-
cising the power of personnel approval and management. The supervision
board is the supervisory institution, exercising the power of personnel ap-
proval and supervision. The relevant working procedures should provide
the basis and guidance for the corporation's operation. The second step
is to appropriately handle the relationship of the board of directors, the
supervision committee with the Party committee, trade union and Inspec-
tion committee in the corporate governance. And the third step is to set up
the organizational framework according to the request of modern enterprise
system and operate according to the international practice.
During this process, it is desirable to create a market for corporate con-
trol. Mergers and acquisitions in the stock markets are important because
they help resolve problems such as duplication of facilities and subopti-
mal plant scale that resulted from denationalization. A viable market for
corporate control would make the threat of potential takeovers of ineffi-
ciently managed firms credible, would strengthen governance and ensure
that state-owned assets are being put to more productive uses. Such a
market could be based on the emerging network of regional property rights
transactions centers.
7.6 Summar y
While it is clear that Chinas economic reforms are responsible for lifting
millions of Chinese out of immiserating poverty, it is also clear that many
of the most difficult reforms are yet to be undertaken. Indeed, there is a
strong case to be made that much of the economic growth during the 1980s
and 1990s was due to an increase in allocative efficiency and increases in
productivity. Many have suggested that China is now entering a second,
and much more difficult stage of reform, one that will not be fully realized
absent a stronger system of private property rights.
The case of SOE reform is illustrative. Since 1984, the Chinese gov-
ernment has quite literally announced every year in some declaration that
the following year was the crucial or would be the breakthrough year in
reforming the stagnant SOE system. However, much work remains to be
done. This chapter finds that there has been some backtracking in the sense
that corporatized SOEs have been able to divert state assets and engage in
spontaneous privatization as evidenced in their post-IPO equity financing
Summary 305
decisions. We conclude that further privatization, denationalization, and
diversification are necessary impetus to speed up China's reform process.
Chapter 8
Accounting Information and Stock
Performance
8.1 Introduction
As more nations develop democratic, market-based economies, adopting
standards based on generally accepted accounting principles (GAAP) is
becoming increasingly important to the soundness of their financial sys-
tems. The adoption of internationally recognized accounting standards can
ease privatization efforts because of the confidence they provide individual
shareholders. Without such standards, the ability of privately held enter-
prises to attract capital from bankers and investors is seriously impaired.
Sound accounting standards support efficient financial management and
reporting systems that provide vital information to creditors and equity
investors so that they can make profitable investments. Investors provide
much needed risk capital and are concerned with the risk inherent in, and
return provided by, their investments. Investors need information that
will help them determine if they should buy, hold, or sell an investment.
Lenders, on the other hand, are interested in information that permits them
to determine whether their loans, and the interest attaching to them, will
be paid when due.
While a nation's choice of accounting standards is necessarily unique,
some consistencies among various nations' standards make them useful to
investors and creditors. The International Accounting Standards Com-
mittee (IASC) and the European Union (EU) both articulate accounting
standards that are widely respected in the major capital markets.
Under these standards, transparency is achieved by requiring firms to
issue financial statements along with footnotes that indicate the accounting
307
308
Accounting Information and Stock Performance
techniques used for certain transactions. The financial statements also pro-
vide information about management's stewardship of enterprise resources
and thus facilitate managerial control. The most common model regarding
the issuance of financial statements calls for the statements to consist of a
balance sheet, a profit and loss statement, and a cash flow statement.
Financial statements portray the effects of transactions and other events
by grouping them into broad classes (or elements) according to their eco-
nomic characteristics. The balance sheet is the primary financial statement
that helps determine an enterprise's financial position-and crucial informa-
tion for potential lenders. Information about financial structure (relative
amounts of capital provided by investors and lenders) is useful in predict-
ing future borrowing needs and how future profits and cash flows will be
distributed among those interested in the enterprise. Financial structure
information also helps predict how successful the enterprise is likely to be in
future financing transactions. Information about liquidity and solvency is
useful for predicting an enterprise's ability to meet financial commitments.
Liquidity relates to short-term availability of cash after taking into consid-
eration short-term financial commitments. Solvency relates to long term
availability of cash to meet commitments as they become due.
The primary financial statement that provides information about an
enterprise's profitability is the income statement, also known as the profit
and loss statement. Information about enterprise profitability is useful to
assess likely changes in the enterprise's control of economic resources and
the ability of the enterprise to generate cash flows from existing resources.
Information about the variability of profits is also useful for predicting
future cash flows. In addition, profitability information helps determine
how effectively an enterprise might employ additional resources.
A lender wants to obtain some confidence an enterprise can generate suf-
ficient cash to meet scheduled debt payments; cash flow information helps
determine this. The cash flow statement reports cash flows from invest-
ing, financing, or operating activities. Lenders are interested in the sources
from which cash is generated, the timing and certainty of its generation,
and how the cash is used. For example, an enterprise's long-term prospects
depend on its ability to generate sufficient cash from profitable operations
to provide returns to investors and other financing sources. If current cash
requirements are instead generated by investing activities such as securi-
ties, or from financing activities such as long-term borrowing or the sale of
Roles of Financial Disclosure for Performance Evaluation 309
stock, this information will be valuable to lenders.
1
Accounting standards also specify supplementary information that ac-
companies financial statements. These items, summarily called footnotes,
should contain additional relevant information about items in the balance
sheet and income statement. This information could relate to the risks
and uncertainties affecting the enterprise, resources and obligations of the
enterprise not recognized in the balance sheet, or breakdowns of figures
by geographic area, customer type, or other category. Footnotes also con-
tain detailed information about the accounting standards a firm employs,
thereby lending transparency to the statements.
2
This chapter provides an empirical study of how useful accounting infor-
mation is in the emerging Chinese stock markets. This chapter is motivated
by recent developments in both research and practice on the valuation ef-
fects of accounting information releases. In Section 8.2, we discussion the
general roles of financial disclosure for evaluating the performance, risk and
valuation of a company. In Section 8.3, we detail the corporate disclosures
made by listed Chinese companies. In Section 8.4, we study how informa-
tive corporate disclosures are by examining whether and to what extent
current stock returns and future earnings changes reflect the degree of vol-
untary disclosures. In Section 8.5, we use event studies to investigate the
stock returns around the earnings announcement. We provide some expla-
nations on why domestic Chinese investors tend to over-react to earnings
releases in Section 8.6.
8.2 Roles of Financial Disclosure for Performance Evalua-
tion
Despite the growing demand for data on companies and an increasing sup-
ply, audited financial statements remain the most important and widely
used source of information on companies. In their current form, the two
1
The cash flow statement is not as universally required as the other statements. The
European Union, for example, does not require it. The cash flow statement is becoming
increasing important, however, in judging management' s performance and assessing an
enterprise' s ability to repay a loan.
2
No one statement tells the whole story. The income statement, for example, provides
an incomplete picture of performance without reference to the balance sheet or the cash
flow statement. As a result, lenders and investors need information from each of t he
statements to make their decisions.
310 Accounting Information and Stock Performance
principal goals of financial statements are to ensure that appropriate mon-
itoring takes place and provide the basis to value securities.
Accounting has evolved from cash accounting, a system for tracking
receipts and payments, to accrual accounting that tracks profit and loss
over time and describes the ability of an enterprise to operate as a going
concern, to describing disposable income or the free cash flow generated
by an operation. The emergence of balance sheets, next profit and loss
and finally cash flow statements can be taken as a rough metaphor for
this evolution. These major additions to financial reporting resulted in a
qualitative improvement in the information available to users.
However, the view of some market analysts is that financial statements,
and in particular the balance sheet, were never designed to provide the
value of a company. Markets value companies, the argument goes, and
the real value of an enterprise can be determined rapidly and simply by
consulting share prices in a newspaper. This view is consistent with con-
temporary valuation models which state that the value of a company is
represented solely by its ability to generate future cash flows and that the
task of financial markets is to value shares incorporating all available data.
Under this model, financial statements provide the information that allows
users-in this case securities analysts-to estimate the size of the cash flow
and any related risk. It is further argued that since many intangible assets
could only be valued based upon the value of the cash flows they generate,
their inclusion in balance sheets would create a circularity of reasoning, i.e.,
statements would no longer be useful for estimating future cash flows since
they already use future cash flows to determine the value of assets. Thus,
the central question remains whether financial statements, and the bal-
ance sheet in particular are supposed to value enterprises or simply provide
elements of a much larger set of data necessary to value enterprises.
While historical data is important in evaluating performance potential,
financial markets rely on future cash flows to determine value. Various parts
of financial statements may provide information of a predictive nature that
give clues to future cash flows. Most probably management's discussion of
operations would include forward-looking information of a strategic nature
such as plans to develop new markets, acquisitions, divestitures, etc. State-
ments, however, rarely include financial projections. Some attempts have
been made to improve the forward-looking content of statements in order
to support the task of projecting performance particularly in the United
States where the SEC has encouraged forward-looking statements since the
Roles of Financial Disclosure for Performance Evaluation 311
1970's and passed additional legislation in 1995.
Despite improvements in financial statements, there has been little ob-
servable increase in the predictive data that companies disclose. On the one
hand the legal protections are insufficient to convince companies to chance
potential legal action against them and many companies believe that the
markets are not sufficiently interested in such data to warrant running the
risk. There is also some doubt about the ability of companies to take a
balanced view given a natural bias to portray positively the results of cur-
rent efforts and programmes. Conceptually, it is uncertain whether it is the
company's role to provide projections or to provide data with which better
projections can be constructed. This may be reflected in current practice
where deep-seated reticence to commit forward-looking data to paper has
been accompanied by increased oral discussions of future performance with
analysts.
Financial risk is generally taken to be the level of uncertainty with re-
spect to potential gain or loss on an investment. By its very nature risk
is forward-looking and thus difficult to quantify. It is often measured by
the past volatility of earnings or the volatility associated with a certain
expected outcome. Better disclosure of risk would promise to improve the
cost of capital to companies, encourage better risk management, improve
governance, the protection of investors and improve the usefulness of fi-
nancial statements overall. As with traditional accounting measures, the
quality of risk related information would be improved if subject to similar
standards of materiality, reliability, comparability and understandability.
Risk disclosure may be useful in several areas, most obviously in the area
of company treasury policy and the use of financial instruments. Companies
with an intensive use of financial instruments may significantly reduce the
information value of the balance sheet. Instruments such as interest rate
swaps (broadly speaking the exchange of differing terms and maturities of
debt instruments) are quickly effectuated and may alter the capital and risk
structure of an enterprise.
Whatever disclosures are ultimately made, it is unlikely that they will
fully satisfy all users since information needs are constantly evolving and
vary across regions.
8.3 Corporate Disclosures Made by Listed Chinese Com-
panies
Accounting reform and the improvements in accounting information have
provided favorable conditions for economic transformation and develop-
ment of the capital market in China. Disclosing accounting information on
stock market listed companies has become an important issue because of
its significant influence on the security market. There is strong evidence
that disclosed accounting profits have conveyed information to the market
with consequent effect. This indicates a kind of rational investment power
based on the nature of capital has been established in the Chinese security
market.
8.3.1 A Brief History of Accounting Dev elopment
The present Chinese accounting regulations and practices have evolved
from a Russian-style macro-economy oriented accounting system adopted
by China in the 1950s. Encouragingly, the accounting reform has speeded
up from the early 1990s to adapt to the market-based economy in the coun-
try.
Since China initiated economic reform from the late 1970s, Chinese ac-
counting, especially accounting for enterprise, has been changing remark-
ably with a view towards accounting standardization and internationaliza-
tion.
A major innovation for SOEs, namely the Management Responsibility
Contract System (MRCS) was introduced in the early 1980s. Government
contracted with each individual SOE establishing performance targets and
specified incentives to achieve or exceed these targets. The liberalization
of the concept of control in accounting took a variety of forms. Within
centralized targets, a greater degree of freedom was given to managers of
state industry concerns. For example, managers were allowed to retain
part of the profit of their concerns to expand production and pay bonuses
to employees; extra goods produced above state plans' targets were allowed
to be sold to market or were directed towards export trade where part of
the foreign exchange earnings could be retained in the relevant concern
to improve technology. Accounting practice became slightly more than a
central government control measure.
To further attract foreign capital, a new accounting theory emerged.
Corporate Disclosures Made by Listed Chinese Companies 313
Accounting conventions and principles in the Accountancy Law for Joint
Ventures fused western accounting theory with Chinese cultural philosophy.
This provided a unique theoretical framework with a special nationalized
accrual accounting and a special notion of revenue recognition/realization
unique to Chinese culture. It embraces a specific shade of accrual account-
ing with a twist in defining recognition and realization of income and costs
as it really occurred within mainland China.
A remarkable change took place in Chinese accounting in 1993. The
Basic Accounting Standards for Enterprise was issued in November 1992
and was brought into practice with 13 new accounting regulations for en-
terprises in different industries from July 1, 1993. The purpose of the
reform is to establish an accounting standards system that not only suits
the socialist market economy in the country but also reconciles Chinese ac-
counting with the internationally accepted accounting practices. The new
accounting standards reflect that by introducing measures to identify more
clearly some accounting parameters that are capable of giving signals to
efficiency and productivity at the microeconomic level of SOEs and also
by allowing outside users of financial statements to assess more accurately
their performance.
3
The first nationwide accounting standard similar to IAS was promul-
gated in 1997. By the end of 1998, a total of eight accounting standards
have been pronounced. Among these standards, seven are applicable only
to listed companies. The new accounting standards are practical standards
as well as conceptual, they relate accounting theory to the new environ-
ment. They are governmental regulations and an accounting conceptual
framework. They consist of basic premises, accounting objectives, general
principles, accounting elements and their determination and measurement,
the content of financial statements and their presentation, and disclosure
of further information.
Overall, Chinese accounting standards have reflected the methods re-
quired for Chinese tax accounting, and have therefore been closer to a cash
or tax system of reporting than international standards. Winkle, Huss and
Xi-Zhu (1994) and Chen, Gui and Sui (1999) provide a summary of the
3
The new accounting st andards introduce the concept of equity capital, which provides
a sharper distinction between product cost and period cost. Equity capital concept is
similar to the capital maintenance concept used internationally. The new accounting
st andards allow state enterprises to accumulate accounting dat a t hat can provide signals
of how economic resources are allocated within firms.
Chinese accounting standards adopted in December 1992. They find that
while these standards parallel U.S. and international practices in many re-
gards, there are a few notable differences. First, in valuing assets, Chinese
standards require strict adherence to historical cost, making no provision
for mark-to- market or lower-of-cost-or-market. This can have a significant
effect on the valuation of inventories, which can be valued at planned prices.
Second, the government has stipulated depreciation rates for capital assets
and bad debt allowances for receivables (0.3-0.5% of gross receivables) re-
gardless of a firms business economics. Third, Chinese standards do not
recognize some of the more complex liabilities, such as contingencies, and
lease obligations. Finally, the level of disclosure is often limited. For exam-
ple, related party sales between listed and holding companies may not be
reported, making sales difficult to interpret.
8.3.2 Financial Reporting Practices
After the corporatization of SOEs, without direct control and without
involvement in day-to-day management, the government authorities and
shareholders have to rely on the SOEs' financial statements to make de-
cisions. However, managers have incentives to manipulate financial infor-
mation. They may inflate profits to their benefit or provide false financial
statements to conceal embezzlement and other irregular behavior. Finan-
cial statements provided by management may not represent the true per-
formance SOEs. In addition, the expansion of the Shanghai and Shenzhen
Securities Exchanges provide direct incentives and pressures for market-
oriented financial disclosure.
Therefore, on August 31, 1994, the Standing Committee of the Na-
tional Congress adopted the Auditing Law of the People's Republic of China,
which was effective on January 1, 1995. The Auditing Law reinforces the
roles and responsibilities of the state audit office. It prescribes the scope
and procedure of auditing of SOEs. Table 8.1 presents the enactment of
new accounting and auditing regimes by the National Congress over time.
The independent audit professions, which can provide unbiased views on
the financial reports are established.
4
The Accounting Law (1993) requires every organization to prepare fi-
4
According to t he requirements from the Ministry of Finance (MOF), three types of
companies in China are required to be audited annually by independent CPAs: listed
companies; companies with foreign investors or partners; and selected SOEs.
C
o
r
p
o
r
a
t
e

D
i
s
c
l
o
s
u
r
e
s

M
a
d
e

b
y

L
i
s
t
e
d

C
h
i
n
e
s
e

C
o
m
p
a
n
i
e
s

X

0
3

a

f
i

o

firms
firms
T
3

-
O

C
D

C
D

f
i

0
3

a

o

<
p

H

<
1

c
o
<
*
*

o

u

a
.

b
C
i

"
S

C
O

^

O

<
!

c
o

pita
0
3

o

C
D

Shar
i

pita
0
3

o

C
D
Shar
C
O

C
O

i
3

U
J

a
)

D

P

a
)

C
O

tC
O

^

_
C
O

J
_
l

(
-
1

+
J

f
H

o

a
)

d

C
D

C
D

C
D

J
;

C
D

<

<
!

t
f



3

o
o

o
o

C
T
>

C
T
J

-
a

C
D

f
i

0
3

s

3

O

u

u

<

C
J

C
T
>

C
O

3

d

0
0

,
-
-
.

J
2

g

^
-



c
o

0
2

C
D

x
i

<

<

_
I

C
D

C
D

C
D

-
f
i

-
f
i

-
C

E
-
i

E
H

E
H

J
3

f
i

r
-
H

C
O

-

f
i

b

b
o

f
i

C
D

o

-
S

O

1
3

o

,

f
i

-
x
i

'

0
3

I

s

f
i

2

0
3

C
O

f
i

>
>

.
f
i

C
O

a

-
s

_

C
S

o
3

S

"
H
I

o
3

f
i

L

0
3

"
O

"
d

0
3

0
3

5
P

c
o

3

W
l

3

c

8

f
i

o

c
f
i

^

>
>

H

H

C
O

b
O

f
i

-
t
-
3

T
)

3

<
l

-
^

f
i

a p u a
[ndep
A

0

C
O

J
%

"
3

P
i

b
O

_
C

ment mple
J
H

o
ulat
b
O

C
D

f
f
j

b
O

f
i

0

O

0
3

f
i

.
g

-
M

0
3

C
D

-
f
l

H

f
i

3

O

a

0
3

C
O

C
D

J
3

E
H

o

C
D

-
f
i

H

h
o

f
i

0
3

,

B

c
s

G
|

S
P

C
O

U

"
1
-
9

<
v

C
O

C
i

S
2

Q

$

&
>

f
i

-

0
3

M
M

C
O

f
i

O

S
f
i

'
>

C
D

C
O

3

I

C
J

o

C
D

r

^

C
O

^

s

s

6
1

5

b
O

-
f
i

<
P

f
i

0
3

a

*
a

o

nancial statements based on its accounting records in accordance with the
accounting system defined by the State. Three main financial statements
need to be prepared by business enterprises, namely the balance sheet, the
income statement and the statement of changes in financial position or cash
flow statement. The three statements should be supplemented by support-
ing schedules, notes, and an explanatory statement on financial conditions.
The Company Law (1993) requires all share capital-based firms to pre-
pare an audited financial report at the end of the year. A limited liability
company should provide the report to its shareholders whereas a share cap-
ital limited company is required to display the report at its registered office
prior to the annual general meeting so that shareholders can consult it. A
public listed firm has to publish its report in a major news media. It must
prepare both an interim report and an annual report within 60 days after
the first 6 months of the year and within 120 days of the end of the year,
respectively. It must submit 10 copies to the CSRC, publish an abstract in
at least one national newspaper approved by the CSRC, and deposit the
reports and abstract in the company's registered office and with the rele-
vant stock exchange and securities trading agents for public consultation.
Board of directors must ensure that there are no fraudulent or seriously
misleading statements contained in their publicly disclosed information, or
any important omissions.
5
The annual report for a public listed company typically consists of the
following sections:
Brief introduction of the company
Three-year summary of accounting and operations data
Chairman or managing director's statement
Directors' report
Financial statements
Statement of material events
Description of related companies
Notice of the annual shareholders' meeting
Major characteristics of the annual report include:
Chinese financial reporting requirements treat domestic and international investors dif-
ferently. Firms issuing A-shares are required to report under Chinese accounting stan-
dards, whereas firms issuing B-shares are required to use IAS, and firms issuing H-shares
are subject to Hong Kong accounting standards or IAS.
(1) the focus is on the company's short term (rather than long term)
objectives and strategies;
(2) there is an emphasis on disclosure of technological innovation;
(3) although discussion is required of the company's immediate indus-
try, the wider political, social, and demographic context is largely
neglected;
(4) an emphasis is placed on reporting material events and legal pro-
ceedings;
(5) business opportunities and risks do not feature as important;
(6) little emphasis is placed on the competitive market forces that in-
fluence the company's businesses;
(7) the requirements for financial statements are relatively simple.
Along with the enforcement of financial reporting practices, the Chinese
accounting profession revived to meet the urgent need for accounting and
auditing services. The first Chinese CPA firm was established in Shanghai
in January 1981, which was followed by a rapid growth of new firms in
other cities. As a result, accounting education has improved remarkably.
6
In the meanwhile, the Chinese auditing market is shared by state auditors,
local CPA firms and international accounting firms.
To further complicate financial reporting^ domestic and international re-
ports tend to be audited by different audit firms. Domestic reports typically
are audited by Chinese auditors that potentially face severe problems in at-
tracting and retaining qualified personnel and maintaining independence.
In addition, Chinese firms still have difficulty enforcing standards for listed
companies that are former SOEs where the government has a strong contin-
uing ownership and an interest in limiting the release of reports that ques-
tion the performance or viability of these entities. Xiang (1998) reports
that both Chinese exchanges are vexed by company hesitancy to follow
corporate disclosure rules and accurate accounting procedures. Only about
half the listed firms submit acceptable annual reports. Xiao, Zhang, and
Xie (2000) report that major features of Chinese auditing include lack of
6
China has been holding nation-wide CPA entrance examinations from 1992 and car-
rying on continuing professional education to enhance the service of accounting. The
CPA entrance examinations are under the supervision of t he Commission on National
CPA Examinations, in t he Ministry of Finance and administered by t he Examination
Committee of the CICPA (the Chinese Institute of Certified Public Accountants). The
CICPA is responsible for preparing, monitoring and grading the national CPA exami-
nations.
audit independence, the shortage of well-qualified auditors, an environment
of extensive corruption, and the existence of many misconceptions about
the audit.
7
Financial reports prepared under IAS are typically audited ei-
ther by Big Five or large Hong Kong firms. These international firms are
less likely to be subject to many of the problems facing local Chinese firms.
However, they are unlikely to be immune to shortage in qualified staff and
to questionable audit practices in their Chinese offices.
Since IAS is often considered a better accounting standard than Chi-
nese GAAP and Big Five auditors compare more favorably in audit quality
to their Chinese counterparts, the existence of a dual reporting and au-
diting mechanism may represent a unique institutional environment which
affects the value relevance of accounting information in the Chinese market.
Consequently, it is more an empirical question of whether accounting infor-
mation is useful to domestic investors in the Chinese market as measured
by the contemporaneous association between accounting numbers and stock
valuation.
8.3.3 Capital Market Infrastructure
Disclosure can only be effective when coupled with an efficient system for
the dissemination of information. A critical challenge in the development
of a modern capital market in a transitional economy is the creation of ef-
fective underlying institutions and infrastructure. These include the afore-
mentioned financial reporting and auditing standards. However, they also
include secondary market efficiency, capital market regulations, legal pro-
tection of private property, maturity of financial intermediaries, and the
financial media in general.
Not surprisingly, China has only begun to develop financial institutions
and capital market infrastructure. The efficiency of secondary market and
accuracy in valuing financial assets are critical for financial reporting, which
in turn affect the pricing of initial public offering. Prior to going public,
SOEs have to prepare financial reports that present data on the historical
cost of capital and current assets, as well as estimates of the current value
of net assets. However, many assets were acquired at planned prices, not
7
In order to audit listed companies, a CPA firm must meet stringent requirements jointly
set by the MOF and the CSRC. Among some 7,000 CPA firms, only 105 have been
authorized to audit around 700 listed companies, as of t he end of 1997 (Yang and
Yang, 1998). As a result, auditing of listed companies is a highly competitive market.
market prices. Prospective issuers obtain independent valuations of their
assets for both setting offer prices and for financial reporting. External
valuation firms (from Hong Kong and elsewhere) are used in the valuations
for international offerings. These valuations can raise sensitive political
questions, since the Chinese government is anxious to ensure that stocks
sold to international investors are not under-priced.
Valuations for domestic issues are typically by Chinese valuation firms.
However, there is a severe shortage of trained professional evaluators, mak-
ing it difficult to assess the quality of these valuations. Responsibility for
the regulation of financial markets in China nominally lies with the CSRC.
8
Another important form of infrastructure for financial reporting and
capital markets is legal protection for investors. Securities laws typically
protect a number of shareholder rights in developed market economies.
These include protections of minority shareholders rights to vote for boards
of directors and to sue management for fraud, misleading disclosures or
misappropriation of funds. Investors are likely to have questions about
the extent of these forms of protection in China. Since the government is
the majority owner of most listed companies, protections of the rights of
minority owners are especially relevant. In addition, international investors
are likely to have concerns about whether they have the same rights as
domestic investors.
Winkle, Huss, and Xi-Zhu (1994) cite claims that PRC- and IAS-based
reports are often released at different times and contain different levels of
information. Typical investor complaints are that A-share investors re-
ceive financial information earlier than B-share investors and that A-share
investors receive a much broader disclosure of the companys performance.
A legal system that protects shareholder rights is also important in
China because the countrys economic system has historically provided job
security and expansive benefits to employees and management, as well as
funding for loss-making businesses. These policies raise several questions
for private investors. First, is it difficult to remove management for good
cause? Second, can funds provided by public issues be diverted from their
intended uses towards meeting pension obligations or financing loss-making
business segments? Another form of infrastructure that has been created
8
Since 1995, t he CSRC has reported directly to the State Council, the Chinese gov-
ernment' s highest-ranking decision-making authority. Previously, it reported to the
Ministry of Finance, the People's Bank of China, and the State Committee for Re-
structuring the Economy.
with the development of a capital market is a community of financial ana-
lysts. In developed market economies financial analysts provide investment
advice and also monitor the performance of management. Management
of stocks that are out-of-favor with analysts because of poor performance
are then likely to be subject to increased shareholder and board scrutiny
or likely to be removed by a hostile acquirer. However, in China analyst
coverage is limited. There is some analyst coverage of H-shares by Hong
Kong investment firms. However, there is relatively low demand for finan-
cial analyst services by international investors since the B-share market is
so thinly traded. Analyst coverage of A-shares listed on the Shanghai and
Shenzhen Exchanges is also limited.
Finally, firms are increasingly sensitive to the economic risks of operat-
ing without paying close attention to public opinion and the public interest.
The role of the news media which can put companies under intense pub-
lic scrutiny is, of course, primary. Moreover, some investment funds have
decided to exercise influence by only investing in the shares of companies
which have a particular policy stance on an issue. Though the influence of
the latter remains limited, the potential impact of public opinion indicates
that it is may not possible to divorce purely financial interests of sharehold-
ers from those of the larger community. There is increasing acceptance of
the notion that the successful companies of the future will be those that
can effectively balance the interests of a variety of stakeholders.
The financial press in China is likely to play a weak role in monitoring
management of public companies. Chinese newspapers provide summaries
of financial information on listed companies every six months. However, it
is not politically feasible for them to publish articles that criticize managers
who, after all, are effectively appointed by the government. In summary,
capital market infrastructure in China is in a formative stage. The types of
professional intermediaries and legal rights that have evolved in developed
economies are still being created in China.
8.4 Stock Returns Around Earnings Releases
From the perspective of information economics, accounting and financial
reporting play a vital role in an efficient capital market. This chapter
provides an empirical study of how useful earnings announcements are in
the emerging Chinese stock markets. This chapter is motivated by recent
Stock Returns Around Earnings Releases 321
developments in both research and practice on the valuation effects of ac-
counting information releases. Numerous studies, including Ball and Brown
(1968), Kross and Schroeder (1984), Easton and Zmijewski (1989) and Gen-
notte and Truemann (1996), find that stock prices respond positively to
announcements of increase in earnings and negatively to announcements
of decrease in earnings for the U.S. firms. Alford, et al. (1993), Amir, et
al. (1993) and Chan and Seow (1996) also provide evidence of the informa-
tional content of earnings announcements in a number of non-U.S. markets.
This chapter extends this line of research to the Chinese stock markets.
8.4.1 Information Content Analysis
The research on information content in the setting of developed and mature
capital markets implicitly assumes some necessary conditions that must pre-
vail in the information environment in order to give credence to any market
tests of information content. Some of these conditions can be stipulated as
follows:
Transaction-related data and source documents of accounting sys-
tems are the outcome of arms-length exchanges.
Transaction data are consistently aggregated in accordance with a
known set of rules.
Effective internal controls are in place to safeguard the company's
assets and to reduce opportunistic behavior.
Internal audits are implemented to ensure the integrity of the in-
formation system, input data and produced information.
Market prices reflect the preferences of market participants.
Capital markets are free of manipulation by regulators, individuals
in power or capricious government interference.
Insiders do not use their informational advantage for their own gain
before allowing equal access to other investors.
The aforementioned conditions are generally satisfied for listed firms in
developed economies where financial and accounting institutions, the pub-
lic, the press and regulators act as external monitors or "watch dogs" that
assure the integrity of the data reported to the public. In addition, enforce-
ment agencies exist for dealing with serious departures from the expected
norms. In that setting, researchers can reasonably assume that investors
vote with their feet, capital markets operate efficiently, and prices reflect
all publicly available information. To make reliable inferences about ac-
counting information content, markets must operate with the least possible
friction, a case that is difficult to make for many emerging capital markets.
There are various reasons why share prices in emerging Chinese capital
markets may not satisfy the conditions listed above. Indeed, the efficiency of
Chinese stock markets has not been established because reported data and
recorded stock prices may not reflect the free expression of the preferences
of all market participants. Unique factors of local customs, bureaucratic
regulation, the absence of a developed accounting and auditing profession,
undue influence of state officials, and loose insider trading constraints do
interact to impact the operation of capital markets.
For example, prior to May 1992, regulators of the stock exchanges im-
posed tight controls over all share prices and trading volume. Although
much of the underlying motives remain behind the scenes, to public ob-
servers these controls were manifested by imposing two constraints on the
trading of A shares, both aimed at reducing volatility. These constraints
were: (1) the maximum allowed daily fluctuation in the price of any listed
share cannot exceed 1%; and (2) trading prices should not change if vol-
ume of trade does not exceed 0.3% of the number of outstanding shares.
On May 5, 1992, the allowed maximum daily fluctuation in prices was in-
creased from 1% to 8%, but restrictions on the minimum volume required
to trigger price changes were retained. On May 21, 1992 both measures of
control were abolished. While the imposition of such regulatory controls
on market forces has prevented share prices from reflecting the true pref-
erences of market participants, it has artificially maintained low levels of
volatility. Upon lifting the restrictions on May 21, 1992, share prices more
than doubled and the SHSE price index increased by 105.3% in one day.
8.4.2 Earnings Announcements and Stock Price Reactions
in China
Using a sample of 183 earnings announcements between 1997 and 1998 for
firms that have issued both A- and B-shares prior to January 1, 1996, we
investigate the stock price reactions to changes in earnings-per-share (EPS).
We find that domestic A-share investors, on average, do not correctly antic-
ipate the EPS changes and do not adjust to the new earnings information
very rapidly in the markets. Excess returns or abnormal returns could
be generated by trading on the earnings information after it is released.
However, international investors can predict changes in EPS better than
domestic investors and there is little or no abnormal announcement-day ef-
fects observed in B-share markets in China. To examine the abnormal stock
returns we choose earnings announcement day as our predictable event day.
We then divide our sample into two groups based on the outcome of the
event. An announcement belongs to Group I if the outcome of the event is
'positive' (actual EPS exceeds last EPS). It belongs to Group II if the out-
come of the event is 'non-positive' (actual EPS is equal to or less than last
EPS). If the market is efficient, security prices should reflect all potential
changes in the event outcomes. That is, if the required rate of return on a
stock around the event day is identical to that of any other random trading
day, one should not be able to make excess returns by trading around all
of these event days. The unconditional expected rate of return of these
announcements should not differ from that on other days. Finding excess
returns, as traditionally measured, around the event day is consistent with
the notion that the required rate of return during an announcement period
is higher than normal and the relevant risk per unit of time during the
event is higher.
Denote the daily abnormal return for any stock as
ARjt = Rjt E(Rjt),
where
t = day measured relative to the event,
t = - 21, - 20, , - 1 , 0 , 1 , -, 20, 21.
The earnings announcement date for each firm is set to be t = 0.
ARjt = abnormal return to stock j for trading day t.
Rjt = return on stock j during trading day t.
E(Rj
t
) = expected rate of return on stock j for trading day t.
Based on the mean adjusted returns model of Brown and Warner (1985),
E(Rjt) is estimated from the time-series of stock j ' s realized returns in the
pre-event window, i.e., t = 122, , 22. The abnormal return for each
stock, ARj
t
, is then calculated as the difference between the actual return
to a stock and the estimated expected return during a pre-event period.
We calculate ARj
t
for all stocks in Group I and II for A- and B-shares.
(8.1)
Average abnormal returns for each trading day within the event window
are calculated by
1
N
where ./V is the number of stocks with abnormal returns during day t. The
cumulative abnormal returns for each stock j , CARj, are formed by sum-
ming average abnormal returns over time as follows:
L
CAR
j<K
,L = X)
AR
i*> (
8
-
3
)
t=K
where the CAR^K,L is for the period from t = day K until t day L.
Average cumulative abnormal returns over the event window from day
K until day L are calculated by
1
N
CAR
KiL
= jjYl
CAR
J,K,L. (8.4)
In particular, a two-day average abnormal return is generated for A- and
B-shares in each group. A two-day excess return is necessary to capture the
effect of an announcement due to its timing relative to the market's trading
hours. Day t = 0 is the day the news of an announcement is published in
the China Securities Daily. In many cases, the news is actually announced
on the previous day, t = 1, and reported the next day. If an earnings
announcement is made before the market close, then the market's response
to the news actually predates the publication by one day. If the earnings
announcement is released after the market closes, the market will respond
the next day and the reaction is indeed on day 0. Thus in reality there is a
two-day announcement period, t = 1 and t = 0. This two-day return for
firm j is
CAR
jt
-
ltQ
= AR
jt
-.! + AR
jfi
, (8.5)
where ARj
t
_i is the abnormal return to stock j on the trading day prior
to a published announcement in the China Securities Daily and ARJQ is
the abnormal return to stock j on the day an announcement is published
in the China Securities Daily.
Stock Returns Around Earnings Releases
Finally, ^-statistics are calculated for CARX,L by
CAR
KtL
*CAR
K
,
L SD
(
CARKiL
y
where SD(CAR
K
,L) = y/TVa,r(AR
t
) + 2(T - 1) x Cov(AR
t
,AR
t+1
),
T = K L + 1, is the standard deviation of CARK,L- Var(ARt) and
Cov(ARt, ARt+i) are estimated from 122 days before the earnings an-
nouncement day to 22 days before the announcement day. The covariance
term adjusts for possible first-order autocorrelation between the abnormal
returns due to non-synchronous trading.
In theory, if the stock market is semi-strongly efficient, i.e., if no one can
earn abnormal returns by trading on the basis of public information, then
the stock prices will reflect changes in the firms' earning power. This implies
that: (1) The stock market will correctly anticipate the earnings changes
before they are announced to the public. Thus, firms with disappointing
EPS (Group I firms) will experience unfavorable downward pressure on
their prices in the days before the actual earnings announcements occur.
Firms with strong EPS (Group II firms) will enjoy upward pressure on the
prices in the days before the announcements to the market. (2) Since the
earnings information is fully incorporated into the stock prices, no abnormal
return exists during and after the earnings release. Thus, CAR_i
i 0
should
not be statistically significant and CARi
t
h+i, h = 1,2,---Ô, should be
random after the announcement date.
Tables 8.2 and 8.3 present the abnormal returns (ARs) and cumulative
abnormal returns (CARs) for the disappointing earnings group (Group I)
and satisfactory earnings group (Group II) for A- and B-shares. The CAR
results are also portrayed in Figures 1 and 2. As shown in these tables
and figures, the A-share CARs for the disappointing earnings group tend
to increase and stay positive before the announcement date while start-
ing to decline two days before the earnings announcement. Such a decline
persists for about 13 days after the announcement date and the A-share
CARs become significantly negative. The average two-day announcement
period A-share abnormal return is -2.27% with a i-statistic of -8.41 for
the disappointing earnings group. On the other hand, the A-share CARs
for the satisfactory earnings group tend to decline 9 days before the an-
nouncement date and become negative, while starting to grow and become
positive two days before the earnings announcement. Such growth con-
tinuously lasts for 21 days after the announcement. The average two-day
325
(8.6)
326
Accounting Information and Stock Performance
Table 8.2 Abnormal Returns and Cumulative Abnormal Returns Around Earnings An-
nouncement Days for A Shares
DAY
-13
-12
-11
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
1
2
3
4
5
6
7
8
9
10
11
12
13
CAR-i,o
t-statistic
N
G roup I
AR (%)
-0.0491
-0.2141
-0.2976
0.1013
0.6307
0.3071
-0.25
0.2053
0.1681
1.5726
0.2193
-0.3391
-1.0391
-1.2315
-0.2883
0.3915
-1.4566
0.3525
-0.1392
0.1558
0.423
-0.4692
-0.0719
-0.9161
0.3289
-0.7876
-1.3216
-
CAR (%)
1.869
1.6549
1.3573
1.4586
2.0893
2.3964
2.1464
2.3517
2.5198
4.0924
4.3117
3.9726
2.9335
1.702
1.4137
1.8052
0.3486
0.7011
0.5619
0.7177
1.1407
0.6715
0.5996
-0.3165
0.0124
-0.7752
-2.0968
2.27
8.41
109
G roup (II)
AR (%)
0.1084
0.6134
-0.6763
0.918
-0.1216
-0.0572
-0.5963
-0.3188
0.4219
-0.6399
0.1823
-0.4025
1.2718
0.8764
-0.9587
0.7353
-0.5888
0.6535
0.128
-0.2927
-0.1976
0.6369
-0.1092
-0.0846
0.1277
-0.3082
1.019
:
'
CAR (%)
0.1902
0.8036
0.1273
1.0455
0.9239
0.8667
0.2704
-0.0484
0.3735
-0.2664
-0.0841
-0.4866
0.7852
1.6616
0.7029
1.4382
0.8494
1.5029
1.6309
1.3382
1.1406
1.7775
1.6683
1.5837
1.7114
1.4032
2.4222
2.15
7.60
74
announcement period A-share abnormal return is 2.15% with a ^-statistic
of 7.6 for the satisfactory earnings group. These results indicate that the
A-share market does not correctly anticipate the earnings changes prior to
the announcements. The two-day announcement period average abnormal
Table 8.3 Abnormal Returns and Cumulative Abnormal Returns Around Earnings An-
nouncement Days for B Shares
Gr oup I Gr oup (II)
DAY
-13
-12
-11
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
1
2
3
4
5
6
7
8
9
10
11
12
13
CAR-i , o
t-statistic
TV
AR (%)
0.6952
0.2618
-0.5316
-0.0983
0.1308
0.285
-0.5149
-0.1138
-0.2896
-0.184
0.4262
-0.1385
-0.2006
-0.0551
-0.1138
0.6215
-0.4127
0.7606
-0.2249
-0.0864
-0.3025
0.4351
0.1692
-0.625
0.5884
0.4266
0.1735
_
-
CAR (%)
0.1914
0.4532
-0.0784
-0.1767
-0.0459
0.2391
-0.2758
-0.3896
-0.6792
-0.8632
-0.437
-0.5755
-0.7761
-0.8312
-0.945
-0.3235
-0.7362
0.0244
-0.2005
-0.2869
-0.5894
-0.1543
0.0149
-0.6101
-0.0217
0.4049
0.5784
0.26
1.26
109
AR (%)
0.1105
-0.2106
0.3799
0.0714
-0.2019
-0.0327
0.212
-0.1518
-0.0527
0.2914
-0.388
-0.0601
0.1153
0.0982
-0.2046
-0.0942
0.1831
0.1065
-0.2286
0.0762
-0.5125
-0.1364
0.2088
-0.1165
0.059
0.2104
-0.1836
CAR (%;
0.4605
0.2499
0.6298
0.7012
0.4993
0.4666
0.6786
0.5268
0.4741
0.7655
0.3775
0.3174
0.4327
0.5309
0.3263
0.2321
0.4152
0.5217
0.2931
0.3693
-0.1432
-0.2796
-0.0708
-0.1873
-0.1283
0.0821
-0.1015
0.21
: L.40
74
returns are significantly negative and positive for the disappointing and sat-
isfactory earnings groups, respectively. The strong downward trend for the
disappointing earnings group and strong upward trend for the satisfactory
earnings group after the earnings information release indicate that A-share
markets failed to adjust instantaneously to the new EPS information. As
a result, abnormal returns can be earned by trading on the EPS informa-
tion release. These results also indicate that Chinese stock markets are not
semi-strong efficient as far as the earnings announcements for the sample
stocks over the time period is considered.
In contrast to the findings on the A-share CARs, the B-share CARs for
the same disappointing earnings group tend to decline 11 days before the
announcement date, while becoming randomly positive and negative after
the earnings announcement. The average two-day announcement period
B-share abnormal return is merely -0.26% with a t-statistic of -1.26 for the
disappointing earnings group. The B-share CARs for the same satisfactory
earnings group tend to rise 15 days before the announcement date, while
also becoming randomly positive and negative after the earnings announce-
ment. The average two-day announcement period B-share abnormal return
is only 0.21% with a t-statistic of 1.04 for the satisfactory group. These
results suggest that B-share investors can predict the changes in earnings
better than A-share investors. The insignificant two-day announcement
period abnormal returns for both groups of firms and the observed ran-
dom behavior of post-announcement returns indicate that B-share markets
are semi-strong form efficient in the sense that B-share prices by and large
reflect the potential changes in event outcome.
Table 3 provides information about the time-series behavior of post-
announcement cumulative abnormal returns for two portfolios ranked in
the highest and lowest CAR_i
>0
quintile. As shown in the table, there
are some evidence of significant differences in CARs subsequent to earnings
announcement for the two portfolios, with A-share portfolios exhibiting
rather different pattern than B-share portfolios. Most of the differences in
CARs for A-share portfolios occur during the first 120 trading days, and
there is little evidence of statistically significant CAR differences beyond
180 trading days. However, most of the differences in CARs for B-share
portfolios occur after 180 days of trading, and there is no evidence of statis-
tically significant CAR differences during the first 180 trading days. Table
3 suggests that there might be incomplete adjustment for risk associated
with earnings administered in A-share markets, and such risk exists only
temporarily.
S
t
o
c
k

R
e
t
u
r
n
s

A
r
o
u
n
d

E
a
r
n
i
n
g
s

R
e
l
e
a
s
e
s

/
%

~

3
?

l
-
H

H

l
-
H

5

3

u

u

1

1

1

,

1

4

1

1

f
^

1

J

'
\

1

\

1

\

1

1

1

J

i

\

/

s
*
s
"
'
^
~

'

i
.

^
^
.

1

\

1

1

1
"

\

1

\

1

J

.
1

\

'

,
'

/

1

1

1

1

c
n

C
h

^

\
f
\

e
n

^
H

T
1

1

1

1

i
'

e
i

1

1

I
1

1

1

i

i

i

o
n

i

i

w
Day
i

e
-
i
Event
i

i

a

a

3

o

c

c

<

s

'
3

S
3

H

o

s

O

O

C

.
a

<
!

3

o

0
0

b
o

f
a

C
n

C
N

<

O

<

(
%
)

u
j
n
q
-
s
y

t
B
T
i
u
o
u
q
y

3
A
T
^
e
-
[
r
u
i
r
r
v
)

3
3
0

A
c
c
o
u
n
t
i
n
g

I
n
f
o
r
m
a
t
i
o
n

a
n
d

S
t
o
c
k

P
e
r
f
o
r
m
a
n
c
e

J
3

W

m

a

3

o

a

a

<

a

T
3

a

3

o

s

o

a

<
!

3

E

(
$
)

u
j
n
^
g
y

f
e
i
m
o
u
q
y

S
A
T

}
B

p
r
o
m
;
)

Why Domestic Investors Over-react to Earnings Release? 331
8.5 Why Domesti c Investors Over-react to Earnings Re-
lease?
We believe that three reasons have contributed to the domestic investors'
over-reaction to earnings announcements: First, government officials and
managers may involve in insider trading of A-shares. Second, most A-share
investors are short-term investors who are more likely to speculate based
on sentimental factors. Third, Chinese stock markets are segmented both
in trading and financial reporting requirements.
8.5.1 Gov ernment and Management as Informed Inv estors
Unless markets are taken to be informationally efficient, the unexpected
variation in market return reflects the joint effect of two hypotheses: mar-
ket efficiency and information content. Testing for the information content
of accounting numbers, or any other news, is contingent on satisfying mar-
ket efficiency. As indicated earlier in this chapter, emerging Chinese capital
markets are not known to be efficient because of the numerous institutional
frictions imposed by regulatory processes and unequal access to informa-
tion. Some of the institutional features that characterize Chinese stock
exchanges are relevant to this issue.
First, the government remains the majority shareholder, and a sizable
percentage of the privatized portions of A-shares are owned by state offi-
cials, party members and other insiders. Because laws restricting insider
trading are not known to be enforced, it is very likely that insiders act on
the information prior to its public release and, as a result, the information
content of disclosure would have been impounded in prices prior to the
public release of annual reports.
Second, given that China is at the "take off" stage of economic de-
velopment and growth, which averaged more than 8% a year since 1979,
financial statements and short-term earnings may not be good indicators or
predictors of future performance. The history of trading shares to absentee
investors in a public forum and of disclosing information about business
firms' performance has been too brief to develop association measures and
rules-of-thumb connecting reported numbers and subsequent events.
The manager of a firm is assumed have inside knowledge or superior
information, i.e., there is asymmetric information concerning various com-
ponents of earnings. Other than macroeconomic variables, the outside in-
vestors are expected to depend on the managers to disclose the information.
If the managers disclosures are informative and revised frequently over the
interim before earnings announcements, there will be no or little earning
surprises or significant stock market reactions. If there is an adequate in-
centive for disclosure, the manager will disclose voluntarily. Otherwise,
regulatory agencies, stock exchanges, the governing body of the accounting
profession have to compel the managers to disclose information involuntar-
iiy-
Without the necessary ingredients for market efficiency and allowing
equal access to information prior to using it in trading, it is not defensible
to conclude that the statistical results obtained for A shares are consis-
tent with market efficiency. These results are consistent with information
being reflected in security prices when any trader, including insiders, acts
on it. Given the information environment described previously, a scenario
can be advanced that insider trading and manipulation of trade by officials
might be responsible for impounding accounting information into share
prices prior to making public disclosure. Another scenario is that the ob-
tained results are a statistical artifact because the findings about A-shares
are consistent across the two stock exchanges. Since these explanations
cannot be ruled out given this analysis, we are unwilling to conclude that
some of the observed findings reflect the reaction to accounting numbers.
8.5.2 Investors' Sentiments
A-share markets lack alternative information sources other than the pub-
lished accounting reports such as earnings forecasts by financial analysts
and high-quality independent audits. The number of established invest-
ment specialists and institutional investors (e.g., mutual funds and pension
funds) is quite small. Most transactions are executed by individual investors
with limited access to information other than publicly available accounting
numbers. As a consequence, price may be less informative and accounting
information may contain more surprises in the A-share markets.
9
9
To test t he hypothesis t hat earnings announcements contain more surprises to A-share
investors t han to B-share investors, we need to compare earnings forecast errors of t he
two groups of investors. Although dat a on earnings forecasts of some Chinese firms by
international financial analysts are available through I / B/ E/ S International, earnings
forecasts by domestic Chinese analysts are not available at t he time this paper is written.
Further research on this topic needs to be conducted.
Why Domestic Investors Over-react to Earnings Release? 333
Most individual A-share holders have a short-term investment horizon.
Long-term investors are few and far between. The average turnover for A
shares in 1996 in the Shanghai and Shenzhen securities exchanges is 45 and
39 trading days, respectively.
10
With short-term horizon, investors tend to
be more concerned with current period accounting earnings and dividends
and may over-react or under-react to accounting information mechanically
to revise their beliefs about future profitability. Therefore, it is more likely
to observe a contemporaneous association of accounting information with
A-share stock returns.
From our data, approximately 50% to 60% of all listed firms in 1994
disclosed accounting reports to the public, with this percentage increasing
to over 90% in 1997, suggesting an improved reporting system. The annual
volume of trade for A-shares was about 20 times that of B-shares in 1997.
At average currency conversion rates of 1997, this volume amounts to about
US$40 billion in 1997 as compared to US$2 billion in December 1994. The
ratio of A-shares to B-shares during 1998 was about 52% and the corre-
sponding ratio in 1997 increased to about 65%. Thus, the number of listed
firms has increased in 1997 over 1994 and A-shares were traded much more
heavily in 1997 as compared to 1994. In addition, the SHSE publication
reports that turnover rates for A- and B-shares are 20 and 6 in 1997 and 18
and 3.5 1998, respectively. These ratios reveal the extent to which B-shares
are thinly traded as compared to A-shares. As a result, analysis of the
association between unexpected returns and accounting information is not
likely to reveal the true linkage between accounting information and prices
of B-shares.
8.5.3 Market Valuation and Transparency
Third, Chinese stock markets are segmented both in trading and reporting
requirements. Ang and Ma (1999) suggest that the extent of transparency
of Chinese stocks is generally quite low. For instance one may suggest
that requirements of the B-share listing have made the B-shares to appear
more transparent than the A-shares, that larger firms have more incentive
and sophistication to voluntarily disclose earnings information. We may
also argue that when there is little information that is made available,
differences in the opinions among market participants may also tend to
10
Source: Annual statistics published by the Shanghai and Shenzhen stock exchanges.
be high. And if less relevant information is available, some investors may
respond to rumors and other unreliable information, resulting in unusually
high price movements and trading volume.
On the other hand, international B-share investors are typically large
institutional investors, such as mutual funds and pension funds. They are
the key force to drive B-share markets up or down. Actions taken by these
institutional investors usually are not based on liquidity reasons, but based
on information revealed by firms that issued securities. Such a case is par-
ticularly true for firms in which financial and operational situations are in
question. Furthermore, international investors require more up-front infor-
mation and that this information is provided, in part, by the publication
of much more detailed and informative balance sheets for B-share investors
than for domestic A-share investors. International investors also have the
capability to conduct more sophisticated analyses.
In summary, Chinese A-share markets are considerably speculative while
B-share markets are relatively more efficient in terms of the reaction to
stock prices during and after earnings announcements. Domestic Chinese
investors do not seem to completely understand the true nature of the
equity market. Following a common trend, most individual investors are
blind to a stock's fundamental or trade stocks without using firms' financial
information correctly. Moreover, some inconsistencies regarding account-
ing standards and information disclosures exist. These suggest that it is
necessary to implement educational programs about securities markets for
individual investors and construct a set of consistent regulations in disclos-
ing accounting, auditing, and other financial information.
Chapter 9
Internationalization of Chinese Stock
Markets
9.1 Foreign Investment in Domesti c B-Share Market
Hoping to attract more foreign investment to fund its economy, China
started issuing B-shares for foreign investors on the Shanghai Securities
Exchange (SHSE) and Shenzhen Securities Exchange (SZSE) soon after is-
suing A-shares for domestic investors.
1
B-shares were issued not by public
offering but solely through international placement by going on road-shows
to meet international fund managers at major financial centers.
Because B-shares were privately placed, most of the shares were in the
hands of institutional buyers. This resulted in two problems: one is a lack of
liquidity in the secondary market since most B-share buyers are big equity
investment funds who tend to hold for long-term gains and their one-time
purchasing lots often exceed a million; the second problem is the incon-
venience in the trading venues and inadequate disclosure of information
about B-share companies. As a result, trading becomes quite thin. The
daily turnover averaged only US$ several millions in SHSE and in SZSE.
The B-share Index is below issuing price, 48.73 on October 21,1996, against
an average of 50.2 on the SHSE; and 89.58 the same day in SZSE, with an
average of 82.669 (close). The Price Earning (P/ E) ratio is about 5-7 times,
much below that of A-shares and H-shares.
2
x
In 1991, t he SHSE and SZSE began to offer B shares, providing foreign investors with
a legal channel to invest in China' s equity markets.
2
Unlike B-shares, A-shares were traded by many individual players and fewer institu-
tional investors. Individuals can be quite speculative in the domestic market. Market
price movement is volatile and market turnover is generally small.
335
336 Internationalization of Chinese Stock Markets
9.1.1 Market Micro structure Issues
B-shares are offered and traded on the SHSE and SZSE, which designate
domestic or overseas securities dealers as specially licensed brokers to accept
foreign investors' consignment for trading. These exchanges began to offer
special seats for B-share trading in 1994 and overseas securities dealers were
allowed to engage in floor trading via the seats. By the end of 1998, 106
companies had issued B shares with a total of 9.589 billion shares issued
and a total of US$ 4.745 billion capital raised.
Any subscription of B-shares must be processed through authorized lo-
cal securities or international financial institutions approved by relevant
authority for the purpose of dealing in B-shares. At the time of subscrip-
tion, the subscriber should by way of evidence present his identity card
or passport or, in case of a legal entity, its registration certificate. If any
holder of B-shares holds more than 5% of the total issued share capital of
the issuer, such holding must be reported to and approved by the People's
Bank of China (PBOC).
All B-shares are centrally listed at the SHSE and SZSE, and transac-
tions are carried out through the computerized trading system. Orders are
matched automatically by computer on price, time and sequence priority.
Foreign investors are required to buy and sell B-shares through authorized
foreign brokers who in turn instruct the approved SHSE and SZSE brokers
actually to initiate trades. Trading outside the exchange is prohibited.
Every investor dealing in B-shares must open a B-share account directly
or through securities companies at the SHSE or SZSE. Foreign investors
shall present their B share account passbooks and their valid certificates
of identity and complete the dealing instruction forms when they deal in
B-shares in China. Foreign investors conducting B-share transactions from
overseas are required to do so through a foreign securities company.
Short selling is prohibited. Newly purchased B-shares cannot be sold
before the settlement and registration procedures for their purchase are
completed.
The rate of exchange between U.S. dollar and RMB to be applied in
connection with any issue or trading of B-shares or payment of dividend
shall be the RMB to US dollar weekly weighted average conversion rate
as quoted by the Shanghai Foreign Exchange Adjustment Center for the
previous week.
Transactions in B-shares are generally effected on the third business day
Foreign Investment in Domestic B-Share Market 337
after the date of transaction, i.e., T+3. A "business day" means any day
on which the B-share clearing banks, the SHSE, the SZSE, and banks in
New York are open for business concurrently. The exchanges may appoint
banks approved by the PBOC to handle the clearance and settlement of
all B-share transactions. However, for each issuer, only one of them will be
involved.
The following transaction costs are levied on each purchase or sale of B
shares in Shanghai and Shenzhen (calculated in RMB but payable in US
dollars only):
In Shanghai, brokerage fees shall be charged at 0.6% of the total trans-
action value. For transactions more than RMB 100,000, the brokerage fees
may be reduced but not less than 0.3%. Transfer fees are set at 0.1% of
the total par value of the shares transferred. 0.3% of the total transaction
value is levied as stamp duty fee.
The brokerage fee for a B share transaction in Shenzhen is 0.6% of
total transaction value if conducted directly by an authorized broker in
Shenzhen, or 0.7% of the transaction value if conducted by an authorized
foreign broker. The stamp duty is 0.3% of the total transaction value. The
exchange transaction levy is 0.1% of total transaction value.
Under the Regulation for Foreign Investment Enterprise (FIE), the For-
eign Enterprise Income Tax Law of the PRC and the Individual Income Tax
Law (the IIT Law), dividends received by a foreign enterprise with no es-
tablishment in the P.R.C. or by a foreign individual are generally subject
to tax at a flat rate of 20%. However, this tax on dividends is exempted
if the payer is a FIE. Pursuant to the Shanghai and Shenzhen joint stock
company regulations, if more than 25% of the total share capital is owned
by foreign investors subsequent to the B-share issue, the company qualifies
for the same privileges as a FIE.
According to existing P.R.C. laws and practices, any capital gains re-
alized on the sale of an interest in a FIE by foreign enterprise with no es-
tablishment in the P.R.C. will generally be classified by the State Taxation
Bureau as income and will be subject to a taxation rate of 20%. However,
the rate of taxation will be reduced to 10% in the case of gains derived from
source in the Special Economic Zones, the Economic and Technological De-
velopment Zones, the Open Coastal Cities and the Coastal Open Economic
Zones.
Pursuant to the Rules and Regulations of Shanghai and Shenzhen on
B Shares, all dividends, distributions and proceeds of sale derived from
338
Internationalization of Chinese Stock Markets
B-shares may be remitted out of China in their entirety. Capital gains
realized on the disposal of the B-shares may be remitted out of China after
payment of tax in accordance with laws. As a matter of procedure, all
the payments are subject to the approval of the State Administration of
Exchange Control (the SAEC).
9.1.2 Risks Involved
The risks faced by foreign investors in B-shares can be generally classified
into two categories. First, by the external risks, including the political,
economic, regulatory environment and foreign exchange risks; secondly, the
internal risks which are directly associated with the operation of the com-
pany, mainly the management, the standard of account and the quality of
the prospectus.
Despite the introduction of a number of market mechanisms into the
Chinese economy, central planning and the state sector still dominate the
economy. This gives the central government tremendous power to intervene
in particular sectors of the economy which may affect the share markets di-
rectly or indirectly. While there are a large number of rules and regulations
on joint stock companies and securities issues, they are not comprehen-
sive enough and sometimes conflict with each other. While the PBOC is
the authority-in-charge, many other institutions overlap in drafting, rule-
making power, and the approving process.
As mentioned above, any transactions for B-shares are calculated in
RMB but payable in US or HK dollars only as the Renminbi is not freely
convertible in international markets. The devaluation of RMB from time
to time has proved to be a significant exchange risk for B-share investors.
The management of listed companies is perhaps the greatest risk that
foreign investors have to face. The concept of "maximizing profits" is very
weak among Chinese managers. Furthermore, the accounts system in China
is different from international accounting standards. Chinese companies are
usually reluctant to disclose information when first issuing B-shares due to
competition. Some of the prospectuses issued are brief and are not capable
of disclosing sufficient information to protect investors.
China's hard currency dominated B-share market has long been slug-
gish because of its own drawbacks since its establishment in the early 1990s,
including its limited scale, low mobility, sharp fluctuations and limited ac-
cess to foreign investors. Anthony Francis Neoh, chief advisor of the CSRC
Overseas Listing of Chinese Stocks 339
revealed that the commission could adopt a qualified foreign institutional
investor pattern (as India's stock markets did), to allow foreign investors
easy access to the domestic A-share market. Foreign investors' involvement
in the Renminbi-b&sed A-share market would soon make the prices of the
listed A- and B-shares the samea prerequisite for the merger of the two
markets.
Before any merger, China's A-share stock markets in Shenzhen and
Shanghai will have to merge first, and the long-awaited high-tech or second-
board stock market, which is designed mainly for small and medium-sized
firms, could be launched in Shenzhen if the merger succeeds. But if it
fails, there would be a possibility of setting up two parallel second boards
at Shanghai and Shenzhen respectively. Nevertheless, the long-awaited
second-board market, to bolster China's high-tech industries, would be
launched before the end of the year or the beginning of 2001. Compa-
nies planning to list in the second market would have much lower require-
ments than their A-share counterparts. Indeed, Unilever, one of the world's
biggest producers of consumer goods, is trying to be the first foreign com-
pany to list domestic A-shares in China. A spokesperson with the Shanghai
headquarters of Unilever (China) said:
We hope that our customers here in China can also be-
come our shareholders, as an incentive for us to upgrade
our services and expand our market share.
But she also admitted that the company was still waiting for policy ad-
justments by the Chinese government, which would clearly map out the re-
quirements for listing. The permission would, however, require the speeding
up of the merging of the A- and B-share markets, which is still proceeding
slowly. Today, most of the A-share stocks are state-owned companies and
foreign investors are still not allowed to list or trade A-shares.
9.2 Overseas Listing of Chinese Stocks
According to the CSRC documents, overseas listing is an important policy
for the reform and opening-up of Chinese enterprises to the outside. From
the listing of Qingdao Beer on the Hong Kong Stock Exchange on July 15,
1993 to Yanzhou Coal IPO and dual listing in Hong Kong and New York in
April 1998, 43 enterprises listed overseas, raising more than US$10 billion.
Of the 43 firms, 31 were listed in Hong Kong, 8 were dual listed in Hong
Kong and U.S., 2 dual listed in Hong Kong and London, 1 was listed solely
in U.S. (see Table 9.1) and 1 solely in Singapore.
Some enterprises, after initial offering, go back to the international cap-
ital markets for re-financing. In 1997, Zhenghai Refinery issued US$ 200
million in convertible bonds, which were listed in London and Hong Kong;
Qingling Motors issued US$ 110 million in convertible bonds, listed on Lux-
emburg Bond Exchange and Hong Kong Stock Exchange (HKSE); Huaneng
International issued US$ 230 million convertible bonds, listed on New York
Stock Exchange and Luxemburg Stock Exchange.
9.2.1 Cross-Listing of H-Shares
Because the original thought from Chinese leaders on cross-listing was to
seek more foreign investment for Chinese state-owned enterprises, they did
not expect to let any Hong Kong registered company or any other coun-
try' s company list in return on Chinese domestic markets. There exists an
unwritten rule of "one-way traffic"only to allow for Chinese companies to
list on the HKSE or other overseas stock exchanges but not the other way
around. On the secondary market, the rule holds as domestic investors are
restricted to A-shares and legal-person shares and prohibited from holding
B- and H-shares.
As mentioned in the previous section, the growth in B-share markets
in mainland China is extremely limited due to illiquidity and deep price
discounts. As a natural step, China started tapping Hong Kong market
by issuing H-shares. Painstaking efforts were made to prepare by various
decision-making government institutions consulting with foreign investment
bankers, law firms, accountants, and especially with the HKSE counter-
parts. In early 1993, the CSRC announced the first batch of 9 Chinese
companies to be listed in the HKSE.
3
After six months of due diligence and
prospectus documentation, Tsingdao Brewery (HKSE: 0168) became the
first H share to list on the HKSE on July 15, 1993. At the same time, the
company offered American Depository Receipt (ADR) in the NYSE. The
dual offerings brought immediate attention of the world's investment bank-
3
The requirements for a listing in the HKSE include a minimum market capitalization of
HK$50 million (U.S.$6.5 million), no less t han three shareholders for each HK$1 million
of the issue, a minimum of a total of 100 shareholders, and the use of international
accounting standard (IAS).
O
v
e
r
s
e
a
s

L
i
s
t
i
n
g

o
f

C
h
i
n
e
s
e

S
t
o
c
k
s

1
3

C
O

C
O

i
n

0
5

O
}

C
M

<
M

C
M

C
D

0
0

C
O

0
5

,
_
,

1
>

1
1

C
O

C
O

C
O

C
M

42.124 785
C
O

1
0

0
0

1
>

C
O

0
0

C
O

0
1

0
0

C
M

8

l
O

C
O

^

_
,

0

C
O

C
M

|
>

O
l

O
l

C
O

C
M

^

C
O

0
1

C
O

8

2

" Comp No. oi
0
1

^

0

^
H

C
O

rocee I POp
0
0

1
0

C
M

T
-
H
ing ffer
O
ional Additi
C
O

Bon rtible Conve
0
5

T
f

O

1
1
ised
c
S

C
3

Capit Total
ing community. As a result, the IPOs of the first 9 P.R.C. companies were
all over-subscribed by hundreds of times. Kunming Machine (HKSE: 0300),
a relatively small H-share issuer with HK$65 million was over-subscribed
by more than six hundred times with over HK$40 billion funds raised. The
issuer thus had unexpected interest income equivalent to the company's
one-year profit. Table 9.2 presents a full list of companies that have issued
H-shares by the end of 1999.
Insofar as the primary market for H-share listings is concerned, a large
number of international investment banks actively underwrite H-share IPOs.
However, because of problems such as poor company disclosure and a lack of
transparent operations, some H-shares experienced significant price drops
during 1994 and 1995. Compared to the Hang Seng Index, the Hang Seng
China Enterprise Index fell far lower. One interesting example is Zhenghai
Refinery, a state-owned company in the second batch of H-shares. When
it started trading on December 2, 1994, its issue price was HK$1.46. The
state gave the company many subsidized loans, with only 4-5% interest
rate while the market rate was about 14-16%. When the state stopped
its subsidy by abolishing the low interest subsidized loans, its share price
nose-dived by more than one-half.
Table 9.3 presents IPO performance for some H-shares calculated using
the closing prices of the first, 30th, 60th, 90th, and 180th days. As shown
in the table, Beijing Yanhua Petrochemical closed the first day at a price
lower than its IPO price. Most firms traded at the same level of the offering
prices. The average IPO initial return was 2.51%, with a standard deviation
of 5.6%. China Eastern Airline had the highest price increase of 15.28% in
the first day of trading, and Beijing Yanhua had the largest price decline
of -4.35%.
The average 30-day holding period return was 0.9%, with a standard
deviation of 11.95%. The highest holding period return was Shanghai
Petrochemical (20.16%) while the lowest was Yanzhou Coal Mining Co.
(-13.14%). The average 60-day holding period return was 13.72%, with a
standard deviation of 46.14%. Beijing Yanhua had an extraordinary 60-
day holding period return of 117.39% as opposed to Yanzhou Coal Mining
Co. whose return was a disappointing -31.37%. The average 90-day hold-
ing period return was 5.37% with the highest of 70.17% in China Eastern
Airline and lowest of -38.1% in Yanzhou Coal Mining Co.. The average
180-day holding period return was -7.03% and the standard deviation was
43.44%. The highest return for 180 holding days was 93.72% in Shanghai
Overseas Listing of Chinese Stocks
343
Table 9.2 Companies Listed on the Stock Exchange of Hong Kong as of December 1999
Company
Tsingtao Brewery
Yizheng Chemical Fiber
Tianjin Bohai Chemical
Dongfang Electrical Mach.
CATIC Shenzhen Holding
Luoyang Glass
Qingling Motors
Zhenghai Refinery
Chengdu Telecommunications
Harbin Power Equipment
Jilin Chemical
Jingwei Textile Machinery
Nanjing Panda Electronics
Guang Shen Railway
Shandong Xinhua Pharm.
China Eastern Airline
Shenzhen Highway Co.
Beijing Datang Power
Beijing North Star
Zhejiang Highway Co.
Jiangxi Cooper
Beijing Yanhua Petrochem.
First Tractor
Jiangsu Highway Co.
Northeast Electrical
Angang Steel
China Southern Airline
Szechuan Highway Co.
Guangzhou Pharmaceutical
Huaneng Power
Yanzhou Coal Mining
Shangdong Intl. Power
Great Wall Technology
List Date
July 15, 1993
March 20, 1994
May 17, 1994
June 6, 1994
July 2, 1994
July 8, 1994
August 17, 1994
December 2, 1994
December 13, 1994
December 16, 1994
May 23, 1995
February 2, 1996
May 2, 1996
May 14, 1996
December 31, 1996
February 5, 1997
March 12, 1997
March 21, 1997
May 14, 1997
May 15, 1997
June 12, 1997
June 12, 1997
June 23, 1997
June 27, 1997
July 6, 1997
July 24, 1997
July 31, 1997
October 7, 1997
October 30, 1997
January 21, 1998
April 3, 1998
June 30, 1999
August 5, 1999
Funds Raised (mil US$)
385
307
53
62
n/ a
118
169
184
58
145
18
30
67
82
35
86
212
466
219
441
216
33
195
491
60
187
43
179
47
n/ a
26
13
58
Petrochemical.
A notable phenomenon is that prices of the some H-shares behave quite
differently from their A-share counterparts. The prices have shown no
sign of convergence. This is probably because that A-shares are issued and
3
4
4

I
n
t
e
r
n
a
t
i
o
n
a
l
i
z
a
t
i
o
n

o
f

C
h
i
n
e
s
e

S
t
o
c
k

M
a
r
k
e
t
s

O

.
o

E
2

o

o

0
0

Q

Q

I
-
.

P
L
,

a

a

o

O

I
M

C
N

C
O

C
O

0
0

C
N

6
?

o

0
0

t
o

1
-
H

d

(
M

6
?

C
O

C
O

C
O

C
O

l
O

C
O

l
O

d

i
n

C
N

o

C
O

C
N

1
^

t
o

l
O

C
O

C
O

C
M

C
O

C
N

1
^

o

C
N

^

K
=

^

<
*

?

-
*

l
O

o

O

d

c
o

o

C
N

o

C
N

0
0

C
O

C
O

d

C
O

/
5

O
i

0
0

l
O

0
0

0
0

t
^

C
N

d

C
N

0
0

C
N

o

0
0

C
O

0
0

0
0

o
o

6
?

O
J

C
O

C
O

O
S

(
3
2

C
O

O
i

C
O

(
C
N

r
-
l

i
n

c
o

C
O

C
O

o
o

C
O

q

o
i

i
l

C
O

f
e
?

i
n

f
e
?

C
O

C
O

6
S

C
O

0
0

d

S
?

t

c
o

i
n

c
s

n

J
>

c
a

o

c
u

i
p

-
"
3

o

J
3
O
J
J
3

S
3
j
3
j
3
.
a

c
u

3

j

.
a

i

a

traded only within China and there are no abitrage across A- and H-shares.
Shanghai Petrochemical serves as a good example. Its different share prices
show no sign of convergence, in the short term or the long term.
4
9.2.2 Red Chips Rising
Besides the H-share, there are two other types of cross-border listed com-
panies from mainland China to the HKSE. One was known as back-door
listing, which is usually conducted as a take-over by a mainland Chinese
company of a Hong Kong listed company in financial distress. After the
acquisition, the mainland company would inject its capital into this Hong
Kong "shell" company. In this way, the Hong Kong "shell" is used as a
listing vehicle by the mainland Chinese company. It had been a major
way for powerful Chinese companies to seek Hong Kong listings before the
relevant Chinese authorities formulated securities regulations. However,
beginning in 1992, any form of Hong Kong listing from mainland China
must go through various official procedures to get an approval.
The second type of cross-border listing takes place within mainland
Chinese companies registered in Hong Kong. Most of these companies have
been operating in Hong Kong for many years and Hong Kong investors are
familiar with their business scope and sales. Because of their strong busi-
ness background with mainland China, Hong Kong investors called these
companies the Red Chips. A subset of the Red Chips is the military-backed
mainland civilian shell companies. Only the top-level People's Liberation
Army (PLA) organizations have the power to set up companies abroad.
These companies are referred to as PLA Red Chips in Hong Kong.
In general, the class of shares that depends on China operations for
growth, but is not backed by a Chinese business or government entity,
is usually labeled as China concept (e.g., red chips). Strictly speaking,
red chips are extensions of governmental organizations operating within
investment and cash-raising guidelines laid down by a holding company
and a controlling government body. Red chips have sought listings in Hong
Kong, sometimes with and sometimes without government consent, and
have grown through a consistent pattern of acquisitions. Red chips that
did not seek formal approval from supervising bodies typically have gained
4
Not e t hat a stronger correlation does exist between two domestic market prices t hat
may show t hat the cointegration movement exists between A-shares at the SHSE and
SZSE (Su and Fleisher, 1999c).
their listing in Hong Kong through a back-door entry into an inactive listed
company. There were a total of 70 red chips by the end of 1997.
The standard approach of a mainland firm making its move for expan-
sion and foreign capital has been to: first, establish a Hong Kong sub-
sidiary; second, borrow funds from banks to finance existing operations of
subsidiaries and associates; and third, find a way to gain listing in Hong
Kong. China Resources Enterprises Ltd. is a model of how red chips
are knitting together the economies of Hong Kong and China. It borrows
money, or raises it on the HKSE, and uses the funds to invest locallyred
chip spending on real estate has buoyed Hong Kong property pricesand
on the mainland. Last year, China Resources Enterprises Ltd. spun off a
subsidiary called China Resources Beijing Land, offering investors in Hong
Kong the opportunity to invest in property in Beijing, and reaped $200
million in profit from that investment.
5
When purchasing red chips, investors in Hong Kong believed they were
buying assets in a Chinese economy that had been growing at double-digit
rates. But in most cases, they were actually buying holding companies
with a range of disparate assets-a piece of highway here, a shrimp farm
there. Yet investors were willing to pay a premium. Some red chips com-
panies performed so well that they were included in Hang Seng Index of
33 major companies.
6
One of them was known as CITIC Pacific, a sub-
sidiary of China International Trust and Investment Corporation (CITIC)
of Beijing. CITIC Pacific was backed by the mainland Chinese govern-
ment which invested heavily in Hong Kong property, infrastructure such as
the new airport project, container port, and Hong Kong airlines. Another
company was China Resources Enterprises Ltd., whose parent was the PRC
Ministry of Foreign Trade and Economic Relations. China Resources Enter-
5
Wi t h H-share listings, however, Beijing's watchful eye monitors the valuations for the
assets of the new company through the auspices of the State Asset Management Bureau
(SAMB). The discipline established by the practice has prevented massive sales of
China' s assets. Therefore, some red chips have sought the approval of the State Council
for new share issues. This is usually considered a necessity for assets held at t he national
level and t hat are above US$30 million. China Resources Enterprises Ltd., for instance,
had to seek approval for the acquisition of a US$44 million stake in a Beijing property
company in November 1994.
6
Although there was no official index on t he Hong Kong Stock Exchange to measure
the red chips, an unofficial index of 35 mainland stocks compiled by Peregrine showed
a 101% gain over 1996. These shares dramatically out-performed t he blue chip Hang
Seng index, which climbed a respectable 33.5% to end the year.
prises' main business is to provide Hong Kong with daily supplies ranging
from drinking water to fresh vegetables. When it went public, its shares
were over-subscribed by an amount equivalent to twice the money supply
in Hong Kong. All these red chips had shown blockbuster performance in
their share prices until the second half of 1997.
In June 1997, trading in red chips drove Hong Kong's Hang Seng index
above 15,000. About 50 red chips made up 11 percent of stock market
value, but had accounted for about 25 percent of trading for the year.
China, for its part, was alarmed at all the confidence demonstrated. It
decided to impose listing restrictions, requiring all share issues of red chips
to be reported to the CSRC. This was intended to impose quotas on stock
issues by Chinese companies in foreign markets. Only companies that kept
their assets at home in China for at least three years would be permitted
to issue new shares. Then only a few months later, on October 23, 1997
(the Black Thursday), Hong Kong's Hang Seng index fell over 1200 points,
plunging below 10,000. That was the largest point drop in its short 14-year
history. The stock market was down 35 percent for the month. Red chips
were down 50 percent for the month. A new red chip, China Telecom,
making its public debut on Black Thursday, failed to meet its issue price.
In today's Hong Kong, red chips account for 8% of the market's capi-
talization. If an investor pastes on shares owned by mainland investors in
local companies and the H-shares' total capital, China may control up to a
fifth of the market. The future offers more of the same. If the China boom
continues, red chips will make the territory a considerably larger financial
market than it is today-10 to 30 times bigger, according to Richard Wong,
Director of Hong Kong University's Center for Economic Research. "At
that point," he says, "the distinction between Hong Kong and China will
diminish."
9.2.3 Ov erseas Listing in the U.S., U.K. and Singapore
Companies seeking a listing overseas must satisfy two requirements. First,
they must qualify for listing according to the standards set forth by overseas
exchanges. Second, they must comply with the registration, accounting
and disclosure requirements stipulated by the relevant securities regulatory
authorities such as the Securities Exchange Commission (SEC) in the U.S.
and the Ministry of Finance in the U.K..
To be listed on the major U.S. stock exchanges, a company is expected
to meet certain minimum qualifications. Table 9.4 presents listing require-
ments for NYSE, NASDAQ, and AMEX. For example, to be listed in NYSE,
a company needs to have at least 5,000 shareholders worldwide with 2.5 mil-
lion shares outstanding. The total market value of the shares should exceed
$100 million. In addition, the total operating cash flow has to exceed $100
million in the past three years and $25 million for each of the most recent
two years.
To get over the accounting standard and other hurdles, a foreign com-
pany can list its shares through Depository Receipt (DR) program. J.P.
Morgan created the first American DR (ADR) in 1927 to allow Americans
to invest in the British retailer Selfridge. An ADR is a stock representing a
specified number of shares in a foreign corporation. ADR's are bought and
sold in the American markets just like regular stocks. An ADR is issued
by a U.S. bank, consisting of a bundle of shares of a foreign corporation
that are being held in custody overseas. The foreign entity must provide
financial information to the sponsor bank. ADRs are listed on either the
NYSE, AMEX, or NASDAQ.
There are various types of ADR programs, including Level I, Level II,
Level III, and Rule 144A/GDR.
Level I ADRs: They are used when the issuer is not initially
seeking to raise capital in the U.S. markets or does not wish to,
or can't, list its ADRs on an exchange. A Level I ADR program
offers an easy and relatively inexpensive way for an issuer to gauge
interest in its securities and begin building a presence in the U.S.
securities markets.
Level I ADRs are traded in the over-the-counter (OTC) market,
with bid and ask prices published daily and distributed by the Na-
tional Daily Quotation Bureau in the pink sheets. Prices may also
be posted on the OTC Bulletin Board (OTCBB).
7
Level I ADR
programs currently require minimal SEC registration: The issuer
seeks exemption from the SEC's traditional reporting requirements
under Rule 12g3-2(b). With that exemption, the company agrees
to send to the SEC summaries or copies of any public reporting
7
Due to t he SEC' s recent permanent approval of the OTCBB system, effective April
1, 1998, all non-U.S. equity securities, including ADRs, must be registered with the
SEC pursuant to Section 12 of the 1934 Securities Exchange Act to remain eligible for
quotation in the OTCBB system.
O
v
e
r
s
e
a
s

L
i
s
t
i
n
g

o
f

C
h
i
n
e
s
e

S
t
o
c
k
s

X

a

o

7
Z
1

s

*

0
3

O

o

o
_

o
"

f
-
6
0

<

Q

C
O

2

1

a

o

_
,

o

O

.

C
M

-
*

i
-
H

9
3

H

C
O

T
3

O

.

=
3

q

i
q

1
1
5

(
N

J
f
i

9
3

6

9
3

f
f
i

;

a

^
s

o

S
3

6

o

o

o

H

H

C
S

H

c
o

der
"
o

J
3

+
3

o

T
3

S

o

v
.

a
;

C
O

j
i

c
o

_
c
j

^
3

=
1

lue
g

3

A
S

0
3

3

_
o

^

d

C
O

9
>

>
>

C
O

4
J

c
o

c
S

<
D

J
5

-
*
>

^
4

s

0

o

a

x

C
O

0
1

f
t


.

4
^

c

C
D

rec
-
*
J

C
O

O

3

C
N

<
D

-
t
-
s

O

-
G

o

C
D

C
D

3

o

o

c

0
3

+
J

i

0
)

1
*
4

o

c
p

-
(
J

C
f
l

S
3

<
D

>
>

C
M

C
O

<
0

a

o

c
o

-
a

O

B

<

o

o

q
=
!

S
B

J
3

-
B

8

o

o

C
D

C
D

-
u

-
u

C
O

C
O

b
O

b
O

<
u

a
)

u

u

b
O

b
o

b
O

b
O

r
t

e
x
,

a
,

<
;

a
,

<
;

<
;

documents required in its home market (including documents for
regulatory agencies, stock exchanges, or direct shareholder commu-
nications). The depository, working with the issuer, also files the
Form F-6 registration statement with the SEC in order to establish
the program.
Level II ADRs: They can be listed and traded on one of the
U.S. securities exchanges (including NYSE, AMEX, and others) or
quoted on NASDAQ, thereby offering higher visibility, more active
trading, and greater liquidity in the U.S. market.
Level ff ADR programs must comply with the full registration and
reporting requirements of the SEC's Exchange Act, which entails:
(1) Form F-6 registration statement; (2) Form 20-F containing de-
tailed financial disclosure about the issuer, including financial state-
ments and a reconciliation of those statements to U.S. GAAP; (3)
Annual reports and any interim financial statements submitted on
a regular, timely basis to the SEC.
Level III ADRs: An issuer floats a public offering of ADRs in the
U.S. and lists the ADRs on one of the U.S. exchanges. This allows
the issuer to raise capital and leads to much greater visibility in the
U.S. market.
Level III ADR programs must comply SEC's full registration and
reporting requirements including Form F-6, Form 20-F, Annual re-
ports, and Form F-l (to register the equity securities underlying
the ADRs that are offered publicly in the U.S. for the first time,
including a prospectus to inform potential investors about the com-
pany and the risks inherent in its businesses, the offering price for
the securities, and the plan for distributing the shares).
Rule 144A ADRs: Many companies seek to raise capital in the
U.S. markets privately by issuing restricted securities under Rule
144A, which do not require SEC review. Rule 144A facilitates the
trading of privately placed securities by sophisticated institutional
investors (also known as Qualified Institutional Buyers who must
own or manage at least $100 million in securities).
Global Deposi tory Receipts (GDRs): GDRs allow issuers to
raise capital in two or more markets simultaneously, thus broaden-
ing their shareholder base. They can be settled outside the U.S.
(using a link connecting the two major European clearance and set-
tlement facilities, Euroclear and Cedel, with the major U.S. clear-
ance and settlement facility, the Depository Trust Company, or
DTC) and can be traded in the Rule 144A private market. Under
SEC Regulation S, securities offered or sold to investors outside the
U.S. are not subject to SEC registration requirements.
The China listing experiment started back in October 1992 when Bril-
liance China Automotive surprised the world by listing directly on the
NYSE through a Bermuda-based shell company. American investors ini-
tially backed the offering, and the share price doubled in the first month
before falling back to the initial offering price. Shanghai Ek Chor and China
Tire came to market with similar hybrid offerings. However, without a do-
mestic investor base to buy stock when fund managers want to sell, or vice
versa, Chinese stocks seem to lose their allure. For example, even with the
innovative N-Share structure used by Huaneng Power International and
Shandong Huaneng Power to list ADRs, because of no underlying shares
traded in China, the problem of liquidity remains.
8
These shares cannot
match the liquidity and performance of companies with ADR dual listings in
Hong Kong and New York. Shanghai Petrochemical's $343 million offering
in 1993, and Jilin Chemical's $200 million offering in 1995, are considered
to be the most successful of China's capital raising.
To seek GDR listing in the London Stock Exchange (LSE), a company
needs to have a trading record of at least three years. The market value
of the GDR must exceed #00,000 (US$1.15 million). At least 25% of the
GDR should be held by shareholders outside the company. A company can
choose U.K. GAAP, U.S. GAAP or IAS as its accounting standard. To
be listed on the main board of the Stock Exchange of Singapore (SES), a
company needs to have a minimum of three years of track record. At least
25% of the issued and paid-up capital must be in the hands of at least 1,000
shareholders. The paid-up capital (shareholders' equity) should be no less
than US$18.5 million. The cumulative pre-tax profits should be at least
US$10 million during the previous three years and a pre-tax profit of at
least US$2 million for each of the two most recent years. A company must
8
The N-share structure was created to assist issuers in developing markets in accessing
international investors while retaining control over ownership in the home market. N-
shares are not listed in the home market and cannot be purchased by local investors. N-
shares are created by the issuer to provide a means of raising capital among international
investors. They are held in custody by the depository banks.
use IAS as its accounting standard. Table 9.5 presents Chinese companies
listed on the NYSE, LSE and SES by the end of 1999.
Signs have shown that China's high-growth firms may expect a promis-
ing future in raising funds from overseas stock markets. The Chinese gov-
ernment is actively seeking to expedite the formation of a venture capital
investment system and set up an alternative stock market for growth firms.
The LSE has launched its Technology Market and Alternative Investment
Market to suit the fund-raising needs of high-growth firms and small com-
panies. In 1999, a number of Chinese Internet and e-commerce companies
listed their shares on the NASDAQ and the HKSE. The successful listing
of these high-tech companies overseas paved the way for more listings by
other growth firms.
9.3 Concerns That Emerged
Overseas listing is an important policy for the reform and opening-up of
Chinese enterprises to the outside. Although the overseas listing had paved
the way for further expansion of capital raising abroad, a strong ideological
barrier still hampered China from listing its socialist companies on a cap-
italist capital market. By the end of 1992, reform in the rural areas had
pushed many SOEs to become profit-losers. They were badly in need of
capital investment to restructure and grow. But they were already heavily
indebted to domestic and foreign lenders.
Zhu Rongji, the vice-premier and governor of the PBOC at the time,
decided to resort to equity financing from overseas capital markets. Begin-
ning in December, 1992, a policy was announced to allow the first 9 state
companies to list on the HKSE as an experiment for overseas listing. This
was a political decision and the principle of selecting companies for listings
was very simple: to choose the ones that are, in term of profitability, "nei-
ther bad nor good" and the ones that are, in term of size, "neither big nor
small". The rationale was that mainland companies with good profitabil-
ity do not need foreign funds and companies with bad profitability would
find it too difficult to raise funds through overseas flotation. Big strategic
companies are usually highly controlled by the state, which does not want
to lose its majority control. Many of international investment bankers were
involved in educating and bringing H-share companies to list on the HKSE.
The period of allowing PRC companies to list overseas was associated
C
o
n
c
e
r
n
s

T
h
a
t

E
m
e
r
g
e
d

C
O

a

o

-
a

<
u

t
o

'
3

T
3

B

5

o
o

^
f

o
o

o

s
o

m

d

o
o

^
-

,
_
,

i
-
H

r
t

O

^

O
)

N

C

C
O

(
O

N

O
)

H

S

H

H

H

C
S

C
S

a

a

b
e

q

a

a

a

H

a

a

a

H

H

n

n

n

.
.

T
O
W
W
C
O
C
O
C
O
C
O
C
O
C
O
W
W
W
O
T

P

T
3

C
D

l
^
j

O
i

c
j
i

2

>

O
l

O
S

H

H

O
l

'

t
o

C
O

^

S
P
-
3

<

o

M

.
&

O

C
O

S

f
N

-

^
"

r
i

C
N

S

3

3

(
j

*

*

-
^

s

s

3

c
S

S

o

O

c

o

o
3

C
D

o

*

g

c
p

I
I

s

j

I

O
i

f
f
l

(

c
S

R

M

t
l
O
-
O

2

c

C

c
S

<
8

S

^
3

J
3

C
O

C
O

f
f
i

.
3

<
o

3

>
.
M

.
5

I

3

1
5

g

i
l
l

R

S

.
S

3

-
R

O
H

<

a

.

t
s

s

^

o

,
.

c
o

R

c
S

:
=
>

-
S

'
5
3

^
3

m

o

o

O

a

I

o

J

a

Q

o

i
s



k
c
S

c
u

>

p
q

8
.

5

B
.
s
s

h
>

N

O

with the domestic enterprise reform carried out in large scale among 400,000
SOEs. The first batch of H-share listings on the HKSE was extremely suc-
cessful. Soon there came the second batch of 22 H-share candidate com-
panies at the end of 1993 and the third batch of 7 in 1995. The listing
venue included not only Hong Kong but NYSE, NASDAQ, and exchanges
in Canada, Australia, U.K., and Singapore. Issues that emerged include
market segmentation, foreign exchange control, and a lack of foreign par-
ticipation and competition in capital markets.
9.3.1 Market Segmentation and Foreign Exchange Control
The most important question that the Chinese government has to resolve
in the development of the stock markets is the role of foreign investors. At
present, the Chinese government attempts to control foreign investment by
effectively segmenting equity markets. Foreign investors are not permitted
to purchase A-shares, but they can invest in B-shares as well as companies
listed in Hong Kong, New York or other foreign markets. This system
has produced tremendous liquidity problems for the B-share market while
denying Chinese people and financial institutions the opportunity to invest
in some of the country's most promising companies.
After nine years of operation, the B-share market has a market capital-
ization of less than US$4 billion. It is so illiquid that most of the companies
can only sell B-shares at 80-90% discount relative to their A-share prices.
In fact, the average price discount is so large that many brokerage firms
believe that a large portion of the B-share demand has come from Chi-
nese citizens purchasing B-shares illegally through established accounts in
Hong Kong. As a result of their poor performance, B-share markets have
not played a very effective role in helping Chinese companies to raise ad-
ditional capital needed for expansion. There were only two new listings in
the Shanghai B-share market during the 1999 period compared to 8 and 35
during the 1997 and 1994 period, respectively. There were no new listings
in the Shenzhen market in 1999 compared to 5-10 during each year of the
early 1990's. Chinese companies predominantly raise capital is the A-share
markets.
9
The problem with the H-share market is that it has experienced extreme
9
For example, in 1999, there were 98 IPO' s and 117 A-share issues raising 87.7 billion
RMB of funds (US$10.5 billion) compared to only US$600 million raised from share
sales in the Hong Kong H-share market.
Concerns That Emerged 355
mood swings because of changing investor perception about the mainland
economy and the quality of companies offered. In 1997, the H-share market
was excited about the flotation of red chip companies by Chinese local
governments in Shanghai and Beijing. However, the shares subsequently
performed very poorly and attention shifted to the telecommunication and
technology sectors in Hong Kong, where prices have risen sharply since
1999. As a result of these mood swings, China has found it difficult to use
the external marketplace to privatize other companies in sectors, such as
energy and heavy industry.
Once H-shares are listed, trading tends to be more volatile than that
of other non-PRC shares on the HKSE. There is no cross-border arbitrage
regardless of price differentials. This is due to the fact that no free cross-
border movement of capital is allowed and most of the H-share companies
are single-listed in Hong Kong without a previous listing in the mainland
stock markets. The problem of market segmentation is fundamental in the
region. Su (2000b) suggests that there were no general linkages between A-
and B-shares or between A- and H-shares due to ownership restrictions and
market segmentation. But he finds that linkages between the two domestic
markets noticeably strengthened over time. He also finds that B-share
returns in the Shanghai market could be forecasted by market returns in
Hong Kong, when policy allowed foreign institutions to trade. The findings
confirmed that the policy-induced market segmentation hindered arbitrage
opportunities in the regional market.
Because of market segmentation, China is significantly under-weighted
in the Morgan Stanley index for global stock markets and thus unable
to attract adequate capital from investors attempting to achieve market
weights for their portfolios. Despite the fact that China's aggregate stock
market capitalization in the domestic markets and Hong Kong is equal to
about 1.5% of global stock market capitalization of US$34 trillion, China's
weight in the Morgan Stanley World Index is only about 0.39%. About
59% of this weighting also comes from the value of China Telecom, alone.
If China had an open market for A-shares as well as the companies listed
offshore, its weighting would probably rise to at least 1.0% of the Morgan
Stanley World Free Index. Meanwhile, China's own citizens have not had an
opportunity to invest in some of the country's most exciting new technology
and telecommunications companies because those companies are listed in
Hong Kong and New York, not on the domestic A-share market.
China has attempted to maintain a rigidly segmented securities market
because of concern about exchange controls. The Chinese government is
not yet prepared to introduce full capital account convertibility. However,
there are many examples of how a country can regulate capital inflows
without completely denying foreign investors access to the local stock mar-
kets. Taiwan and India require foreign institutional investors to register
with the government and report their transactions as they occur. Some
Indian companies also have parallel listings in New York and the off-shore
GDR market, where the shares are trade at a premium to their local coun-
terparts. Singapore, Korea and Thailand have official restrictions on the
level of foreign ownership of companies but permit foreign investors to pay
a premium in order to purchase shares from each other. Singapore Airlines,
for example, has a two tier market in which foreign investors typically pay
a large premium for the shares compared to local investors. The Singapore
government occasionally adjusts the ownership limits in response to mar-
ket conditions. However, the current two tier system has worked for several
years because it has been predictable and has not been widely circumvented
by either local or foreign investors.
Until recently, South Africa had a two tier exchange rate system which
permitted the rand for financial transactions, such as foreign portfolio in-
vestment, to trade at a 25-30% discount to the rand used in commercial
transactions. The existence of a separate financial rand helped to boost
foreign investment in South African equities at a time when the govern-
ment wanted to promote foreign investment without losing control over
the exchange rate used to denominate trade. The financial rand has been
abolished but South Africa retains exchange controls on outward invest-
ment. Britain also had a dual exchange rate system between 1939 and
1979. Under the system, foreign investors had complete access to the LSE
but British investors seeking to purchase foreign securities had to purchase
dollars from a restricted pool of investment currency which was sold at a
fluctuating premium to the pound's commercial exchange rate.
China's current method of regulating foreign investors is inferior to the
aforementioned alternative forms of control. It has produced a subopti-
mal marketplace in which Chinese investors pay a large premium to the
prices offered to foreign investors, Chinese investors are unable to purchase
the promising companies offered on off-shore markets, and Chinese com-
panies are able to obtain foreign capital only if their sector is in fashion
among global investors. If China had a stock market more integrated with
the global marketplace, the valuation discrepancies would diminish, there
Concerns That Emerged 357
would be more liquidity from both domestic and foreign investors to re-
duce price volatility, and the marketplace would be more open to private
companies, not just state enterprises. Although China likes the model of
one country/two systems for reconciling the tension between capitalism and
communism, the current market segmentation policy is producing the worst
features of both capitalism and a command economy: illiquidity, volatility,
price anomalies and increasing corruption as investors attempt to capitalize
on the B-share discounts through circumvention of the rules. A new system
of one unified equity market with ownership ceilings for foreign investors
would produce far superior outcomes for all parties than the current system
of one country/multiple markets.
9.3.2 A Lack of Foreign Participation and Competition in
Financial Markets
Foreign banks have generally been restricted to hard currency operations,
although the Chinese government, in its bid to join the World Trade Organi-
zation (WTO), has announced its intention to partially open local currency
business to foreign banks. Since 1985, foreign banks have been allowed to
set up branches and local subsidiaries and to establish joint-venture banks
with Chinese partners in selected cities and Specialized Economic Zones
(SEZs). But their activities have been limited to wholesale banking and a
limited number of foreign exchange transactions such as foreign exchange
deposits and loans for joint-venture enterprises, foreign exchange invest-
ments and guarantees, and the settlement of import and export accounts.
Extension of foreign bank branches to Shanghai was made possible in 1990
when the Pudong New Area was designated as a development zone. The
PBOC approved the applications of seven foreign banks to open branches
in Shanghai in 1991, and that number increased to 30 by late 1994.
10
As
of the end of 1996, there were 538 representative offices, 131 foreign bank
branches, 6 joint-venture banks, and 5 wholly owned foreign banks. More-
10
To establish a branch, a bank must have total assets over US$20 million, have operated
a representative office in China for at least three years, and, for branches in Shanghai,
have US$30 million in registered capital. In addition, US$10 million in capital must be
deposited in cash with t he PBOC, essentially as a minimum reserve requirement. In
late 1994, foreign branching restrictions were somewhat eased: t he minimum number
of years of operation of a representative office was lowered to two, and t he capital
requirement is renminbi 100 million, of which 30 percent must be deposited with the
PBOC.
over, from December 1996 to April 1997, the PBOC issued licenses to 9
foreign banks to conduct renminbi business in Shanghai.
11
Foreign non-bank financial institutions in China include six finance com-
panies and six fully licensed insurance companies. To obtain a PBOC-issued
insurance license, a foreign company generally has to spend three years go-
ing through the application process. PBOC has traditional placed heavy
emphasis on solvency and a conservative investment style, which discour-
ages foreign companies from diversifying risk and limit their sources of
income beyond issuing individual insurance policies. Sixty unlicensed for-
eign overseas insurance companies have representative offices in China, but
they are regulated to operating in a few coastal areas on a trial basis.
Foreign securities firms may establish representative offices, but not lo-
cal branches or subsidiaries. They may purchase seats on the SHSE and
SZSE at the same price as that charged to domestic securities firms, but all
their transactions require the partnership of a local broker. Foreign securi-
ties houses are limited to B-share transactions and, subject to government
approval, to underwriting domestic securities. Twenty-three foreign bro-
kers have seats on the SHSE and eight foreign brokers have seats on the
SZSE. Foreign securities firms are prohibited from introducing new domes-
tic financial products, although they may offer off-shore derivatives, which
have been an attractive source of business. Repatriation of profits require
government approval, while dividend payments on B-shares do not.
12
Off-
shore listings of mainland companies (the H-, N-, L- and S-share markets)
are traded only outside China. They are less volatile than B-shares and
therefore offer better returns.
Foreign financial institutions are subjected to a certain degree of inten-
tional discriminatory treatment in China. For example, each account held
by a joint-venture enterprise with a foreign financial institution requires of-
ficial approval, which could take up to two years. In contrast, such accounts
11
These banks face stringent operating requirements, including keeping two separate
reserves (in local and foreign currency) equal to 18 percent of their renminbi deposits
and limiting local currency liabilities to less t han 35 percent of their foreign exchange
liabilities. Domestic banks are not subject t o these requirements, and, as such, t he
newly licensed foreign banks are not likely to create significant competition for domestic
banks yet.
12
Since 1990, foreigners have been permitted to own equity in Chinese companies through
B-shares. The B-share market developed without national guidelines, which may ex-
plain why, after an initial positive reaction to their introduction, the demand for
B-shares has dropped such t hat they frequently trade below issuance price.
China Moving Toward World Capital Market: Strategic Issues and Options 359
with the PBOC or its subsidiaries require only one-time approval, giving the
PBOC and its subsidiaries a competitive advantage over foreign financial
institutions. In addition, foreign financial institutions are subject to the
annual PBOC' s evaluation of their contributions to China, and the criteria
for these assessments lack transparency and often are politically motivated.
As is the case for all foreign companies operating in China, the staff for a
foreign financial institution must be recruited through a Chinese govern-
ment agency, which retains 55 percent of an employees gross salary. This
arrangement hinders recruitment of highly qualified investment personnel.
9.4 China Moving Toward World Capital Market: Strategic
Issues and Options
9.4.1 Enhancing the Interest of Ov erseas Inv estors
To be eligible for a listing overseas, a company needs to maintain a cer-
tain amount of cumulative pre-tax profit over a period of three most recent
years according to the rules of most foreign exchanges. This leads to an
interesting phenomenon among Chinese SOEs, which may list via "finan-
cial packaging", i.e., businesses put together just for the purpose of listing.
Therefore, a three-year profit record has little meaning. At a more destruc-
tive level, financial packaging makes it much easier to hide problematic
issues when there is no common thread in the track record. When there is
so much to be gained in a listing, there is bound to be much temptation
to be economical with the truth, or to resort to "creative accounting" or
downright fraud.
In order to mitigate the problems caused by packaging, extensive re-
search has been undertaken by the CSRC on a series of policy measures
to promote the standard of operation and strengthen the framework of
regulation on overseas listings. The CSRC had carefully summarized the
experience of overseas listings and existing problems, kept improving the
standardized operation of overseas-listed companies and regulatory mea-
sures of in-depth reform in order to bring into full play the CSRC regula-
tory functions and active roles of all relevant government agencies in the
promotion of standardization operation and in-depth reform of overseas-
listed companies, thus protect the interests of investors, promote the con-
tinued responsibility of overseas-listed companies after listing, strengthen
the investors' confidence in Chinese companies, and improve the image of
overseas-listed companies.
In 1998, the CSRC and the State Economic Commission issued Proce-
dures on Promoting Standardized Operations and In-depth Reform of Over-
seas Listed Companies, raising requirements for overseas listed companies
with regard to standardized operation, and by increasing transparency and
deepening restructuring. A breakthrough agreement has been reached to
promote standardized operations and in-depth reform to reflect the respon-
sible attitude to the investors by improving disclosure and enhancing trans-
parency.
The CSRC has paid special attention to the regulatory suppression of
overseas-listed companies, narrowing the gap between overseas-listed com-
panies and market investors, as well as improving investor relation. By
inviting staff from the Hong Kong Securities and Futures Commission and
the HKSE to lecture about information disclosure requirements regarding
listing rules, mergers and acquisitions, CSRC conducted training to relevant
government officials to facilitate their support of overseas-listed informa-
tion disclosure. CSRC also held training sessions for the board chairmen,
CEOs and board secretaries of overseas-listed companies for their voluntary
information disclosure. Additionally, the CSRC has formulated, after stud-
ies, several measures on improvement of information disclosure by overseas
listed companies.
In September and November 1998, the CSRC and the Ministry of Per-
sonnel jointly organized both the senior training program for board chair-
men and CEOs of Chinese overseas-listed companies as well as the seminar
for board secretaries. Tutoring management and managerial personnel will
assist directors and senior management of companies coming to list to un-
derstand their responsibilities. To this end, the CSRC hopes to conform
operation decisions to the standardization of listed companies, to develop
strategic and risk concepts, to improve management and decision-making,
and to train senior management who understand the law and regulations.
The companies are expected to make returns for investors through greater
accomplishments and the radiant glow of an improved international image.
To better prevent financial risks, the CSRC and other relevant gov-
ernment agencies required overseas-listed companies to implement internal
audit and take adequate preventive actions in mid 1998. The CSRC, in
conjunction with relevant government agencies, conducted an overall check
on the deposits status of 43 overseas-listed companies, and organized two
field inspection teams to look into the claims outstanding for these compa-
nies. The time will come when international investors are confident that the
investor protection mechanisms in Chinese financial markets are sufficiently
robust.
9.4.2 Improving the Role of Financial Intermediaries
China is still in the very early stages of financial deepening, as the practice
of central planning to allocate credit is not effective in channeling funds
from savers to borrowers. The financial reforms begun in 1979 delegated
some economic decisions to the microeconomic level and so necessitated
the development of financial intermediaries. However, developments on
the stock market are still based more on speculation than on investment,
indicating that securitization of financial assets still has a long way to go.
As Chinese companies proceed to ensure wider access to domestic and
world capital markets, the securities regulatory authorities must begin to
consider ways by which the pool of investors could be increased. At present,
Chinese stock markets are essentially populated by individual investors,
most with limited means and often with very little understanding of risk.
There some 45 million such investors by the end of 1999. It is therefore
important that the market should have a group of investors who have the
means to make investments with longer time horizons as well as the means
to make informed judgments about the underlying company, industry and
general economic outlook. Investment corporations have begun to operate
but they are few in number (in fact only 23).
In terms of financial intermediation, China has four state banks and
eleven commercial banks. The state banks were created in 1984, when
specialized banks were transformed into commercial banks. The largest
state bank is the Industrial and Commercial Bank of China, which extends
working capital loans to SOEs for fixed-asset investment. The Agricultural
Bank of China finances services in rural areas; the Peoples Construction
Bank of China provides medium- and long-term financing for capital con-
struction; and the Bank of China is the main international and foreign
exchange bank. State banks are responsible for implementing the PBOC' s
credit plan.
13
They hold the bulk of the SOEs bad debt, and are therefore
13
St at e banks have a network of branches, newly created affiliates, and special depart-
ments. This network allows them to hold the bulk of all deposits outside the credit
plana large proportion of these deposits flow into t he semi-formal and informal chan-
nels t hat are frequently of a speculative nature.
estimated to have negative net worth.
14
Non-bank financial institutions are generally more autonomous and
profit driven than state banks and commercial banks, and have introduced
a limited degree of competition into the financial system. For example, the
state-owned People's Insurance Company of China (PICC) was a monopoly
insurer until 1988. In 1993, it still handled over 95 percent of Chinas total
insurance business. The new insurance law of 1995 limited the PICC to
commercial insurance business and transferred its social insurance business
to the Ministry of Labor. While the PICC and several government financial
authorities own 17 regional life insurers, there are now three other regional
insurers as well as two independent national insurers.
15
To promulgate non-bank financial institutions, the China International
Trust and Investment Corporation (CITIC) was established and awarded
industry-level status. As the countrys major consulting firm and invest-
ment bank, it invests and lends funds raised from overseas capital markets.
Its subsidiary, the China Investment Bank (CIB), invests and lends funds
borrowed from international financial institutions, such as the World Bank.
It is responsible for raising construction funds from abroad. During the
1990s, many international trust and investment corporations (ITICs) were
set up, offering consulting and investment banking services similar to those
offered by the CITIC although not necessarily linked to the latter.
Two important decisions were made in August 1999. The first was to
allow SOEs to invest in the stock markets on a non-leveraged basis and if
they had held stocks for a period of at least six months. The second was to
allow life insurance companies to allocate up to 15 percent of their assets
portfolio for investment in the stock market. As the national pension funds
regulatory system comes into relatively stable operation, the time will come
when pension fund portfolios will be able to invest in the stock markets in
14
To free t he newly formed commercial state banks from policy burdens, the policy
lending previously extended by the three specialized banks was transferred to three
newly created, special purpose policy banks. The State Development Bank, with access
to funding from postal deposits, focuses on infrastructure lending. The Agricultural
Development Bank provides agricultural procurement and infrastructure funding, and
the Export-Import Bank provides export-import financing for key sectors.
15
Chi nas market for life insurance and household casualty insurance is still small and
most casualty insurance is purchased by corporate customers. Most assets have to be
deposited with domestic banks in interest-bearing accounts, and other investments are
required to be spread among safe investments and are limited to short-run commit-
ments.
addition to the rest of the capital markets.
Life insurance and pension funds have few short-term liabilities because
of contractual relationship they have with young generation. Young work
force has much longer investment time horizons than other investors. Ac-
cordingly, they would make excellent long-term investors in the markets.
Foreign investors constitute yet another set of long-term investors. For-
eign investors have tended to be institutional investors. Those who buy
B- and H-shares in the primary market are usually foreign pension funds
or mutual funds. H-shares are traded in Hong Kong and are therefore di-
rectly exposed to the international marketplace and are regulated by the
Hong Kong authorities who have a long track record of regulating accord-
ing to international standards. While both H- and B-shares have enabled
the country to raise substantial capital from outside, the fact that foreign
investors cannot invest in the domestic markets have denied most Chinese
companies of a further source of capital and a further source of institutional
investors.
Taiwan and India have demonstrated that a system of Qualified Foreign
Institutional investors (QFIIs) can successfully be instituted to attract for-
eign capital into their domestic stock markets. The system requires QFIIs
to be first registered and then money brought in should first be changed into
the local currency. The money is kept in specified accounts and the shares
bought kept with specified custodians, so that both money and shares can
be audited at all times. Before money is taken out, taxes will have to be
paid and remittance be cleared by the central bank. QFIIs have proven
a phenomenal success not only in attracting capital to Taiwan and India
but have in fact become a potent force in making investment choices and
putting management on their toes. Investors in these markets tend to pay
close attention to the investment choices of QFIIs. Chinese financial mar-
kets can potentially benefit from these practices.
9.4.3 Introducing Insurance Capital to Securities Market
As China's securities market and insurance industry expand, methods for
introducing insurance capital into the securities market are proliferating.
There are at least five ways we can think of to jointly develop insurance
and securities markets.
Direct securities investment is common worldwide. The insurance com-
pany's capital utilization and management departments invest a portion
of the capital in stock and bond markets. Many countries allow insurance
companies to act as investment institutions and receive commissions from
other organizations and investors to manage their capital and engage in
direct securities investment. In China, it is still not possible for insurance
companies to act as investment institutions and directly participate in secu-
rities investment, or for insurance companies to receive commissioned work.
The reason is that the insurance companies do not have securities invest-
ment capabilities such as risk management and control. Because no other
investment institutions are involved and insurance regulatory agencies are
unable to effectively monitor insurance companies' securities investment
activities, mechanisms of insurance control and restraint are not in place.
Consequently, investment risks are great. Furthermore, Chinese insurance
companies' investment institution status in securities investment activities
conflicts with related industry separation regulations in the current P.R. C.
Securities Law.
The second method is for insurance companies to entrust their capital
with investment institutions, which would be responsible for the manage-
ment of the investment portfolio. This method is essentially a type of com-
missioned capital management, which is something that Chinese insurance
companies can look forward to in the future. There are two reasons for the
lack of capital management firms in China. First, Chinese laws have not
clearly outlined the limits of commissioned capital management. If man-
agement is inadequate, commissioned capital management would become
a set of improper activities such as capital solicitation. Second, commis-
sioned capital management offers a great deal of flexibility in areas such
as investment amounts, investment methods and information disclosure.
Insurance companies usually cannot exercise proper supervision over in-
vestment institutions, nor can regulatory agencies effectively monitor and
manage commissioned capital management.
In a third method, insurance companies may be given a purchasing
quota of securities investment funds but also a time limit on their stock
holdings. For example, a maximum of two years may be imposed, af-
ter which insurance companies are allowed to sell but not buy. However,
China's securities investment funds were originally geared toward meeting
the indirect securities investment needs of small-to-medium investors. As-
signing a large portion of the funds to insurance company quotas would
contravene the original intention. After the time limit runs out, questions
concerning when and how much insurance companies would be allowed to
sell remain unanswered. Moreover, whether or not insurance companies
would be able to complete these investment activities independently is still
unknown.
Insurance capital can also be introduced into the securities market
through the buying and selling of securities investment funds in the sec-
ondary market, but in practice this can also be difficult. Because publicly
listed securities investment funds remain small in number as well as scale,
the amount of insurance capital they can absorb is limited. In addition,
choosing the right securities investment fund is no easier than choosing the
right stocks and bonds, and insurance companies are not prepared to do
the job satisfactorily at present.
The fifth method involves insurance companies using a portion of the
capital to set up a securities investment insurance fund, which would be
given to specialized fund management firms for securities investment man-
agement. Since insurance companies entrust their securities investment
insurance funds to the fund management firms, they usually do not di-
rectly participate in the management of the fund. In the future, however,
insurance companies may become initiators and participate directly in fund
management. Securities investment insurance fund management is a kind
of commissioned capital management that is structured and standardized.
It meets the current needs of Chinese insurance companies, and seems re-
alistic and feasible.
Recently, the CSRC issued the Notice on the Improvement of Stock Is-
suance Methods. This notice defined the strategic investor as a judicial
unit closely associated with the issuing firm and intending to hold issued
stocks for the long term. Thus insurance companies need to obtain ap-
proval from relevant regulatory authorities before they can receive stock
purchasing quotas as strategic investors. Furthermore, insurance compa-
nies must strengthen their management capabilities after the stock-holding
time limits expire.
9.4.4 Strengthening China Investment Funds
The first investment fund in China was born in 1991. Despite this late
start, China's investment funds have grown rapidly, and their number and
size are expected to expand greatly in the coming years. The funds have
developed in two stages demarcated by the national regulation promulgated
in 1997. Before that time, funds were at the stage of self-development. A
total of 75 funds and 47 fund-like certificates had been launched with the
accumulated capital of more than 7 billion renminbi yuan (approximately
US$900 million) as of September 1997. These funds are often referred to
as "old funds" as opposed to the new funds established after the national
regulation.
The small size of the old funds, however, made them unable to play
a significant role in China's securities market. Their erratic management
and poor risk control also highlighted the necessity of official regulation.
The Provisional Measures for the Administration of Securities Investment
Funds, promulgated by the CSRC in November 1997, was the first national
regulation governing the domestic fund industry. Under the Provisional
Measures, China's investment funds began a new era of development, and
22 new funds had been set up with a total market capitalization of renminbi
50.5 billion (approximately US$6 billion) by the end of December 1999, the
amount of which accounted for more than 6% of the capitalization of the
SHSE and SZSE. It was estimated that the total number of individual in-
vestors owning shares in securities investment fund was around 1.5 million.
Among the 22 new funds, the first 4 funds established in 1998 were
growth funds. The 18 funds launched subsequently were balanced funds.
Although all these funds are contractual closed-end funds with a 15-year
maturity, the balanced funds typically allocate 50% of their assets to trading
the market index, with the remainder in various types of growth stocks.
Because most Chinese listed companies currently pay out little dividend,
the conditions were not ripe for income funds. At year-end 15 out of the 17
funds that had been active in 1999 out-performed the stock market's 12.3%
growth.
A significant step in enhancing the reputation of the investment fund
industry has been the restructuring and consolidation of old funds. The
CSRC has actively steered the securities market towards more rational in-
vestment by encouraging the development of institutional investors who
will be allowed to invest in the market through securities investment funds.
During 1999, under instructions from the CSRC, work began on consolidat-
ing 43 old funds. Although only 1 of these old funds had been restructured
by the end of the year, the process of restructuring the others is likely to
be completed before the end of 2000.
To enhance the performance of securities investment funds, the Govern-
ment announced in August 1999 that fund management companies were
allowed to enter the inter-bank market, which greatly expanded their access
to capital to support their operations and enabled them to make better use
of it. Then the CSRC expanded fund management companies' privileged
access to IPOs. The Supplementary Notice on Improving New Stock Allo-
cation to Securities Investment Funds, which was issued in November 1999,
stipulated that investment funds could henceforth hold up to 30% of their
assets in IPOs (up from the previous 15%) and gave them unlimited access
to the IPOs of companies with more than 400 million registered shares.
At the same time, the CSRC promulgated the Provisional Regulations on
Qualification of Fund Practitioners, which further strengthened the legal
framework for fund administration by improving qualification requirements
on practitioners in fund management companies and fund custodians. In
parallel, the National People's Congress embarked on the preparation of
an important new piece of legislation-the PRC Investment Fund Law. The
preparatory team, which began work in March 2000, has already completed
the first draft of the Law.
As the volume of assets managed by investment funds increased, a wider
fund management industry began to take shape. Custodian banks are ser-
vicing the new funds and brokers are providing special trading seats for
funds. Both the custodian banks and the brokers are looking for ways
to develop their business as the industry expands, especially when fund
management companies are allowed to operate open-ended funds. At the
same time, some investment funds have been listed in major international
markets. In the meanwhile, there are 7 China funds listed on the Stock
Exchange of Hong Kong and some more elsewhere. Together with those
unlisted China funds in Hong Kong, an estimated managed funds totaling
HK$4.2 billion has been amassed by Hong Kong institutions for investment
in equity and venture-capital market in China. Table 9.6 presents a list of
China funds listed in international marketplace.
The events over the previous years have shown that China's securities
market needs the securities investment funds and those funds are begin-
ning to reshape the securities market. In 1999, the securities investment
fund industry has done well. It is, however, still in the initial phase of its
development. There will inevitably be trials and challenges ahead as the
industry continues to expand.
3
6
8

I
n
t
e
r
n
a
t
i
o
n
a
l
i
z
a
t
i
o
n

o
f

C
h
i
n
e
s
e

S
t
o
c
k

M
a
r
k
e
t
s

J
3

O

J
3

(
3

Q

a

3

O

a

<

f
t

a

(
8

f
t

a

o

Q

b
o

a

o

h
o

a

o

W

I
P

"

1

p

.
3

a
g

f
t

a
i

I
D

S
-

f
t
^

b
O

C
L

C
O

o

c
3
~

*

C
U

W

*
H

"

a

-

e
>

c
u

3

T
3

c
o

H

i
s

P

a

o

b
e

b
O

o

o

T
3

W

t
o

b
o

"
7

a

a

o

o

v

P

K

Z

O
H

i
-
,

T
3

t
o
"

a

J
S

c
S

c
p

w
2

a

r
t

5

o

o
o

i
n

C
O

t
-
H

/
>

/
?

e
e

C
O

C
O

C
O

p

p

p

f
l

C
U

C
D

r
t

c
u

c
u

C
D

C
O

C
O

o
,
,
o

_
o

o

o

o

-
a

a

g

P

o
3

M
M

b
O

^
^

c
S

P

g

4
J

C
S

o

-
2

a

C
U

C
S

.
3

1
3

a

J
3

c
o
.
S
O

3

1

&

2

a

*
.

t
o

i
r
!

^

J
3

c
u

t
u

c
o

O

5
?

b
O

b
o

7
3

.
3

.
3

c
d

c
o

c
o

P

0
3

P

o

c
o

P

-
o

a

T
3

C
O

P

3

a

J
-
i

C
P

u

P

o

a

3

s

g
o

o

g

1
3

O

3

O

P
H

!

c
u

f
t

c
o

u

&

c
u

3

o

s

3

3

2

a

o

-
a

c
u

c
u

c
o

^

S

P

s

3

_

8
,

O

<
D

b
o

>

a

3

3

b
o

>

3

a

c
?

3

o

O

3

W

O

b
o

a

c
u

C
O

8

S

S
3

a

C
O

-
o

a

-
o

9
>

-
a

a

c
c
i

a

O

a

c
u

o

C
U

a

a

o

C
M

-
a

a

c
u

a

o

-
o

a

"
?

3

C
O

a

o

J
3

J
S

5
1

O

O

P

Island
J
4
Coo
lative
3

cum
Kong
b
O

Hon
a

a
i

0
0

&
$

C
O

-
a

a

c
u

c
u

3

c
u

b
O

0
3

a

(
8

s

a

O

-
*
-
>

a

F
H
Tho
-
a

a

r
3

U
i

a

a

2

Q

!>
>

c
u

3

War
9.4.5 Reforming Pension System
China has a rapidly aging population which will require the development
of pension funds in order to finance retirement. At present, Chinese state
enterprises provide retirement income to elderly people on a pay-as-you-go
basis. But the number of elderly is expected to double by 2020. In 2045,
more than half of China's population will be over the age of 50. There-
fore pay-as-you-go system cannot survive indefinitely. The development of
pension funds will require a parallel growth of capital markets to provide
an outlet for the investment of retirement savings. China already has 45
million individual shareholders, but if pension funds develop, hundreds of
millions of people would have exposure to the equity market. China's other
financial intermediaries also need a more active capital market to serve
as a repository for their funds. The government, for example, recently
opened the door to insurance companies investing 10-30% of their assets in
A-shares of Chinese companies. Under the new rules, both domestic and
foreign insurance companies will be allowed to purchase A-shares. Such
a development is significant because until recently, Chinese stock markets
had no institutional investors except for a small number of investment trust
companies launched during the mid-1990's.
Generally speaking, further opening of financial markets provides a
means to develop a private sector safety net to provide for retirement, health
care and additional future claims on income. In China, pensions and re-
payment of banking system represent substantial claims on future income.
Pensions have become an increasing financial strain on SOEs. Despite po-
litical objectives of maintaining low levels of unemployment, government
revenues have declined. Burdened with a rapidly aging population, Chi-
nese government faces an enormous hurdle. Furthermore, Chinas one child
policy has created a distortion between generations. China faces a huge
under-funded pension liability. Pensions are currently indexed to wages,
which is problematic given that wages appear to be growing faster than
productivity. The World Bank estimates that wages in China will grow
about 7% or less per year. Returns on pension funds will need to meet if
not exceed the wage growth.
16
16
Somewhat surprisingly, rates of return on pension balances were below t he inflation
rate between years 1993 and 1995. In 1997, Chinese government began to unify locally
administered pension plans into a national model. The government plans to assign
"social security numbers" to allow workers to change jobs and move cities without
losing benefits.
A new pension system was launched in 1984 and dramatically reshaped
in 1991. It became mandatory for all employees and employers immediately.
New pension plans had emphasized savings over entitlement. In 1998, the
Ministry of Labor and Social Welfare was established to administer pension
plans. As of 1998, old retirees still receive full salaries until death, but
younger workers must fund their own pensions, i.e., the nation has been
shifting to an investment-based system of individual accounts.
Under the new system, the government requires employers to contribute
20% of a companys total wage or salary to retirement funds. 80% of the de-
fined contribution must be in bonds while 20% in safe deposits. Employees
and employers must contribute a total of 11% of the workers monthly wage
to the pension. The plan covers 96% of SOEs, 60% of collectively-owned
companies, and over 95% of retirees. A potential threat to the success of
this model is whether these funds will achieve the necessary growth given
Chinas recent low and even negative real interest rates.
The pension system would benefit from an expansion in investment
opportunities. The development of a stable pension system depends on the
creation of a modern banking system and securities market. The Trans Asia
Group, a direct investment specialist, estimates that 10,000 Chinese foreign-
funded companies have set up supplementary pension and health funds with
$2 billion under management.
9.5 Liberalization of Capital Movements
International capital and domestic savings are valuable components of sus-
tainable, long-term economic growth. However, recurring problems such as
the debt crisis of the 1980s, the Mexican currency crisis in late 1994, and
the current East Asian crisis, demonstrate that over dependence on foreign
portfolio investment, particularly mobile and short-term investment instru-
ments, can threaten the economic stability of emerging markets. This un-
derscores that liberalization carries risks as well as benefits and has major
implications for the policies that governments will find it feasible and de-
sirable to follow. With the onset of macroeconomic stability and the legal
and regulatory reforms that have taken place over the last decade, institu-
tional investors in China are beginning to play a more important role in
Liberalization of Capital Movements 371
channeling domestic savings to investment opportunities. These investors,
which include pension funds, insurance companies and mutual funds, can
also finance infrastructure and other long-term investments, and they can
encourage the development of mortgage markets.
9.5.1 What the Theory Says
The flows of capital debt, portfolio equity, and direct and real estate invest-
ment between one country and others are recorded in the capital account
of its balance of payments. According to the International Monetary Fund
(IMF), outflows include residents' purchases of foreign assets and repayment
of foreign loans; inflows include foreigners' investments in home-country fi-
nancial markets and property and loans to home-country residents. Freeing
transactions like these from restrictions, i.e., allowing capital to flow freely
in or out of a country without controls or restrictions, is known as capital
account liberalization.
Classic economic theory argues that international capital mobility allows
countries with limited savings to attract financing for productive domestic
investment projects, that it enables investors to diversify their portfolios,
that it spreads investment risk more broadly, and that it promotes intertem-
poral tradethe trading of goods today for goods in the future. Capital
mobility thus enables investors to achieve higher risk-adjusted rates of re-
turn. In turn, higher rates of return can encourage saving and investment
that deliver faster economic growth.
The currency crises afflicting Asia in 1997 and 1998 re-opened the de-
bate on the wisdom of capital controls and liberalization of the capital
account for emerging market economies. Generally speaking, the existence
of investors, including hedge funds, who are willing to take risks, provides
the liquidity to the market and allows easier reallocation of market risks. In
a world awash in "hot money", where capital, or financial claims to capital,
can be transferred across borders instantaneously with very low transac-
tion costs, there is growing concern that investors and analysts can wreak
havoc on developing economies. Still, over the long term, most agree (in-
cluding the Chinese government) that liberalization of the capital account
is beneficial for several reasons:
The improved ability to tap savings globally (at lower cost than
using only domestic savings);
Allowing domestic economic agents freedom to diversify risk by
choosing how and where to borrow, invest, or exchange assets;
Improvements in resource allocation through increased competition
for financial resources;
Increased availability of resources to support investment and to
finance trade and other significant economic entities; and
Imposition of macro-economic discipline on national governments.
Theoretically, speculation performs a market stabilizing function when,
for example, someone buys an asset when its price is low and sells it when
the price is high in a situation where there is only actual demand. In this
case, the speculator can make profits, but prices tend to be equalized since
there would be additional demand when prices are low and there would
be additional supply when prices are high. In this connection, the IMF
gives a positive evaluation on the function of hedge funds; while a mutual
fund that enjoys high returns may attract new investors and be bound
by its prospectus to buy more of the recently appreciated asset and other
institutional investors may be forced to cut their losses by their internal
controls and to sell into a falling market, hedge funds are better able to ride
out these fluctuations because their investors are locked in for substantial
periods and thus hedge fund should stabilize the market.
Contrary to the view, others would argue that hedge funds sometimes
have an extremely strong influence or monopolistically distorting power on
a specific market due in some part to herd behavior, bringing about an equi-
librium far removed from the economic fundamentals of a country, which
would not have occurred but for their speculative activities. They argue
that hedge funds fall into either one of the following models: (1) "Self-
realizing monetary crisis model", which maintains that a monetary crisis
is not necessarily caused by the aggravation of economic fundamentals but
rather it becomes actual crisis because investors make a speculative attack
on a currency in anticipation of a devaluation of the currency. (2) "Multi-
ple equilibria model", which maintains that although in an economic model
there is only one equilibrium guided by initial conditions or exogenous vari-
ables, there could be more than one equilibrium in a market like a foreign
exchange market, where a price forecast for the future often determines the
price for the moment. According to these arguments, in a small economy
with a fixed exchange rate regime, a massive short-selling of the country's
currency in expectation of a shift from the fixed exchange rate regime could
cause a plunge in the value of the currency first on the futures market and
then on the spot market and it could make it impossible to restore the
original equilibrium, even though its policy management is sound.
Furthermore, if there is a lack of transparency and no restrictions or
supervision are in place on international institutional investors, there is a
greater possibility that trading partners (counter parties in domestic coun-
try) of such investors will not be fully aware of the risks involved and the
final risk-taking burden will shift to the trading partner or the whole sys-
tem. In this sense, benefits and costs of international institutional investors
(such as hedge funds) have to be evaluated in relation to the market size
of an emerging market country. If a market is small, international institu-
tional investors can exert a greater influence through fluctuations in market
prices. On the other hand, what is most important for international insti-
tutional investors' investment strategy is liquidity of the market. If there is
not enough liquidity, it is difficult to hedge risks for investments or liquidate
positions but there exists far-reaching effects on the whole market when it
decides to liquidate.
9.5.2 China's Experience
China has had enormous success in attracting external capital over the last
decade and this has contributed to its growth in a variety of ways. Figure
9.1 presents GDP growth rates and foreign direct investment (FDI) growth
rates in China over the years from 1978 to 1995.
In December 1996, China liberalized its current account but its capital
account remains closed. This allows only those enterprises with necessary
trade documents to buy foreign exchange from designated foreign exchange
banks. The rules stipulate that Chinese enterprises must sell all hard cur-
rency earnings to designated foreign exchange banks or hold them in for-
eign currency bank accounts after receiving approval to open an account.
Foreign currency earnings cannot be held outside of China. Meanwhile,
Chinese residents can hold foreign currency in banks or sell them to the
banks on a voluntary basis but no private trading is permitted.
Foreign investors can remit profit, dividend and interest earnings out of
China from their foreign currency accounts or can convert yuan earnings
into hard currency at designated banks before sending money from China.
Foreign investors can only invest in B-shares on the stock exchanges in
China, which are separated from the primary security market, the A shares,
3
7
4

I
n
t
e
r
n
a
t
i
o
n
a
l
i
z
a
t
i
o
n

o
f

C
h
i
n
e
s
e

S
t
o
c
k

M
a
r
k
e
t
s

V

1

^
^
*
s

^
^
^

1

1

i

1

J
T
^

1

9 6 6 1
1994 1992 /l990 1988 \ 1986 1984 /l982
C
O

C
P

C
O

K
-

1

Year
o

I

o

C
M

I

(
%
)

H
I
*
"
*
)

which can be traded only by domestic investors.
This opening of the current account but closing of the capital account
has served China well in some respects, notably in that it prevents specu-
lators from instigating an exchange crisis. So far China has absorbed the
surge of external capital by avoiding most of the problems that less devel-
oped countries (LDCs) in similar circumstances often encounter.
17
Over
the course of the past five years, the government has allowed the currency
to fluctuate but within a very narrow band. Until recently, China also suc-
cessfully avoided an undesirable appreciation of its exchange rate following
the large inflow of external capital by making prudent interventions, e.g.,
the accumulation of reserves to offset the large flows in 1997.
For China the maintenance of its share of foreign capital should be easier
than for most LDCs since most of its capital inflow consists of FDI, bor-
rowing from private financial institutions being the distant second source.
As long as China is able to maintain its international competitiveness and
its domestic market continues to grow, FDI is likely to continue to flow in.
Although China's share of official development assistance is substantial, of-
ficial development assistance as a proportion of capital inflow in China is
low (about 12%). A decline in the volume of official development assistance
is therefore unlikely to have a serious effect on China.
Even so, prudence dictates that China be careful about increasing its
dependence on external capital. Its debt burden, although still well below
danger level, has increased rapidly and is significantly above that of South
Korea and Malaysia (and Taiwan). In addition, as a country with a very
high rate of domestic saving, China should be able to finance much of its
investment needs from its own resources.
However, the volatility of private capital is quite distinct in China. As
the recent Mexican experience showed, LDCs can hardly be stoical about
the volatility of foreign funds, making use of these inflows when they come
and not lamenting their departure for other pastures at other times. This is
because unsustainable surges can inflict serious economic damage by caus-
ing an appreciation in the exchange rate above the level that is warranted
by a healthy current account. As the exchange rate keeps appreciating
and the current account sinks deeper and deeper into deficit, the owners
17
At the most recent accounting, at the end of 1996, the present value of China' s future
debt service stood at 19.6% of GDP and 81.3% of exports. These ratios are higher t han
for South Korea (13.9% and 46.2%) but considerably lower t han for almost any other
major LDC including India (29.1% and 227.6%) and Indonesia (58.5% and 194.6%).
of private foreign capital suddenly decide that the country is no longer a
viable place for investment. A dramatic and traumatic flight of private
capital follows with crippling effect on output and employment of the re-
cipient economy. All this happens because, unlike a current transaction in
international trade, capital flows involve returns in the future about which
information is incomplete and hence capital mobility is anything but per-
fect.
Despite the government's tight controls, there is increasing evidence that
Chinas capital account is increasingly subject to leakage. Most of the illegal
capital flight is being conducted by SOEs, either through under-reporting of
export earnings or over-reporting of import expenses. By under-reporting
exports, companies can keep significant foreign exchange revenues that they
would otherwise have had to hand over to the State Administration of
Foreign Exchange (SAFE), the government organ that manages Chinas
US$158 billion in reserves. By over-reporting imports, companies are able
to secure permission to change renminbi into US dollars. The foreign trade
partner will deposit the over-paid portion of the contract into a foreign bank
account. Affiliates of state companies may also over-report or over-pay
for raw materials and other inputs, falsely inflating their foreign currency
expenses. Consequently, foreign affiliates of Chinese companies tend to
perform suspiciously poorly.
18
Consequently, it is clear that despite the
best intentions of capital controls, the Chinese government can only stem
or slow down capital flight at best.
Since capital flight from China is now primarily conducted through di-
verting investment and through the fudging of import and export docu-
ments, expanded trade and investment opportunities will continue to foster
these opportunities. Clearly one impact of Chinas accession to the WTO
will be to accelerate and expand options for individuals who want to in-
vest capital outside of the country illegally. This makes capital account
liberalization all the more necessary.
9.5.3 Policy Issues
In general, international capital flows and institutional investors should be
kept under control for three reasons, i.e., to protect investors; to ensure the
18
One SAFE study of 2,202 affiliates in 1998 found t hat only 36.6 percent were either
making a profit or breaking even (Kynge, 2000).
soundness and integrity of the market; and to maintain the stability of the
financial system.
First of all, it is generally thought that tighter controls, including more
information disclosure, on international institutional investors are neces-
sary to protect investors. But it is important to realize that if the degree
of information disclosure to investors is paramount, information disclosure
will make progress under the market mechanism. Secondly, ensuring the
soundness and integrity of the market is an important policy objective. If
such things as market manipulation by a big player, circulation of rumors
and insider trading become rampant, the market would lose its integrity
and would eventually lose its function to effectively allocate resources. But
we have to keep in mind that excessive restrictions will disrupt free transac-
tions and efficiency of the market. Thirdly, maintaining the stability of the
financial system is also an important policy objective. If a major interna-
tional institutional investor incurs a huge loss, it would cause deterioration
of assets at creditor financial institutions and this would worsen market
sentiment and cause a flight to liquidity, even leading to a collapse of the
financial system.
For a country like China whose financial markets are still in the de-
velopment stage, it is vital that the government takes preventive measures
against a huge, abrupt shift of capital and promote the stability of its own
currency and financial market. It is important for the government to take a
sound macroeconomic policy and strengthen the financial supervisory sys-
tem by thoroughly implementing prudent rules for financial institutions.
But at the same time, the government should explore the possibility of in-
troducing regulations from the standpoint of stabilizing the financial system
and keeping the market integrity in the same manner as discussed above.
The conventional economic wisdom dictates that there are several pre-
conditions that should be met first before a country liberalizes its capital
account. Now an international consensus exits that the liberalization should
be proceeded in a well-sequenced manner assuring whether certain precon-
ditions are in place. The preconditions to promote the liberalization of
capital transactions include: (1) a firmly established market economy sys-
tem; (2) progressive liberalization of trade; (3) an economy that is mature
and integrated into the world economy to a certain extent; and (4) a strong
financial sector and a highly efficient supervisory system. Regarding the
last precondition, it is essential, for example, to monitor the risks involved
in exposing foreign currency-denominated loans and maturity mismatches
through appropriate implementation of prudent regulations. Some also sug-
gest that the liberalization of the more stable long-term direct investment
should precede that of short-term capital transaction.
19
Given the current huge and drastic movement of capital in the world
economy, what kind of exchange rate regime Chinese government should
decide to choose is very important. Of course, what would be the most
suitable exchange rate regime for China depends on the composition of its
trading partners, the composition of its major trading items, the degree
of capital liberalization and its inflation experience. But the Asian cri-
sis has taught us that pegging a country's currency to any one particular
currency is highly risky. Generally speaking, China can flexibly cope with
exchange fluctuations by pegging its currency with, or referring to, a bas-
ket of currencies of some advanced countries with which it has the closest
interdependent relationship in terms of trade and investment. In any case,
there is no simple formula. Therefore, it should be stressed here that it is
important for China to choose a suitable exchange rate regime consistent
with appropriate macroeconomic policies.
In any event, a full convertibility of renminbi will be inevitable. Full
convertibility of renminbi could represent a dramatic expansion of the range
of financial assets available to Chinese households, pose the threat of finan-
cial disintermediation, and thus undermine the financial viability of domes-
tic banks. This could potentially lead to a disastrous run on the Chinese
banking system, which would also entail the closing of roughly one-third of
Chinese SOEs. While some might argue that this is a good thing over the
long-term, there is a legitimate and palpable fear that such a result would
lead to chaos and turmoil in the society. If such a situation occurred,
Chinese banks could either create a massive hyper-inflationary spiral by
printing money, or by restricting or limiting customer withdrawals. The
first option is disastrous; the second option, while preferable to the first,
19
It is essential to be aware t hat the direction of short-term capital may change sud-
denly from an inflow to an outflow and t hat holding foreign currency-denominated
debts contains foreign exchange risk. During the Asian financial crisis, the amount of
capital t hat once flew in and then flew out of Asian countries in one year amounted
to more t han 10% of t he combined GDP of the countries involved. Therefore, it is
important to make t he best use of market friendly controls on capital inflow in a way
t hat would not distort the market. Such capital controls would include higher reserve
ratios for deposits received from non-residents and stricter prudent regulations on bor-
rowing from abroad and securities issuance by residents (in particular, by financial
institutions).
would erode confidence even further and result in more illegal capital flight.
A palatable solution to this problem is to distinguish between different
types of capital flows such as borrowing and lending, securities, and FDI.
As noted above, foreign direct investment is largely already liberalized. For
securities market, China should rapidly expand the number of companies
listed on the B-Share markets, and gradually phase out the segmentation
between A- and B-share markets. This will also serve the purpose of market
recapitalization. At present, a practical step is to allow foreign investors to
purchase both A- and B-shares and foreign companies to issue both types
of shares. Chinas accession to the WTO will accelerate this process and
further open up of its capital account by freeing outward remittance of
investment returns by foreign-invested funds, as well as distribution of such
returns to overseas shareholders. New changes will occur to the structure
of Chinas securities markets.
There is a saying in China that things will happen when the conditions
are right (a fruit is the best when it is ripe, guashu diluo). It is time for
bolder thinking, however, in the light of the reform experience that the
state is still firmly entrenched in many parts of the economy. The fruit will
likely never get "ripe" to continue the metaphor, absent a clear signal from
the top as to when China will liberalize its capital account. Unfortunately,
at this time, there are appears to be little movement on the top of this
issue. Chinese Premier Zhu Rongji has publicly declared that there is no
firm timetable for reform. In short, he subscribes to the belief that the
closed capital account saved China from recent economic turmoil. While
true in the past, China should recognize that prolonging capital account
liberalization would only increase the aggregate costs over the long-term.
Thus, while Michael Camdessus of the IMF believes that capital account
liberalization must be "bold in its vision, cautious in its implementation",
there is a strong case in China that it should be bold in both vision and
implementation.
Appendi x A
Securities Law
A. l Introduction
The new P.R.C. Securities Law, which was promulgated by the Standing
Committee of the 9th National Peoples Congress (NPC) on December 29,
1998 and came into force on July 1, 1999, is the national legislation gov-
erning the securities market in mainland China. The new law by and large
completes the national legal framework for regulation of the Chinese stock
market, a process originally begun with the passage of the P.R.C. Company
Law in 1993.
The Securities Law has attracted much attention both within and out-
side of China. Among the reasons for the attention is that the law received
an unusual amount of support from the countrys leaders. Following the
National Financial Working Conference of the Chinese Communist Party
and the State Council held in October 1997 (which had focused on the
importance of reducing financial risks in the market), the China Securities
Regulatory Commission (CSRC) began in earnest to promote the enact-
ment of the new Law. Before it was tabled at the Standing Committee of
the NPC at the end of 1997, Li Peng, Chairman of the NPC, personally led
a special mission to Shenzhen to discuss comments on certain outstanding
issues. At the Standing Committee meeting it was reported that not one of
the 138 delegates voted against the law, an event quite rare in China since
it began using the secret ballot in voting.
A second reason that the law has drawn much attention is that the
draft had received wide consultation amongst academics, ministry officials,
professional regulators and market practitioners during the five years since
381
382 Securities Law
it was first considered. It had also gone through as many as four previous
reviews by the Standing Committee. This ensured that all interested parties
had an opportunity to express their views on what has been perceived as a
vitally important law.
Moreover, by promoting the involvement of foreign securities law experts
in the consultation process, it has managed to attract significant interest
overseas. A series of international conferences dealing with provisions of
the law were held, the major one being the "International Symposium on
Securities Law" held in Beijing in 1996 and co-sponsored by the United
Nations Development Program. Experts from the U.K. Securities and In-
vestment Board, the U.S. NYSE, Hong Kongs Securities and Futures Com-
mission, together with practitioners and academics from Japan, the U.S.
and Hong Kong, made presentations on their respective areas of exper-
tise. A collection of essays has since been published by the CSRC. In 1997,
another international conference was held where experts from the U.S. Se-
curities and Exchange Commission, in addition to experts from Japan and
Hong Kong, gave presentations on selected topics concerning the Securities
Law. Because of this open approach adopted by the Chinese authorities,
many experts in the securities industry had themselves been involved in the
drafting stage and have since remained keenly interested in witnessing the
progression of the draft into law.
Finally, and perhaps most importantly, the law would seem to indicate
a consensus view among the Chinese authorities that:
The nature of the stock market and its role and function within the
Chinese socialist market economy has now been firmly entrenched;
any effective regulation of the stock market must have a strong legal
foundation; and
given the existing realities of the stock market against the back-
ground of the Asian financial crisis in 1998, the authorities believe
that legal regulation of the stock market cannot be further delayed.
The Law thus arguably indicates that, since the stock market is now
an integral part of the Chinese socialist economy, the use of securities as
a tool of capital formation will be encouraged and the quality of the stock
market will thereby improve with time. In other words, it provides a firm
basis for the belief that greater business opportunities, both domestically
and overseas, will be created in the Chinese capital market.
The Role of the Stock Market 383
A. 2 The Role of the Stock Market
In a preface to a study guide on securities knowledge published by the
CSRC in 1998, the Chairman, Zhou Zhengqing, reiterated the four main
benefits of the securities market from the Chinese perspective:
Through exercising a controlling interest in listed companies, state
or collectively owned units not only retain their control over pro-
duction assets, but also manage to attract surplus funds from the
public in the form of investments. By participating in asset valu-
ation in the course of the listing exercise and issuing shares at a
premium, the capital of these units will increase in value as they
become listed companies. Hence, the market is a useful tool for
strengthening the central position of public ownership in the so-
cialist economic system.
As China's only source of funds in a planned economy is from bank
loans, the rate of indebtedness of state-owned enterprises has been
very high. This is neither good for the enterprises' own development
nor for management of the country's financial risk. Establishing
a stock market and finding new sources of direct capital funding
is compatible with China's current policy of developing a modern
financial and corporate structure. Moreover, it offers additional
source of foreign capital investment.
As state-owned enterprises must be restructured for the purpose
of listing, the process requires an enterprise to reform itself, estab-
lish a corporate management structure, increase the level of trans-
parency, maintain a continuous disclosure system and submit to
the supervision of its business by its shareholders, the regulators
and the public. By facilitating such enterprises in the restructuring
of their operational mechanism and developing a modern corpo-
rate structure, listings on the stock market should be considered
complementary with China's policy of enterprise reform.
By regulating the types of businesses approved for listings, the
authorities have implemented China's policy of enhancing certain
types of production assets and achieving a more efficient allocation
of resources amongst different industries. Significant funds have
been raised from the stock market and channeled to preferred indus-
trial sectors, such as infrastructure development, high-technology
384 Securities Law
production and important national pillar industries. Moreover, ac-
quisitions of loss-suffering enterprises by listed companies will en-
able these enterprises to be preserved and used in a more positive
way.
The acceptance of these benefits has shaped the present form of the
Securities Law, which has 12 chapters and 214 provisions. The general
principles are contained within Chapter 1. Article 3 states the basic prin-
ciples of transparency, fairness and integrity, while under Articles 7-9, a
centralized approach to regulation is emphasized.
Article 2 provides that the Law shall govern the issuance and trading,
within China, of shares, corporate debentures and other securities defined
by the State Council. This provision defines the scope of the conduct
regulated, the territorial scope and the products regulated under the Law.
The Law specifically provides that it does not apply to treasury bonds.
At this stage, the Law applies to corporate equity securities and corporate
bonds, but the open-ended nature of Article 2 envisages that other securities
may be acceptable for regulation in future.
A. 3 Regulating the Chinese Stock Market
China has approximately 40 million investors and 2,400 securities brokerage
outlets across the country. As at the end of August 1999, there were 448
companies listed in Shenzhen and 470 in Shanghai, of which more than 100
have issued B shares. The total market capitalization is close to renminbi
2.8 trillion (see Shenzhen Securities Exchange Market Statistics, September
1999 issue; SZSE Monthly Statistics, September 1999 issue), which repre-
sents about 30% of Chinas GDP.
Prior to the enactment of the Law, the governing legislation was the
State Council Provisional Regulations on Share Issuance and Trading (22
April 1993). The Chinese stock market took the same road of trial im-
plementations as has been the case with other innovations in the Chinese
socialist economy. For example, the first issuance of shares occurred in
1983 (Shenzhen Bao An Enterprise Group Limited); the stock exchanges
in Shenzhen and Shanghai were established in 1990; and the CSRC was
created in pursuant to State Council Directive in 1992.
The Law has made an important difference by establishing a centralized
Regulating the Chinese Stock Market 385
regulator for the market and providing a legal anchor for all of the regula-
tions that apply in the securities area. According to official figures, there
were close to 250 piecemeal provisions. A collection of these provisions
dating from 1992-1998 has been published by the CSRC. The securities
regulator, the CSRC, is not specifically named in the Law, but is referred
to as the securities regulator under the State Council.
The Law gives specific powers to the CSRC, as well as imposes respon-
sibilities on it by way of checks and balances. The enactment of the Law
indicates that the Chinese market has entered into a more mature phase.
As Chairman Zhou of the CSRC noted in an interview on December 30,
1998, "with the creation of a comprehensive legal system for the market,
a fundamental improvement in the quality of regulation is expected to oc-
cur". "Regulation", he said, "is the prerequisite for growth" (see report
published in the CSRC Official Bulletin, Vol 12, 1998).
The Law gives clear legal backing to the work of the CSRC. Article 167
specifies 8 regulatory functions for the CSRC.
To make regulations governing the market and to exercise the power
of approval and ratification;
to govern the issuance, trading, registration, custody and settle-
ment of securities;
to govern securities issuers, listed companies, the stock exchange,
securities companies, securities registration and settlement agen-
cies, fund management agencies, investment advisory agencies, credit
rating agencies as well as law firms, accounting firms and valuation
firms that engage in the securities business;
to stipulate standards as to qualifications and conduct for persons
engaging in the securities business;
to govern and inspect the disclosure of information in relation to
securities issuance and trading;
to instruct and govern the activities of the securities association;
to investigate breaches of securities law; and
other functions defined under other laws and regulations.
The regulator is entrusted with delegated authority to make rules gov-
erning the market (see Article 167(1), Articles 109, 111, 154, 158). This
flexibility recognizes the need for the regulator to respond to rapidly chang-
ing market conditions and is consistent with the approach used in other
jurisdictions.
386 Securities Law
The regulator has power under Article 167(7) to investigate violations
of the law. New specific investigative powers are given in Article 168 and
include the power to:
Enter the premises where the law has been broken for investiga-
tions;
question the parties and related persons;
examine and make copies of trading records, registration records,
accounting reports and other materials belonging to the parties or
related persons; and
examine the cash accounts and securities accounts of the parties and
related persons. Where there is evidence of transfer or dissipation
of assets, the regulator may apply to court to have the assets frozen.
These powers are necessary in practice to enable the investigator to
obtain evidence. On the other hand, they serve to limit the power of the
regulator in the sense that it shall not have the power to make demands
that are not particularized in this provision.
The Law adopts a centralized approach to regulation. This can be
seen as encompassing 3 aspects: (i) regulation of the securities market is
undertaken by the central government and not the provinces; (ii) the CSRC
is the paramount regulator of the securities market vis-a-vis other central
government ministries; and (iii) the exchanges are subordinated to, and
supervised by, the central regulatory body. As such, the responsibility of the
regulator is enormous. Recognizing the power bestowed on the regulator,
the Law also provides checks and balances. Article 9 requires the regulators
work to be supervised by the State Audit Bureau. Article 209 requires
confiscated gains and fines to be paid to the State Treasury.
The Law gives a party grieved by a decision of the regulatory body
the right to apply for review or begin an action in the Peoples Court. A
new measure of accountability is stipulated in the Law. For example, the
listing committee is required to give reasons for rejecting an application for
a listing of securities under Article 16 and the regulator is required to make
public all of its regulations, as well as its decisions, on the imposition of
penalties for violations of the law under Article 172. The Law explicitly
provides that the CSRC staff must observe duties of confidentiality, fairness
and integrity.
Capital Formation 387
A. 4 Capital Formation
Chapter 2 of the Securities Law deals with the issuance of shares. The
Law provides that a public offering of shares must comply with the condi-
tions of the Company Law and be first ratified by the securities regulator.
Moreover, a public offering of corporate bonds must be approved by a body
authorized by the State Council. The P.R. C. Company Law, which as noted
earlier was enacted on December 29, 1993, contains most of the substantive
requirements concerning the issuer (e.g., Article 152 of the Company Law
lays down the minimum requirements for a listed company). By contrast,
the Securities Law contains mostly procedural and administrative require-
ments for a share issuance or listing. The authorities have carefully avoided
overlap between the two laws, but if there is conflict, Article 2 of the Securi-
ties Law implies that it shall override the Company Law, i.e., the Company
Law will only apply to matters that are not specifically prescribed under
the Securities Law.
Article 10 lays down the general principle that a prospective issuer of
shares to be offered to the public must comply with the minimum require-
ments laid down by law and no person may offer securities to the public
unless it has been ratified or approved by the securities regulator. Article
16 requires a decision on an application to be made within three months.
Article 20 provides that a new issuance of shares may take the form of
a public offering or a rights issue. In both cases, the issuer must ensure
that the proceeds are used in accordance with the proposal stated in the
offering document, except where the change of use has been approved at a
shareholders' meeting (Article 20). Underwriting is required for all public
offerings (Articles 21-28). A major innovation under this Law is the issuance
committee, to be comprised of internal and external experts appointed to
approve applications for share issuance (Articles 14-15).
Article 10 also contains one of the most enigmatic provisions in the Law.
It states that no public offer of securities is allowed unless it is ratified or
approved by the CSRC or other ministries authorized by the State Council.
This represents a major step forward from the previous method that had
relied entirely on state approval for share issuance. The new ratification
process places more emphasis on the participation of an external issuance
committee and the issuer's objective fulfillment of the listing requirements.
Some commentators have sought to differentiate ratification and approval
by describing the former as a procedural review and the latter as a sub-
388 Securities Law
stantive review. A procedural review process is one where the regulator
ensures that all matters required to be disclosed are contained in the is-
suance document and it makes no judgment on the substantive merits of
the application. Hence, it is often described as a disclosure-based process.
A substantive review, on the other hand, is one where the regulator makes
a judgment on the substantive merits of the application and approves only
those it deems fit for listing. It is often described as a merit-based process.
Under the Law, the approval procedure for share issuance is a two-tier
process. First, the application will be examined by the issuance committee
comprised of regulatory staff and external professional experts and, pur-
suant to Article 14, the committee members shall vote to express a decision
whether to approve the application. The application must also be delivered
to the regulator for ratification by its staff. Under Article 15, the regulator
is required to ratify the application in accordance with conditions laid down
by law and must do so in an open manner. In this way, the Chinese author-
ities have managed a compromise between the substantive and procedural
approaches. It also allows external participation and market input into its
share issuance mechanism, while maintaining ultimate control over it.
The Law, moreover, makes no reference to the pre-selection (quota)
system for approving enterprise listings. This confirms what Mr. Zhou
said during a speech in Hong Kong in April 1999 regarding the abolition
of the pre-selection system for overseas listings. The new approach is to
approve company listings as and when they become ready.
A. 5 Securities Trading and Information Disclosure
Chapter 3 is entitled Securities Trading and contains four distinct parts
governing General Principles (Part 1), Securities Listings (Part 2), Disclo-
sure Obligations (Part 3) and Prohibited Conduct (Part 4). The general
principles in Part 1 stipulate the basic rules of the market, e.g., that only
lawfully issued securities may be traded and that securities may only be
transacted at a securities exchange by way of open, centralized competitive
price bidding. Article 34 provides that the securities may be evidenced in
paper form or in other forms stipulated by the securities regulator. Article
41 requires a shareholder who holds 5% or more of the issued share capital
of the company to disclose that fact to the company, the regulator and, if
the shares are listed, the exchange.
Securities Trading and Information Disclosure
389
Part 2 (Articles 43-57) governs the listings of shares and corporate
bonds. Part 3 (Articles 58-66) prescribes obligations for disclosure of infor-
mation. Part IV (Articles 67-77) identifies insider trading, the acquisition
of unlawful interest or the shifting of risk and the dissemination of false
information as undesirable conduct.
The Law stresses the importance of continuing disclosure obligations,
which are provided under Articles 58-66. They relate to issuance docu-
ments (prospectuses), interim reports, annual reports and reports on ma-
terial events. There is an obligation imposed on shareholders holding 5%
of the issued share capital of the listed company to disclose their interests
(Articles 41 and 79). Article 59 requires that the information be authentic,
accurate and complete, and may not contain any false record, misleading
statement or major omission. Article 63 states that the issuer, underwriter
and/or their officers may incur legal liability for any loss to investors aris-
ing from false or misleading statements in the disclosure reports. Article
177 provides that an issuer who fails to make adequate or true disclosures
may be ordered to make corrections and pay fines in the region of ren-
minbi 300,000-600,000, whereas responsible officers may be ordered to pay
personal fines ranging from renminbi 30,000-300,000. Article 189 provides
that persons making false statements or producing misleading information
in the course of securities trading may be ordered to make corrections or
to pay fines in the region of renminbi 30,000-200,000. Since intent is not
an essential ingredient, the Law effectively imposes strict liability for in-
adequate disclosures. The Law also provides that, in both cases, criminal
liability may ensue if the acts or omissions constitute a crime.
The Chinese market has suffered from a lack of transparency result-
ing from listed companies making inadequate disclosures. In a Hong Kong
survey conducted in 1994 after a huge fall of H-share prices, 90% of the
fund managers surveyed expressed concern over the lack of transparency
of H-share companies; 76.9% thought H-share companies did not faithfully
disclose their financial and business information; and 80% felt that H-share
companies did not understand what information ought to be disclosed, es-
pecially in relation to substantial transactions. This concern has been some-
what alleviated by the efforts of the CSRC, the HKSE and the Institute
of Directors in Hong Kong, who have conducted seminars on disclosure in
the past year for officers of such companies. Yet lack of transparency will
remain a problem if there is no clear statutory requirement for disclosure
390 Securities Law
backed by adequate sanctions. However, the strengthening of disclosure
standards in the new Law is expected to improve the situation.
Article 43 requires a company applying for listing of its shares on an
exchange to submit an application to the securities regulator for ratification.
Article 43(2) empowers the regulator to delegate this function to the stock
exchanges. Once an application is ratified for listing, the stock exchange in
question will be required under Article 46 to effect the company's listing
within six months of receiving all the prescribed documentation.
A. 6 Acqui si t i ons
The important objective of the restructuring and reforming of state en-
terprises has been noted above. Since company acquisitions can facilitate
reforms, mergers and restructuring of companies and state enterprises by
encouraging competition and discovering unrealized potential, it is no sur-
prise, therefore, that the rules relating to company acquisitions are given a
prominent position in the Law. Chapter 4 (Articles 78-94) recognizes two
types of offers: acquisition by tender offer and acquisition by agreement.
Following the U.S. statutory model rather than Hong Kong's voluntary
code model, the Law provides a step-by-step guide on how a tender offer
should be made.
A tender offer applies to shares traded on the exchange. Article 79
provides that an investor who has reached a share-holding of 5% of the
issued share capital of a listed company should, within 3 days, report the
holding to the CSRC, the exchange and the listed company and refrain from
further purchases during that period. Subsequently, if the investor increases
or reduces its holding by 5%, it shall make a similar report within two days
and shall refrain from further purchases during that period. Article 81 then
provides that an investor wishing to acquire further shares in a company
after accumulating 30% of its shares is required to make a general offer
to shareholders. The general offer must take the form of a written offer
document, which must first be approved by the CSRC. Article 83 prescribes
that an offer shall be announced 15 days after the date of submission of the
document, and that the offer period must not be shorter than 30 days or
longer than 60 days. Once the offer is announced, it cannot be withdrawn,
although the investor may change its terms subject to the approval of the
CSRC and the relevant exchange. The investor cannot trade in the shares
Acquisitions 391
during the offer period other than under the terms of the offer.
Article 85 provides that all terms stated in a tender offer shall apply
equally to all shareholders of a company. Article 86 states that the company
will be de-listed if the investor succeeds in acquiring 75% or more of the
share capital. Under Article 87, if an investor is found to have acquired
90% or more of the share capital, the shareholders who did not accept the
initial offer and are still holding shares in the company will be given the
right to sell their shares to the investor on the terms stated in the initial
offer.
The regulation of private acquisitions is, by comparison, less detailed.
Article 89 states that an investor can enter into an agreement with the
shareholders of the target company to effect a transfer of their shares. The
investor is required to submit a report of the agreement to the CSRC and
make a public announcement within three days. The agreement cannot
take effect until it is publicly announced.
As at the end of August 1999, the combined market capitalization in
China exceeded renminbi 2 trillion, but only about 30% of is comprised
of liquid shares. The Law did not address the status of state shares, legal
person shares and employee shares indicating perhaps that the issue is still
unresolved. Given the poor extent of liquidity of shares in the market, it is
difficult to expect take-over activities to flourish in the market. However,
Article 89 envisages that acquisitions by agreement can be governed by
State Council regulations. It is possible that further regulations will be
enacted to govern this topic when the government has clarified its policy in
this respect.
In addition to the open provisions on acquisitions by agreement, it
should be noted that the Law gives the CSRC a discretion to exempt an in-
vestor from complying with the mandatory offer requirement under Article
81. This emphasis on flexibility enables the regulator to modify the statu-
tory take-over requirements to address the needs of specific transactions. It
remains to be seen whether the regulator will, in time, develop guidelines
as to how this discretion will be exercised. However, this waiver, together
with the flexible provisions on acquisitions by agreement, will no doubt be
wide enough to facilitate the reform of state enterprises by encouraging the
re-organization of assets in an orderly and commercial fashion.
392 Securities Law
A. 7 Market Infrastructure
China is, and shall remain certainly for the foreseeable future, a semi-closed
market. Article 138 provides that only Chinese citizens and legal persons
can open accounts for investing in the A-share market.
In addition, the Law contains numerous prohibitions:
Article 35 states that only spot trading is permitted. Forward con-
tracts and derivative futures contracts on securities are considered
not suitable for trading at this stage.
Article 36 states that margin trading and securities borrowing are
prohibited. Brokers are not allowed to extend credit to clients,
either by way of providing them with cash loans or lending them
securities.
Article 39 prohibits trading and prescribes minimum holding pe-
riods for securities for persons involved in the issue of the audit
report, asset valuation report, legal opinion and other documents.
This is designed to discourage insider trading.
Article 42 states that shareholders are not allowed to keep profits
earned within six months of previous sale or purchase, and directors
are personally made responsible for enforcing this provision. This is
designed to curtail short-swing profits and reduce market volatility
and price fluctuations.
Article 76 explicitly prohibits state enterprises from speculating in
shares. This is designed to curtail speculation as well as prevent
state assets from being exposed to market risk.
Article 106 prohibits day trades. This is designed to reduce market
volatility and price fluctuations.
Article 133 states that bank funds are prohibited from being in-
vested in the stock market. This is designed to limit the injec-
tion of "hot money" into the market. It is also recognized that
banks have an unfair advantage over the ordinary investor because
they have more available funds and consequently have the ability
to destabilize the market.
Despite the good intentions behind the prohibitions, an inevitable result
will be a reduction of liquidity arising from a credit crunch. This has
not gone unnoticed, as stated by the CSRC Chairman in an interview on
December 30, 1998:
Market Infrastructure 393
"I heard some rumors that the proposed enactment of the
Securities Law had dampened confidence in investing in
the market, as if the Law would be the cause of a falling
market. This view is completely groundless and is a serious
misrepresentation to investors'. It certainly appears that
the authorities are aiming to create a more stable market
for the long term, with better protection for investors, and
the restrictions are the result of a deliberate policy choice
to reduce leverage and speculation in the market. "
Chapter 5 deals with securities exchanges. Article 95 defines an ex-
change as a place for trading in securities at which the price is fixed cen-
trally by competition. Only authorized exchanges are allowed to operate,
and at present there are two: in Shenzhen and Shanghai. Only firms, not
individuals, are allowed to be members of the exchanges (Article 103).
Chapter 6 deals with securities companies. They are the main type of
intermediary in the market. Only limited liability companies and compa-
nies limited by shares may apply to become securities companies (Article
118). Under Article 119, companies are divided into two types according
to their business: general companies and agency companies. The mini-
mum registered capital for a general firm is renminbi 500 million and for
an agency firm it is renminbi 50 million. General companies are allowed to
trade on their own account (proprietary trading) as well as act for clients.
There are new requirements as to the qualifications and conduct of per-
sons engaged in the securities business in chapter 6. Article 122 envisages
a licensing regime that requires persons engaged in the management and
operation of the securities company to have the requisite standards. It
also requires companies to have appropriate trading facilities with sound
management systems in place.
Concerning financial resources requirements, Article 124 states as a mat-
ter of principle that the external liabilities of a securities company must not
exceed a certain percentage of its net assets and the liquid liabilities must
not exceed a certain percentage of its liquid assets. It further provides that
the regulator may set the specific percentages and the procedures for gov-
erning this, which gives room for adoption of risk-based capital adequacy
standards that are common in other jurisdictions.
The Law also provides detailed requirements concerning the treatment
of clients. The issuance of client agreements, statements of account and
394 Securities Law
verification statements are now mandatory (Articles 139-140). Client ac-
counts must be segregated under Article 138. Brokers are prohibited from
offering discretionary services (Article 142) or guaranteeing profits (Article
143) to clients. Under Article 145, a securities company shall remain fully
responsible for its staff in any of their misconduct in securities trading.
Proprietary trading is the term used to describe a company's investment
of its own funds in the securities market. The Law makes a distinction
between proprietary trading and agency trading. The main risks of propri-
etary trading are, first, that a company might expose itself to conflicts of
interest with its client as far as trading opportunities are concerned and,
secondly, that a company which engages in proprietary trading may incur
risks to its capital. The Chinese authorities recognize the benefits of, as
well as the difficulty of regulating, proprietary trading. A compromise has
been struck by allowing only those companies with a substantial capital
base and an adequate system for segregation of assets to do so as general
companies, on top of their agency business. The minimum capital require-
ment for general companies permitted to engage in proprietary trading is
renminbi 500 million. Article 132 requires that the two types of businesses
in a general company must be strictly segregated. As industry practitioners
are generally in favor of proprietary trading, this requirement is expected
to encourage mergers among lesser players so as to combine their businesses
to reach the above threshold.
Chapter 7 regulates securities registration and settlement agencies. They
are recognized as forming an integral part of the clearing and settlement
system of the Chinese market. These agencies are non-profit making units,
which provide centralized registration, custodian and settlement services in
support of the trading of securities. Article 150 requires all securities to
be deposited with such agencies before they can be traded. Under Article
151, securities holders obtain evidence of title from these agencies. Bro-
kerage services and custodian services are segregated under This system,
minimizing the risk of loss from misuse of client assets.
Chapter 8 deals with intermediary service providers, which refer to three
types of companies: the investment adviser, the credit rating agency and
the professional adviser. Under Article 159, the investment adviser is not
permitted to act for the client in securities trading, so as to avoid potential
conflict. Under Article 161, professional advisers are required to assume
liability for their advice and official reports.
Investor Protection 395
The Law envisages a form of self-regulation, which is encouraged in
other financial jurisdictions. Chapter 9 establishes a securities industry
association to be comprised of securities companies. The association is
characterized as a self-regulatory body whose scope of activities includes
providing training to members, supervising their compliance with the law,
raising business standards and mediating disputes.
When the stock market began in China, the power to grant licenses to
intermediaries was vested with the People's Bank of China. This power
was recently transferred to the CSRC and is now formally entrenched in
Chapter 4 and Article 167(4) of the Law. The quality of the intermediaries
in a market reflects the quality of the market. It is important for effective
regulation of intermediaries that the same body governs the entry and exit
points for them and has the authority to supervise the daily conduct of
their business. With the introduction of the clear authority of the securities
regulator in this area, the discipline and quality of the market is expected
to improve.
A. 8 Invest or Pr ot ect i on
In order to create a fair and orderly market, the prohibition of fraudulent
activities must be one of the major goals of the Law. The importance of this
aspect is indicated by the prominent position in which the Law has placed
as "securities crimes". Article 5 in the General Principles section defines
the securities crimes of fraud, insider trading and market manipulation,
which are all prohibited.
The detailed provisions concerning these activities are contained in Part
4 of Chapter 3 (Articles 67-77). Insider trading is defined by Article 67
to mean the activity of an insider using insider information to trade in
securities. The scope of an insider and insider information are defined.
Insiders who are aware of insider information are prohibited from trading
in securities, divulging the insider information or advising others to trade
in the relevant securities. The penalty for this offence includes a mandatory
sale of the securities in question, confiscation of the illegal gains and a fine
of up to five times the amount of illegal gains or the value of the securities
in question, whichever is lower (Article 183). The Law states that a heavier
penalty will be imposed on securities regulators who violate this provision.
In addition, the Law introduces further offences of the acquisition of
396
Securities Law
unlawful interest or the shifting of risk (Article 71) and the dissemination
of false information (Article 71). Article 71 defines the former offence as
follows:
To, singly or in concert with others, make use of one's
competitive edge over capital, share-holding or infor-
mation to collectively or continuously transact securi-
ties and manipulate their prices;
to intentionally conspire with others to engage in secu-
rities transactions or mutually buy and sell securities
that they did not own, according to a pre-determined
time, price and manner, with a view to affecting the
prices or volume of securities trading;
to use oneself as the opposite party in a transaction so
as to be both the buyer and the seller, with no change
in beneficial ownership, with a view to affecting the
prices or volume of securities trading; or
other means of manipulating the prices of securities
trading.
A similar provision governing market manipulation in futures trading
is found in Article 62 of the P.R.C. Provisional Regulations on Futures
Trading, which came into force on September 1, 1999.
Article 72 prohibits the fabrication or dissemination of false information
concerning securities trading. Government employees and media reporters
are specifically mentioned. In addition, this provision prohibits personnel
working in the various bodies regulated under the law from making false or
misleading statements.
In terms of actual enforcement, the Law strengthens the regulators
power to impose administrative penalties, which are particularized in Chap-
ter 11. These penalties may take the form of a warning, restitution, confis-
cation and fines. Added to that is the provision of administrative liabilities
for state officials who violate the Law.
The Law specifically preserves the possibility of criminal liability in se-
rious cases, which can arise under the P.R.C. Criminal Law. The 1997
amendment to the Criminal Law included a variety of commercial crimes
that may apply to unlawful conduct in the securities market, such as mis-
representation in prospectuses and disclosure documents, insider dealing,
market manipulation, activities relating to illegal use of foreign exchange
Overseas Markets 397
and causing instability to the financial system. Chapter 3 deals with new
economic crimes that undermine the order of the socialist market economy
and contains eight sections. Sections 3-5 (Articles 158-200) relate to finan-
cial crimes such as "undermining the management order in a company oi
enterprise", "disrupting the financial order" and "financial fraud".
A. 9 Overseas Mar ket s
At present, there are two main types of overseas-related listings. The firsl
type is where the listing of a company's securities takes place on the twc
domestic stock exchanges, but the shares are issued to overseas investors
These shares may only be traded by non-PRC nationals and companies anc
are transacted in foreign currency. They are known as overseas-invested do-
mestic shares (or B-shares in short). The other type is where the listing o
a company's securities takes place on an overseas exchange and the secu-
rities are offered to overseas investors. Where they are listed on the Stocl
Exchange of Hong Kong, they are known as overseas issued or listed share;
(or H-shares in short).
Since the Law's main focus is on the domestic market, it contains little
provision for overseas-related listings. What it does provide is the gen-
eral principle that all overseas related offerings (e.g., B-shares) and listings
(e.g., H-shares) must first be approved by the regulator. Article 29 states
that where a domestic issuer wishes to issue or list its securities overseas,
whether directly or indirectly, it must receive the approval of the securities
regulator. The reference to direct or indirect activities gives the regulator
the power to regulate all sorts of overseas capital-raising activities of com-
panies, including back-door listings. In the past, the regulator's ambition
over these areas has been the subject of debate.
Article 213 states that specific measures will be enacted by the State
Council to govern the offering of securities by domestic companies for sub-
scription by overseas investors. Currently, the State Council's Regulations
for the Listing of Foreign Invested Domestic Shares promulgated on Febru-
ary 1, 1995 apply to B-shares. The governing provisions for H-shares are
contained in the State Council's Regulations for the Overseas Issuance and
Listing of Shares promulgated on April 8, 1994.
The absence of more definitive remarks in the Law was seen as disap-
pointing for some foreign investors eager to tap the potential of the Chinese
398 Securities Law
market. From an optimistic point of view, however, the silence may suggest
that the authorities are satisfied with the status quo and will continue to
allow flexibility in the governance of this area by the use of administrative
rules.
A. 10 The Role of Lawyers
Regarding the role of lawyers, it was recently reported in the China Securi-
ties Weekly that the mainland legal community welcomed the promulgation
of the Securities Law because practitioners believed that the improvements
to the legal framework and the substantive contents of the governing provi-
sions would reduce the level of difficulty in this area of practice. At present,
lawyers who wish to practice in the securities field are required to obtain an
authorization to that effect from the CSRC. This is not expected to change
under the Law.
The Securities Law recognizes the legal opinion issued by a law firm
as one of the essential documents required for approval (Article 45) and
imposes certain responsibilities and liability on legal advisers. Article 13(2)
provides that a professional body that issues documentation for the purpose
of a share issuance must strictly discharge its statutory function and ensure
the truthfulness, accuracy and completeness of these documents. According
to one Mainland practitioner, the role of lawyers is not merely to act as an
intermediary service provider or the agent of the issuer or underwriter. It
is also a defender of an orderly market and an enforcer of the law to the
extent that it discharges statutory functions in issuing legal opinions that
will be relied upon by the authorities in approving an application. As such,
a law firm has a duty of care in providing the legal opinion.
A. 11 Summary
With the implementation of the Securities Law in July 1999 and the Provi-
sional Regulations on Futures Trading in September 1999, it can be argued
that the mainland's securities and futures markets have entered a new phase
of development and steady growth may be expected in the coming years.
In comparison to the fledgling futures market, it is clear that the securities
market has progressed significantly further and benefited from the clear pol-
icy support of the authorities. Chinese authorities is dedicated to, at least
Summary 399
from the enactment of the Law, the protection of investors, the reduction of
systematic risk and the ensuring of a fair, efficient and transparent financial
market. As the securities market matures, it is expected that the Chinese
authorities will take full advantage of the various windows left open in the
law to expand the scope of market activities and welcome increased outside
participation. Securities practitioners around the world can no doubt look
forward to this with a high degree of optimism.
Appendix B
Summary of P. R. C. Taxes
Individual Income Tax
The personal allowance for an expatriate is RMB 4,000 per month. There
are no additional allowances for spouses or dependent children. Table B.l
shows that individual income tax is assessed monthly at progressive tax
rates which vary depending on the types of income.
Foreign Enterprise Income Tax
Foreign Enterprise Income Tax is levied on taxable income derived by for-
eign investment enterprises (FIE) and establishments of foreign enterprises
Table B. l Income Tax Rate for Monthly Salary k. Wages
Salary range
less than RMB 500
RMB 501 - RMB 2,000
RMB 2,001 - RMB 5,000
RMB 5,001 - RMB 20,000
RMB 20,001 - RMB 40,000
RMB 40,001 - RMB 60,000
RMB 60,001 - RMB 80,000
RMB 80,001 - RMB 100,000
greater than RMB 100,000
tax rate
5%
10%
15%
20%
25%
30%
35%
40%
45%
401
402 Summary of P.R.C. Taxes
in China. The standard tax rate is 33%. This is comprised of state tax of
30% and local tax rate of 3%.
The state tax of 30% may be reduced to 15% or 24% if the enterprises are
located in one of the specially designated zones in China and/or engaged in
some prescribed businesses or projects. The local tax of 3% may be waived
or reduced by the local governments.
A FIE that engages in production for a period of ten years or more will
be exempt from foreign enterprise income tax during the first two years and
is eligible for a 50% reduction in tax during the subsequent three years,
starting from the first profit-making year.
A foreign investor of a FIE who directly reinvests its profits into the
FIE or establishes another FIE may obtain a refund of 40% of the state
tax paid. If profits are invested in an export-oriented company or high-tech
company, the foreign investor may obtain a 100% tax refund.
With respect to depreciation allowances, the residual value of fixed as-
sets should not be less than 10% of the original cost. Depreciation is gener-
ally calculated using the straight line method. The minimum depreciation
periods are as follows: 20 years for buildings and structures, 10 years for
trains, ships, machinery, apparatus and other production equipment, and
5 years for electronic equipment, means of transport other than trains and
ships, as well as appliances, tools, furniture related to production or busi-
ness operations.
Capital Gains Taxes
Capital gains are treated as income and taxed accordingly. Both Chinese
residents and non-residents are taxed at 20% with respect to capital gains.
The rate is reduced to 10% for capital gains realized within specially des-
ignated zones in China.
Wi thhol di ng Taxes
Withholding tax is imposed on interest, rent, royalties and capital gains
received by non-residents. The statutory rate of withholding tax is 20%.
It is reduced to 10% for income derived from specially designated zones in
China. It could also be reduced under avoidance of double taxation treaties.
Dividends from a FIE to a foreign shareholder are exempt from withholding
taxes.
Summary of P.R.C. Taxes 403
Table B.2 Business Taxes Schedule
Taxable services
Construction
Cultural activities and sports
Post and telecommunications
Transportation
Finance and insurance
Sale of immovable properties
Transfer of intangible assets
Entertainment
Other services
Tax rate
3%
3%
3%
3%
5%
5%
5%
5 - 20%
5%
Value Added Tax
All individuals and units which sell taxable goods, provide processing or
repair services in China or which import goods into China are subject to
value added tax (VAT). The standard VAT rate is 17%. However, certain
items are subject to a reduced rate of 13%. Small scale taxpayers pay VAT
at the rate of 6% but will not be eligible for input VAT tax credits.
Consumpti on Tax
Consumption Tax is levied on the importation, production and processing
of luxury or high-end goods in China. There are 11 categories of goods
which are subject to the consumption tax, including tobacco, alcoholic
drinks, cosmetics, skin and hair care products, jewelry, fireworks, gaso-
line, diesel oil, tires, motorcycles and automobiles. The tax rate ranges
from 3 to 45%. Some items are taxed at a fixed amount per unit of volume.
Business Tax
All firms and individuals who provide services listed in Table B.2 are re-
quired to pay business taxes calculated on the gross amount received.
404 Summary of P.R.C. Taxes
Table B.3 Stamp Duty Schedule
Type of Document
Purchase and sales
Engineering projects
Storage and custody of goods
Properties leasing
Loans
Processing agreements
Construction & Installation
Properties insurance
Transportation
Technology
Tax Rat e
0.03%
0.05%
0.10%
0.10%
0.005%
0.05%
0.03%
0.003%
0.05%
0.03%
Customs Dut y
Customs duty is levied on the importation of goods into China. Before April
1, 1996, FIEs importing machinery, equipment and materials required for
producing exports were free of customs duty, VAT and consumption taxes.
This preferential tax treatment has been withdrawn for FIEs which were
approved after March 1996. However, a grace period is granted to FIEs
who have been approved before April 1, 1996.
Stamp Dut y
Stamp duty is levied on stipulated documents which are effective under
and protected by the Chinese Laws. The applicable duty rates for various
contracts are contained in Table B.3:
Appendix C
Major Events in Chinese Stock
Markets
1990
January 4: Shenzhen Branch of the People's Bank of China issued:
Some suggestions on setting up the Shenzhen Securities Exchange
February 7: Single trading system was changed into counter concen-
tration trading system in Shenzhen Securities Corporation. Agree-
ments on selling and buying were signed to adopt the rule of price
priority and time priority.
May 28: To strengthen the regulation over securities market and to
demolish the illegal trading outside the market, Shenzhen municipal
government stipulated that trading need had to be processed via
securities intermediaries.
May 29: Shanghai Branch of the People's Bank of China announced:
The bidding price should not be 10% higher or lower than the price
of the previous day's closing price.
June 24: Radio Shenzhen started to broadcast daily stock quota-
tion.
July 15: Shenzhen Telecom, launched the stock information chan-
nel.
August 25: Shanghai Securities Daily started to publish stock quo-
tations.
September 28: Securities trading automatic quotation (STAQ) sys-
tem was launched.
November 27, Shanghai municipality government published the rules
for regulation over securities market.
December 1: After one year of preparation, Shenzhen began exper-
405
406 Major Events in Chinese Stock Markets
imental stock trades over the counter.
December 18: People's Bank of China announced that shares could
be issued publicly all over China but could only traded in Shenzhen
or Shanghai.
December 19: Opening ceremony for the Shanghai Securities Ex-
change was held with the presence of Zhu Rongji, Mayor of Shang-
hai, Liu Hongru, vice director of State System Reform Commission
and Zhou Zhengqing, vice governor of People's Bank of China.
1991
April 3: The Shenzhen Securities Exchange published the SZSE
Composite Index with April 3, 1991 as the starting date and 100
as the starting point.
April 11: The Shenzhen Securities Exchange was formally estab-
lished with the approval of People's Bank of China.
April 22: There was no transaction on the Shenzhen Securities Ex-
change, marking the historical record of the lowest trading volume.
July 26: Ceremony was held for using computer transaction net-
work system in the Shanghai and Shenzhen securities exchanges.
August 12: First convertible bond was issued by the Hainan New
Energy Group.
August 28: Council for Securities Research was established in Bei-
jing-
November 21: Premier Li Peng visited the Shanghai Securities Ex-
change.
November 29: First B-share was successfully issued by the Shanghai
Vacuum Electronics.
1992
January 2: the Shenzhen Securities Exchange began using the 5-
day work schedule following the international standard.
February 28: B-shares issued by the Shanghai Vacuum Electron-
ics and Shenzhen Nanbo were traded in the secondary exchanges,
marking the establishment of B-share market in China.
April 7: Shenzhen Branch of People's Bank of China started the
investigation into the finance of Shenzhen Yuanye.
Major Events in Chinese Stock Markets 407
May 21: The daily price change limit was removed in the Shanghai
Securities Exchange, causing the market index to be more than
doubled in a single trading day.
May 26: First brokers started to trade stocks in Shenzhen.
September 28: Huaxia, Guotai, and Nanfang securities corpora-
tions started business as three of the largest brokers with the ap-
proval of People's Bank of China.
October 8: China Securities Daily published its first issue in Bei-
jing-
October 26: Xinhua News Agency announced: With the estab-
lishment of the CSRC, Zhu Rongji was appointed concurrently as
the director of the State of Council and Liu Hongru was appointed
the chairman of CSRC, symbolizing the development of securities
market in a standard way.
November 20: Wuhan Department Store was listed on the Shenzhen
Securities Exchange as the first non-local company got listed.
1993
January 20: Wang Xiyi, governor of Shenzhen Branch of the Peo-
ple's Bank of China said that the Shenzhen Securities Exchange
will no longer be managed by the bank.
March 1: Trading on employee shares of Feile Accoustics began on
the Shenzhen exchange, marking the first employee shares being
transacted.
April 22: The State Council issued Provisional Regulations on Stock
Issues and Transactions.
May 9: The State Council reiterated that no government official
was supposed to receive stocks as gifts or compensations.
July 7: The State Council issued Provisional Regulations on Secu-
rities Exchanges.
August 15: The CSRC issued Provisional Regulations on Fraud
Prevention in Securities Transactions.
August 20: With the approval of the People's Bank of China, Zibo
Mutual Fund, the first mutual fund in China, was established.
October 25: Trading began on the government Treasury Futures in
the Shanghai Securities Exchange.
December 29: P.R.C. Company Law was promulgated by the Na-
tional People's Congress.
1994
January 25: The CSRC issued regulations on information disclos-
ing and shares issuing, requesting all listed companies to disclose
annual report in the required form.
February 4: The State Council set a quota in the amount of 5.5
billion renminbi yuan for the stock issues in that year.
March 19: Shenzhen Branch of People's Bank of China and Shen-
zhen Securities Regulation Office made the annual inspection over
securities registration offices, brokers, department offices, trading
offices, and mutual funds. Those that operated illegally were closed.
April 29: A memorandum was signed in Beijing between t he CSRC
and U.S. Securities and Exchange Commission, marking the begin-
ning of cooperation between the two countries.
May 12: The CSRC reiterated that dividends were subject to 20%
personal withholding tax.
June 27: Auction method was first used in setting the IPO prices
for Haisuibao (renminbi 4.6) and Hainan Jinpan (renminbi 6.68).
July 2: Association of Shenzhen Securities was established.
August 4: Shangdong Huaneng was listed in the NYSE, marking
the first Chinese stock listed in the U.S..
August 30: A satellite statistic network was launched in Shenzhen.
Investors across the nation were able to buy and sell stocks listed
in the Shenzhen Securities Exchange.
October 26: The CSRC approved 10 futures exchanges in China.
October 31: The CSRC published details for Shares subscription
application and information disclosing. The State Council required
that state-owned shares, legal-person shares, subscription shares
and dividend shares were not allowed to be traded.
November 14: Financial Times reported that social insurance fund
would not allowed into Chinese securities market in the near future.
November 25: The CSRC restricted the number of local securities
institutions that could be established.
December 16: Shanghai service center for the Shenzhen Securities
Exchange went into operation.
1995
January 5: The Shenzhen Securities Exchange decided to revise its
composite index with 1000 as the starting point and July 20 as the
starting date.
January 9: People's Bank of China and Ministry of Finance issued
a joint statement requiring the separation of financial securities
institutions from Ministry of Finance and commercial banks.
February 23: Unusual activity was spotted in Shanghai Bond and
Future market. Wanguo Securities operated illegally resulting in
327 illegal transactions.
March 31: Mr. Zhou Daojong was appointed as the chairman for
the CSRC.
May 10: Shenzhen Securities Registration Corporation announced
partial floating of state-owned shares and legal person shares in the
two official exchanges in China.
May 17: The CSRC halted the trading of bond and futures saying
that there were serious illegal activities in domestic bond and future
market. It announced that China was not ready for establishing
markets for bond and futures.
June 29: Effective on July 3, the trading time was set as from
9:30-11:30am and l:00-3:00pm
August 11: China Finance Corporation was established as the first
Chinese investment bank.
September 21: The Ministry of Inspection and the CSRC published
the investigation result for Wanguo Securities case.
October 6: Shenzhen Securities Registration Corporation issued
unified registration card for securities transactions.
1996
January 5: The CSRC implemented a new format of annual report
to allow for sufficient information disclosing.
March 25: To strengthen securities supervision, the CSRC estab-
lished 24 branches across the nation.
April 1: The People's Bank of China announced, effectively from
that day, that there would be no fixed interest rate for savings
deposits. The central bank would adjust interest rate according to
market demand and supply for loanable funds.
April 8: The State Council and the CSRC issued 8 principles in fur-
ther developing China's securities markets, namely, strengthening
the regulation over the securities market, speeding up the legislation
over securities market, improving the macro control over securities
market, balancing the role of securities market in economic reform,
allowing the demand and supply of capital to work their way, im-
proving the quality of the listed companies, developing financial
institutions and exploring overseas securities market.
April 19: Following the international practice, Shanghai and Shen-
zhen securities exchanges started to categorize all the listed com-
panies in terms of industry.
May 27: China Securities Daily and Securities Times jointly pub-
lished an index for the best-performed companies in the stock mar-
kets, using May 24 as the starting date and 1000 as the starting
point.
May 29: Dow Jones began publishing Dow Jones China Index, Dow
Jones Shanghai Index and Dow Jones Shenzhen Index.
July 1: The CSRC strengthened the regulation and supervision
over B-share registration and transaction.
October 3: The CSRC launched an investigation against Bohai
Group, one of the most reputable Chinese firms, on the grounds of
illegal accounting practices.
October 18: The Shanghai Securities Exchange improved the online
subscription of shares. Subscription period was shortened to three
or four days, with T+0 for application, T+l for fund transfer, T+2
for fund approval, T+3 for lot-draw and T+4 for removing the hold
on funds.
November 1: The CSRC announced procedures to prevent market-
making.
December 15: The CSRC reimposed a price change limit of 10% on
share transactions.
1997
January 14: The People's Bank of China closed the Agricultural
Entrust Investment Corporation for its illegal operation that had
led to enormous amount of financial loss. The Construction Bank
of China was designated to take over the equity and debt of China
Agricultural Entrust Investment Corporation.
January 20: The CSRC fined Zhangjiajie for illegal transaction of
management shares and Hunan Securities Exchange for its illegal
financing of renminbi 750,000 yuan for Zhang Jiajie.
February 20: The Shenzhen Composite Index rose by 9.54% at the
news of Deng Xiaoping's death.
March 4: The CSRC announced Provisional Regulations of Prohi-
bition Into Securities.
May 22: The Securities Commission of the State Council, the Peo-
ple's Bank of China, the State Economic and Trade Promotion
Commission jointly announced that state-owned companies and
listed companies are prohibited from trading stocks.
June 6: The People's Bank of China announced the prohibition of
bank funds into stock market.
October 1: The new P.R.C. Criminal Law went into effect. The
CSRC announced that rigging the market will be punished accord-
ing to the Criminal Law.
November 17: Zhou Zhengqing, chairman of the CSRC signed the
Provisional Regulations over Securities Fund.
1998
March 17: Kaiyuan and Jintai Securities Funds were approved to
go public on the two official exchanges.
March 30, the CSRC certified the first 74 stock and futures con-
sulting institutions and 487 financial consultants.
April 3: The People's Daily pointed out it was important to boost
the securities market and to serve the reconstruction for the state-
owned companies. Investors took the news as a signal of near-term
government support of the securities market, and the stock markets
in Shanghai and Shenzhen surged.
April 22: The Shanghai and Shenzhen Securities Exchange pub-
lished a "black list" of firms whose financial performance had been
among the worst during the previous three years.
June 15: Case of Hainan Minyuan was investigated in court in
Beijing.
July 23: Guotai and Jinan Securities, two of the largest securities
investment firms in China, announced a merger with each other.
September 14: Trading on some companies was suspended by the
exchanges until the companies could earn net profit in the next
three years.
November 20: The CSRC looked into the illegal performance of
Hongguan.
December 18: The CSRC required further licensing of securities
institutions on laws and regulations of securities industry.
December 29: The 9th National People's Congress promulgated the
P.R. C. Securities Law.
1999
January 5: The CSRC set up new rules for rights issues as a method
of floating seasoned equity offerings.
April 19: Zhou Zhengqing, the chairman of the CSRC, announced
that he would phase out the tedious procedural requirements for
companies to raise funds from overseas exchanges.
May 19: Shares listed in the Shanghai and Shenzhen Securities
Exchanges surged because of high-tech fever.
July 1: The P.R.C. Securities Law was formally published. But
the Shenzhen and Shanghai Securities Exchanges fell 7.61% and
6.28%, respectively, breaking the single-day share-dropping record
since 1996. Nearly 500 shares fell to the lowest prices of the year.
July 12, Zhongguancun got listed to replace the original Hainan
Minyuan. The listed price was renminbi 34.09 yuan, about 45%
higher than the price of Hainan Minyuan as of its last day of trad-
ing.
September 8: The CSRC announced that state-owned enterprises
and listed companies could invest in the secondary market but the
hold period for each buying and selling of the same share must
exceed 6 months. The Shanghai and Shenzhen composite indices
rose by more than 6% that day.
October 13, the CSRC issued detailed rules on listing in Hong Kong
stock market by domestic companies and clarified that all compa-
nies could apply for overseas listing but high-tech companies would
be given the priority.
October 18: Zhou Zhengqing published an essay in People's Daily
to develop the securities market and to boost the high-tech industry.
November 12: Provisional Regulations on Fund Operators was is-
sued by the CSRC. Thus the management system for securities,
futures and investment funds came into being.
November 26: The Ministry of Inspection punished Daqing Lianyi
Corporation Ltd. for the fraudulent financial practice committed
by the management.
December 27: Dongfang Electronic listed 0.23 billion employee
shares with a market value of renminbi 3.7 billion. It was the
largest flotation of employee shares in China.
Bibliography
Alford, A., Jones, J., Leftwish, R. and Zmijewski, M. (1993) "The relative in-
formativeness of accounting disclosures in different countries", Journal of
Accounting Research 31, 183-223.
Allen, F. and Faulhaber, G. R. (1989) "Signaling by underpricing in the IPO
market", Journal of Financial Economics 23, 303-323.
Amihud, Y. and Mendelson, H. (1987) "Trading mechanisms and stock returns:
An empirical investigation", Journal of Finance 42, 533-553.
Amir, E., Harris, T. and Venuti, E. (1993) "A comparison of the value-relevance
of U.S. versus Non-U.S. GAAP accounting measures using form 20-F rec-
onciliations", Journal of Accounting Research 31, 230-264.
Andersen, T. G. (1994) "Stochastic autoregressive volatility: A framework for
volatility modeling", Mathematical Finance 4, 75-102.
Andersen, T. G. (1996) "Return volatility and trading volume: An information
flow evidence", Journal of Finance 51, 169-204.
Anderson, T. W. (1984) Probability and Mathematical Statistics, John, Weiley &
Sons, New York.
Ang, J. S. and Ma, Y. (1999) "Transparency in Chinese stocks: A study of
earnings forecasts by professional analysts", Pacific-Basin Finance Journal
7 129-155.
Bailey, W. and Jagtiani, J. (1994) "Foreign ownership restrictions and stock prices
in the Thai market", Journal of Financial Economics 36, 57-87.
Baillie, R. T. and Bollerslev, T. (1989) "The message in daily exchange rates: A
conditional variance tale", Journal of Business and Economic Statistics 7,
297-305.
Ball, R. and Brown, P. (1968) "An empirical evaluation of accounting income
numbers", Journal of Accounting Research 6, 159-178.
Banz, R. (1981) "The relationship between return and market value of common
stocks", Journal of Financial Economics 9, 3-18.
Basu, S. and Li, D. D. (1998) "Corruption in transition", University of Michigan
415
416 Bibliography
Working Paper.
Beatty, R. P. and Ritter, J. R. (1986) "Investment banking, reputation, and the
underpricing of initial public offerings", Journal of Financial Economics
15, 213-232.
Bekaert, G. and Harvey C. R. (1997) "Emerging equity market volatility", Journal
of Financial Economics 43, 29-78.
Bodurtha, J. N., Jr., and Mark, N. C. (1991), "Testing the CAPM with time-
varying risks and returns", Journal of Finance 46, 1485-1505.
Bollerslev, T. (1986) "Generalized autoregressive conditional heteroskedasticity"
Journal of Econometrics 31, 307-327.
Bollerslev, T. (1987) "A conditional heteroskedastic time series model for specu-
lative prices and rates of return", Review of Economics and Statistics 69,
542-547.
Bollerslev, T., Chou, R. Y. and Kroner, K. F. (1992) "ARCH modeling in finance:
A review of the theory and empirical evidence", Journal of Econometrics
52, 5-59.
Brown, S. and Warner, J. (1985) "Using daily stock returns: The case of event
studies", Journal of Financial Economics 14, 3-31.
Camerer, C. (1989) "Bubbles and fads in asset prices: Review of theory and
evidence", Journal of Economic Survey 3, 3-41.
Campbell, J. Y., Lo, A. W. and MacKinlay, A. C. (1997) The Econometrics of
Financial Markets, Princeton University Press, New Jersey.
Carlton, D. W. and Perloff, J. M. (1994) Modern Industrial Organization, Harper-
Collins, New York.
Cecchetti, S. G., Lam, P. S. and Mark, N. C. (1990) "Mean reversion in equilib-
rium asset prices" American Economic Review 80, 398-418.
Chakravarty, et. al (1998)
Chan, K.C., Chen, N. F. and Hsieh, D. A. (1985) "An exploratory investigation
of the firm size effect", Journal of Financial Economics 14, 451-471.
Chan, K. and Seow, G. (1996) "The association between stock returns and for-
eign GAAP earnings versus earnings adjusted to U.S. GAAP, Journal of
Accounting and Economics 21, 139-158.
Chari, V., Jagannathan, R. and Ofer, A. (1988) "Seasonalities in security returns:
The case of earnings announcements", Journal of Financial Economics 21,
101-121.
Chau, C. T., Ciccotello, C. S. and Grant, C. T. (1998) "Role of ownership in
Chinese privatization: Empirical evidence from returns in IPOs of Chinese
A-shares, 1990-1993", Advances in Financial Economics 4, 124-139.
Chemmanur, T. J. (1993) "The pricing of initial public offerings: A dynamic
model with information production", Journal of Finance 48, 285-304.
Chen, Z. and Knez, P. J. (1995) "Measurement of market integration and arbi-
trage", Review of Financial Studies 8, 287-525.
Chopra, N., Lakonishok, J., and Ritter, J. (1992) "Measuring abnormal perfor-
mance: Do stocks overreact?", Journal of Financial Economics 31, 235-268.
Bibliography 417
Chowdhry, B. and Sherman, A. (1996) "International differences in oversubscrip-
tion and underpricing of IPOs", Journal of Corporate Finance 2, 359-381.
Clark, P. K. (1973) "A subordinated stochastic process model with finite variance
for speculative prices", Econometrica 41, 135-155.
Copeland, T. E. and Weston, J. F. (1988) Financial Theory and Corporate Policy,
Addison-Wesley, New York.
Cross, F. (1973) "The behavior of stock prices on Fridays and Mondays", Finan-
cial Analysts Journal, 29 67-69.
Datar, V. and Mao, D. (1998) "Initial public offerings in China: Why is under-
pricing so severe?", Seattle University working paper.
De Santis, G. and Imrohoroglu, S. (1995) "Stock returns and volatility in emerging
financial markets" Journal of International Money and Finance 16, 561-
579.
Dewenter, K. L. and Malatesta, P. H. (1997) "Public offerings of state-owned
and privately-owned enterprises: An international comparison", Journal of
Finance 52, 1659-1679.
Dickey, D. A. and Fuller, W. A. (1979) "Distribution of the estimators for autore-
gressive time series with a unit root" Journal of the American Statistical
Association 74, 427-431.
Dickey, D. A. and Fuller, W. A. (1981) "Likelihood ratio statistics for autoregres-
sive time series with a unit root", Econometrica 49, 1057-1072.
Domodaran, A. (1993) "A simple measure of price adjustment coefficients", Jour-
nal of Finance 48, 387-400.
Domowitz, I., Glen, J. and Madhavan, A. (1997) "International cross-listing,
foreign ownership restrictions and order flow migration: Evidence from
Mexico", Journal of Finance 52, 1059-1085.
DuMouchel, W. H. (1975) "Stable distributions in statistical inference: 2. Infor-
mation from stably distributed samples", Journal of the American Statis-
tical Association 70, 386-393.
Easton, P. and Zmijewski, M. (1989) "Cross-sectional variation in the stock mar-
ket response to the announcement of accounting earnings", Journal of Ac-
counting and Economics 11, 117-142.
Eckbo, B. and Masulis, R. (1992) "Adverse selection and the rights offer paradox",
Journal of Financial Economics 32, 293-332.
Engle, R. F. (1982) "Autoregressive conditional heteroskedasticity with estimates
of the variance of U.K. inflation", Econometrica 50, 987-1008.
Engle, R. F. and Granger, C. W. J. (1987) "Cointegration and error correction:
Representation, estimation and testing", Econometrica 55, 251-276.
Engle, R. F. , and Ng, V. K. (1993) "Measuring and testing the impact of news
on volatility", Journal of Finance 48, 1749-1778.
Epps, T. W., and Epps, M. L. (1976) "The stochastic dependence of security
price changes and transaction volumes: Implications for the mixture-of-
distribution hypothesis", Econometrica 44, 305-321.
Errunza, V. and Losq, E. (1985) "International asset pricing under mild segmen-
418 Bibliography
tation: Theory and test" Journal of Finance 40, 105-124.
Eun, C. and Janakiramanan, S. (1986) "A model of international asset pricing
with a constraint on the foreign equity ownership", Journal of Finance 41,
897-914.
Fama, E. F. (1991) "Efficient capital markets: II", Journal of Finance 46, 1575-
1617.
Fama, E. F., Fisher, L., Jensen, M. and Roll, R. (1969) "The adjustment of stock
prices to new information", International Economic Review 10, 1-21.
Fama, E. F. and French, K. R. (1988) "Permanent and temporary components
of stock prices", Journal of Political Economy 96, 246-273.
Fama, E. F. and French, K. R. (1992) "The cross-section of expected stock re-
turns", Journal of Finance 47, 427-465.
Fama, E. F. and MacBeth, J. D. (1973) "Risk, return and equilibrium: Empirical
tests", Journal of Political Economy 81, 607-636.
Fama, E. F. and Roll, R. (1971) "Parameter estimates for symmetric stable dis-
tributions", Journal of The American Statistical Association 66, 331-338.
Fang, J. (1991), "Shanghai: Stock market re-established", Beijing Review 34,
38-40.
Fernald, J. and Rogers, J. H. (1998) "Puzzles in Chinese stock markets", Board
of Governors of the Federal Reserve System working paper.
Ferson, W. E., Kandel, S., and Stambaugh, R. F. (1987) "Tests of asset pricing
with time-varying expected risk premiums and market betas", Journal of
Finance 42, 201-220.
French, D. (1980), "Stock returns and weekend effect", Journal of Financial Eco-
nomics 8, 55-69.
French, K. R., Schwert, G. W. and Stambaugh, R. F. (1987) "Expected stock
returns and volatility", Journal of Financial Economics 19, 3-30.
Fuller, W. A. (1976) Introduction to Statistical Time Series, John Wiley and Sons,
New York.
Gallant, A. R., Rossi, P. E. and Tauchen, G. E. (1992) "Stock prices and volume",
Review of Financial Studies 5, 199-242.
Garfmkel, J. (1993) "IPO underpricing, insider selling and subsequent equity
offerings: Is underpricing a signal of quality?", Financial Management 22,
74-83.
Gennotte, G. and Truemann, B. (1996) "The strategic timing of corporate dis-
closures" , Review of Financial Studies 9, 665-690.
Ghysels, E., Harvey, A. C. and Renault, E. (1996) "Stochastic volatility", in
Maddala, G. S. and Rao, C. D. Handbook of Statistics 14, 119-191.
Grinblatt, M. and Hwang, C. Y. (1989) "Signaling and the pricing of unseasoned
new issue", Journal of Finance 44, 393-420.
Grinblatt, M., Masulis, R. and Titman, S. (1984) "The valuation effect of stock
splits and stock dividends", Journal of Financial Economics 13, 461-490.
Grossman, S. and Hart, O. (1982) "Corporate financial structure and managerial
incentives", in: McCall, J. The Economics of Information and Uncertainty,
Bibliography 419
University Press, Illinois.
Grossman, S. and Hart, O. (1986) "The costs and benefits of ownership: A theory
of vertical and lateral integration", Journal of Political Economy 94, 691-
719.
Groves, T., Hong, Y., McMillan, J. and Naughton, B. (1994) "Autonomy and
incentives in Chinese state enterprises", Quarterly Journal of Economics
109, 183-209.
Gultekin, M., Gultekin, N. B. and Penati, A. (1987) "Capital controls and inter-
national capital market segmentation", Journal of Finance 44, 849-870.
Hamilton, J. D. (1994) Time Series Analysis, Princeton University Press, New
Jersey.
Hansen, L. P. (1982) "Large sample properties of generalized methods of moments
estimators", Econometrica 50, 1029-1054.
Harris, L. (1986a) "A transaction dat a study of weekly and intradaily patterns
in stock returns", Journal of Financial Economics 16, 99-117.
Harris, L. (1986b) "Cross-security tests of the mixture of distributions hypothe-
sis", Journal of Financial and Quantitative Analysis 21, 39-46.
Harris, L. (1987) "Transaction dat a tests of the mixture of distributions hypoth-
esis", Journal of Financial and Quantitative Analysis 22, 127-141.
Harris, M. and Raviv, A. (1990) "Capital structure and the information role of
debt", Journal of Finance 45, 321-349.
Harris, M. and Raviv, A. (1991) "The theory of capital structure", Journal of
Finance 46, 297-355.
Hart, O. (1995) Firms, Contracts, and Financial Structure, Oxford University
Press, London, United Kingdom.
Hart, O. and Moore, J. (1990) "Property rights and the nature of the firm",
Journal of Political Economy 98, 1119-1158.
Hart, O. and Moore, J. (1998) "Cooperatives vs. outside ownership", Harvard
University Working Paper.
Harvey, C. R. (1995) "Predictable risk and returns in emerging markets", The
Hietala, P. (1989) "Asset pricing in partially segmented markets: Evidence from
the Finnish market", Journal of Finance 44, 697-718.
Heinkel, R. and Schwartz, E. (1986) "Rights versus underwritten offerings: An
asymmetric information approach", Journal of Finance 41, 1-18.
Hirshleifer, D. and Thakor, A. V. (1992) "Managerial conservatism, project choice
and debt", Review of Financial Studies 5, 437-470.
Hsieh, D. A. (1989) "Modeling heteroskedasticity in daily foreign exchange rates",
Journal of Business and Economic Statistics 7, 307-317.
Huang, Q. (1997) "The underpricing and aftermarket performance of new public
equity offerings by recently privatized firms: An international test", New
York University doctoral dissertation.
Ingersoll Jr., J. (1987) Theory of Financial Decision Making, Rowman & Little-
field, Maryland.
420 Bibliography
Jagadeesh, N., Weinstein, M. and Welch, I. (1993) "An empirical investigation
of IPO returns and subsequent equity offerings", Journal of Financial Eco-
nomics 34, 153-175.
Jefferson, G. H. and Rawski, T. G. (1994) "Enterprise Reform in Chinese Indus-
try", Journal of Economic Perspective 8, 446-473.
Jensen, M. C. (1986) "Agency costs of free cash flow, corporate finance and
takeovers", American Economic Review 76, 323-329.
Jensen, M. C. and Meckling, W. H. (1976) "Theory of firm: Managerial behavior,
agency cost and capital structure", Journal of Financial Economics 3, 305-
360.
Johnson, D. L. (1995) "Barriers to international investment: Stock premiums in
Sweden", University of Wisconsin Working Paper.
Jorion, P. (1988) "On j ump processes in the foreign exchange and stock markets",
Jung, K., Kim, Y. and Stulz, R. (1996) "Timing, investment opportunities, man-
agerial discretion, and the security issue decision", Journal of Financial
Economics 42, 159-185.
Keloharju, M. (1993) "The winner's curse, legal liability, and the long run price
performance of initial offerings in Finland", Journal of Financial Economics
34, 251-277.
Kernen, A. (1997) "State enterprises in Shenyang: Actors and victims in the
transition", China Perspectives 12, 26-32.
Kim, D. (1995) "The errors in the variables problems in the cross-section of
expected stock returns", Journal of Finance 50, 1605-1634.
Kinge, J. (2000) "Capital flights from China: Twice estimates", Financial Times
Jul y 9.
Koh, F., and Walter, T. (1989) "A direct test of Rock's model of the pricing of
unseasoned issues", Journal of Financial Economics 23, 251-271.
Koutmos, G. (1998) "Asymmetries in the conditional mean and the conditional
variance: Evidence from nine stock markets", Journal of Economics and
Business 50, 277-290.
Kross, W. and Schroeder, D. (1984) "An empirical investigation of the effect
of quarterly earnings announcement timing on stock returns", Journal of
Accounting Research 22, 153-176.
Lardy, N. R. (1998) China's Unfinished Economic Revolution, The Brookings
Institution, Washington, DC.
Lamoureux, C. G. and Lastrapes, W. D. (1990) "Heteroskedasticity in stock
return data: Volume versus GARCH effects", Journal of Finance 45, 221-
229.
Lamoureux, C. G. and Lastrapes, W. D. (1994) "Endogenous trading volume and
momentum in stock-return volatility", Journal of Business and Economic
Statistics 12, 253-260.
Lee, C. S., Shleifer, A., and Thaler, R. H. (1991) "Investor sentiment and the
closed-end fund puzzle", Journal of Finance 46, 75-109.
Bibliography 421
Liesenfeld, R. (1998) "Dynamic bivariate mixture models: Modeling the behavior
of prices and trading volume", Journal of Business and Economic Statistics,
16, 101-109.
Liesenfeld, R. (1999) "Trading volume and the short and long-run components of
volatility", Eberhard-Karls University Working Paper.
Lo, A. W. and MacKinlay, A. C. (1988) "Stock market prices do not follow ran-
dom walk: Evidence from a simple specification test", Review of Financial
Studies 1, 41-66.
Loughran, T. and Ritter, J. R. (1995) "The new issues puzzle", Journal of Finance
50, 23-51.
Loughran, T., Ritter, J. R. and Rydquist, K. (1994) "Initial public offerings:
International insights", Pacific-Basin Finance Journal 2, 165-199.
MacDonald, R. and Kearney, C. (1987) "On the specification of Granger-causality
tests using the cointegration methodology", Economics Letters 25, 149-153.
McCulloch, J. H. (1997) "Measuring tail thickness to estimate the stable index
a: A critique", Journal of Business and Economic Statistics 15, 74-81.
MacKinnon, R. (1991) "Critical values for cointegration tests", in Engle, R. F.
and Granger, C. W. J. Long-Run Economic Relationships 267-276, Oxford
University Press, London, United Kingdom.
Merton, R. C. (1971) "Optimum consumption and portfolio rules in a continuous-
time model", Journal of Economic Theory 3, 373-413.
Merton, R. C. (1973) "An intertemporal asset pricing model", Econometrica 41,
867-887.
Merton, R. C. (1980) "On estimating the expected return on the market: An
exploratory investigation", Journal of Financial Economics 8, 323-361.
Merton, R. C. (1995) "A functional perspective of financial intermediation", Fi-
nancial Management 24, 23-41.
Michaely, R. and Shaw, W. (1994) "The pricing of initial public offerings: Tests
of adverse selection and signaling theories", Review of Financial Studies 7,
279-319.
Milgrom, P. and Stokey, N. (1982) "Information, trade, and common knowledge",
Journal of Economic Theory 26, 17-27.
Mok, H. M. K. and Hui, Y. V. (1998) "Underpricing and aftermarket performance
of IPO in Shanghai, China", Pacific-Basin Finance Journal 6, 453-474.
Myers, S. C. and Majluf, N. (1984) "Corporate financing and investment deci-
sions when firms have information that investors do not have", Journal of
Financial Economics 13, 187-221.
Naughton, B. (1987) "Macroeconomic policy and response in the Chinese econ-
omy: The impact of the reform process", Journal of Comparative Eco-
nomics 11, 342-368.
Nelson, D. B. (1991) "Conditional heteroskedasticity in asset returns: A new
approach", Econometrica 59, 347-370.
Nelson, D. B. and Cao, C. Q. (1992) "Inequality constraints in the univariate
GARCH model", Journal of Business and Economic Statistics 10, 229-235.
422
Bibliography
Newey, W. K. and West, K. D. (1987) "A simple, positive semi-definite, het-
eroskedasticity consistent covariance matrix", Econometrica 55, 703-708.
Perotti, E. C. (1995) "Credible privatization", American Economic Review 85,
847-859.
Poterba, J. and Summers, L. (1988) "Mean reversion in stock prices: Evidence
and implications", Journal of Financial Economics 22, 27-59.
Perron, P. (1988) "Trends and random walks in macroeconomic time series: Fur-
ther evidence from a new approach", Journal of Economic Dynamic and
Control 12, 297-332.
Phillips, P. C. B. (1987) "Time series regression with a unit root", Econometrica
55, 277-301.
Phillips, P. C. B. and Loretan, M. (1994) "Testing the covariance stationarity
of heavy-tailed time series: An overview of the theory with applications to
several financial datasets". Journal of Empirical Finance 1, 211-248.
Phillips, P. C. B. and Perron, P. (1988) "Testing for a unit root in time series
regression", Biometrika 75, 335-346.
Ritter, J. R. (1984) "The 'hot issue' market of 1980", Journal of Business 57,
215-240.
Ritter, J. R. (1991) "The long-run performance of initial public offerings", Journal
of Finance 43, 789-822.
Rock, K. (1986) "Why new issues are underpriced?", Journal of Financial Eco-
nomics 15, 187-212.
Rogalski, R. (1984) "New findings regarding day-of-the-week returns over trading
and non-trading periods", Journal of Finance 39 1603-1614.
Scholes, M. and Williams, J. (1977) "Estimating betas from nonsynchronous
data", Journal of Financial Economics 5, 309-327.
Shanken, J. (1992) "On the estimation of beta-pricing models", Review of Finan-
cial Studies 5, 1-33.
Shiller, R. (1984) "Stock prices and social dynamics", Brookings Papers on Eco-
nomic Activity 2, 457-498.
Shiller, R. (1990) "Speculative prices and popular models", Journal of Economic
Perspective 4, 55-65.
Shleifer, A. and Vishny, R. (1986) "Large shareholders and corporate control",
Journal of Political Economy 94, 461-488.
Shleifer, A. and Vishny R. (1994) "Politicians and firms", Quarterly Journal of
Economics 109, 995-1025.
Smith, C. (1977) "Alternative methods for raising capital: Rights versus under-
written offerings", Journal of Financial Economics 5, 273-307.
Stulz, R. (1990) "Managerial discretion and optimal financing policies", Journal
of Financial Economics 26, 3-27.
Stulz, R. M. and Wasserfallen, W. (1995) "Foreign equity investment restrictions,
capital flight, and shareholder wealth maximization: Theory and evidence",
Su, D. (1998) "The behavior of Chinese stock markets", in Doukas, J. and Choi,
Bibliography 423
J. Emerging Capital Markets: Financial and Investment Issues 253-273,
Greenwood Publishing Group, Connecticut.
Su, D. (1999a) "Ownership restrictions and stock prices: Evidence from Chinese
markets", Financial Review 34, 37-56.
Su, D. (1999b) "Market segmentation and foreign share discount in China", Asia-
Pacific Financial Markets 6, 283-309.
Su, D. (2000a) "Earnings announcements and stock returns in emerging Chinese
markets", Emerging Markets Quarterly 4, 55-61.
Su, D. (2000b) "Asset pricing in a segmented market", Journal of Applied Eco-
nomics 3, 201-226.
Su, D. (2000c) "Leverage, insider ownership, and the underpricing of IPOs in
China", Working Paper.
Su, D. (2000d) "Adverse-selection and signaling: Evidence from Chinese IPO
pricing", Working Paper.
Su, D. and Fleisher, B. M. (1998a) "Risk, return and regulation in Chinese stock
markets", Journal of Economics and Business 50, 239-256.
Su, D. and Fleisher, B. M. (1998b) "What explains the high IPO returns in
China?", Emerging Market Quarterly 2, 1-16.
Su, D. and Fleisher, B. M. (1999a) "An empirical investigation of underpricing
in Chinese IPOs", Pacific-Basin Finance Journal 7, 173-202.
Su, D. and Fleisher, B. M. (1999b) "Why does return volatility differ in Chinese
stock markets?", Pacific-Basin Finance Journal 7, 557-586.
Su, D. and Fleisher, B. M. (1999c) "Efficiency of Chinese stock markets: Some
preliminary evidence", Accounting and Business Review 6, 171-187.
Su, D. and Fleisher, B. M. (1999d) "Explaining IPO underpricing in China", in
Chen, B., Dietrich, J. K., and Feng, Y. Financial Market Reform in China:
Progress, Problems and Prospects 243-260, Westview Press, Colorado.
Tauchen, G. E. and Pitts, M. (1983) "The price variability-volume relationship
on speculative markets", Econometrica 51, 485-505.
Viard, A. D. (1995) "The asset pricing effects of fixed holding costs: An upper
bound", Journal of Financial and Quantitative Analysis 30, 43-59.
Welch, I. (1989) "Seasoned offerings, imitation costs and the underpricing of
initial public offerings", Journal of Finance 44, 421-449.
World Bank (1995) China: The Emerging Capital Market I and II, New York.
Xiang, B. (1998) "Institutional factors influencing Chinas accounting reforms and
standards", Accounting Horizons 12, 105-119.
Xiao, J. Z., Zhang, Y., and Xie, Z. (2000) "The making of independent auditing
standards in China", Accounting Horizons 14, 69-89.
Yan, F. (1997) "Influences on stock returns from issuers and investors: Evidence
from a segmented market", Yale University Working paper.
Yang, J. W. and Yang, J. L. (1998) The Handbook of Chinese Accounting (in
Chinese), Oxford University Press.
Zakoian, J. M. (1994) "Threshold heteroskedastic models", Journal of Economic
Dynamics and Control 18, 931-995.
Index
/3-coefficient, 139, 140, 142, 146
A-share, 8, 49, 76, 86, 104, 125, 128,
135, 151, 193, 200, 207, 209, 210,
221, 252, 322, 325, 332, 354
abnormal return, 291, 292, 294, 323,
325, 328
accounting, 52, 66, 279, 299, 313
information, 307, 312, 318, 332, 333
standard, 307, 309, 313, 351
Accounting Law, 67, 314
acquisition, 65, 269, 304, 360, 390
adverse-selection models, 227, 263
agency
costs, 277, 282, 284, 286, 295
trading, 394
Alford, A., 321
Allen, F., 216, 235, 236
allowance, 55
American Depository Receipts
(ADRs), 9, 159, 340, 348, 350, 351
American Stock Exchange (AMEX),
111, 348
Amihud, Y., I l l
analysis of variance, 82
Andersen, T., 84, 170, 171, 172, 203
Ang, J. S., 333
annual
meeting, 52
report, 316, 331, 350, 389, 408
arbitrage, 355
Arbitrage Pricing Theory (APT), 75
assets, 55, 68
asymptotic, 88
auction, 62, 220, 249, 250
auditing, 52, 58, 299, 317
Auditing Law, 314
autocorrelation, 78, 81, 87, 98, 111,
172, 175, 180, 186, 325
autonomy, 45
autoregressive process, 114, 170, 210
B-share, 8, 27, 50, 58, 65, 76, 104,
125, 128, 135, 153, 154, 164, 166,
193, 200, 207, 209, 210, 218, 251,
252, 263, 323, 328, 333, 336, 354,
358, 406
backbone industries, 24
bad debt, 30, 279, 314
Bailey, W., 123, 140, 169
Baillie, R. T., 95
Ball, R., 321
bankruptcy, 73, 282, 283, 285, 286
Basu, S., 215
Beatty, R. P., 227, 229
Bellman equation, 150
Bera-Jarque test, 100
bid-ask spread, 170
425
426 Index
black market, 35
board of directors, 269, 282, 283,
303, 316
Bollerslev, T., 94, 95
bribery, 216, 246-248
brokerage, 52, 58
fee, 62
Brown, P., 321
Brown, S., 323
Brownian motion, 148
bubble, 56
business tax, 70, 403
Calton, D. W., 141
Camerer, C, 256
Campbell, J., 160
Cao, C. Q., 94
capital
account, 354, 371, 376, 377, 379
gains tax, 70, 402
management, 364
market, 2, 31, 42, 55, 69, 340, 352,
354, 359
mobility, 371, 376
requirement, 357, 394
structure, 24
Capital Asset Pricing Model
(CAPM), 75, 123, 151, 154
cash, 62
dividend, 156
flow, 28, 133, 236, 278, 282, 284,
308, 310
flow statement, 69
Cauchy distribution, 96
causality, 76
central government, 56
Chan, K., 321
Chemmanur, T. J., 235, 236
chief financial officer, 51
China Insurance Regulatory
Commission (CIRC), 26
China International Trust and
Investment Company (CITIC), 4,
346. 362
China Securities Regulatory
Commission (CSRC), 26, 36, 41,
43, 52, 55, 57, 59, 64, 106, 110, 132,
133, 217, 316, 319, 359, 360, 365,
367, 381, 385, 407
Chinese Security Exchange
Commission (CSEC), 36
Chopra, N., 256
Chowdhry, B., 225, 227, 230, 234
Clark, P. K., 171
clearing, 59, 62, 66, 71
closed-end fund, 366
cointegration tests, 76, 90, 121
collective enterprises, 39, 66, 370
commercialization, 44, 267, 278
commission, 63, 71
Company Law, 49, 53, 72, 268, 314,
387, 407
conclusions, 62, 63
conditional mean, 76, 93, 103, 112,
115
conditional variance, 76, 94, 111, 112
conglomerates, 24, 45, 54
constrained maximum likelihood
(CML), 97
consumption tax, 403, 404
control rights, 274, 278, 289
convertibility, 356, 378
cooperative shareholding system
(CSS), 274, 277, 279
corner solution, 144
corporate
bond, 9, 37, 38, 384
control, 304
disclosure, 309, 312
financing, 274
governance, 40, 267, 268, 270, 283,
285, 299, 301, 304
corporatization, 23, 44, 279, 282, 314
corruption, 45, 357
cost of capital, 35, 75, 318
Cournot competition, 133
covariance, 257, 325
stationary, 94
Index 427
Cross, F., 82
cross-
border, 53, 345, 355
correlation, 122
listing, 340
section, 124, 159, 161, 163-165,
167, 200, 202, 204, 205, 227,
229, 255
validation, 186
cumulative, 256
abnormal return (CAR), 324,
325, 328
current account, 373, 375
custom duty, 404
day-of-the-week effect, 82, 83
dealers, 63, 71
debt, 230
delivery, 62, 66, 220
denationalization, 301
depository receipt (DR), 348
detrend, 172, 180, 185, 186
diagnostic tests, 115
Dickey-Fuller test, 91
diffusion process, 148
disclosure, 51, 60, 72, 309, 360,
377, 389
distribution function, 297
diversification, 302, 303
dividend, 53, 62, 125, 282, 337,
373, 408
growth model, 133
growth rate, 135
tax, 70, 402
Domodaran, A., I l l
double-entry method, 69
Dow Jones China Index, 410
Dow Jones Industrial Average
(DJIA), 82
dual listing, 125, 339, 351
e-commerce, 352
earnings, 203, 229, 309, 311
announcements, 320-322, 324,
328, 331
forecasts, 332
surprises, 332
earnings-per-share (EPS), 322, 323,
325, 328
Eastern, P., 321
Eckbo, B., 287
efficiency, 20, 23, 30, 34, 38, 45,
276, 377
elasticity, 240
embezzlement, 59
employee shares, 10, 34, 217, 240,
407, 413
endogeneity, 240, 249
Engle, R. F., 90, 115
entrustment, 62
envelope condition, 150
Epps, M. L., 171
Epps, T. W., 171
equity, 255
error
correction model, 90
generation process, 93, 95
in variable (EIV), 159-161, 164,
167, 238, 243
Errunza, V., 123
Eun, C, 123
European Union (EU), 307
ex ante uncertainty, 228-230, 234, 263
excess
return, 161, 165, 323
volatility, 106, 169, 172, 193
exchange rate, 356, 372, 375
regime, 372, 378
exogenous variable, 227, 243, 372
expected
return, 124, 151, 154, 243, 265,
323, 333
underpricing, 228, 234
explanatory power, 243-245
external financing, 294
428
Index
factor productivity, 275
Fama, E. F., 291
Faulhaber, G. R., 216, 235, 236
Federal Reserve, 2
financial
disclosure, 309, 350
distress, 345
institution, 354, 358, 375, 377, 410
intermediary, 361, 369
management, 307
packaging, 359
report, 67, 69, 314, 318
risks, 27, 360
statement, 307, 308, 310, 316,
331, 350
structure 300, 308
firm
commitment, 221, 260
size, 124, 165, 203-205
value, 247, 277
Fisher, L., 291
Fleisher, B. M., 75, 93, 148, 169, 172,
200, 345
float, 58, 350, 355, 412
foreign
banks, 34, 357
currency, 51, 376, 397
direct investment (FDI), 373, 375,
379
exchange, 312, 338, 354, 357, 372,
373, 376
investment, 51, 71, 337, 340, 354,
356, 401
share, 133, 251, 255
Fourier transformation, 186
French, D., 82
fulfillment method, 286
Gallant, A. R., 171, 180
Gaussian
kernel, 185
normal distribution, 95, 96
go public, 56, 230, 411
Generalized
Autoregressive Conditional
Heteroskedasticity
(GARCH), 93, 97, 98, 102,
172, 200
Error Distribution (GED), 95, 115
Least Squares (GLS), 160, 163,
205, 231
Method of Moments (GMM), 170,
173
generally accepted accounting
principles (GAAP), 307, 318,
350, 351
Gennotte, G., 321
global depository receipts (GDRs),
350, 351, 356
government, 270, 289
interference, 271, 321
ownership, 218, 294, 297
shares, 217, 240
subsidies, 279
Granger, C. W., 90
Greenshoe, 52
Grinblatt, M., 216, 235, 236, 239,
292, 296
Gross Domestic Product (GDP), 1, 5,
31, 43, 279, 282, 373
Grossman, S., 278
Group of Thirty (G30), 66
Groves, T., 279
Growth Enterprise Market (GEM), 26
H-share, 8, 27, 53, 58, 65, 218, 340,
342, 355, 389
Hang Seng Stock Exchange, 54
Hansen, L. P., 173, 193
Hansen's J-statistic, 193
harmonic mean, 149
Harris, L., 82, 171
Hart, O., 278
hedge fund, 372, 373
Heinkel, R., 287
herd behavior, 372
Index
429
heteroskedasticity, 87, 231
Hietala, P., 123
homoskedasticity, 84
Hong Kong Stock Exchange (HKSE),
8, 18, 54, 157, 340, 352, 355, 360
Hong, Y., 279
Hotelling's T
2
-statistic, 83
Hsieh, D. A., 95
Hwang, C. Y., 216, 235, 236, 239
hybrid offering, 351
idiosyncratic variance, 164
imitation expenses, 236
income
statement, 69, 308
taxes, 71, 401
incorporation, 39
individual
investors, 27, 256, 332, 334,
361, 366
shares, 218
inefficiency, 45
inflation, 45
information asymmetry, 205, 209,
216, 234, 263
informed
investors, 227, 228, 331
trading, 200, 202
infrastructure, 24, 217, 318, 319,
383, 392
initial public offerings (IPOs), 38, 56,
57, 125, 215, 218, 219, 232, 237,
242, 257, 279, 367
return, 215, 224, 225, 227, 230,
231, 238, 247, 250, 255
market, 228
price, 220, 227, 231, 237, 408
proceeds, 221, 229, 234, 236, 238
underpricing, 215, 216, 224, 225,
227, 229, 230, 234, 235, 240,
242, 244, 245, 248, 249, 251,
254, 263, 297
insider, 216, 229, 286, 321
information, 60, 395
ownership, 263
trading, 58, 322, 331, 332, 389,
392, 395
institutional
investors, 19, 27, 221, 229, 294,
301, 332, 334, 350, 363, 372,
373, 376
ownership, 297
instrumental variable (IV), 227, 243
insurance
capital, 363, 365
company, 362, 364, 365, 369
intangible, 69
integration, 121
inter-bank, 29, 366
interior solution, 142, 144
internal financing, 294
international accounting standard
(IAS), 58, 313, 318, 319, 338, 351
International Monetary Fund (IMF),
371, 372, 379
Intertemporal CAPM (ICAPM), 124,
148, 159, 164, 165, 167
intrinsic value, 113, 156, 218, 238, 240
investment
barrier, 123
corporation, 361
expenditure, 287
fund, 34, 365, 367, 412
institutions, 364
opportunity, 124, 265, 287, 294,
295, 376
Jagadeesh, N., 219
Jagtiani, J., 123
Janakiramanan, S., 123
Jefferson, G., 276
Jensen, M. C, 275, 282, 291
Johnson, D. L., 123
joint
distribution, 170, 210
hypotheses, 229, 234
430
Index
stock company, 49, 69, 279, 282,
285, 338
test (JT), 116, 118, 203
venture, 133, 231, 300, 357
Jorion, P., 95
Kearney, C, 90
kernel estimation, 185, 186
Kernen, A., 284
key industry, 56
Koutmos, G., I l l , 113
Kross, W., 321
kurtosis, 78, 79, 98, 175, 186
L-share, 9
Lakonishok, J., 256
Lamoureux, C. G., 171
Laplace distribution, 115
Lastrapes, W. D., 171
latent information, 170, 172
Lee, C. S., 256
legal person shares, 10, 28, 110, 217,
282, 340, 408
legislation, 45
lemon's problem, 263
leptokurtotic, 81, 95, 115, 121
leverage, 230, 234, 263, 287
effect, 112
liabilities, 68
Li, D. D., 215
likelihood ratio (LR) test, 104, 107
liquidity, 40, 121, 287, 308, 351,
357, 373
trading, 204, 212
listing, 49, 60, 352, 359, 383, 389
fees, 71
Ljung-Box portmanteau statistic, 78,
92, 98
Lo, A. W., 87, 160
logit model, 242, 252
London Stock Exchange (LSE), 9, 82,
351, 352
long-term
debt, 287, 288
investment, 287, 288, 298
investor, 333, 363
performance, 256, 260, 297
productivity, 290
Loretan, M., 186
Losq, E., 123
lottery, 220, 221, 230, 231, 246, 249,
250, 254
Loughran, T., 215, 256
Macau, 51
MacDonald, R., 90
MacKinlay, A. C., 87, 160
MacKinnon, R., 91
macroeconomic
policy, 378
variable, 331
Majluf, N., 286
management, 56, 267, 274, 278, 282,
290, 319, 360
ownership, 294
responsibility contract system
(MRCS), 278, 279, 312
manager, 51, 58, 60, 231, 256, 268,
277, 284, 286, 301, 331
managerial
compensation, 284
discretion, 278, 282, 284
fee, 21
incentive, 285
ownership, 297
rights, 45
manipulation, 41, 61, 332, 396
margin trading, 59, 392
market
betas, 124
capitalization, 9, 43, 76, 119, 165,
218, 240, 251, 282, 355,
366, 384
efficiency hypothesis, 76, 86, 90
feedback hypothesis, 238, 242, 244,
245, 255, 263
Index
431
liberalization, 106, 107, 110, 121
microstructure, 58, 170, 336
segmentation, 122, 123, 167, 354,
357
market-to-book ratio, 287, 288, 294,
295
Masulis, R., 287, 292, 296
Ma, Y., 333
McCulloch, J. H., 96, 100
McMillan, J., 279
Meckling, W. H., 275
Memorandum of Understandings
(MoUs), 42, 53
Mendelson, H., I l l
merger, 69, 269, 282, 301, 304, 339,
360, 390
Merton, R. C, 20, 81, 150
Ministry of Finance (MOF), 4, 28, 55,
67, 69, 314, 409
Ministry of Foreign Trade and
Economic Cooperation, 34
Ministry of Labor and Social Welfare,
370
minority shareholder, 319
mixing variable, 171, 210
mixture of distribution hypothesis
(MDH), 169, 170, 201
modern enterprise system, 22, 268,
304
modified MDH (MMDH), 170, 171,
180, 193, 200, 201, 209, 210, 212
monetary
crisis, 372
markets, 29
monitor, 276, 285, 320
Monte Carlo, 100
Morgan Stanley Composite Index
(MSCI), 77, 157
moving average, 83, 85, 180
multinomial logit model (MNL), 296,
297
municipal government, 56
mutual fund, 21, 27, 221, 332, 372,
407
Myers, S. C, 286
N-share, 9, 218, 351
National Association of Securities
Dealers Automatic Quotation
System (NASDAQ), 111, 157, 348,
350, 352
National Electronic Trading System
(NETS), 217, 282
National People's Congress, 22, 24,
39, 42, 48, 67, 314, 367, 381, 407,
412
Naughton B., 5, 279
negative size bias test (NSBT), 115,
118
Nelson, D. B., 94, 95
New York Stock Exchange (NYSE),
9, 28, 111, 157, 218, 340, 348, 350,
351, 352, 408
Newey, W. K., 175, 193
Ng, V. K., 115
noise traders, 29, 200, 204
non-
bank financial institution, 362
parametric, 185
synchronization of trading, 102,
160, 325
stationary, 180
off-shore
derivatives, 34, 358
listing, 358
market, 356
open-end fund, 367
operating expenses, 279
ordinary least squares (OLS), 224
Organization for Economic
Cooperation and Development
(OECD), 20
orthogonality condition, 173
over-allotment option, 52, 53
over-identifying restrictions, 174, 193
432
Index
over-reaction, 256, 260, 263, 331
overseas, 53, 410
listing, 339, 347, 352, 359, 360, 397
over-subscription, 221, 249, 255
over-the-counter (OTC) markets, 5,
66, 159, 348
owner's equity, 68
ownership
concentration, 296
reform, 46
restrictions, 123, 125, 146, 355
rights, 289
structure, 10, 44, 276
parallel listing, 356
partial adjustment model, 111, 113,
118
partial equilibrium, 135
pecking order, 286, 289, 294
pension
fund, 299, 332, 363, 369
plan, 370
People's Bank of China (PBOC), 3,
4, 36, 59, 217, 336, 338, 352, 357,
358, 359, 406
People's Insurance Company of China
(PICC), 362
Perloff, J. M., 141
Perotti, E., 232
persistence, 81, 121, 200
Phillips-Perron test, 91, 180
Pitts, M., 171
Poisson distribution, 171
political
control, 267
costs, 277, 282, 284, 286, 295
politician, 247, 275, 277
pooling equilibrium, 237, 238, 247,
263
positive size bias test (PSBT), 116,
118
price
depression effect, 247
discount, 131, 142, 209, 354
discrimination model, 124, 148,
155, 159, 165
earning ratio, 133, 255, 335
premium, 129, 142, 159, 356
priority, 63, 405
subsidies, 270
principal-agent problem, 275, 277
private
capital, 375, 376
enterprises, 67, 133
investors, 294, 319
privatization, 23, 216, 232, 267, 279,
282, 299, 300
profit-per-share, 202-205, 230, 294
property rights, 45, 68, 268, 304
proprietary trading, 394
pro-rata, 221, 249
provisions, 51
public
information, 325
offerings, 50, 350, 387
ownership, 22
quintile, 328
quorum, 53
quota, 33, 52, 55, 132, 218, 265, 408
random walk hypothesis, 86, 88
Rawski, T., 276
Red Chips, 9, 53, 345, 355
reform, 30, 44
registration, 58, 62, 347, 350, 394
reorganization, 23, 69, 283, 286, 391
reserve, 375
requirement, 29, 357
residual rights, 278, 279, 285
restructure, 55, 267, 302, 352
rights issues, 156, 286, 387, 412
rigidity, 45
risk, 75, 133, 230, 231, 236, 295, 328,
338, 361
Index
433
aversion, 137-139, 147, 149, 151,
154, 155, 161, 163
free asset, 135, 138, 139
management, 311, 364
premium, 75
Ritter, J. R., 215, 225, 227, 229, 230,
234, 256, 257
Rock, K., 227
Rogalski, R., 82
Roll, R., 291
Rossi, P. E., 171, 180
royalty, 70, 402
Rydquist, K., 215, 256
S&P, 82, 500
savings, 35
Scholes, M., 160, 161
Schroeder, D., 321
Schwartz, E., 287
seasonal component, 85
season equity offerings (SEOs), 219,
225, 235, 237, 241, 244, 247, 248,
255, 263, 279, 412
secondary markets, 5, 245, 250
securities, 2, 40
crimes, 395
taxes, 70
Securities Exchange Commission
(SEC), 347, 350, 351, 408
Securities Law, 49, 53, 61, 65,
381, 412
Securities Trading Automatic
Quotation System (STAQS), 7,
217, 282, 405
semi-strong form efficiency, 325, 328
sentiment, 56, 255, 332
Seow, G., 321
separating equilibrium, 216, 235, 236,
240, 248, 250, 263
service fees, 71
settlement, 66, 71, 357, 394
SGARCH, 97-102, 106, 107, 111
Shanghai Securities Exchange
(SHSE), 1, 6, 61, 63, 72, 105, 110 ,
125, 217, 322, 333, 335, 358, 406
Shanken, J., 161
shareholding, 49, 53, 64, 390
Sharpe ratio, 77, 79
shell company, 54, 351
Shenzhen Securities Exchange
(SZSE), 1, 6, 21, 62, 72, 105, 110,
125, 217, 234, 335, 358, 406
Sherman, A., 225, 227, 230, 234
Shiller, R., 256, 275
Shleifer, A., 256, 278, 283
shortage, 216
short-selling, 66, 372
signaling models, 216, 232, 235, 241,
246, 263
sign bias test (SBT), 115, 118
simultaneous equations, 231
skewness, 78, 79, 96, 98, 175, 186
Smith, C, 287
social
responsibilities, 279
security, 23, 278, 369
soft budget constraint, 30, 276, 278,
283, 285, 287
solvency, 308, 358
Special Economic Zones (SEZs), 3,
337, 357
speculation, 41, 361, 372, 392
sphericity test, 84
spill-over effect, 76, 113, 118-120
spin-off, 55
spontaneous privatization, 30, 304
stable distribution, 95, 96, 213
stake, 45
stamp duty, 62, 337, 404
State Administration of Foreign
Exchange (SAFE), 59, 376
State Asset Management Bureau
(SAMB), 28, 282, 284
State Council, 21, 33, 43, 51, 58, 269,
381, 407
state
owned banks, 31, 278, 302, 361
434 Index
owned enterprises (SOEs), 5, 22,
26, 30, 44, 56, 67, 215, 218,
267-269, 271, 275, 276, 278,
279, 282, 284, 286, 290,
299-302, 304, 314, 354, 370,
412
owned holding corporation, 301
ownership, 303
shares, 28, 408
stationary, 172, 186
statistics, 77
stochastic, 170, 172, 210
stock, 3, 7
dealer, 58
dividend, 286, 292, 294, 296, 297
split, 156
Stock Exchange of Singapore (SES),
351, 352
structural equation, 227
Stulz, R., 123, 132, 133, 135, 282
Su, D., 75, 93, 148, 169, 172, 200,
230, 345, 355
supervision, 54, 269, 275, 373, 410
suspend, 60, 72
swap, 55
System regression, 244, 245
Taiwan, 51
takeover, 64, 301, 304, 335, 391
target company, 64
Tauchen, G. E., 171, 180
taxes, 70, 297, 401
temporal dependence, 95
tender offer, 390, 391
terminate, 60, 72
term structure spread, 97, 103
TGARCH, 97-100, 102
Thaler, R. H., 256
Threshold GARCH, 115
time
priority, 63, 405
series, 77, 93, 102, 110, 135, 156,
180, 209, 328
varying, 81, 115, 121
Titman, S., 292, 296
township and village enterprises, 267
trading
holidays, 62
volume, 170, 171, 180, 185, 193,
200, 204, 210
transaction
cost, 63, 337
data, 321
transfer fee, 62, 72
transparency, 40, 307, 309, 333, 360,
373, 384, 389
Treasury bond, 5, 157, 291
triangular debt, 279, 299
Truemann, B., 321
turnover ratio, 9, 97
two stage least squares (2SLS), 227,
240, 243, 250
uncertainty, 265, 311
unconditional
moments, 174
variance, 106, 107
underwriter, 52, 230, 247, 297, 389
underwritten offers, 286, 289, 296, 297
uninsured rights, 286, 288, 292, 294,
296, 297
unit root test, 91, 180
value
added, 70, 403
function, 150, 153
weighted, 77
variance, 237, 240
ratio, 76, 87
venture capital, 16, 352, 367
Vishny, R., 275, 278, 283
volatility, 29, 75, 79, 105, 170, 172,
193, 208, 209, 356, 375
asymmetry, 76, 111
clustering, 94
spikes, 109
transmission, 118
voucher privatization, 300
Wald confidence limits, 105
Warner, J., 323
warrants, 51, 53
Wasserfallen, W., 123, 132, 133, 135
weak form efficiency, 93
wealth relative, 256, 257
weekend effect, 82
weighting matrix, 173, 175, 193
Weinstein, M., 219
Welch, I., 216, 219, 235, 236
West, K. D., 175, 193
Wiener process, 148
Williams, J., 160, 161
winner's curse, 228, 229
withholding, 70, 402, 408
World Bank, 28, 249, 362, 369
World Trade Organization (WTO),
16, 24, 31, 357, 362, 376, 379
Xiang, B., 317
Xiao, J. Z.,, 317
Xie, Z., 317
Yule-Walker equation, 193
Zakoian, J. M., 115
Zhang, Y., 317
Zmijewski, M., 321
Dongwei Su was born in December 1
(
'7() in Xiamen,
China. After obtaining his 15.A. in economics from
Xiamen University in July l')')2. Dongwei Su attended
Ohio State University (USA) and obtained his Ph.D.
in economics from Ohio State in June 1
n<
)7. Since
then. Dongwei Su had been an assistant professor of
economics with University of Akron (USA) and
assistant professor of finance with Morgan State
University (USA).
Professor Su lias published extensively in die area of
Chinese financial markets. I le was named a research
fellow by I he Sandra Ann Morsilli Pacific-Basin
Capital Markets Research Center.
The exponential growth of China' s stork markets in the past decade
has at t ract ed global at t ent i on from academi cs and practitioners.
The pr act i t i oner ' s i nt er es t in Chi nes e mar ket s s t ems from
corporations; investors and financial institutions foresee substantial
benefits from investing in China in the long run. However, the
academi c literature on t he development of securities market s and
reform of st at e ent erpri ses in Chi na is still in its i nfancy and
fragmented. Thi s handbook aims to bridge that gap by presenting
a wide spectrum of research in the forefront of financial applications.
It i nt egrat es theory and pract i ce with st at e-of-t he-art statistical
techniques and provides numerous insights into the main challenges
c onf r ont i ng Chi ne s e ma r k e t s in t he new mi l l e n n i u m.

Dongwei Su Chinese Stock Markets A Research Handbook 2003

Uploaded by

Copyright:

Available Formats

Dongwei Su Chinese Stock Markets A Research Handbook 2003

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Dongwei Su Chinese Stock Markets A Research Handbook 2003

Uploaded by

Copyright:

Available Formats

H I N E S E

You might also like