Asia Pacific Management Review 24 (2019) 61e71

H O S T E D BY Contents lists available at ScienceDirect

Asia Pacific Management Review

journal homepage: www.elsevier.com/locate/apmrv

Financial performance and corporate social responsibility: Empirical evidence

from Taiwan
Lin Lin a, Pi-Hsia Hung a, De-Wai Chou b, Christine W. Lai b, *
a
Department of Banking and Finance, National Chi Nan University, Nantou County, 545, Taiwan
b
Graduate Institute of Management, National Taiwan Normal University, Taipei City, 106, Taiwan

a r t i c l e i n f o a b s t r a c t

Article history: This study, using data provided by CSRHub, examines the effect of CSR dimensions on the financial
Received 5 May 2017 performance of firms in Taiwan. Specifically, we examine whether CSR in employment exhibits a
Received in revised form signaling effect and results in financial benefits to firms in Taiwan. We find that allocating resources to
19 June 2018
diversity, labor rights, treatment of unions, compensation, benefits, training, health, and worker safety
Accepted 1 July 2018
Available online 29 September 2018
can be beneficial to a firm's value creation. Using sustainability ratings by CSRHub, we are able to cover a
broader range of companies in size, geography, and industry type than those previously studied.
© 2018 College of Management, National Cheng Kung University. Production and hosting by Elsevier
JEL classification:
G14
Taiwan LLC. All rights reserved.
D21
L21

Keywords:
Corporate social responsibility
Financial performance
CSRHub
Asia
Taiwan

1. Introduction Stakeholder theory (Freeman, 1984) suggests that a socially

responsible firm devotes attention simultaneously to the interests
The literature suggests that there are significant differences in of all appropriate stakeholders. The link of CSR to corporate
corporate social responsibility (CSR, hereafter) initiatives across financial performance can further be established once investors
Western and Eastern countries (Baughn, Bodie, and McIntosh, and key stakeholders reward firms that are sensitive and accouns to
2007; Hah & Freeman, 2014; Welford, 2005; Wokutch, 1990; stakeholders' concerns. The purpose of this study is to investigate
Yang & Rivers, 2009). One way to empirically explore CSR practices how investors reward firms with good image/reputation across the
in Western/Eastern countries is to examine how the subsidiaries of following four CSR dimensions: Governance, Community, Em-
multinational enterprises align the CSR approaches of their parent ployees, and Environment. Data is drawn from the CSRHub data-
firms with local practices in emerging markets. Another approach base for companies listed in Taiwan.
to exploring these differences is to investigate how CSR works for Taiwan companies are a good laboratory for testing the rela-
Asian firms under globalization, in which industries and firms are tionship between CSR and corporate financial performance under
moving toward practices common around the world. The current globalization in Asian countries. On one hand, Taiwan companies
study takes the second approach by examining the link between have been exposed to the values associated with Asian business,
CSR and corporate financial performance for Taiwan firms. such as the cultivation of special relationships and a substantial
distinction between ‘insiders’ versus ‘outsiders’ (Ang & Leong,
2000), and weaker policies associating CSR with employment
* Corresponding author. 162, Section 1, Heping E. Rd, Taipei City, 106, Taiwan.
(Welford, 2005). On the other, Taiwan companies are exposed to
E-mail addresses: [email protected] (L. Lin), [email protected] globalization, which shapes organizational and firm competitive-
(P.-H. Hung), [email protected] (D.-W. Chou), [email protected] (C.W. Lai). ness. In particular, exports in Taiwan account for around 70 percent
Peer review under responsibility of College of Management, National Cheng of total GDP, predominantly by industrial goods (98% in 2017,
Kung University.

https://doi.org/10.1016/j.apmrv.2018.07.001
1029-3132/© 2018 College of Management, National Cheng Kung University. Production and hosting by Elsevier Taiwan LLC. All rights reserved.
62 L. Lin et al. / Asia Pacific Management Review 24 (2019) 61e71

including electronics and information & communication products). for employment issues are rewarded by market investors and key
Taiwan's main exports partners include the USA, Europe, Japan, stakeholders in Asian companies, at least in a more developed
ASEAN countries, and China & Hong Kong.1 According to the 2017 emerging market, such as Taiwan.
Investment Climate Statements issued by the US Department of The remainder of this study is as follows. Section 2 reviews the
State, Taiwan is one of the world's top 25 economies in gross do- related literature and provides hypothesis development. Section 3
mestic product (GDP), the United States' 10th largest trading describes the data and methodology used in this study. Section 4
partner, a key link in global supply chains, and a major center for addresses the empirical results, while Section 5 presents the
advanced research and development (R&D). According to the most conclusions.
recent U.S. Department of Commerce data, the total stock of U.S.
Foreign Direct Investment (FDI) in Taiwan reached USD 15 billion, 2. Literatures review and hypothesis development
while U.S. private commercial services exports to Taiwan totaled
over USD 12 billion in2015.2 Therefore, globalization has brought 2.1. CSR and financial performance
sophisticated suppliers/buyers/investors to Taiwan, who act as a
new audience evaluating the reputations of local firms from the Studies show that CSR adds value to a firm's market perfor-
perspective of global norms and expectations. In addition, Taiwan mance. One way to look at the stock performance of CSR firms is to
firms face intense competition for critical/skilled employees in examine the performance of socially responsible mutual funds.
global supply chains. Socially responsible mutual funds are those created by securities
Research shows that for Asian companies, the effects of CSR on investors involving companies they believe are committed to so-
financial performance are mixed.3 One reason for the inconclusive cially responsible activities that will earn competitive returns. Hill,
results may be inadequate measurement issues and insufficient Ainscough, Shank, and Manullang (2007) study socially responsible
data (Lee, 2008). To overcome the measurement issues and insuf- mutual funds in Asia, Europe, and U.S., finding a significantly pos-
ficient data, this study, the first of its kind, uses data provided by itive return in all three countries using a 10-year time horizon.
CSRHub over the period from 01/2009 to 06/2014 to examine the One underlying reason that CSR adds value to a firm's market
effect of CSR dimensions on the financial performance of corpora- performance is the screening process of CSR. Strong CSR firms
tions in Taiwan. There are hardly any comprehensive and easily might gain higher screening scores and become more secure when
accessible CSR databases that incorporate a variety of sustainability undergoing financial crises or environmental difficulties. Several
ratings. Previous studies have often relied on the Kinder, Lyden- studies show that CSR screening intensity has a significant effect on
berg, and Domini's KLD Research & Analytics database as a source financial performance. Galema, Plantinga, and Scholtens (2008)
for sustainability ratings. Currently, CSRHub provides a web-based find that the score of employee relationships generates a signifi-
tool combining over 93 million detailed data items from more than cant positive effect on excess returns. Kempf and Osthoff (2007) use
480 data sources on sustainability into a consistent set of ratings. the Socially Responsible Investing (SRI) ratings of KLD Research &
Using sustainability ratings by CSRHub, we are able to cover a Analytics to compare stock portfolios with high and low SRI ratings,
broader range of companies in size, geography, and industry type finding that the higher ratings portfolio performs better than the
than previously studied. lower ratings portfolio.
We find that allocating resources to diversity, labor rights, However, some scholars argue that the disadvantages of socially
treatment of unions, compensation, benefits, training, health, and responsible activities will erode a firm's financial performance
worker safety can be beneficial to a firm's value creation in a more (Barnett & Salomon, 2006). According to Freeman’s (1984) stake-
developed emerging market, such as Taiwan. The contribution of holder theory and Cornell and Shapiro (1987), CSR firms should not
the current study is to provide more recent evidence to update only serve their owners by realizing value-maximization but also
previous findings in Asian emerging markets. Welford (2005) stakeholders' goals in order to reach their potential value. However,
compared large companies' written policies of CSR in fifteen the interests of various stakeholders often conflict with each other.
countries across Asia, North America and Europe, and found that Stakeholder theory indicates that managers need to make tradeoffs
Asian companies were less likely to have a policy about CSR in to take into account all of the interests of the stakeholders in a firm.
employment. Crane, Matten, and Spence (2013) document that Jensen (2002), however, argues that stakeholder theory increases
Asian companies usually have a legacy of CSR, including employ- the agency costs and weakens the internal control systems of firms,
ment, not so much as a result of voluntary corporate policies, but since its performance measures are only vaguely defined. A new
more as a response to the regulatory and institutional environment way of measuring value, such as long-run maximization of the
of business. Using a more comprehensive data and diverse CSR value of the firm, could resolve this conflict. Renneboog, Ter Horst,
measurements, we find that over years the lack of CSR in employ- and Zhang (2008) state that in socially responsible investments,
ment for Asian companies may have been improved to some extent, portfolio managers pursue both financial goals and social objec-
at least in economic entities, such as Taiwan, which are exposed to tives, and this multi-focused nature may weaken fund managers'
globalization. We do find that corporate sustainability ratings in incentives to pursue high risk-adjusted returns and increase po-
employment, including labor rights, treatment of unions, tential agency costs. Barnea and Rubin (2010) find that insiders
compensation, benefits, training, health, and worker safety, will (managers and large blockholders) who are affiliated with the firm
affect Book-to-Market (BM) ratio and Tobin's q for firms listed in may want to over-invest in CSR for their private benefit since it
Taiwan. Our results indicate that CSR in employment can become a improves their reputation for being good global citizens.
capability differentiator for Asian companies. Firms which account
2.2. Signaling effect of CSR and Asian emerging markets

CSR may create a good signaling effect and earn certification for
1
Information available at https://tradingeconomics.com/taiwan/exports. its reliability, becoming a form of certification for a company that
2
Investment Climate Statements issued by the US Department of State and the
may provide superior reliability and reputation. A good reputation
FDI are available at https://www.state.gov/e/eb/rls/othr/ics/2017/eap/269854.htm.
3
CSR in Asia has been positively (Cheung, Tan, Ahn, & Zhang, 2010; Oh, Chang, &
can attract capital, quality employees, and good investors. Turban
Martynow, 2011), negatively (Li & Zhang, 2010), and non-significantly (Cao, 2011) and Greening (1997) find that an outstanding reputation makes
related to performance. the company attractive to highly-qualified employees, giving it a
 L. Lin et al. / Asia Pacific Management Review 24 (2019) 61e71 63

competitive advantage. In addition, an increase in perceived social welfare, health care, training, and retirement. Further, there are
responsibility may improve a firm's reputation and permit it to costs associated with labor rights, treatment of unions, and worker
exchange costly explicit claims for less costly implicit charges safety. In addition, firms incur long-term costs, including explicit
(McGuire, Sundgren, and Schneeweis, 1998; Robinson, Kleffner, and monetary costs and implicit management costs, when adopting
Bertels, 2008). In addition, companies involved in CSR activities CSR practices and establish sustainability ratings in employment. If
could experience reduced information asymmetry between man- an Asian firm is willing to consistently allocate reasonable re-
agers and investors (Cui, Jo, & Na, 2012). For example, disclosure of sources to maintain a sustainable relationship with its employees,
CSR gives investors more information, mitigating agency problems this endeavor can be important since it not only helps to strengthen
through corporate governance, which will be reflected in market the firm's access to critical employee resources, but is costly for
prices. Bauer, Guenster, and Otten (2004) show a significant posi- low-capability competitors to imitate. In addition, CSR practices in
tive relation between corporate governance and firm value. Lopatta, employment can contribute to a firm's sustainable long-term
Buchholz, and Kaspereit (2016) conclude that companies with growth since demonstrated commitment to employee welfare
higher sustainability rankings (defined as CSR dimensions) are can help a company to attract and retain good employees as well as
more proactive in disclosing information. Good relationships be- encourage them to invest in firm-specific human capital. Human
tween CSR firms and direct stakeholders (employees, customers, capital is quite important when Asian emerging markets face
retailers, producers, suppliers) can contribute to a decrease in intense competitions for critical/skilled employees in the global
agency costs. CSR involves environmental, social, and govern- supply chains. Therefore, we posit that CSR practices in employ-
mental activities which allow firms to echo the anticipation of in- ment may be a signal that reveals additional information to em-
direct stakeholders (social communities, charities, legislative ployees and other stakeholders, especially in Asian emerging
organization, and government) and win their faith and trust. economies. We form the following hypothesis:
For emerging markets, Su, Peng, Tan, and Cheung (2016) Hypothesis 1: CSR sustainability ratings in employment are
investigate what signals firms in emerging markets send to stake- positively related to firm financial performance in emerging
holders when they adopt CSR practices. They find a positive rela- economies, such as Taiwan.
tionship between CSR practices and financial performance. The For other CSR dimensions, Webb (2006) states that the CSR
financial benefits of CSR practices are also more salient in low in- debate in the Asian countries focuses on the issues of corporate
formation diffusion markets than in the high information diffusion governance and transparency. One way to increase corporate
markets. transparency is through the adoption of global norms such as In-
A further question is which CSR dimensions, Employment, ternational Financial Reporting Standards (IFRS). In Taiwan, all
Governance, Environment or Community, are more likely to firms, including listed and unlisted firms, are required to prepare
generate signaling effects in emerging markets and result in financial statements in accordance with Taiwan-IFRS starting from
financial benefits. The literature suggests that there are significant January 1, 2015. In addition, appointment of independent directors
differences in CSR practices across different countries (Baughn, on company boards is required by corporate law, and there is more
Bodie, and McIntosh, 2007; Hah & Freeman, 2014; Welford, 2005; awareness of the need for related party transactions disclosure in
Wokutch, 1990; Yang & Rivers, 2009). In particular, the US business corporate regulations in Taiwan. Hence, in a more developed
system is rooted in an American society characterized by a high emerging market, such as Taiwan, it is not so difficult to commu-
appreciation of individual freedom and fairly unregulated markets. nicate with stakeholders about the quality of a firm's corporate
Consequently, many social issues, such as the environment, governance when the local institutional infrastructure is estab-
employment and corporate community contributions, have been at lished. Hence, in a more developed emerging market, such as
the core of CSR in U.S. companies (Brammer & Pavelin, 2005). Taiwan, whether firm engagement in CSR practices in corporate
However, in Asian countries there has always been a stronger governance can signal more unobservable attributes to stake-
tendency to address social issues through governmental policies holders and whether stakeholders value these unobservable attri-
(Crane et al., 2013). Many Asian companies have a legacy of CSR, butes and provide premiums to these firms remain unknown, and
including employment, benefits, social services, and healthcare, not we leave them as empirical questions.
so much as a result of voluntary corporate policies, but more as a Similarly, for CSR practices involving the environment, compli-
response to the local regulatory regime. Hence, in general, the CSR ance with certain environmental standards (ISO 14001 environ-
policies of Asian companies are more likely to be affected by mental management system certification) is often driven by supply
mandatory policies and provide less differentiable information chains from the global markets. Companies facing pressure from
across firms within an economic entity. However, we argue that supply chains are required to adopt certain CSR practices involving
firms can distinguish themselves in emerging economies by the environment. In addition, for Asian companies, CSR practices
adopting CSR practices in employment for the following reasons. associated with the community tend to be responses to regulatory
First, Asian companies tend to focus on CSR's economic ratio- regimes. Therefore, whether CSR practices for the environment or
nale, but ignore issues such as human rights for employees. Welford the community can become a capability differentiator for Asian
(2005) compares large companies' written CSR policies in 15 companies in a more developed emerging market remains un-
countries across Asia, North America and Europe, and finds that known, and we also leave it as an empirical question.
Asian companies are less likely to have a policy related to working
hours, maximum overtime, and fair wage structures, and are less 3. Data and methodology
committed to freedom of union or association, not to mention staff
development. By comparison, Welford (2005) does not find sig- 3.1. Data
nificant difference in the frequency of CSR policies related to
external stakeholders (suppliers or community) between Asian In this study, we use the monthly data ratings obtained from
companies and European and North American. Therefore, volun- CSRHub over the period from 01/2009 to 07/2014 to analyze the
tary CSR practices in employment may become signals to investors relationships between CSR and financial performance in Taiwan
in differentiating firm quality in emerging Asian economies. firms. The reasons are twofold. First, to the best of our knowledge,
Second, a comprehensive employee benefits package is costly, no related study with a similar research design using Taiwan data
including a competitive salary, insurance, paid time off from work, exists. Second, the CSRHub uses data sources from various socially
64 L. Lin et al. / Asia Pacific Management Review 24 (2019) 61e71

responsible investing firms, well-known indexes, publications, Since we primarily use panel data for this study, 83 firms with
crowdsources and government agencies and provides CSR ratings adequate data are analyzed. Table 1 presents the sample's industry
of more than 5000 companies from 65 countries. Some sources distribution. Each industry listed in Table 1 contains fewer than five
provide numerical scores while others use relative rankings or firms, except “Chemicals, Plastics & Rubber”, “Computers & Pe-
signs (“þ” or “-“). By aggregating and normalizing the information ripherals” and “Semiconductor & Other Electronic”, which have
from these sources, CSRHub provides a data format that uses a nine to fifteen firms. The distribution is consistent with that of
single value ranging from 0 to 100 for each category/subcategory, listed firms in Taiwan. The list of firms used in the current study is
which makes the interpretation easier and more meaningful based in the Appendix. Table 2 presents the data description for the rat-
on the relationships found in the analysis. ings. The ratings under each major category as well as each sub-
Previous studies have often relied on the Kinder, Lydenberg, and category are provided. The ratings range from 12 to 82. There are
Domini's KLD Research & Analytics database as a source for CSR 875 firm-month observations.
sustainability ratings. The KLD Research & Analytics database in-
cludes the following six dimensions: community involvement,
corporate governance, diversity, employee relations, environment, 3.2. Methodology
and product. KLD uses multiple criteria to evaluate firms, using
positive, indicating strength, and negative, indicating weakness, The current study focuses on the relation between financial
screens. Each screen is a binary variable, 1 and 0. Unlike the clas- performance and CSR dimensions. Three financial performance
sification in KLD Research & Analytics, CSRHub classifies all di- measures are used, including excess stock returns, book-to-market
mensions into four categories: Community (COMM), Employee ratios, and Tobin's qs. Excess stock return is defined as the monthly
(EMP), Environment (ENV), and Governance (GOV). COMM covers return of a stock that exceeds the risk-free rate (proxied by one-
activities and concepts related to human rights, supply chain, month T-bills rate) in the corresponding month. We use the
product quality and safety, product sustainability, community excess returns to examine the relationship between CSR di-
development and philanthropy. EMP covers diversity, labor rights, mensions and returns after allowing for firm characteristics.
treatment of unions, compensation, benefits, training, health, and Therefore, consistent with Brammer, Brooks, and Pavelin (2006),
worker safety. ENV covers environmental policy, environmental we control for the firm's systematic beta risk, market capitalization,
reporting, waste management, resource management, energy use, book-to-market ratio, past average returns, share turnover, and age
climate change policies and performance. GOV covers leadership as performance attribution factors in the excess returns
ethics, board composition, executive compensation, transparency regressions.
and reporting, and stakeholder treatment. Each of the four cate- Galema et al. (2008) investigate the impact of SRI on the value of
gories is further divided into three sub-categories. The three sub- firm using book-to-market regressions. They find that SRI impacts
categories for Community (COMM) include: (1) Community stock returns by lowering the book-to-market ratio, but not by
Development and Philanthropy, (2) Product, and (3) Human Rights generating positive alphas. Their analyses are based on different
and Supply Chain. The three sub-categories for Employee (EMP) dimensions of socially responsible performance classified by KLD
include: (1) Compensation and Benefits, (2) Diversity and Labor Research & Analytics. Consistent with Galema et al. (2008), we
Rights, and (3) Training, Safety, and Health. The three sub- measure the book-to-market ratio of a firm as the logarithm of the
categories for Environment (ENV) include: (1) Energy and Climate book-to-market ratio of equity value. However, the major differ-
Change, (2) Policy and Reporting, and (3) Resource Management. ence between our study and that of Galema et al. (2008) lies in the
The three sub-categories for Governance (GOV) include: (1) Board, data values we use. Our scores range from 1 to 100, allowing us to
(2) Leadership Ethics, and (3) Transparency and Reporting. generate a more meaningful interpretation of the data. However,
CSRHub uses its own methodology to collect information from data values of 1 and 0 do not enable us to infer the extent of the
more than 175 resources and then follows five steps to generate the impact on each subcategory. Consistent with Galema et al. (2008),
ratings for each category and subcategory. The first step is to map our second set of model specifications investigates the impact of
the data to a central schema. CSRHub divides CSR performance into CSR dimensions on the value of the firm using book-to-market
twelve subcategories, which we roll up into four categories: Com- regressions.
munity (COMM), Employee (EMP), Environment (ENV), and Our third performance measure is Tobin's q. Guenster, Derwall,
Governance (GOV). Each element of data received is mapped into Bauer, and Koedijk (2011) focus on the environmental aspect of CSR
one or more subcategories. In the second step, each data item from and investigate the impact what they term its “relative eco effi-
the data sources is converted into a rating on a 0 to 100 scale ciency” on Tobin's q. Harjoto and Jo (2011) also apply Tobin's q as a
(100 ¼ positive rating). The third step is normalizing the data. financial performance measure to investigate the effects of CSR in
Comparisons for the same company between scores from different corporate governance on firm value. We measure firm value with
data sources is performed. CSRHub makes adjustments in order to Tobin's q, which is defined as the ratio of the sum of market equity
remove bias and create a more consistent rating based on the value and liabilities to its corresponding total asset book value. A
variation between sources. The data are then aggregated in the firm displaying Tobin's q greater than unity is considered as using
fourth step. CSRHub weights each source based on estimate of its scarce resources effectively, and those with lower Tobin's q (or even
credibility and value and then combines all of the available data on less than unity) as using resources poorly.
a company and generates base ratings at the subcategory level.
CSRHub then aggregates these ratings to the category level. Finally,
CSRHub drops ratings when it does not have enough information. 3.2.1. Excess returns regressions
Detailed explanations of the major CSR categories COMM, EMP, To establish the relationship between returns and CSR at the
ENV, and GOV, along with the sub-categories COMMsub1- level of individual stocks, we investigate the direct impact of
COMMsub3, EMPsub1-EMPsub3, ENVsub1-ENVsub3, and GOV- CSRHub scores on excess returns. This allows us to identify whether
sub1-GOVsub3 are available in the company's website.4 different dimensions have confounding effects on the relation be-
tween CSR and return. Using the CSRHub scores defined above
along with a host of control variables, we perform the following
4
https://www.csrhub.com/content/csrhub-data-schema/. regression:
 L. Lin et al. / Asia Pacific Management Review 24 (2019) 61e71 65

Table 1
Sample Industry Distribution.
This study analyzes the effect of CSR dimensions on the financial performance of firms in Taiwan. We use monthly ratings obtained from the CSRHub over the period from 01/
2009 to 07/2014. Table presents industry distribution of our sample in Taiwan.

Industry Frequency Percent Cumulative frequency Cumulative percentage Percent

Architectural, Engineering 1 1.20 1 1.20

Audio & Video Equipment Manufacturing 3 3.61 4 4.82
Banking 4 4.82 8 9.64
Brokerage & Capital Markets 1 1.20 9 10.84
Business Support Services 1 1.20 10 12.05
Chemicals, Plastics & Rubber 9 10.84 19 22.89
Communications Equipment Manufacturing 1 1.20 20 24.10
Computers & Peripherals 10 12.05 30 36.14
Conglomerates 2 2.41 32 38.55
Construction Materials 1 1.20 33 39.76
Containers & Packaging Manufacturing 2 2.41 35 42.17
Diversified Financial Services 4 4.82 39 46.99
Electrical Equipment Manufacturing 3 3.61 42 50.60
Electronic Equipment & Instrument 5 6.02 47 56.63
Food Products 2 2.41 49 59.04
Hardware Manufacturing 2 2.41 51 61.45
Leisure Equipment Manufacturing 1 1.20 52 62.65
Manufacturing 2 2.41 54 65.06
Mechanical Component Manufacturing 1 1.20 55 66.27
Motor Vehicle Manufacturing 1 1.20 56 67.47
Office Machinery Manufacturing 1 1.20 57 68.67
Passenger Airlines 1 1.20 58 69.88
Residential Building Construct 1 1.20 59 71.08
Semiconductor & Other Electronic 15 18.07 74 89.16
Supermarket, Food & Beverage 1 1.20 75 90.36
Telecommunications 1 1.20 76 91.57
Textiles & Apparel 2 2.41 78 93.98
Water Transportation 3 3.61 81 97.59
Wholesale Trade 1 1.20 82 98.80
Wireless Telecommunications 1 1.20 83 100.00

COMMsub3, employee (EMP)/EMPsub1-EMPsub3, environment

Ri;t  RFt ¼ b0 þ b1 CSRRatingi;t1 þ b2 Betai;t þ b3 Xi;t þ εi;t ; (ENV)/ENVsub1-ENVsub3, and governance (GOV)/GOVsub1-
(1) GOVsub3.Xi;t is the vector of control variables including: the loga-
rithm of return on equity at the end of month t (ROE) and the
where Ri;t is the monthly return of stock i in month t and RFt is the logarithm of age as measured at the end of month t (Age). ROE has
return on one-month T-bills in month t. The vector CSRi;t1 includes been winsorized to exclude the 1% smallest and largest
the scores of four CSR categories (twelve subcategories variables) at observations.
the end of month t-1. Betai;t is a stock's post-ranking beta estimated
using the traditional method of Black (1972). Xi;t is a vector of 3.2.3. Tobin's q regressions
control variables including the natural logarithm of firm i's market We measure firm value with Tobin's q, which is defined as the
capitalization at the end of month t (SIZE), the logarithm of the ratio of the sum of market equity value and liabilities to its
book-to-market ratio of stock i at the end of month t (BM), a firm's
simple average of returns during the past 12 months (AveRet), the
one-month lagged monthly average of daily share turnover in firm i Table 2
Sample Characteristic Description.
(TurnOver), and the natural log of a company's age measured at the
We use monthly ratings obtained from CSRHub over the period from 01/2009 to 07/
end of month t (Age). εi;t is the error term for firm i in month t. 2014. Table presents data description for ratings. The ratings under each major
category as well as sub-category are provided. There are 875 firm-month
observations.
3.2.2. Book-to-market regressions
Book-to-market (BM) ratio, defined as the ratio of book value to Variable N Mean Minimum Maximum
market value, is used to proxy for the firm's growth opportunity COMM 875 49.04 25 69
and bankruptcy risk. High BM is mainly seen as a high bankruptcy EMP 875 50.93 26 72
risk for equity investors. Hence, it has a higher than expected re- ENV 875 50.32 23 72
GOV 875 43.13 26 69
turn. Low BM, sometimes, may reflect the success of managers in COMMsub1 875 45.99 16 70
overseeing strong operating performance and growth in net assets COMMsub2 875 46.63 22 75
of the firm. In this study, we examine the relationship between BM COMMsub3 875 54.53 12 77
and CSR dimensions. The regression is as follows. EMPsub1 875 48.67 15 70
EMPsub2 875 53.27 23 74
BMi;t ¼ b0 þ b1 CSRRatingi;t1 þ b2 Xi;t þ εi;t ; (2) EMPsub3 875 50.83 23 77
ENVsub1 875 49.78 18 75
ENVsub2 875 49.60 18 78
where BMi;t is the logarithm of the book-to-market ratio of firm i at ENVsub3 875 51.56 19 77
the end of month t. CSRRatingi;t1 is the vector containing the GOVsub1 875 35.15 14 60
(lagged) ratings of four major categories/twelve sub-major cate- GOVsub2 875 52.29 22 82
GOVsub3 875 41.87 13 74
gories at end of month t-1: community (COMM)/COMMsub1-
66 L. Lin et al. / Asia Pacific Management Review 24 (2019) 61e71

corresponding total asset book value. The regression is as follows. management of its products and services. Surprisingly, none of the
other CSR components show statistically significant results at any
TobinQi;t ¼ b0 þ b1 CSRRatingi;t1 þ b2 Xi;t þ εi;t ; (3) conventional level. In general, the results are consistent with the
literature that finds few significant relations between CSR di-
where TobinQi;t is the logarithm of the ratio of the sum of market mensions and stock returns.
equity value and liabilities to its corresponding total asset book value
for firm i at the end of month t. CSRRatingi;t1 is the vector containing 4.2. Book-to-market and CSR dimensions
the (lagged) ratings of four major categories/twelve sub-major cat-
egories at end of month t-1: community (COMM)/COMMsub1- Book-to-market (BM) ratio, defined as the ratio of book value to
COMMsub3, employee (EMP)/EMPsub1-EMPsub3, environment market value, is used to proxy for firm's growth opportunity and
(ENV)/ENVsub1-ENVsub3, and governance (GOV)/GOVsub1- bankruptcy risk. High BM is mainly seen as a high bankruptcy risk
GOVsub3.Xi;t is the vector of control variables including: the loga- for equity investors. In the finance literature, high BM firms are
rithm of return on equity at the end of month t (ROE) and the log- often regarded as “value firms”. Low BM, on the other hand, largely
arithm of age as measured at the end of month t (Age). ROE has been reflects the success of managers in overseeing strong operating
winsorized to exclude the 1% smallest and largest observations. performance and growth in net assets of the firm. Firms with low
BM are often regarded as “growth firms”. Tables 5 and 6 show the
4. Results results in which we examine the association between BM ratio and
CSR dimensions. No single major CSR category shows statistical
4.1. Excess returns and CSR dimensions significance when tested. EMP has a negative and significant as-
sociation with BM at the 0.05 level in model (5). This is also
Tables 3 and 4 show the results in which we examine the as- confirmed in model (2) and (5) of Table 6. This result indicates that
sociation between a firm's excess returns with CSR ratings across growth firms allocate more resources to issues related to workplace
different categories and control variables. In Table 3, COMM and policies and practices covering fair and non-discriminatory treat-
ENV show a negative association with the firm's excess returns, ment of employees, and to diversity policies. The results are
while EMP and GOV show a positive association. However, none is consistent with Hypothesis 1. In addition, COMMsub3 is positively
statistically significant at any conventional level. and significantly associated with BM. This indicates that value firms
We next explore whether the aggregate CSR categories may allocate more resources to the firm's capacity to reduce environ-
have confounding effects because their sub-categories may intro- mental costs, create new market opportunities through new sus-
duce conflicts that affect excess returns. For example, news related tainable technologies or processes, and produce or market goods
to philanthropy can positively affect firm's stock returns, whereas and services that enhance the health and quality of life for con-
news related to reducing environmental costs may negatively affect sumers. Under the ENV category, ENVsub3 shows a positive and
the firm's stock returns due to rising production costs. We apply the significant association with BM. This means that value firms put
sub-categories in the analysis. In Table 4, only two sub-categories effort into managing how efficiently resources are used in
under COMM show statistical significance at the 0.1 level. COMM- manufacturing and delivering products and services, including
sub1 is positively and significantly associated with the firm's stock those of a company's suppliers.
returns. This indicates that the market recognizes efforts made in
Community Development & Philanthropy. COMMsub3 is nega- 4.3. Tobin's q and CSR dimensions
tively and significantly associated with the firm's stock returns. This
indicates that investors are concerned about the impact of the re- Tobin's q, defined as the ratio of the sum of market equity value
sponsibility a company assumes for the development, design, and and liabilities to its corresponding total asset book value, is used to

Table 3
Panel Data Analysis-The Effects of CSR Ratings on Stock Returns (Major Categories).
Table presents results in which we use panel data analysis to examine the association between firm's returns and CSR major category ratings along with control variables. The
regression is as follows. Ri;t  RFt ¼ b0 þ b1 CSRRatingi;t1 þ b2 Betai;t þ b3 Xi;t þ εi;t , where Ri;t is the monthly return of stock i in month t and RFt is the return on one-month T-
bills in month t. CSRRatingi:t1 is the vector containing the (lagged) ratings of four major categories/twelve sub-major categories at end of month t-1: community (COMM)/
COMMsub1-COMMsub3, employee (EMP)/EMPsub1-EMPsub3, environment (ENV)/ENVsub1-ENVsub3, and governance (GOV)/GOVsub1-GOVsub3. Betai;t is a stock's post-
ranking beta estimated using the traditional method of Black (1972). Xi;t is a vector of control variables including the natural logarithm of firm i's market capitalization at
the end of month t (SIZE), the logarithm of the book-to-market ratio of stock i at the end of month t (BM), a firm's simple average of returns during the past 12 months (AveRet),
the one-month lagged monthly average of daily share turnover in firm i (TurnOver), and the natural log of a company's age measured at the end of month t (Age). εi;t is the error
term for firm i in month t. ***,** and*, indicate significance at the 1%, 5%, and 10% levels, respectively.

Model (1) Model (2) Model (3) Model (4) Model (5)

coefficient t-stat. coefficient t-stat. coefficient t-stat. coefficient t-stat. coefficient t-stat.

Intercept 2.44*** 9.23 2.39*** 9.16 2.40*** 9.22 2.41*** 9.23 2.50*** 9.36
COMM 0.04 0.91 0.10 1.46
EMP 0.01 0.19 0.03 0.60
ENV 0.02 0.46 0.02 0.37
GOV 0.06 1.14 0.11 1.56
BETA 0.14*** 4.19 0.14*** 4.31 0.14*** 4.22 0.14*** 4.43 0.14*** 4.30
SIZE 0.12*** 8.45 0.12*** 8.37 0.12*** 8.41 0.12*** 8.39 0.12*** 8.51
BM 0.06*** 3.98 0.05*** 3.89 0.06*** 3.93 0.05*** 3.84 0.06*** 3.96
AveRet 0.07 0.67 0.07 0.70 0.07 0.68 0.07 0.69 0.07 0.66
TurnOver 0.01 0.93 0.01 0.97 0.01 0.98 0.01 1.02 0.01 0.94
Age 0.04 1.18 0.04 1.11 0.03 1.10 0.05 1.43 0.06* 1.70
Year fixed Yes Yes Yes Yes Yes
Industry fixed Yes Yes Yes Yes Yes
N Cross-sections 83 83 83 83 83
 L. Lin et al. / Asia Pacific Management Review 24 (2019) 61e71 67

Table 4
Panel Data Analysis-The Effects of CSR Ratings on Stock Returns (Sub Categories).
Table presents results in which we use panel data analysis to examine the association between firm's returns and CSR sub-categories ratings along with control variables.
Equation and all variables are as previously defined as in Table 3. ***,** and*, indicate significance at the 1%, 5%, and 10% levels, respectively.

Model (1) Model (2) Model (3) Model (4) Model (5)

coefficient t-stat. coefficient t-stat. coefficient t-stat. coefficient t-stat. coefficient t-stat.

Intercept 2.52*** 8.96 2.43*** 9.08 2.48*** 9.19 2.43*** 8.91 2.61*** 8.92
COMMsub1 0.09* 1.82 0.06 0.84
COMMsub2 0.03 0.84 0.06 1.33
COMMsub3 0.08* 1.80 0.08 1.64
EMPsub1 0.02 0.36 0.03 0.62
EMPsub2 0.02 0.48 0.06 1.07
EMPsub3 0.01 0.28 0.01 0.11
ENVsub1 0.01 0.19 0.03 0.69
ENVsub2 0.02 0.52 0.01 0.11
ENVsub3 0.00 0.04 0.02 0.39
GOVsub1 0.08 1.30 0.09 1.29
GOVsub2 0.03 0.78 0.03 0.42
GOVsub3 0.06 1.28 0.02 0.32
BETA 0.14*** 3.92 0.15*** 4.49 0.14*** 4.32 0.13*** 4.06 0.15*** 4.09
SIZE 0.12*** 8.03 0.12*** 8.24 0.12*** 8.33 0.12*** 8.17 0.12*** 7.86
BM 0.05*** 3.40 0.05*** 3.79 0.06*** 3.95 0.05*** 3.78 0.05*** 3.31
AveRet 0.11 1.01 0.07 0.70 0.07 0.72 0.06 0.55 0.09 0.85
TurnOver 0.01 0.87 0.01 1.03 0.01 0.98 0.01 1.13 0.01 0.95
Age 0.05 1.53 0.04 1.20 0.04 1.26 0.05 1.53 0.07** 2.03
Year fixed effects Yes Yes Yes Yes Yes
Industry fixed effects Yes Yes Yes Yes Yes
N Cross-sections 83 83 83 83 83

Table 5
Panel Data Analysis-The Effects of CSR Rtings on Book-to-Market Ratios (Major Categories).
Table 5 presents results in which we use panel data analysis to examine the association between book-to-market ratios and CSR major category ratings along with control
variables. The regression is as follows. BMi;t ¼ b0 þ b1 CSRRatingi;t1 þ b2 Xi;t þ εi;t , where BMi;t is the logarithm of the book-to-market ratio of firm i at the end of month t.
CSRRatingi;t1 is the vector containing the (lagged) ratings of four major categories/twelve sub-major categories at end of month t-1: community (COMM)/COMMsub1-
COMMsub3, employee (EMP)/EMPsub1-EMPsub3, environment (ENV)/ENVsub1-ENVsub3, and governance (GOV)/GOVsub1-GOVsub3. Xi;t is the vector of control variables
including: the logarithm of return on equity at the end of month t (ROE) and the logarithm of age as measured at the end of month t (Age). ROE has been winsorized to exclude
the 1% smallest and largest observations. ***,** and*, indicate significance at the 1%, 5%, and 10% levels, respectively.

Model (1) Model (2) Model (3) Model (4) Model (5)

coefficient t-stat. coefficient t-stat. coefficient t-stat. coefficient t-stat. coefficient t-stat.

Intercept 10.42*** 7.31 10.43*** 7.33 10.28*** 7.16 10.40*** 7.31 10.20*** 7.10
COMM 0.01 0.05 0.13 0.56
EMP 0.19 1.24 0.43** 2.02
ENV 0.10 0.69 0.13 0.73
GOV 0.12 0.61 0.27 1.14
ROE 2.61*** 9.10 2.60*** 9.09 2.59*** 9.03 2.61*** 9.11 2.58*** 8.97
Age 0.27** 2.22 0.28** 2.32 0.28** 2.26 0.26** 2.13 0.29** 2.28
Year fixed Yes Yes Yes Yes Yes
Industry fixed Yes Yes Yes Yes Yes
N Cross-sections 83 83 83 83 83

proxy for the firm's value creation. Tables 7 and 8 show the results results support Hypothesis 1 that CSR sustainability ratings in
in which we examine the association between Tobin's q and CSR employment are positively related to firm financial performance in
dimensions. Again, no single major CSR category shows statistical emerging economies, such as Taiwan.
significance when tested. However, COMM shows a negative and There are two contradictory impacts of ENV on Tobin's q. ENV-
significant association with Tobin's q at the 0.05 level and EMP sub2 shows a positive and significant association at the 0.1 level,
shows a positive and significant association with Tobin's q at the while ENVsub3 shows a negative and significant association with
0.01 level. When we investigate further in Table 8, none of the sub- Tobin's q at the 0.05 level. The former covers “a company's policies
categories under COMM is significant. The negative sign is consis- and intention to reduce the environmental impact of a company
tent with that in Table 7, which indicates that the negative impact and its value stream to levels that are healthy for the company and
dominates the association with Tobin's q. In other words, efforts or for the environment, now and in the future.” The data includes the
resources spent on human rights, supply chain, product quality and company's environmental reporting performance, adherence to
safety, community development, and philanthropy have a negative environmental reporting standards such as the Global Reporting
impact on the firm's creation of market value. Conversely, positive Initiative, and compliance with investor, regulatory and stake-
impacts dominate the association with Tobin's q under the EMP holders' requests for transparency. This indicates that transparency
category. EMPsub2 and EMPsub3 are both positively and signifi- is beneficial to a firm's market value. The latter covers “how effi-
cantly associated with Tobin's q. This indicates that efforts related ciently resources are used in manufacturing and delivering prod-
to activities or concepts, such as diversity, labor rights, and benefits, ucts and services, including those of a company's suppliers. This
have a positive impact on the firm's creation of market value. The subcategory includes environmental performance relative to
68 L. Lin et al. / Asia Pacific Management Review 24 (2019) 61e71

Table 6
Panel Data Analysis-The Effects of CSR Ratings on Book-to-Market Ratios (Sub Categories).
Table 6 presents results in which we use panel data analysis to examine the association between book-to-market ratios and CSR sub-categories ratings along with control
variables. Equation and all variables are as previously defined as in Table 5. ***,** and*, indicate significance at the 1%, 5%, and 10% levels, respectively.

Model (1) Model (2) Model (3) Model (4) Model (5)

coefficient t-stat. coefficient t-stat. coefficient t-stat. coefficient t-stat. coefficient t-stat.

Intercept 10.47*** 7.30 10.24*** 7.18 10.45*** 7.25 10.39*** 7.25 10.14*** 6.98
COMMsub1 0.05 0.39 0.05 0.26
COMMsub2 0.19 1.49 0.16 1.11
COMMsub3 0.24* 1.70 0.33** 2.08
EMPsub1 0.17 0.98 0.05 0.30
EMPsub2 0.27* 1.77 0.40** 2.19
EMPsub3 0.03 0.18 0.14 0.78
ENVsub1 0.04 0.32 0.14 1.01
ENVsub2 0.21 1.43 0.18 1.08
ENVsub3 0.34** 2.57 0.54*** 3.06
GOVsub1 0.13 0.61 0.19 0.84
GOVsub2 0.08 0.53 0.21 1.16
GOVsub3 0.06 0.39 0.22 1.12
ROE 2.62*** 9.13 2.57*** 8.94 2.60*** 9.04 2.60*** 9.06 2.54*** 8.83
Age 0.26** 2.09 0.28** 2.31 0.25** 2.02 0.26** 2.06 0.25** 1.96
Year fixed effects Yes Yes Yes Yes Yes
Industry fixed effects Yes Yes Yes Yes Yes
N Cross-sections 83 83 83 83 83

Table 7
Panel Data Analysis-The Effects of CSR Ratings on Tobin's q (Major Categories).
Table presents results in which we use panel data analysis to examine the association between Tobin's q and CSR major category ratings along with control variables. The
regression is as follows. TobinQi;t ¼ b0 þ b1 CSRRatingi;t1 þ b2 Xi;t þ εi;t , where TobinQi;t is the logarithm of the ratio of the sum of market equity value and liabilities to its
corresponding total asset book value for firm i at the end of month t. All other variables are as previously defined in Table 5. ***,** and*, indicate significance at the 1%, 5%, and
10% levels, respectively.

Model (1) Model (2) Model (3) Model (4) Model (5)

coefficient t-stat. coefficient t-stat. coefficient t-stat. coefficient t-stat. coefficient t-stat.

Intercept 6.22*** 7.03 6.27*** 7.12 6.21*** 6.97 6.25*** 7.07 6.10*** 6.87
COMM 0.07 0.63 0.33** 2.29
EMP 0.19** 2.03 0.45*** 3.44
ENV 0.03 0.31 0.02 0.19
GOV 0.02 0.13 0.14 0.97
ROE 1.50*** 8.43 1.49*** 8.40 1.49*** 8.36 1.49*** 8.42 1.49*** 8.38
Age 0.08 1.09 0.09 1.19 0.08 1.05 0.08 1.02 0.11 1.47
Year fixed Yes Yes Yes Yes Yes
Industry fixed Yes Yes Yes Yes Yes
N Cross-sections 83 83 83 83 83

Table 8
Panel Data Analysis- The Effects of CSR Ratings on Tobin's q (Sub Categories).
Table presents results in which we use panel data analysis to examine the association between Tobin's q and CSR sub-categories ratings along with control variables. Equation
and all variables are as previously defined as in Table 7. ***,** and*, indicate significance at the 1%, 5%, and 10% levels, respectively.

Model (1) Model (2) Model (3) Model (4) Model (5)

coefficient t-stat. coefficient t-stat. coefficient t-stat. coefficient t-stat. coefficient t-stat.

Intercept 6.29*** 7.06 6.22*** 7.03 6.26*** 7.01 6.25*** 7.03 6.00*** 6.65
COMMsub1 0.05 0.62 0.08 0.73
COMMsub2 0.05 0.56 0.06 0.70
COMMsub3 0.05 0.61 0.15 1.55
EMPsub1 0.01 0.13 0.04 0.40
EMPsub2 0.11 1.17 0.24** 2.15
EMPsub3 0.06 0.58 0.19* 1.72
ENVsub1 0.01 0.09 0.03 0.39
ENVsub2 0.20** 2.22 0.19* 1.87
ENVsub3 0.20*** 2.41 0.22** 2.02
GOVsub1 0.11 0.83 0.10 0.74
GOVsub2 0.11 1.14 0.08 0.72
GOVsub3 0.02 0.20 0.15 1.25
ROE 1.51*** 8.45 1.48*** 8.32 1.48*** 8.32 1.50*** 8.44 1.46*** 8.19
Age 0.07 0.94 0.09 1.19 0.06 0.85 0.09 1.09 0.11 1.38
Year fixed effects Yes Yes Yes Yes Yes
Industry fixed effects Yes Yes Yes Yes Yes
N Cross-sections 83 83 83 83 83
 L. Lin et al. / Asia Pacific Management Review 24 (2019) 61e71 69

production size and is monitored by the production-related Eco perspective of BM, two sub-categories: Product, and Resource
Intensity Ratios (EIRs) for water and energy defined as resource Management, tend to be valued by high BM firms, the value firms,
consumption per produced or released unit.” This indicates that while growth firms tend to value Diversity and Labor Rights. Last,
firms need to be more cautious when consuming resources during from the perspective of Tobin's q, two sub-categories under EMP
the production process, which may reduce produc competitiveness show consistent and positive effects on Tobin's q, while the oppo-
and the firm's market value. Surprisingly, the majority of sub- site effect is seen in two sub-categories under ENV. Therefore, firms
categories under GOV show a negative association with Tobin's q. that allocate resources to diversity, labor rights, treatment of
However, none of the sub-categories under GOV is statistically unions, compensation, benefits, training, health, and worker safety
significant. experience benefits for the firm's value creation. The results indi-
cate that CSR polices/practices in employment can have signaling
5. Conclusions effects in differentiating firm quality in emerging Asian economies,
at least in a more developed Asian emerging market such as
Do firms that implement CSR dimensions see any benefits to the Taiwan, and that investors value these attributes and provide pre-
bottom line? In this study, we use the data provided by CSRHub to miums to these firms.
examine the relationship between CSR dimensions and stock
returns, book-to-market ratios, and Tobin's q for firms listed in Acknowledgements
Taiwan.
The results in this study have three implications. First, from the Funding: The author gratefully acknowledges the financial
perspective of stock returns, we observe two opposite effects under support from the Ministry of Science and Technology, Taiwan (105-
the COMM categories: Community Development and Philanthropy 2420-H-003-002-MY2).
is positive while the Product subcategory is negative for stock
returns. However, these effects disappear when all sub-categories Appendix. List of Firms
are considered together. Thus, no single CSR sub-category has a
significant association with stock returns. Second, from the

Company Exchange IPO Year Industry

Taiwan Cement Corp. TSE 1962 Cement

Tatung Co., Ltd TSE 1962 Electric & Machinery
Chang Hwa Commercial Bank, Ltd. TSE 1962 Financial
Nankang Rubber Tire Corp., Ltd. TSE 1963 Rubber
Formosa Plastics Corp. TSE 1964 Plastics
Nan Ya Plastics Corp. TSE 1967 Plastics
Walsin Lihwa Corp. TSE 1972 Electric Appliance & Cable
Taiwan Glass Ind Co., Ltd. TSE 1973 Glass & Ceramics
Teco Electric & Machinery Co., Ltd. TSE 1973 Electric & Machinery
China Steel Corp. TSE 1974 Iron & Steel
Yulon Motor Co., Ltd. TSE 1976 Automobile
Ruentex Industries Limited TSE 1977 Trading & Cons.
LCY Chemical Corp. TSE 1977 Chemical
TSRC Corporation TSE 1982 Rubber
United Microelectronics Corp. TSE 1985 Semiconductor
Oriental Union Chemical Corp. TSE 1987 Chemical
Uni-President Enterprises Corp. TSE 1987 Foods
Tung Ho Steel Enterprise Corp. TSE 1988 Iron & Steel
Tainan Spinning Co., Ltd. TSE 1989 Textiles
Pou Chen Corporation TSE 1990 Others
U-Ming Marine Transport Corp. TSE 1990 Shipping & Trans.
Hon Hai Precision Ind. Co., Ltd. TSE 1991 Other Electronic
China Petrochemical Development Corp. TSE 1991 Plastics
Compal Electronics, Inc. TSE 1992 Computer & Peripherals
Yang Ming Marine Transport Corp. TSE 1992 Shipping & Trans.
Ruentex Development Co., Ltd. TSE 1992 Others
Siliconware Precision Ind. Co., Ltd. TSE 1993 Semiconductor
CTCI Corporation TSE 1993 Others
Eternal Materials Co., Ltd. TSE 1994 Chemical
Standard Foods Corporation TSE 1994 Foods
Giant Manufacturing Co., Ltd. TSE 1994 Others
Macronix International Co., Ltd. TSE 1995 Semiconductor
Winbond Electronics Corp. TSE 1995 Semiconductor
Synnex Technology International Corp. TSE 1995 Elec. Products Dist.
KGI Securities Co., Ltd. TSE 1995 Financial
Wan Hai Lines Ltd. TSE 1996 Shipping & Trans.
Qisda Corp. TSE 1996 Computer & Peripherals
Acer Inc. TSE 1996 Computer & Peripherals
Asustek Computer Inc. TSE 1996 Computer & Peripherals
President Chain Store Corp. TSE 1997 Trading & Cons.
Taiwan Business Bank TSE 1998 Financial
Taiwan Fertilizer Co., Ltd. TSE 1998 Chemical
Vanguard International Semiconductor Corp. OTC 1998 Semiconductor
(continued on next page)
70 L. Lin et al. / Asia Pacific Management Review 24 (2019) 61e71

(continued )

Company Exchange IPO Year Industry

Realtek Semiconductor Corp. TSE 1998 Semiconductor

Wintek Corporation TSE 1998 Optoelectronic
Chicony Electronics Co., Ltd. TSE 1999 Computer & Peripherals
Quanta Computer Inc. TSE 1999 Computer & Peripherals
Everlight Electronics Co., Ltd. TSE 1999 Optoelectronic
Taiwan Semiconductor Co., Ltd. OTC 2000 Semiconductor
Sino-American Silicon Products Inc. OTC 2001 Semiconductor
Transcend Information, Inc. TSE 2001 Semiconductor
Mediatek Incorporation TSE 2001 Semiconductor
Catcher Technology Co., Ltd. TSE 2001 Other Electronic
Chunghwa Picture Tubes Ltd. TSE 2001 Optoelectronic
EVA Airways Corporation TSE 2001 Shipping & Trans.
Simplo Technology Co., Ltd. OTC 2001 Computer & Peripherals
Cathay Financial Holding Co., Ltd. TSE 2001 Financial
Yuanta Financial Holding Co., Ltd. TSE 2002 Financial
Taishin Financial Holding Co., Ltd. TSE 2002 Financial
Shin Kong Financial Holding Co., Ltd. TSE 2002 Financial
Sinopac Financial Holdings Co., Ltd. TSE 2002 Financial
Novatek Microelectronics Corp. TSE 2002 Semiconductor
Unimicron Technology Corp. TSE 2002 Elec. Parts & Comp.
Tripod Technology Corp. TSE 2002 Elec. Parts & Comp.
Taiwan Mobile Co., Ltd. TSE 2002 Comm. & Internet
First Financial Holding Co., Ltd. TSE 2003 Financial
Wistron Corp. TSE 2003 Computer & Peripherals
Richtek Technology Corp. TSE 2003 Semiconductor
E Ink Holdings Inc. OTC 2004 Optoelectronic
Powertech Technology Inc. TSE 2004 Semiconductor
Phison Electronics Corp. OTC 2004 Computer & Peripherals
Far EasTone Telecommunications Co., Ltd. TSE 2005 Comm. & Internet
WPG Holdings Limited TSE 2005 Elec. Products Dist.
Inotera Memories, Inc. TSE 2006 Semiconductor
Nan Ya Printed Circuit Board Corp. TSE 2006 Elec. Parts & Comp.
PixArt Imaging Inc. OTC 2006 Semiconductor
Radiant Opto-Electronics Corp. TSE 2007 Optoelectronic
Young Fast Optoelectronics Co., Ltd. TSE 2009 Optoelectronic
Pegatron Corporation TSE 2010 Computer & Peripherals
TPK Holding Co., Ltd. TSE 2010 Optoelectronic
MStar Semiconductor Inc. (Cayman) TSE 2010 Semiconductor
Zhen Ding Technology Holding Limited TSE 2011 Elec. Parts & Comp.
Hermes Microvision, Inc. OTC 2012 Semiconductor

References Freeman, R. E. (1984). Strategic management: A stakeholder approach (Vol. 1). Boston:
Pitman.
Galema, R., Plantinga, A., & Scholtens, B. (2008). The stocks at stake: Return and risk
Ang, S. H., & Leong, S. M. (2000). Out of the mouths of babes: Business ethics and
in socially responsible investment. Journal of Banking & Finance, 32(12),
youths in Asia. Journal of Business Ethics, 28(2), 129e144.
2646e2654.
Barnea, A., & Rubin, A. (2010). Corporate social responsibility as a conflict between
Guenster, N., Bauer, R., Derwall, J., & Koedijk, K. (2011). The economic value of
shareholders. Journal of Business Ethics, 97(1), 71e86.
corporate eco-efficiency. European Financial Management, 17(4), 679e704.
Barnett, M. L., & Salomon, R. M. (2006). Beyond dichotomy: The curvilinear rela-
Hah, K., & Freeman, S. (2014). Multinational enterprise subsidiaries and their CSR: A
tionship between social responsibility and financial performance. Strategic
conceptual framework of the management of CSR in smaller emerging econo-
Management Journal, 27(11), 1101e1122.
mies. Journal of Business Ethics, 122(1), 125e136.
Bauer, R., Guenster, N., & Otten, R. (2004). Empirical evidence on corporate gover-
Harjoto, M., & Jo, H. (2011). Corporate governance and CSR nexus. Journal of Business
nance in Europe: The effect on stock returns, firm value and performance.
Ethics, 100(1), 45e67.
Journal of Asset Management, 5(2), 91e104.
Hill, R. P., Ainscough, T., Shank, T., & Manullang, D. (2007). Corporate social re-
Baughn, C. C., (Dusty) Bodie, N. L., & McIntosh, J. C. (2007). Corporate social and
sponsibility and socially responsible investing: A global perspective. Journal of
environmental responsibility in Asian countries and other geographical regions.
Business Ethics, 70(2), 165e174.
Corporate Social Responsibility and Environmental Management, 14(4), 189e205.
Jensen, M. C. (2002). Value maximization, stakeholder theory, and the corporate
Black, F. (1972). Capital market equilibrium with restricted borrowing. Journal of
objective function. The Journal of Applied Corporate Finance, 14, 8e21.
Business, 45(3), 444e455.
Kempf, A., & Osthoff, P. (2007). The effect of socially responsible investing on
Brammer, S., Brooks, C., & Pavelin, S. (2006). Corporate social performance and stock
portfolio performance. European Financial Management, 13(5), 908e922.
returns: UK evidence from disaggregate measures. Financial Management, 35(3),
Lee, M. (2008). A review of the theories of corporate social responsibility: Its
97e116.
evolutionary path and the road ahead. International Journal of Management
Brammer, S., & Pavelin, S. (2005). Corporate community contributions in the United
Reviews, 10(1), 53e73.
Kingdom and the United States. Journal of Business Ethics, 56(1), 15e26.
Li, W., & Zhang, R. (2010). Corporate social responsibility, ownership structure, and
Cao, X. (2011). Does it pay to be green? An integrated view of environmental marketing
political interference: Evidence from China. Journal of Business Ethics, 96(4),
with evidence from the forest products industry in China. PhD Thesis. University of
631e645.
Washington.
Lopatta, K., Buchholz, F., & Kaspereit, T. (2016). Asymmetric information and
Cheung, Y., Tan, W., Ahn, H., & Zhang, Z. (2010). Does corporate social responsibility
corporate social responsibility. Business & Society, 55(3), 458e488.
matter in Asian emerging markets? Journal of Business Ethics, 92(3), 401e413.
McGuire, J. B., Sundgren, A., & Schneeweis, T. (1988). Corporate social responsibility
Cornell, B., & Shapiro, A. C. (1987). Corporate stakeholders and corporate finance.
and firm financial performance. Academy of Management Journal, 31(4),
Financial Management, 16, 5e14.
854e872.
Crane, A., Matten, D., & Spence, L. J. (September 1, 2013). Corporate social re-
Oh, W. Y., Chang, Y. K., & Martynow, A. (2011). The effect of ownership structure on
sponsibility in a global context. In Chapter in: Crane, A., Matten, D., and Spence,
corporate social responsibility: Empirical evidence from Korea. Journal of
L.J., 'Corporate Social Responsibility: Readings and Cases in a Global Context', 2/e.
Business Ethics, 104(2), 283e297.
Abingdon (pp. 3e26). Routledge. Available at: SSRN: https://ssrn.com/
Renneboog, L., Ter Horst, J., & Zhang, C. (2008). Socially responsible investments:
abstract¼2322817.
Institutional aspects, performance, and investor behavior. Journal of Banking &
Cui, J., Jo, H., & Na, H. (2012). Does corporate social responsibility reduce informa-
Finance, 32(9), 1723e1742.
tion asymmetry? Journal of Banking and Finance, 1e43 .
 L. Lin et al. / Asia Pacific Management Review 24 (2019) 61e71 71

Robinson, M., Kleffner, A., Bertels, S., Street, C., & Arbor, A. (2008). The value of a Webb, T. (2006). Is Asian corporate governance improving? Ethical corporation, may,
reputation for corporate social responsibility: Empirical evidence. (Unpublished 25e26.
paper). Michigan: University of Michigan. Welford, R. (2005). Corporate social responsibility in Europe, North America and
Su, W., Peng, M. W., Tan, W., & Cheung, Y. L. (2016). The signaling effect of corporate Asia: 2004 survey results. The Journal of Corporate Citizenship, 17, 33e52.
social responsibility in emerging economies. Journal of Business Ethics, 134(3), Wokutch, R. E. (1990). Corporate social responsibility Japanese style. The Academy of
479e491. Management Executive, 4(2), 56e74.
Turban, D. B., & Greening, D. W. (1997). Corporate social performance and organi- Yang, X., & Rivers, C. (2009). Antecedents of CSR practices in MNCs' subsidiaries: A
zational attractiveness to prospective employees. Academy of Management stakeholder and institutional perspective. Journal of Business Ethics, 86(2),
Journal, 40(3), 658e672. 155e169.