The Effect of Financial Secrecy and IFRS Adoption On Earnings Quality: A Comparative Study Between Indonesia, Malaysia and Singapore

International Journal of Innovation, Creativity and Change. www.ijicc.
net
Volume 5, Issue 2, Special Edition, 2019
The Effect of Financial Secrecy and

IFRS Adoption on Earnings Quality:
A Comparative Study between
Indonesia, Malaysia and Singapore
Evita Puspitasaria, Citra Sukmadilagab, Handiani Suciatic, Rianno Febriano Bahard,
*Erlane K Ghanie, a,b,c,dUniversitas Padjadjaran Indonesia, eFaculty of Accountancy
Universiti Teknologi MARA Malaysia.
a
[email protected] [email protected]
c
[email protected], d [email protected],
*Corresponding Author:*[email protected]
This study examines the effect of financial secrecy and IFRS adoption
on earnings quality among three countries namely, Indonesia,
Malaysia, and Singapore. This study utilizes the financial statements
of 71 companies listed on the Indonesian Stock Exchange (IDX),
Bursa Malaysia and the Singapore Stock Exchange (SGX) over a 6
year period, consisting of 426 observations. A regression analysis was
used to analyse the data with discretionary accruals as the earnings
quality and using a secrecy index produced by the Tax Justice
Network as the secrecy proxy. This study uses control variables such
as investor protection, total sales, leverage, sales growth ratio, plant
assets growth rate, operating cash flow, and loss for the period and
industry types as dummy variables in order to provide more robust
findings. The results of this study show that although companies in
Indonesia, Malaysia, and Singapore have adopted IFRS, they produce
different earnings quality. This study shows that the earnings quality
among the countries is not the same. This study provides evidence that
secrecy is an important factor influencing earnings quality. In other
words, a higher secrecy level would lead to lower earnings quality.
The findings of this study provide a new contribution to the financial
reporting literature and a further understanding to academics and
practitioners about the impact of financial secrecy and IFRS adoption
on earnings quality.
863
International Journal of Innovation, Creativity and Change. www.ijicc.net
Key words: earnings quality, discretionary accrual, IFRS adoption, secrecy,

financial statements.
Introduction
It is generally known that different countries would have different accounting standards due
to their unique conditions such as economics view, culture, politics, and social conditions.
However, voluminous business transactions coupled with globalisation have led to the need
to implement one standardised international financial accounting standard in order to achieve
standardisation and harmonisation that represent two important qualitative characteristics of
financial reporting. In 2001, the International Accounting Standard Board (IASB) which
consists of four international organisations namely, the European Commissions (EC),
International Organisation of Securities Commission (IOSOC) and International Federation
of Accountants (IFAC), issued the International Financial Reporting Standard (IFRS). IASB
aims to develop one global accounting financial reporting standard in order to increase the
quality of financial reporting. The international accounting standards are expected to fulfil
public need of information in order to increase the trust, improvement and stabilisation of
long term global finance (Pacter, 2015). To date, there are more than 120 countries that have
adopted and implemented IFRS (Deloitte, 2014).
The IFRSs are developed on principle based standards instead of rule based standards with
the expectation of an increase in the quality of the financial reporting. However, studies that
have examined the impact of principle based standards have often provided mixed findings.
For example: Daske and Gunther (2006) found that principle based standards elevate earnings
quality. Barth, Landsman and Lang (2008) conducted a study on earnings quality before and
after IFRS adoption using 327 companies as the sample from 21 different countries. They
found that earnings management level decreases after IFRS adoption. They also found that
the value relevance of accounting information has elevated and the loss was recognised
earlier after IFRS adoption. Similarly, Morais and Curto (2008) showed that income
smoothing practices in Portuguese companies have decreased IFRS adoption. Other studies
however, showed contrasting findings. For example: Jeanjean and Stolowly (2008) found that
earnings management activities in Australia, France, and UK for the period 2005-2006 did
not decrease after IFRS adoption. In other words, earnings management is still excessive.
Other studies showed that the adoption of high accounting standards does not directly
increase quality of accounting information (Ball, Robin and Wu, 2003; Prochazka, 2017).
864
Their study concluded that the incentive and motivation of the financial information preparer
(management) along with auditor also influence the quality of financial information. In other
words, there are other factors influencing financial information quality. The inconsistent
findings are presumed to be caused by the non-existence of direct measurement over the
financial statement quality (Barth e.al, 2008; Schipper and Vincent, 2003; Cohen,
Krishnamoorthy and Wright, 2004; Nicholas and Wahlen, 2004; Nilwala, Gunawardana &
Fernando 2017). According to Houqe, Monem, Tareq and van Zijl (2015), it is probable that
the implication of the IFRS implementation to the earnings quality varied in every region,
depending on the culture and financial secrecy.
This study aims to examine earnings quality in the relation to financial secrecy. Specifically,
it aims to examine factors influencing IFRS implementation in creating qualified information
in different regions, especially Indonesia, Malaysia, and Singapore. The findings of this study
provide a new contribution to the financial reporting literature and further understanding to
the academics and practitioners on the importance of secrecy to earnings quality. The next
section presents the literature review in Section 2. This is followed by the research design in
Section 3. Section 4 presents the results and discussion. The last section, Section 5 presents
the summary and conclusion.
Literature Review
Earnings Quality
Research on earnings quality over the past decades has increased dramatically which led
researchers to providing various definitions of earnings quality. These definitions vary
according to the context used in their studies. Dechow, Ge and Schrand (2010) argued that
earnings quality is contextual as it provides different things to different users of financial
reporting. The context can range from equity valuation, debt contracts, managerial
compensation and/ or to internal use. However, they defined earnings quality as a
measurement of firm performance where it reflects current operating performance, an
indicator of future operating performance and accurately annuitizes the company’s intrinsic
value. As stated in Dechow et al. (2010, p.344):
“Higher quality earnings provide more information about the features of a

company’s financial performance that are relevant to a specific decision made
by a specific decision-maker”
Dechow et al (2010) further defined high earnings quality as demonstrating more honesty in
representing features of the company's fundamental profit processes that are relevant for
865
certain decisions and helps predict future earnings, taking into account the stability and
persistence of earnings. Lev (1989) on the other hand, defined earnings quality as decision
usefulness that is used in the context of making decisions on equity evaluation. Dichev,
Graham and Rajgopal (2012) stated that management would often view earnings are of high
quality if they are sustainable with backup from actual cash flows. That is, current earnings
would be considered high quality if they serve asa good guidance to the company’s long term
profit (Santi, Puspitasari & Ghani, 2017; Gideon, Puspitasari, Ghani & Gunardi, 2018;
Nguyen, 2018). Earnings quality has often been underlined by the stakeholders, including the
internal and the external users of financial statement and according to Dechow et al (2010),
represents the features of the fundamental process of earnings and relevance in supporting the
decision making and also the ability to predict future earnings.
An analysis of earnings quality is important to assist investors to judge the congruence

between the company’s earnings and the integrity of the financial statements. It helps the
investors to rely on more reliable numbers. There are several ways of measuring earnings
quality namely, persistence, predictability, value relevance, timeliness, and conservatism
(Francis and Wang, 2008; Dechow et. al, 2010). Another form of measurement is the accrual
which is consistent with what Dechow and Dichev (2002) defined earnings quality, the
magnitude of estimation errors in accruals. In addition, Jones (1991) proposed a way of
measuring earnings quality by using discretionary accrual. According to Wysocki (2004) and
Meuwissen, Moers, Peek and Vanstraelen (2005), the new Jones measurement of earnings
quality is more appropriate to evaluate the international data. Jones (1991) explained that
accrual is the difference between cash flow and net profit (loss). It is described as a very
operational transaction that influences the operational cash flow within on period of
accounting.
This study assesses earnings quality by using accrual quality as a proxy. The reason being
that profits arranged on an accrual basis can actually better show the economic implications
of existing transactions and events. However, in its preparation, profits on accrual-based
accounting cannot be separated from estimates, assumptions, choices of accounting policies
determined by management considerations that contain subjectivity. Management's flexibility
is also feared to be deliberately utilized by management to manipulate earnings (earnings
management) because of certain motives and incentives from the management. Easley and
O'hara (2004) noted that the company's accounting treatment of earnings and disclosure can
affect the company's information environment which then impacts on information risk and
cost of capital.
Accrual could be classified as non-discretionary and discretionary. Non-discretionary accrual

866
is the normal accrual which is caused by the normal condition of a company, and exists in the
company management policy. Meanwhile, the discretionary accrual is the abnormal accrual
which is caused by the ‘management choices’ in electing the accounting method and
estimation in order to achieve certain numbers. Discretionary accrual can distort the quality
of the financial statements. That is, the higher the discretionary accrual, the lower the quality
of financial statements (Dechow et al, 2010). Dechow et al (2010) presented the mechanism
to separate accruals into normal accrual and abnormal accrual. Normal accrual relates to the
fundamental profit whilst abnormal accrual does not relate to the company’s fundamental
profit. Abnormal accruals are generally used as approaches to detect the possibility of
earnings management in the company which will ultimately affect the quality of earnings.
The smaller the abnormal accrual value, the better the quality of earnings. Abnormal accruals
themselves are defined as actual accruals minus expected accruals. Abnormal accruals
themselves are generally influenced by the manipulation of sales, depreciation and estimation
of bad debt expenses.
According to Jones (1991), earnings quality can be measured using discretionary accruals and
this method can detect indications that managers increase profits illegally, with large
discretionary accruals indicating a greater likelihood of opportunistic activities carried out by
managers and higher earnings quality. While Jones measurement model is considered less
practical for the calculation of abnormal accruals with international data, the reasoning being
that a small sample of industries in each country makes this model less reliable and with this
method the company can identify its own abnormal accruals (Wysocki 2004; Meuwisse et al,
2005).
Financial Secrecy
Secrecy is one of the accounting values that most influences the information presented in
corporate financial reporting, especially information disclosure. Besides having a negative
effect on earnings quality, secrecy also has a negative effect on the comparability of financial
statement. Gray (1988) states that secrecy is an accounting value that chooses to maintain
information and does not provide disclosures about a business activity, secrecy has a
tendency to prevent disclosure to external parties. Radebaugh and Gray (1990) also stated
that the preference for secrecy is an activity that is consistent with strong uncertainty
avoidance that prevents disclosure to external parties, thereby reducing competition and
conflicts that can reduce company security. Secrecy or financial secrecy produces some
positive effect, however there are other problems arising from financial confidentiality
resulting in an increasing number of financial crimes such as fraud, tax fraud, money
867
laundering, and the release of a jurisdiction from international financial regulations, which
involve accruing wealth at the expense of the general public.
The Tax Justice Network in 2015 issued a Financial Secrecy Index (FSI), which contained a
secrecy index for 92 countries, where if the secrecy score is above 50 then it is said to be
high, whereas if it is equal to 50 or below it is categorized as low. Countries that have high
secrecy values tend to try and hide or not fully disclose relevant financial information (Geiger
& Smith, 2010). The criteria for selecting these countries are based on the possibility of them
being the centre of the offshore economy and the centre of the world economy. A high degree
of secrecy is deemed less transparent in reporting or disclosing financial information to local
government authorities, and less compliant in adhering to international standards, particularly
in combating financial crime, so high secrecy jurisdiction tends to be the key to placing funds
of illegal origin. In conducting research related to this secrecy index, the Tax Justice Network
uses several methodologies used in calculating this secrecy index namely, banking secrecy,
recorded company ownership, public company accounts, country by country reporting,
efficiency of tax administration, anti-money laundering, automatic information exchange and
international transparency commitments.
According to the Tax Justice Network (2015), the high secrecy jurisdictions tends to hide
information less transparent when reporting and disclosing financial information. Further,
they have less enforcement in preventing the money laundering and corruption causing higher
illegal funds placement. Further, high secrecy also produces high asymmetry information to
the investors which trigger the irrelevant/incorrect decisions. Meanwhile, IASB has been
developing IFRS comprising of accounting methods and approaches to improve financial
information accountability, transparency, and comparability. IFRS has been developed using
the principle-based over the rule-based standard (Gideon et al., 2018). Hence, the use of the
fair value measurement has been massively escalated. There are arguments that the use of the
fair value measurement can increase the accounting information quality since it can reflect a
company’s real economic condition.
International Financial Reporting Standard (IFRS)
IFRS is an international standard issued by the International Accounting Standard Board

(IASB). The International Accounting Standard Board (IASB), formerly called the
International Accounting Standard Committee (IASC), is an independent institution formed
to develop accounting standards. This organisation has the aim of developing and compiling
accounting standards, encouraging the use of high-quality, understandable and comparable
global accounting standards (Choi, Frost, Carol and Meek, 1999; Sadique and Sheikh, 2013;
868
Hussain et al., 2018). These accounting standards are prepared by four major world
organizations, namely the International Accounting Standard Board (IASB), the European
Commission (EC), the International Organisation of Securities Commissions (IOSOC), the
International Federation of Accountants (IFAC). Most of the contents of this standard are
part of the International Accounting Standard (IAS), which later IASB continued to develop
into a new standard known as IFRS. In addition, the fair value used in the IFRS rules has the
advantage that the assets and liability posts held reflect more the actual value at the time of
financial reporting.
One effort to reduce earnings management is to make corrections to accounting standards.
The improvement in accounting standards currently being carried out is with the adoption of
IFRS. The purpose of the IASB establishes international accounting standards is to simplify
various alternative accounting policies that are permissible and can limit management policy
considerations (management's discretion) to earnings manipulation so as to improve earnings
quality and present comparable financial reports in the eyes of the international community.
So far, IFRS has been used in many countries to become international accounting standards,
including Europe, Japan, Hong Kong, Australia, Malaysia and Singapore, Pakistan, Russia,
Turkey, South Africa and GCC countries (Cooperation Council for the Arab States of The
Gulf). To date, IFRS has been used by more than 150 countries and 57% or 85% of them
have required the use of IFRS as a financial reporting standard for domestic companies or
listed companies. Most countries that are members of the G-20 have also adopted IFRS, one
of which is Indonesia.
Before adopting IFRS, accounting used the historical cost basis to measure transactions (Siala
Bouaziz & Jarboui, 2019). Using the historical cost basis, the items in the financial
statements are measured at the cost of the transaction. The costs recorded in the financial
statements in accordance with the transaction will be the basis for reporting the size of a post
for the next period. The use of historical cost has the advantage that it can be easily proven
because the amount is in accordance with the costs recorded during the transaction. However
the disadvantage is that this value does not reflect changes in value, for example when there
is a decline or increase in value in the market due to inflation or deflation and so on. So the
historical cost cannot reflect the value of a post at the time of reporting
Research Methodology
Sample
The public listed companies on the Indonesian Stock Exchange (IDX) under the LQ-45 index
are chosen as the sample study. This study also relies on the public listed companies in the
Bursa Malaysia under the KLCI, and the Singapore Stock Exchange (SSE) under the Straits
869
Times Index. The LQ-45, KLCI, and the Straits Times Index consists of companies that are
listed in the capital market with highest market capitalization. This study only utilised
companies with the consideration that these companies could work as representative samples
for the implementation of IFRS adoption in Indonesia, Malaysia and Singapore. The financial
reporting quality before the IFRS adoption was based on financial statements over a two year
period from 2009-2010. The financial reporting quality after IFRS implementation was
represented by the financial statements over a two year period from 2012 to 2014. This study
excludes financial statements of 2011 with the consideration that the period is the starting
point of IFRS adoption in the countries. Therefore, using the Slovine sample, the total
companies used for this study are 71 companies over a period of six years, resulting in 426
observations.
Variable Measurement
Independent Variables
Secrecy is the first independent variable. Secrecy is proxied by the financial secrecy index
issued in 2015. The financial secrecy index is measured by the methodologies issued by the
Tax Justice Network, the results of which range from 0-100. If the index is above 50 then it is
said to be high, whereas if it is equal to 50 or smaller then it is categorized as low. However,
in its research, the Tax Justice Network did not include scores for Indonesia, so another proxy
was needed to get the secrecy score. Gray (1988) stated that countries with high levels of
uncertainty avoidance and power distance have high secrecy rates. Countries with a high
level of uncertainty avoidance have a tendency to hold or not disclose accounting information
as a whole (full disclosure), this is done to reduce conflicts and competition that will be faced
by companies, while countries with high power distance levels allow accounting information
to be known only by top management, thereby increasing the likelihood of secrecy occurring
within the company. According to Desender, Castro, Leon and Escamilla (2011)
individualism has a significant and negative relationship to earnings management such that
the secrecy score can be obtained by summing the uncertainty avoidance (UA) and power
distance (PD) scores reduced by the score of individualism. Furthermore, from the Hofstede
index, a country with a Hofstede index similar to Indonesia that had similar comparison
criteria was Mexico.
Adoption of IFRS
Adoption of IFRS is proxied by a dummy variable. Dummy variables are variables used to
quantify qualitative variables. The following is the proxy for IFRS Adoption:
870
0 = companies that do not implement IFRS. It was said that it did not implement
IFRS, namely in the year when the company had not carried out IFRS adoption at
all in stages or in full.
1 = a company that implements IFRS, is said to implement IFRS in the year when the
company has carried out IFRS adoption in stages or in full.
Dependent Variable
The dependent variable in this study is earnings quality. The measurements using
Discretionary Accruals by Jones (1995) with large Discretionary Accruals indicate a greater
likelihood of opportunistic activities carried out by managers and smaller earnings quality.
The Jones 1991 measurement model is considered less practical for the calculation of
abnormal accruals with international data because a small sample of industries in each
country makes this model less reliable and with this method the company can identify its own
abnormal accruals (Wysocki 2004; Meuwisse et al, 2005).
Control Variables
Investor Protection
This study uses a tool developed by Houqe et al (2011) comprising five measuring
instruments that can be used to measure protection for investors. The five instruments
included independence of the supervisory board, protection of minority shareholders, capital
market law, irregular payment and bribery scores, and financing activities originating from
the local equity market score (Zandi and Haseeb, 2019). For investor protection, the index
issued by the World Bank is proxied, with scale 1-10.
Total Sales
The size of the company can also be marked by the total sales made by the company.
Previous research has found that large scale companies have a tendency to have lower
accruals than companies with smaller scales and the value of total sales variables is obtained
in the following ways:
Total Sales = Total Sales

871
Leverage
Leverage is the ability of a company to use assets or funds that have a fixed burden to
increase the level of income for the owner of the company. The leverage variable is used as a
control variable because leverage is one mechanism that can be used to reduce opportunistic
management. Leverage is calculated by comparing the total debt that the company has with
total assets in carrying the risk.
𝑇𝑜𝑡𝑎𝑙 𝐷𝑒𝑏𝑡
Debt Ratio = 𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡
Sales Growth
Accrual development is also followed by developments in the company's operating activities,

one of which is sales growth. The sales growth ratio is used as a proxy for measuring annual
sales growth variables. Sales growth has a positive relationship to accruals, the higher the
sales growth of the company, the higher the company's accruals. Earnings quality is also said
to be low if sales growth is lower than growth on net operating assets (Ohlson, 2015). Sales
growth can also predict the possibility of corporate profits in the future. Earnings
management is dominated by accrual manipulation functions. The greater the total accrual as
a percentage of assets, the greater the likelihood that the quality of earnings is low, although
there is also the possibility that accruals are estimates of the business that will occur in the
future. It is difficult to determine which are manipulating accruals or not, especially for
companies that do have high accruals. Sales growth was calculated using the following
formula:
(𝑇𝑜𝑡𝑎𝑙 𝑠𝑎𝑙𝑒𝑠 𝑦𝑒𝑎𝑟 𝑡 − 𝑡𝑜𝑡𝑎𝑙 𝑠𝑎𝑙𝑒𝑠 𝑦𝑒𝑎𝑟 𝑡 − 1)

𝑆𝑎𝑙𝑒𝑠 𝐺𝑟𝑜𝑤𝑡ℎ =
𝑡𝑜𝑡𝑎𝑙 𝑠𝑎𝑙𝑒𝑠 𝑦𝑒𝑎𝑟 𝑡 − 1
Property, Plant and Equipment (PPE)
Gross increase in Plant, Property and Equipment (PPE) is considered to have an effect
on the increase in the number of accruals for companies that also affect the increase in
abnormal accruals of companies. While growth from gross PPE was obtained by the
following formula:
872
(𝐺𝑟𝑜𝑠𝑠 𝑃𝑃𝐸 𝑡−𝐺𝑟𝑜𝑠𝑠 𝑃𝑃𝐸 𝑡−1)

Gross PPE Growth 𝑅𝑎𝑡𝑒 = 𝐺𝑟𝑜𝑠𝑠 𝑃𝑃𝐸 𝑡−1
Operating Cash Flow
The amount of cash flows arising from operating activities is an indicator that determines
whether from the company's operations it can generate sufficient cash flow to repay loans,
maintain the company's operating capability, and make new investments without relying on
external funding sources. Cash flows from operating activities were mainly derived from the
company's main income activities. The value of operating cash flow was obtained from net
income less than the total accrual of the year.
Loss
This variable is a dummy variable that is worth 1 for companies that have a loss in the x
period and are worth 0 for the opposite. The loss variable (LOSS) can be indicated through
negative net income. In previous literature referenced, this loss variable becomes a control
variable in testing earnings quality because it indicates there is a potential difference between
a company that experiences a loss (negative net income) and a company that experiences
profit (profit) in determining the amount of accruals (Choi et al., 2010) This variable is
expected to be negatively related to earnings quality.
Fixed Effect
Fixed effect model using fixed parameters for each individual as individual heterogeneity.
Industrial type is used as a control variable because the type of industry is thought to be able
to strengthen the relationship between profitability, leverage and inventory. Industrial type
and year are used as control variables and are proxied by dummy variables (Hussain et al.,
2018).
The Earnings Quality Model
This study utilises the modified Jones model to determine the discretionary accrual. This
study calculated the total accruals using the cash flow approach:
TACC it = NIit - CFO it (1)
Where:
873
TACC it = Total accrual of company i for period t

NIit = Net income of company i for period t
CFO it = Operating cash flow of company i for period t
To determine the coefficient of the accrual regression model, this study used discretionary
accrual as the difference between the total accrual (TACC) with the non-discretionary accrual
(NDACC). In order to find the non-discretionary accrual, this study performed the following
regression model:
Where:
TACC it = Total accrual of company i for period t
TA it-1 = Total assets of company i for period t
∆REV it = The revenue changing of company i from period t-1 to period t
PPE it = Property, plant, equipment of company i for period t
∆REC it = The net receivable changing of company i from period t-1 to period t
e = Error
The regression model from equation (2) produced the coefficients β1, β2, and β3.
These coefficients were then utilised to predict the nondiscretionary accrual through the
following equation:
Where:
NDACC it = Nondiscretionary accrual of company i for period t
Data Analysis
The objective of this study to find empirical findings on the earnings quality in relation to
financial secrecy. To accomplish this objective, this study performed a data panel regression
analysis. The model used in this study is as follows:
Ab_Accrualsit = αₒ + αıSEC + α2IFRS + α3SEC*IFRS + α4INV + α5LN_SALESit +

α6F_LEVit + α7S_GWTHit + α8ΔPPEit + α9CFOit + α10LAGLOSSit + fixed effects
874
Where:
Ab_Accruals = Signed abnormal accruals company x in year t

SEC = Financial secrecy Index produced by Tax Justice Network
IFRS = is a dummy variable that gives a value of 1 when adopting IFRS in
that year and 0 if the opposite
INV = Investor protection being measured based on five components:
(i) JUD - Judicial Independence score from World Economic Forum (2012)
(ii) MIN - Protection to the minority shareholders interest score form World
Economic Forum (2012)
(iii) RSE - Regulation of securities laws score from World Economic Forum
(2012)
(iv) IIPB - Irregular payment and bribery scores from World Economic
Forum (2012)
(v) FTEM - Financing of local equity market scores from World Economic
Forum (2012)
LN_Sales = Company size measured by natural logarithm of the total sales of

company x in year t
F_LEV = Leverage measured by total debt divided by shareholders equity of
company x in year t
S_GWTH = Sales growth rate measured by total sales year t less sales in year t-
1 divided by sales year t-1
ΔPPE = Growth rate of gross PPE in t-1 compared to the gross PPE in year t
CFO = Operating cash flows of company x in t with total asset.
LOSS = dummy variable 1 if the company in year t reporting negative
income before extraordinary item and 0 if otherwise.
While the fixed effects are:
Industry dummy = industry sector membership dummy variable

Year dummy = year dummy variable
Country dummy = Country dummy variable.
Results and Discussion

Descriptive Analysis
875
This section presents the results of the descriptive statistics before and after IFRS adoption in
Indonesia. Table 1 below shows the mean score of financial secrecy and the earnings quality
before and after IFRS adoption in Indonesia. The results show that an average quality of
earnings before IFRS adoption in Indonesia is -0.027 with a standard deviation of 0.084
which shows the majority of companies have good earnings quality with the lowest
discretionary accrual ratio reaching -0.225 owned by Astra International in 2010, while the
highest reaches 0.187 owned by Antam in 2010. Furthermore, the average quality of earnings
after IFRS adoption was -0.005 with a standard deviation of 0.086 which showed that the
majority of companies had better earnings quality with discretionary accrual ratio lowest
reaching -0.257 owned by Astra International in 2012, while the highest value reached 0.266
owned by Antam in 2013.
Table 1: Descriptive Analysis for Indonesia

Before IFRS adoption After IFRS adoption
DACC DACC
SEC
(Earnings (Earnings SEC (Secrecy)
(Secrecy)
Quality) Quality)
Observations 38 38 76 76
Cross Sections 19 19 19 19
Mean -0,027 47,000 -0,005 46,340

STDV 0,084 0,000 0,086 0,582
Maximum 0,187 47,000 0,266 47,000
Minimum -0,225 47,000 -0,257 45,000
The average secrecy index before IFRS adoption in Indonesia is 47,000 with a standard
deviation of 0,000. Furthermore, the average secrecy index after IFRS adoption is 46.340
with a standard deviation of 0.582 and there is a decrease in the value at the secrecy level.
Table 2 below shows the mean score of the financial secrecy and the earnings quality before
and after IFRS adoption in Malaysia. The results show that the average earnings quality
before IFRS adoption in Malaysia was 0.046 with a standard deviation of 0.063 which
showed that the majority of companies had poor earnings quality with the lowest
discretionary accrual ratio reaching -0.202 owned by DIGI in 2010, while the highest reached
0.159 owned by Maxis in 2009. Furthermore, the average quality of earnings after IFRS
adoption is -0.003 with a standard deviation of 0.057 which shows the majority of companies
876
have better earnings quality with discretionary accrual ratio lowest reaching -0.183 owned by
DIGI in 2012, while the most value the high reached 0.180 was owned by Hong Leong Bank
in 2012. Table 2 also shows the average secrecy index before IFRS adoption in Malaysia is
67,000 with a standard deviation of 0,000. Furthermore, the average secrecy index after IFRS
adoption is 64.750 with a standard deviation of 0.435 and there was a decrease in the value at
the secrecy level.
Table 2: Descriptive Analysis for Malaysia

DACC DACC
SEC
(Earnings (Earnings SEC (Secrecy)
(Secrecy)
Quality) Quality)
Mean 0,046 67,000 -0,003 64,750

STDV 0,063 0,000 0,057 0,435
Maximum 0,159 67,000 0,180 65,000
Minimum -0,202 67,000 -0,183 64,000
Table 3 below shows the mean score of financial secrecy and the earnings quality before and
after IFRS adoption in Singapore. The results show that on the average, quality of earnings
before IFRS adoption in Singapore is 0.020 with a standard deviation of 0.093 which shows
the majority of companies have poor earnings quality with the lowest discretionary accrual
ratio reaching -0.206 owned by Starhub in 2010, while the highest reaches 0.260, owned by
Hong Kong Land in 2010. Furthermore, the average quality of earnings after IFRS adoption
was 0.012 with a standard deviation of 0.061 shows the majority of companies have better
earnings quality with discretionary accrual ratio lowest reaching -0.156 owned by Starhub in
2012, while the highest reached 0.211 owned by Yangziang in 2013. The average secrecy
index before IFRS adoption in Singapore was 79,000 with a standard deviation of 0,000.
Furthermore, the average secrecy index after IFRS adoption is 69.750 with a standard
deviation of 0.4354 and there is a significant decrease in value at the secrecy level.
Table 3: Descriptive Analysis for Singapore

DACC SEC DACC SEC (Secrecy)
877
(Earnings (Secrecy) (Earnings

Quality) Quality)
Mean 0,020 79,000 0,012 69,750

STDV 0,093 0,000 0,061 0,4354
Maximum 0,260 79,000 0,211 70,000
Minimum -0,206 79,000 -0,156 69,000
Data Analyses and Hypotheses Testing
The statistical method used to test the conceptual hypothesis proposed is panel data
regression analysis. In conducting panel data regression analysis, there are three types of
models that can be used, namely the common effect (PLS), fixed effect (FEM) and random
effect (REM) models. To determine the regression model used, first a chow test and a
Hausman test were performed. The chow test was used to compare whether panel data is
better estimated using fixed effects or a common effect. The hypotheses tested in the chow
test are as follows:
Ho: Common Effect Model

Ha: Fixed Effect Model
If the probability of cross-section F produced is less than 0.05 (α) then the test decision is to
reject Ho and accept Ha or in other words, panel data is better estimated using fixed effects.
The summary of the test results is presented in Table 4 below.
Table 4: Chow Test

Cross Section F Prob. F Result
Chow Test 4,285 0,000 Fixed Effect
Based on the results of the chow test above, it can be decided that panel data is better
estimated using fixed effects because the probability value generated by the two regression
models is 0,000 and is much smaller than 0.05. Since the chosen model is a fixed effect, then
a Hausman test is performed to determine the fixed effect with random effects and reveal
whether the fixed effect model is better to use than the random effect model. The hypothesis
formed in the Hausman test is as follows:
878
Ho: Random Effect Model

Ha: Fixed Effect Model
If the resulting chi-square is smaller than 0.05, this indicates that the decision from the
results of the Hausman test is to reject Ho and accept Ha or in other words, panel data is
better estimated using fixed effects. The summary of the test results is presented in Table 5
below.
Table 5: Hausman Test

Chi-Sq (χ2) Prob. Chi-Sq Result
Hausman Test 38,738 0,000 Fixed Effect
Based on the results presented, it can be concluded that in this study both panel data models
are better estimated using fixed effects, because the test results show prob. χ2 which is
smaller than 0.05. The calculation results from panel data regression using the EViews 8
program are presented in Table 6 below.
Table 6: Results of Multiple Regression Analysis

Dependent Variable: DACC?
Method: Pooled Least Squares
Sample: 2009 2014
Included observations: 6
Cross-sections included: 88
Total pool (balanced) observations: 426
Variable Coefficient Std. Error t-Statistic Prob.
C -1.191975 0.311534 -3.826149 0.0001
IFRS? -0.017369 0.007120 -2.439334 0.0076
SEC? 0.001442 0.000836 1.723985 0.0427
IFRS*SEC? -0.006523 0.003044 -2.142755 0.0164
INV? 0.029283 0.021749 1.346417 0.0895
SALES? 0.039424 0.011118 3.546118 0.0002
LEV? -0.019996 0.007922 -2.524127 0.0060
CFO? -3.26E-15 9.00E-16 -3.625275 0.0002
LOSS? -0.086448 0.021601 -4.001999 0.0001
Effects Specification
Cross-section fixed (dummy variables)
R-squared 0.621429 Mean dependent var -0.005018
879
Adjusted R-squared 0.536031 S.D. dependent var 0.076853

S.E. of regression 0.052348 Akaike info criterion -2.895887
Sum squared resid 1.178355 Schwarz criterion -2.103517
Log likelihood 862.5143 Hannan-Quinn criter. -2.585691
F-statistic 7.276813 Durbin-Watson stat 2.076799
Prob(F-statistic) 0.000000
Based on the results of regression calculations, it is believed that the panel data regression
model that is formed is the fixed effect model or in other words the difference in the
characteristics of each company is considered to have an effect on the quality of earnings.
The regression equation formed is as follows:
DACC = -1,191 – 0,017 IFRS + 0,001 SEC – 0,006 IFRS*SEC + 0,029 INV + 0,039 SALES
– 0,019 LEV + 0,003 GROWTH + 0,0001 PPE – 3,26E-15 CFO – 0,086 LOSS
Table 7: Comparative Results

Indonesia Singapore Malaysia
β Sig β Sig β Sig
(Constant) -0,904 0,0000 -1,377 0,000 -0,239 0,000
IFRS -0,067 0,0297 -0,025 0,008 -0,011 0,010
Secrecy 0,003 0,0068 0,004 0,003 0,005 0,045
IFRS_Secrecy -0,026 0,022 -0,013 0,002 -0,013 0,029
Investor -0,011 0,0383 0,227 0,037 0,034 0,045
Sales 0,029 0,0000 0,004 0,022 -0,012 0,004
Leverage -0,028 0,0281 0,002 0,019 -0,022 0,002
CFO -4E-15 0,0000 -8E-12 0,000 -6E-12 0,000
Loss -0,061 0,0438 -0,087 0,028 -0,023 0,047
The simultaneous hypothesis formulations to be tested are as follows:

Ho: βi = 0 Simultaneously IFRS, secrecy implementation, interaction between IFRS and
secrecy and investors, total sales, sales growth, property, plant & equipment, operating cash
flow and losses as control variables did not significantly influence earnings quality.
Ha: βi ≠ 0 Simultaneously IFRS, secrecy, interaction between IFRS and secrecy and
investors, total sales, sales growth, property, plant & equipment, operating cash flows and
losses as control variables have a significant effect on earnings quality.
The significance level (α) used is 5%.
880
The summary of the test results is presented in Table 8 below. The results show that with a
probability value F that is smaller than 0.05 and with a confidence level of 95%, it was
decided to reject Ho and accept Ha. These results indicate that the implementation of IFRS,
secrecy, interaction between IFRS and secrecy and investors, total sales, growth in sales,
property, plant & equipment, operating cash flows and losses as control variables
simultaneously have a significant effect on earnings quality.
Table 8: Effect of IFRS, Secrecy and Investor Implementation, Total Sales, Growth in Sales,
Property, Plant & Equipment, Operating Cash Flow and Losses as Control Variables on
Profit Quality
F hitung F tabel Prob.F Α Result Summary
7,276 1,849 0,000 0,05 Ho rejected Significant
The T-Test (Partial) was used to show how far the influence of one independent variable can
influence the dependent variable. Decision making in this t test by comparing the probability
value (Prob.T) with the level of significance was used to test H1, testing the effect of IFRS
implementation on profit quality. The results are shown in Table 9 below. Table 9 shows that
with a probability value F that is smaller than 0.05 and with a confidence level of 95%, it can
be decided to reject Ho and accept Ha, which means that the implementation of IFRS has a
significant negative effect on earnings quality.
Table 9: Effect of IFRS Implementation on Profit Quality

Model t hitung t tabel Prob. T Α Result Summary
X1 → Y -2,439 -1,648 0,007 0,05 Ho rejected Significant
Table 10 below shows that with a probability value F that is smaller than 0.05 and with a
confidence level of 95%, it can be decided to reject Ho and accept Ha, which means that
secrecy has a significant positive effect on earnings quality.
Table 10: Effect of Secrecy on Earnings Quality

Model t hitung t tabel Prob. t Α Result Summary
X2 → Y 1,723 1,648 0,042 0,05 Ho rejected Significant
This study also examines whether there is an effect of interaction between IFRS
implementation and secrecy on earnings quality. Table 11 below shows the results. From the
table above, with F probability values smaller than 0.05 and with a confidence level of 95%,
881
it can be decided to reject Ho and accept Ha, which means that the interaction between IFRS
implementation and secrecy has a significant negative effect on earnings quality. These
results indicate that the implementation of IFRS is able to weaken the influence of secrecy on
earnings quality which later results in better quality of earnings or in other words IFRS is a
moderating variable that weakens the effect of secrecy on earnings quality.
Table 11: Effect of Interaction between IFRS Implementation and Secrecy on Earnings
Quality
Model t hitung t tabel Prob. T Α Result Summary
X 1 *X 2 →
-2,142 1,648 0,016 0,05 Ho rejected Significant
Y
Overall Determination Coefficient
The coefficient of determination was used to determine the contribution of the influence
given by the independent variable on the dependent variable. In Tables 4-7 above, it is shown
that the Adjusted R-squared value obtained is 0.536. These results indicate that the
implementation of IFRS, secrecy, interactions between IFRS and secrecy and investors, total
sales, growth in sales, property, plant & equipment, operating cash flows and losses as a
simultaneous control contributed 53.6% to earnings quality, while the remaining 46.4% is a
large contribution to the influence given by other factors not examined.
From Table 12 below, it can be seen that the coefficient of determination for Indonesia
obtained is 37.79%. This shows that IFRS, Secrecy, IFRS & Secrecy Interaction and all
control variables contributed to the quality of earnings by 37.79%, while the remaining
63.21% was contributed by other variables not examined. The coefficient of determination
for Singapore obtained is 60.37%. This shows that IFRS, Secrecy, IFRS and Secrecy
Interaction and all control variables contribute to the quality of earnings by 60.37% whilst the
remaining 39.63% is a contribution from other variables not examined. The coefficient of
determination for Malaysia obtained is 62.75%. This shows that IFRS, Secrecy, IFRS &
Secrecy Interaction and all control variables contribute to the quality of earnings of 62.75%
whilst the remaining 37.25% is a contribution from other variables not examined.
According to Leonidas (2011), a decrease in discretionary accrual occurs because initially the
quality of financial reporting before adopting IFRS is weak so that when using IFRS which is
considered a good quality standard, earnings management becomes lower and earnings
882
quality improves.
Table 12: Coefficient of Determination per Country

Country Adjusted R Square %
Indonesia 0,3779 37,79%
Singapore 0,6037 60,37%
Malaysia 0,6275 62,75%
Summary and Conclusion
This study aimed to determine the effect of IFRS adoption and financial secrecy on earnings
quality. The control variables used are investor protection, total sales, leverage, sales growth,
operating cash flow, gross PPE, loss. Research samples are companies listed on the
Indonesia, Singapore and Malaysia Stock Exchanges with 2009-2014 as the target
observation years. The sample used involved 71 companies so that the sample total was 426
observations. Sampling was based on Slovine sampling techniques. Data processing was
completed using multiple linear regression equations, the main model being Eviews 8.0. The
data analyses performed using multiple linear regression show that simultaneous adoption of
IFRS and secrecy have a significant effect on earnings quality. While in part the adoption of
IFRS and secrecy have a significant effect on earnings quality and control variables such as
investor protection, the total sales, leverage, operating cash flow and loss simultaneously
have a significant effect, sales growth and gross PPE do not significantly influence earnings
quality.
This study provides evidence that even though each country has implemented IFRS, it does
not indicate that the level of earnings quality being reported would be the same. The earnings
quality would be induced by investigation of the respective cultural factors, especially the
concept of secrecy. Further, stakeholders need to be cautioned when making financial
statements comparisons between companies in different countries, since fundamental
dissimilarity could still occur.
Acknowledgment
We wish to thank Faculty of Accountancy and the Institute of Research Management and
Innovation of Universiti Teknologi for their support and funding.
REFERENCES
883
Ball, R., Robin, A. & Wu, J. S. (2003). Incentives versus standards: properties of accounting
income in four East Asian countries. Journal of accounting and economics, 36(1),
235-270.
Barth, M., Landsman, W & Lang, M. (2008). International Accounting Standards and
accounting quality. Journal of Accounting Research, 46(3), 467-498.
Cohen, J. R., Krishnamoorthy, G., & Wright, A. (2004). The corporate governance mosaic
and financial reporting quality. Journal of accounting literature, 87-152.
Choi, F, Frost, D.S; Carol, A. & Meek, G.K (1999) “International Accounting”, Prentice
Hall, Upper Saddle River, NY.
Daske, H & Gunther, G. (2006). International Financial Reporting Standards and Experts
Perceptions of Disclosure Quality. Journal of Accounting Finance and Business
Studies. New York: Willey Online Library.
Dechow P & Dichev, I (2002), The quality of accruals and earnings: The role of accrual
estimation errors, The Accounting Review, 77(Supplement), 35-59
Dechow, P., Ge, W., & Schrand, C. (2010). Understanding earnings quality: A review of the
proxies, their determinants and their consequences. Journal of Accounting and
Economics, 50(2), 344-401.
Deloitte. (2014). Use of IFRS by jurisdiction. Retrieved from

http://www.iasplus.com/Plone/en/resources/use-of-ifrs.
Desender, K.A., Castro, C. E., Leon, D & Escamilla, S.A. (2011). Earnings management and
cultural values. American Journal of Economics and Sociology, 70(3), 639-670.
Dichev, I; Graham, J & Rajgopal, S (2012), Earnings quality: Evidence from the field,
Working Paper,
https://faculty.fuqua.duke.edu/~charvey/Research/Working_Papers/W108_Earnings_quality_
evidence.pdf
Easley, D & O’Hara, M (2004), Information and the cost of capital, The Journal of Finance,
IIX (4), 1153-1583
884
Francis, J. R., & Wang, D. (2008). The joint effect of investor protection and Big 4 audits on
earnings quality around the world. Contemporary accounting research, 25(1), 157-
191.
Gideon, B.D; Puspitasari, E; Ghani, E.K & Gunardi, A (2018), Earnings quality: Does
principles standards versus rules standards matter? Journal of Applied Economic
Sciences, XIII (2), 586-594
Gray, S. J. (1988). Towards a theory of cultural influence on the development of accounting

systems internationally. Abacus, 24(1), pp.1-15.
Houqe, N., Monem, R., Tareq, M. & van Zijl, T. (2015). Secrecy and mandatory IFRS
adoption on earnings quality. Working paper, Victoria University of Wellington,
Hussain, H.I., Shamsudin, M.F., Anwar, N.A.M., Salem. M.A. & Jabarullah, N.H. (2018).
The Impact of Shari’ah Compliance on the Adjustment to Target Debt Maturity of
Malaysian Firms, European Research Studies Journal, 21 (2), 48 – 61.
Jeanjean,T.,Stolowy,H. ,2008.Do accounting standards matter?An exploratory analysis of

earnings management before and after IFRS adoption. Journal of Accounting and
Public Policy, 27(1), 480-494.
Jones, J. (1991), Earnings management during import relief investigations. Journal of

Accounting Research 29, 193–228.
Lev, B. (1989), On the usefulness of earnings and earnings: Lessons and directions from two
decades of empirical research, Journal of Accounting Research, 27, 153-192.
Meuwissen, R., Moers, F., Peek, E. & Vanstraelen, A. (2005). An evaluation of abnormal
accruals measurement models in an international context. Working Paper, University
of Maastricht and University of Antwerp.
Morais, A.I & Curto, D. (2008). Accounting Quality and the Adoption of IASB Standards:
Portuguese Evidence. Revisa Contabilidade & Financas. Sao Paolo: Scielo.
Nicholas, C.D & Wahlen, J,M (2004), How do earnings numbers relate to stock returns? A
review of classic accounting research with updated evidence. Accounting Horizons,
18(4), 263-286
885
Nguyen, A. T. (2018). The Relationship among Economic Growth, Trade, Unemployment,

and Inflation in South Asia: A Vector Autoregressive Model Approach. Asian Journal
of Economics and Empirical Research, 5(2), 165-172.
Nilwala, N., Gunawardana, K., & Fernando, R. L. S. (2017). Scale for Measuring
Transformational Leadership in Public Sector Organizations in Sri Lanka: With
Special Reference to Ministries of Western Provincial Council.
Ohlson, J. A. (1995). Earnings, book values, and dividends in equity valuation.

Contemporary Accounting Research, 11(2), 661-687.
Pacter, P. (2015). The Global Reach of IFRS is Expanding.

https://www.iasplus.com/en/news/2014/01/pacter-ifrs-adoption
Procházka, D. (2017). The Unintended Consequences of Accounting Harmonization in a

Transition Country: A Case Study of Management Accounting of Private Czech
Companies. Contemporary Economics, 11(4), 443-457.
Radebaugh L. H. and Gray S. J. (1997). International Accounting and Multinational

Enterprises. IV ed., New York: Wiley
Sadique, M. S., & Sheikh, M. A. R. (2013). Investor reaction to strategic emphasis on

earnings numbers: An empirical study. Contemporary Economics, 7(3), 51-64.
Santi, M; Puspitasari, E & Ghani, E.K (2017), Analysis of unconditional conservatism and
earnings quality on financial reporting practices in Indonesia upon IFRS convergence,
Management and Accounting Review, 16(1), 75-87
Schipper, K., & Vincent, L. (2003). Earnings quality. Accounting Horizons, 17, 97-110.
Siala Bouaziz, S., & Jarboui, A. (2019). The moderating effect of audit quality on the relation
between shareholder activism and earnings management: Evidence from France.
Contemporary Economics, 13(1), 63-78.
Wysocki, P. 2004. Discussion of ultimate ownership, income management, and legal and
extra-legal institutions. Journal of Accounting Research 42, 463–474.
Zandi, G., & Haseeb, M. (2019). The role of trade liberalization in carbon dioxide emission:
Evidence from heterogeneous panel estimations. International Journal of Financial
Research, 10 (5), 228-240.
886

The Effect of Financial Secrecy and IFRS Adoption On Earnings Quality: A Comparative Study Between Indonesia, Malaysia and Singapore

Uploaded by

Copyright:

Available Formats

The Effect of Financial Secrecy and IFRS Adoption On Earnings Quality: A Comparative Study Between Indonesia, Malaysia and Singapore

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

The Effect of Financial Secrecy and IFRS Adoption On Earnings Quality: A Comparative Study Between Indonesia, Malaysia and Singapore

Uploaded by

Copyright:

Available Formats

International Journal of Innovation, Creativity and Change. www.ijicc.

The Effect of Financial Secrecy and

Key words: earnings quality, discretionary accrual, IFRS adoption, secrecy,

“Higher quality earnings provide more information about the features of a

An analysis of earnings quality is important to assist investors to judge the congruence

Accrual could be classified as non-discretionary and discretionary. Non-discretionary accrual

International Financial Reporting Standard (IFRS)

IFRS is an international standard issued by the International Accounting Standard Board

Total Sales = Total Sales

Accrual development is also followed by developments in the company's operating activities,

(𝑇𝑜𝑡𝑎𝑙 𝑠𝑎𝑙𝑒𝑠 𝑦𝑒𝑎𝑟 𝑡 − 𝑡𝑜𝑡𝑎𝑙 𝑠𝑎𝑙𝑒𝑠 𝑦𝑒𝑎𝑟 𝑡 − 1)

Property, Plant and Equipment (PPE)

(𝐺𝑟𝑜𝑠𝑠 𝑃𝑃𝐸 𝑡−𝐺𝑟𝑜𝑠𝑠 𝑃𝑃𝐸 𝑡−1)

Operating Cash Flow

The Earnings Quality Model

TACC it = NIit - CFO it (1)

TACC it = Total accrual of company i for period t

Ab_Accrualsit = αₒ + αıSEC + α2IFRS + α3SEC*IFRS + α4INV + α5LN_SALESit +

Ab_Accruals = Signed abnormal accruals company x in year t

LN_Sales = Company size measured by natural logarithm of the total sales of

While the fixed effects are:

Industry dummy = industry sector membership dummy variable

Results and Discussion

Table 1: Descriptive Analysis for Indonesia

Mean -0,027 47,000 -0,005 46,340

Table 2: Descriptive Analysis for Malaysia

Mean 0,046 67,000 -0,003 64,750

Table 3: Descriptive Analysis for Singapore

(Earnings (Secrecy) (Earnings

Mean 0,020 79,000 0,012 69,750

Data Analyses and Hypotheses Testing

Ho: Common Effect Model

Table 4: Chow Test

Ho: Random Effect Model

Table 5: Hausman Test

Table 6: Results of Multiple Regression Analysis

Adjusted R-squared 0.536031 S.D. dependent var 0.076853

Table 7: Comparative Results

The simultaneous hypothesis formulations to be tested are as follows:

Table 9: Effect of IFRS Implementation on Profit Quality

Table 10: Effect of Secrecy on Earnings Quality

Overall Determination Coefficient

Table 12: Coefficient of Determination per Country

Summary and Conclusion

Deloitte. (2014). Use of IFRS by jurisdiction. Retrieved from

Gray, S. J. (1988). Towards a theory of cultural influence on the development of accounting

Jeanjean,T.,Stolowy,H. ,2008.Do accounting standards matter?An exploratory analysis of

Jones, J. (1991), Earnings management during import relief investigations. Journal of

Nguyen, A. T. (2018). The Relationship among Economic Growth, Trade, Unemployment,

Ohlson, J. A. (1995). Earnings, book values, and dividends in equity valuation.

Pacter, P. (2015). The Global Reach of IFRS is Expanding.

Procházka, D. (2017). The Unintended Consequences of Accounting Harmonization in a

Radebaugh L. H. and Gray S. J. (1997). International Accounting and Multinational

Sadique, M. S., & Sheikh, M. A. R. (2013). Investor reaction to strategic emphasis on

You might also like