Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
Proceedings of the Second Asia-Pacific Conference on Global Business, Economics, Finance
and Social Sciences (AP15Vietnam Conference) ISBN: 978-1-63415-833-6
Danang-Vietnam, 10-12 July, 2015 Paper ID: V535
1
www.globalbizresearch.org
The Effects of Corporate Governance on Idiosyncratic Risk:
Evidence from Taiwan Financial Institutions
Tsun-Jen Wei,
National Kaohsiung First University of Science and Technology, Taiwan.
Email: [email protected]
Hsien-Ming Chen,
Department of Finance,
Chang Jung Christian University, Taiwan.
Email: [email protected]
Chu-Hsiung Lin,
Department of Finance,
National Kaohsiung First University of Science and Technology, Taiwan.
Email: [email protected]
Jui-Heng Kang,
National Kaohsiung First University of Science and Technology, Taiwan.
Email: [email protected]
____________________________________________________________________
Abstract
We use the data of Taiwanese financial institutions from 2006:Q1 to 2012:Q4 to examine the
effects of corporate governance mechanisms on idiosyncratic risk. Our results show that the
firms with better corporate governance mechanisms (including more independent board,
better transparency) tend to have a lower idiosyncratic risk. However, firms with higher
foreign ownership appear to have a higher idiosyncratic risk.
___________________________________________________________________________.
Key Words: dynamic panel regression, corporate governance, idiosyncratic Risk.
Proceedings of the Second Asia-Pacific Conference on Global Business, Economics, Finance
and Social Sciences (AP15Vietnam Conference) ISBN: 978-1-63415-833-6
Danang-Vietnam, 10-12 July, 2015 Paper ID: V535
2
www.globalbizresearch.org
1. Introduction
The financial industry is an indicator of national economic development and plays an
important role in economic activities. If the financial industry is mismanagement, financial
institutions will lose financial intermediation functions, and affects the development of other
industries. In addition, if a financial crisis occurs, the crisis would seriously affect the
financial order and economic development. Therefore, some series of financial industry
problems, like the Iceland financial crisis and the Lehman Brothers bankruptcy, raised the
global financial crisis. Thus, in the academic field, find out the mechanisms of avoiding
malpractice and reducing risk of financial industry is critical.
Since the opening to establishing private banks in Taiwan in 1991, Taiwan has adopted a
series of financial reforms, which have substantially increased the number of financial
institutions in Taiwan. Consequently, harsh competition has occurred between financial
institutions, and credit quality have deteriorated, bank profits have decreased, and
non-performing loan ratios have increased, thereby increasing the risk and damaging the
rights of stakeholders. Therefore, how to decrease the risk and avoid financial crisis of
Taiwan is an important issue.
BASEL III suggested to strengthened corporate governance to prevent the risk occurring
from the financial industry. Besides, previous studies also indicate that corporate governance
serve as a type of a mechanism, protect minority shareholders and stakeholders, and enhance
the wealth of shareholders. Lin, et al. (2010) specified that through the design of the corporate
governance mechanism could reduce the agency problems and decrease idiosyncratic risk.
Firms with better corporate governance mechanisms have fewer agency problems. The
idiosyncratic risk of the firm and capital costs would be reduced, thereby enhancing corporate
performance and shareholder wealth. Hence, if financial institutes establish better corporate
governance mechanisms can reduce the risk to improve the financial industry environment
and to avoid malpractice of the financial industry.
However, the literatures regarding the effects of corporate governance quality on the risk
of financial institutions is lack. Furthermore, after reviewing the literature, numerous studies
have focused on exploring the relationship between partial corporate governance mechanisms
and firm performance. Few studies have explained the relationship between corporate
governance and risk. Thus, to make up the gap in the literatures, this study provides direct
empirical evidences of the effects of corporate governance quality on risk. This paper follows
Proceedings of the Second Asia-Pacific Conference on Global Business, Economics, Finance
and Social Sciences (AP15Vietnam Conference) ISBN: 978-1-63415-833-6
Danang-Vietnam, 10-12 July, 2015 Paper ID: V535
3
www.globalbizresearch.org
Lin et al. (2010) to use idiosyncratic risk as the proxy for the level of risk in financial industry.
The idiosyncratic risk represents the risk link with how the financial institutes operate their
own business and systems.
In this study, the financial holding industry, banking industry, and securities industry in
Taiwan were the research subjects. Empirical evidence was used to analyze the relationship
between financial corporate governance mechanisms and idiosyncratic risk. Flannery and
Hankins (2013) indicated that dynamic panel data regression has become increasingly vital in
the corporate finance field. In addition, if explained variables of lag periods are included in
independent variables, dynamic panel data regression must be used to avoid biased parameter
estimates. Thus, this paper modified the empirical model by Lin et al. (2010) and used
dynamic panel data regression in this study. This study also referred to Arellano and Bond
(1991) and conducted generalized method of moments (GMM) regression to estimate the
regression parameters. Furthermore, the Sargan test was used to examine the effectiveness of
the instrumental variables adopted by the dynamic panel data regression.
This study explored the effects of corporate governance quality on idiosyncratic risk.
Corporate governance involves ownership structure, board structure, executive incentive, and
information disclosure. The results show that when the high proportion of independent
directors and supervisors in the board associated with lower idiosyncratic risk. Moreover,
higher information transparency indicates less idiosyncratic risk. The main contribution of
this study is the comprehensive investigation on the effects of the corporate governance
mechanism on idiosyncratic risk in financial industry.
The literature mainly focuses on exploring the effects of the corporate governance
mechanism on operating performance or conceptually explains the influence of parts of the
corporate governance mechanism on idiosyncratic risk. These studies have failed to examine
the effects of the entire corporate governance mechanism on idiosyncratic risk. Only Lin et al.
(2010) used general industry as the research subject and comprehensively focused on the
effect of internal and external corporate governance mechanisms on idiosyncratic risk.
However, they did not examine the financial industry. Because the financial sector is a
franchise industry and is closely related to the public, a firm must possess a strong corporate
governance mechanism. We conducted comprehensive analysis to determine how the
corporate governance mechanisms influence idiosyncratic risk. This study can serve as a
reference for government agencies and financial institutions in promoting corporate
Proceedings of the Second Asia-Pacific Conference on Global Business, Economics, Finance
and Social Sciences (AP15Vietnam Conference) ISBN: 978-1-63415-833-6
Danang-Vietnam, 10-12 July, 2015 Paper ID: V535
4
www.globalbizresearch.org
governance so that the essence of corporate governance can be implemented to maintain
stakeholder interests. Thus, corporate organizations and operations can develop steadily.
Section II presents a literature review; Section III introduces data sources, variable definitions,
and the research model; Section IV shows the empirical analysis results; and Section V the
conclusions.
2. Literature Review
2.1. Financial Institutions and Corporate Governance
The financial industry is the primary industry in a nation. However, Taiwan’s financial
industry lacks industrial competitiveness. Since 1980, to adapt to the globalization and
liberalization trends in the global financial markets, the government has gradually relaxed the
financial regulatory measures and reduced the regulatory thresholds for establishing banks.
The government hopes that by enabling fully competition in the financial industry, the
industry could improve efficiency and establish fair competition in the financial system.
Because of the special nature of the financial industry, poor operations affect the firm role
as funding agencies and the national economic development. Chen (2005) presented
characteristics of the financial industry and the necessity of strengthening corporate
governance. First, the financial industry is the economic lifeline of a nation. Corporate
borrowing, fund collection, and international trade are dependent on the financial sector. If
the financial sector possesses poor corporate governance, the funding agency function would
be compromised and would affect the economic sector. Major funding in the financial
institution is obtained from the public. Specifically, banking funds are primarily obtained
from the community. Although banks possess low equity funds, they operate large-scale
businesses. With this operation from a high financial leverage, corporate governance must be
implemented to safeguard the rights and interests of depositors. Regarding the financial
industry, integrity and trust are essential; therefore, the managerial style and ethical standards
determine the stability of a bank and bank performance. Taiwan’s financial institutions have
focused on a personal-network and family-oriented business model in financial institutions,
which is a substantial barrier to corporate governance. Taiwan’s financial institution requires
a stable corporate governance mechanism to reduce the risks and problems in moral issues.
To implement a stable corporate governance system and promote the healthy
development of the financial market, the Taiwanese government has made several major
Proceedings of the Second Asia-Pacific Conference on Global Business, Economics, Finance
and Social Sciences (AP15Vietnam Conference) ISBN: 978-1-63415-833-6
Danang-Vietnam, 10-12 July, 2015 Paper ID: V535
5
www.globalbizresearch.org
revisions to the legal regulations and initiated a series of complementary measures and reform
items.
2.1.1 Improving the independency of board
To prevent the board from becoming a formality, the Securities and Futures Commission
of Ministry of Finance, beginning in February 22, 2002, implemented an independent director
and supervisor system in two stages. During the first stage, initial public offering (IPO) and
over the counter (OTC) firms must disclose in their annual report whether the board of
directors comply with crucial resolutions made by directors and supervisors and the opinions
of both parties. In addition, to apply for becoming a listed or OTC firm, firms must establish
at least two independent directors and one independent supervisor. If these firms failed to
follow these requirements, they could not be listed. The second stage involved publically
promoting these regulations to encourage all the listed and OTC firms to implement these
regulations.
2.1.2 Strengthen the information transparency
Financial holding corporations should disclose all net operating income. This included a
financial holding corporation’s banks, insurance, securities, and investment firms; current
regulations do not specifically require that corporations reveal all of their businesses and
regions of operation. Consequently, corporate profit sources are vague to the public.
Regarding corporate expansion and globalization, financial institutions must revise disclosure
items to respond to global trends. Domestic banking businesses focus on lending, but the
competent authority has not regulated the disclosure lending structure. Consequently,
investors are unable to obtain the lending policies of various banks or subsequently assess
potential credit risks.
2.1.3 The Best-Practice Principles of Corporate Governance
To implement a corporate governance system, the Taiwanese government released the
Corporate Governance Best-Practice Principles for TSEC/GTSM Listed Companies, which
was approved by the Securities and Futures Commission, in October 2002. The content
contains provisions regarding protecting shareholder equity, strengthening board
responsibilities, exerting the supervisor functions, respecting stakeholder rights and interests,
and enhancing information transparency.
Proceedings of the Second Asia-Pacific Conference on Global Business, Economics, Finance
and Social Sciences (AP15Vietnam Conference) ISBN: 978-1-63415-833-6
Danang-Vietnam, 10-12 July, 2015 Paper ID: V535
6
www.globalbizresearch.org
2.2 The Corporate Governance and Risk in Financial Industry
The majority of the literature explores only the relationship between partial corporate
governance mechanisms and risk. Regarding internal corporate governance mechanisms,
empirical studies have indicated that strong corporate governance could reduce the capital
costs of firms, but they failed to explain the relationship between corporate governance and
the idiosyncratic risk of capital cost. Himmelberg et al. (1999) stated that when managers
possessed high shareholding ratio idiosyncratic risk was reduced. Regarding the external
governance mechanism, Jin and Myers (2006) conducted an empirical study from a national
perspective. They found that firms possessing less information transparency exhibited high
idiosyncratic risk. However, they did not research the quality of firm-level governance and its
effects on idiosyncratic risk. Moreover, Gasper and Messa (2006) used data obtained from
CRSP Compustat to analyze the effects of product market competition on idiosyncratic risk.
The study showed that highly competitive product markets exhibited increased idiosyncratic
risk. Ferreira and Laux (2007) explored the effects of the market for corporate control on
idiosyncratic risk. Their results indicated that firms that possessed numerous anti-takeover
provisions had low idiosyncratic risk. Unlike previous studies, Lin et al. (2010) examined the
effects that comprehensive corporate governance, which involved internal and external
mechanisms, has on idiosyncratic risk. The results indicated that when the shareholding ratio
by external blockholders, ratio of independent directors and supervisors on boards, and
shareholding ratio by managers were high, and when information was obtained in a timely
manner, then idiosyncratic risk was reduced. In other words, improved internal corporate
governance mechanisms effectively reduce idiosyncratic risk. Legal regulations and product
market competitiveness have no substantial effects on idiosyncratic risk, thereby indicating
that external corporate governance mechanisms cannot reduce idiosyncratic risk.
Regarding the financial industry, scholars have mostly focused on parts of corporate
governance mechanisms and their effects on system risk or corporate governance mechanisms
and their influence on partial idiosyncratic risk. Saunders et al. (1990) examined the
relationship between bank ownership structure and risk taking. The results indicated that total
risk, non-system risk, and the shareholding ratio of operators are significantly positively
correlated. In addition, the non-significant relationship between system risk and the
shareholding ratio of operators indicated the importance of idiosyncratic risk. Chen (2003)
examined factors from 1996 to 2001 (i.e., the period of Taiwanese bank recession) that
Proceedings of the Second Asia-Pacific Conference on Global Business, Economics, Finance
and Social Sciences (AP15Vietnam Conference) ISBN: 978-1-63415-833-6
Danang-Vietnam, 10-12 July, 2015 Paper ID: V535
7
www.globalbizresearch.org
influenced bank risk-taking behavior. They found that corporate governance mechanisms had
a significant effect on bank credit risks and overall risks. Chen et al. (1998) used 302 banks
from 1988 to 1993 as their sample. The study indicated that the shareholding ratio of
management (including managers and directors) was negatively correlated with the risk proxy
variables in two-factor market models.
In other words, when the shareholding ratio of management increased, risk aversion
behaviors also increased, thereby supporting the relative risk aversion hypothesis. Cebenoyan
et al. (1995) found that when institution investors possessed high shareholding ratios, the
risk-taking rate of the bank was reduced, thereby supporting the efficient monitoring
hypothesis. However, Li (2002) showed that high shareholding ratios by institution investors
increased bank credit risk, market risk, and overall risk. This relationship supported the
conflict of interest hypothesis. Kan (2003) indicated that no significant correlation was
observed between the shareholding ratio by legal personalities of institutions and the
nonperforming loan ratio of a bank.
3. Methodology
3.1 The Data
Financial institutions issued by the Taiwan Stock Exchange were recruited as research
subjects, including independent banks of the listed and OTC firms, financial holding banks,
and securities industry. Research data included the Taiwan stock index, firm stock price, and
financial reports. All data were obtained from the Taiwan Economic Journal (TEJ) and
Market Observation Post System based on public-issued listed and OTC firms.
Because the insurance industry lacks information disclosure mechanisms, these firms
were not included in this study. For the research sample, 33 firms were selected including
nine independent banks, 14 financial holding banks, and 10 securities firms. Prior to 2006,
TEJ only collected semiannual reports of the financial industry firms and not quarterly reports;
therefore, this study began its examination from the first quarter of 2006 to the fourth quarter
of 2012; the period was 7 years, overall. The study examined the data of the quarterly reports
for each year.
3.2 The Variable Definitions
3.2.1 Ownership structure
This study used the shareholding ratio by external block holders and by institutional legal
personalities as the proxy variables for the ownership structure. This study defined the
Proceedings of the Second Asia-Pacific Conference on Global Business, Economics, Finance
and Social Sciences (AP15Vietnam Conference) ISBN: 978-1-63415-833-6
Danang-Vietnam, 10-12 July, 2015 Paper ID: V535
8
www.globalbizresearch.org
external block holder ownership (BOR) as the shareholding ratio of block holders who were
not identified as directors and managers. The institutional ownership (IOR) was defined as the
sum of the ratio of foreign legal personality ownership from the investment sample firms
(FOREIGN), investment trust and consulting ownership (ITCS), and the dealer shareholding
ratio (DEALERS). This study predicted that when the external block holder shareholding
ratio and the institutional legal-personality shareholding ratio were high, then the supervising
ability of a firm would be high and idiosyncratic risk would be low.
3.2.2 Managerial incentives
This study used managerial ownership ratio (MOR) as the proxy variable of the
managerial incentive mechanism. We predicted that when the managerial shareholding ratio
was high, the interests of the managers and shareholders would be consistent and
idiosyncratic risk would be low.
3.2.3 Board composition
This study defined the independent director and supervisor ratio to the overall director
and supervisor seats (INDR) as the number of seats of independent directors and supervisors
of the sample firm divided by the total seats of the board of directors and supervisors. This
study predicted that a high ratio of independent directors and supervisors to the director and
supervisor seats would elicit a highly independent board. Thus, the managerial supervisory
capacity would be strong and the idiosyncratic risk low.
3.2.4 Information transparency
This study used information timeliness and disclosure rating as proxy variables of
information transparency. This study predicted that when the information transparency is high,
the quality of corporate governance is strong and therefore idiosyncratic risk is low.
3.2.4.1 Information timeliness
Timeliness of information (TIMELINESS) was used as the first proxy variable of
information transparency (Ashbaugh et al., 2006). The regression equation is established as
follow:
, 0 1 2 3 4× Δі τ і,τ і,τ і,τ і,τ і,τ і,τRET = β β NIBE β LOSS β NIBE LOSS β NIBE ε (1)
where ,τiRET represents the average stock return of firm i in quarter τ ; ,τiNIBE
denotes the quarterly net income of firm i in quarter τ divided by the shareholder equity
Proceedings of the Second Asia-Pacific Conference on Global Business, Economics, Finance
and Social Sciences (AP15Vietnam Conference) ISBN: 978-1-63415-833-6
Danang-Vietnam, 10-12 July, 2015 Paper ID: V535
9
www.globalbizresearch.org
market cap at the beginning of the quarter; ,τiLOSS represents a dummy variable. When
,τiNIBE is a negative number, ,τiLOSS is 1; otherwise, ,τiLOSS is 0; ( ,τΔ iNIBE )
represents the quarterly net income change of firm i in quarter τ divided by the
shareholder equity market cap at the beginning of the quarter. Regression analysis on data of
quarter τ of all firms is conducted using (1). The regression residual i resulting from the
regression analysis is squared and multiplied by 1 . The product is information timeliness
( ,τiTIMELINESS ) of firm i in quarter. When TIMELINESS is high, the data respond to
return in a timely manner; therefore, firm information transparency level is high.
3.2.4.2 Disclosure rating
Information disclosure assessment (INF) was used as the second proxy variable
representing information transparency. To measure the degree of information disclosure, this
study cited the assessment results of the information disclosure and transparence ranking
system provided by the Taiwan Securities and Futures Institute. The following paragraphs
present the assessment ranks converted into numeral codes for measurements, as shown in
Table 1.
Table 1: The rating scale of information transparency
The Rating Scale Codes
A+ 5
A 4
B 3
C 2
C- 1
3.2.5 The measurement of idiosyncratic risk
This study follows the direct decomposition method by Xu and Malkiel (2003) to
estimate idiosyncratic risk. In addition, by establishing the market model, we estimated the
volatility sequence of idiosyncratic and systemic risks. To solve the heteroscedasticity and
heavy-tailed distribution patterns concerns that the sequence of returns possessed, when
estimating idiosyncratic risk, we used a GARCH model to modify the direct decomposition
method by Xu and Malkiel (2003).
Proceedings of the Second Asia-Pacific Conference on Global Business, Economics, Finance
and Social Sciences (AP15Vietnam Conference) ISBN: 978-1-63415-833-6
Danang-Vietnam, 10-12 July, 2015 Paper ID: V535
10
www.globalbizresearch.org
і,τ і і m,t і,tr α β r ε (2)
, , , / , ,i t i i m t i mv t i b v t i tr r r r (3)
(2) Is the market model, where α i and β i represent parameters to be estimated; ,i tr
denotes the excess return of stock i on day t ; ,m tr depicts the excess return of the market
portfolio on day t , and ,εi t represents the residuals. The equation (3) is three factor model
proposed by Fama and French (1993), wherei
r ,i
k represent parameters to be estimated;
,mv tr represent the market size factor,
/ ,b v tr and represents the book to market value factor.
In general, (2) and (3) disregards the issue that data in financial asset time series
possesses heteroscedasticity, which leads to inefficient estimations of the parameters. Thus,
Xu and Malkiel (2003) used rolling methods to estimate the idiosyncratic risk of individual
stocks to solve conditional heteroscedasticity. In (2) and (3), the residuals appeared to possess
a GARCH-model effect. Because of this effect, the idiosyncratic risk of market factors was
estimated.
2
1 (0, )і,t t i,tε |ψ ~ N h (4 )
2 2 2
, 0 1 , 1 2 , 1εi t i t i th h (5)
where 1ψt represents the total information collection prior to period 1t ; 0 , 1 , 2
are parameters that are not negative numbers, and 0 1 < 1 ; 2
, 1i th denotes the estimate of
idiosyncratic risk of stock i at period t ; 2
1i,tε represents the residual square of
idiosyncratic risk of stock i during period 1t ; if 2
i,th were calculated using market
model; 2
,i thF were calculated using three factor model.
Because relevant financial variables could only be obtained from quarterly data reports,
we converted the other research variables into variables representing quarterly data. Daily
Proceedings of the Second Asia-Pacific Conference on Global Business, Economics, Finance
and Social Sciences (AP15Vietnam Conference) ISBN: 978-1-63415-833-6
Danang-Vietnam, 10-12 July, 2015 Paper ID: V535
11
www.globalbizresearch.org
idiosyncratic risk 2
i,th and 2
,i thF were converted into quarterly idiosyncratic risk by adding
all trading days in that quarter. The converted idiosyncratic risks are denoted by ,ij τIV and
,ijFIV
.
3.2.6 Control variables
This study establishes 8 control variables in the study model. The following control
variables were converted into quarterly data. The 8 control variables were firm size
( і,τLNSIZE ), market-to-book ratio ( і,τMTB ), leverage rate ( і,τLEV ), stock turnover ratio
( і,τTURN ), capital expenditure ratio ( і,τCE ), return on assets ( і,τROA ), non-performing
loans ( і,τNPL ), and bank of international settlement ratio ( і,τBIS ).
3.3 Empirical Model
Because the data in this study were panel data that involved cross-sectional and
time-series data of the listed firms in Taiwan’s financial industry from 2006:Q1 to 2012:Q4,
the data were suitable for constructing a panel-data model for statistics analysis, thereby
reducing the collinearity problem. Flannery and Hankins (2013) indicated that when a lagged
period of an explained variable was included in the explanatory variables, a dynamic panel
data-regression model was used to conduct empirical analysis to avoid deviated parameter
estimates. Thus, the dynamic panel data-regression model was established as follows:
,τ ,τ 1 ,τ
2
β β ε , 1,..., , τ 1,...,K
i oi k ki i
k
Y X i N T
(6)
where ,τiY represents the idiosyncratic risk ( ,τiIV or ,ij
FIV
) of firm i in quarter τ ;
,τkiX denotes the K th explanatory variable of firm i in quarter τ ; 0 1β ,β ,...,βK
represent the parameters to be estimated; and ,τεi denotes a random error item.
4. Results
4.1 Descriptive Statistics
This study used the Taiwan financial holdings, banks and securities firms from 2006:Q1
to 2012:Q2 as samples. Descriptive statistical analysis was conducted to obtain the mean,
Proceedings of the Second Asia-Pacific Conference on Global Business, Economics, Finance
and Social Sciences (AP15Vietnam Conference) ISBN: 978-1-63415-833-6
Danang-Vietnam, 10-12 July, 2015 Paper ID: V535
12
www.globalbizresearch.org
standard deviation, median, and quartiles of the various research variables. Table 2 presents
the descriptive statistical analysis of the overall sample.
The results presented in Table 2 are the results of the descriptive statistics of all sample
variables. First, by observing the block holder shareholding ratio, the mean shareholding ratio
of the external block holders of the sample firms was established as 4.08% (SD 0.0833). The
first quartile and the third quartile were 0, which indicated that the external block holder
shareholding ratio in the sample firms was generally low and the differences were not
significant. Regarding the shareholding ratio of institutional legal personalities, foreign
ownership possessed the highest shareholding ratio, with a mean of 16.12% and a median of
10.55%. The results indicated that foreign ownership was relatively strong compared with
other types of legal-personality ownership.
In addition, the shareholding ratio presented a negative skew, thereby indicating that
foreign ownership had a relatively high shareholding ratio in specific firms. The results in the
table show that the mean value of the managerial shareholding ratio was 0.25%. The data
indicated that in over half of the sampled firms, the managerial shareholding ratio was 0%. If
these firms do not have a comprehensive and stable supervising mechanism or transparent
information disclosure policy, then these firms have severe agency problems and information
asymmetry concerns.
Regarding board composition, the mean value was 12.44 (SD 0.1001), and the sample
firms showed only a slight difference, thereby indicating that employing independent
directors and supervisors was common in the sample firms. Regarding operating performance,
the mean value of the return on assets was 0.53% and the standard deviation was 0.0133,
which indicated that the average performance of the financial industry from 2006 to 2012 was
relatively poor.
Table 2: Descriptive Statistics
Average St. Dev. Q1 Median Q3
і,τIV 1.6953 0.2465 1.5525 1.6855 1.8896
,ijFIV
1.7121 0.2634 1.5888 1.7045 1.9108
і,τBOR 0.0408 0.0833 0.0000 0.0000 0.0000
і,τIOR 0.1701 0.1488 0.0332 0.1189 0.2793
і,τFOREIGN 0.1612 0.1521 0.0411 0.1055 0.2753
і,τITCS 0.0065 0.0071 0.0005 0.0042 0.0098
Proceedings of the Second Asia-Pacific Conference on Global Business, Economics, Finance
and Social Sciences (AP15Vietnam Conference) ISBN: 978-1-63415-833-6
Danang-Vietnam, 10-12 July, 2015 Paper ID: V535
13
www.globalbizresearch.org
і,τDEALERS 0.0038 0.0099 0.0000 0.0009 0.0031
і,τINDR 0.1244 0.1001 0.0000 0.1266 0.2000
і,τMOR 0.0025 0.0040 0.0004 0.0015 0.0048
і,τTIMELINESS -0.1421 0.0555 -0.1233 -0.1167 -0.1132
і,τINF 3.9788 0.8221 4.0000 4.0000 4.0000
і,τLEV 0.8121 0.1634 0.6905 0.9091 0.9532
і,τCE 0.0252 0.0233 0.0122 0.0171 0.0320
і,τMTB 0.0201 0.0048 0.0082 0.0108 0.0129
і,τROA 0.0053 0.0133 0.0012 0.0031 0.0080
і,τLNSIZE 24.2555 1.3434 23.2134 24.1279 25.4434
і,τTURN 0.0045 0.0049 0.0011 0.0025 0.0050
Note: Q1 and Q3 are represented the firth and third quartile. ,τiIV is represented the
idiosyncratic risks of firm i at quarter τ . ,τiBOR ,
,τiIOR , ,τiMOR are represented
the outside block-holder ownership, institutional ownership, managerial ownership of
firm i at quarter τ . ,τiINDR is represented the proportion of independent
supervisor/director in the board of firm i at quarter τ . We take ,τiTIMELINESS for
estimating the information timeliness of firm i at quarter τ . і,τLNSIZE ,
і,τMTB ,і,τLEV ,
і,τTURN , і,τCE and
і,τROA are the control variables.
4.2 Correlation Coefficient Analysis
Before a regression model could be established, high degrees of similarities between
independent variables must be prevented from influencing the study results. We conducted
Pearson correlation coefficient analysis to explore variables related to corporate governance
regarding the extent of relationships and the trend of idiosyncratic risk. These variables were
as follows: an external block-holder shareholding ratio; shareholding ratio by institutional
ownership; shareholding ratio of foreign ownership; shareholding ratio by securities
investment trust and consulting representatives; shareholding ratio by dealers; the number of
independent directors and supervisors; shareholding ratio by managers; information
timeliness; information transparency; and disclosure assessments.
Correlation coefficient analysis indicated that the current idiosyncratic risk and the
idiosyncratic risk for the following period were positively correlated. Regarding internal
corporate governance variables, the shareholding ratio by institutional leg personalities,
shareholding ratio of foreign ownership, shareholding ratio by dealers, and the cost and price
differences were all positively correlated with idiosyncratic risk. However, the correlations
were not significant. All other variables showed significant correlations with idiosyncratic
Proceedings of the Second Asia-Pacific Conference on Global Business, Economics, Finance
and Social Sciences (AP15Vietnam Conference) ISBN: 978-1-63415-833-6
Danang-Vietnam, 10-12 July, 2015 Paper ID: V535
14
www.globalbizresearch.org
risk. When the external block holder shareholding ratio, securities investment trust and
consulting representatives, the ratio of independent directors and supervisors, and the
shareholding ratio by managers were high, then idiosyncratic risk would also be high. In
addition, the two variables related to information transparency (i.e., information timeliness
and information disclosure assessment) were negatively correlated with idiosyncratic risk.
Table 3: Pearson Correlation Coefficients
A B C D E F G H I J K
IVt+1 A 1.00
IV B 0.69 1.00
BOR C 0.09 0.07 1.00
FOREIGN D 0.00 0.02 -0.22 1.00
ITCS E 0.12 0.06 -0.14 0.39 1.00
DEALERS F 0.05 0.06 0.03 0.02 -0.06 1.00
IOR G 0.01 0.02 -0.22 0.99 0.43 0.08 1.00
INDR H 0.15 0.08 0.21 0.14 0.07 -0.02 0.14 1.00
MOR I 0.13 0.10 -0.18 -0.13 -0.17 -0.05 -0.14 -0.12 1.00
TIMELINESS J -0.27 -0.23 -0.08 0.09 0.14 -0.05 0.09 -0.01 -0.07 1.00
INF K -0.19 -0.19 -0.09 -0.03 .217** 0.02 -0.01 0.00 -0.24 0.12 1.00
Note: The definitions of variables are reported in table 2. Bold-faced coefficients are significant at the
1% and 5% level, respectively.
4.3 Empirical Results
The results presented in Table 4 showed that the entire current idiosyncratic risk of the
model and the idiosyncratic risk for the following period were positively correlated. These
results indicated that the idiosyncratic risk of a firm could change over time. In addition, the
effects of partial corporate governance on idiosyncratic risk were significant. One exception
was that the shareholding ratio by block holders was positively but not significantly correlated
with idiosyncratic risk. The shareholding ratio by managers and was negatively, but not
significantly, correlated with idiosyncratic risk. Reaching more than the 5% level of
significance, the other variables, such as the ratio of independent directors and supervisors on
the board, information timeliness, and information disclosure assessments, were negatively
correlated to idiosyncratic risk. This trend showed that high information transparency levels
indicate low idiosyncratic risk in financial institutions. When the ratio of independent
directors and supervisors on the board was high, the idiosyncratic risk of the firm is low.
These results were consistent with the study predictions. The shareholding ratio by
Proceedings of the Second Asia-Pacific Conference on Global Business, Economics, Finance
and Social Sciences (AP15Vietnam Conference) ISBN: 978-1-63415-833-6
Danang-Vietnam, 10-12 July, 2015 Paper ID: V535
15
www.globalbizresearch.org
institutional legal personalities and idiosyncratic risk were positively correlated, thereby
indicating that when the shareholding ratio by institutional legal personalities was high, the
idiosyncratic risk of the financial institution was also high. This result was inconsistent to our
predictions.
To comprehensively understand the relationship of the shareholding ratio by institutional
legal personalities to idiosyncratic risk, this study defined the shareholding ratio by
institutional legal personalities separately as the shareholding ratio of foreign ownership,
shareholding ratio by securities investment trust and consulting representatives, and
shareholding ratio by dealers.
The results shown in Table 5 indicate that the current idiosyncratic risk and the
idiosyncratic risk of the following period were positively correlated, thereby suggesting that
idiosyncratic risk change over time. By reaching a level of significance of more than 5%, the
ratio of independent directors and supervisors on the board and information transparency
were negatively correlated to idiosyncratic risk. This indicated that a high number of
independent director and supervisors on the board along with high information transparency
lowers the idiosyncratic risk of financial institutions. The shareholding ratio by foreign
investors and idiosyncratic risk were positively correlated. This indicated that when the
shareholding ratio by foreign investors was high, the idiosyncratic risk was high. Regarding
the control variable results, Tables 4 and 5 indicate that at a 5% level of significance, firm
size and idiosyncratic risk were negatively correlated. A large-scale firm experiences low
idiosyncratic risk. In addition, at a 5% level of significance, leverage ratio and idiosyncratic
risk were positively correlated, thereby indicating that when high leverage ratios increase
idiosyncratic risk increases.
This study also used idiosyncratic risk that was estimated by the three-factor model to
verify the stability. The results shown in Tables 4 and 5 are consistent. Therefore, financial
institutions must strengthen corporate governance quality to reduce idiosyncratic risk and
protect the rights and interests of all stakeholders.
Previous studies have focused on exploring whether the corporate governance
mechanisms in general industry could enhance corporate operating performance and
shareholder wealth. These studies have rarely focused on the effects of the quality of
corporate governance in the financial industry on idiosyncratic risk. In addition, relevant
studies have been limited to the influence of partial corporate governance mechanism on
Proceedings of the Second Asia-Pacific Conference on Global Business, Economics, Finance
and Social Sciences (AP15Vietnam Conference) ISBN: 978-1-63415-833-6
Danang-Vietnam, 10-12 July, 2015 Paper ID: V535
16
www.globalbizresearch.org
idiosyncratic risk. These studies were not comprehensive investigations on the effect of
corporate governance mechanisms in the financial industry on firm idiosyncratic risk. Only
Lin et al. (2010) used general industry as the research subject and comprehensively explored
the effect of the corporate governance mechanism on idiosyncratic risk.
Due to the financial sector is a franchise industry and is closely related to the public, a
financial institution with a strong corporate governance mechanism is essential. This study
comprehensively analyzed the factors of corporate governance that influenced the
idiosyncratic risk in financial industry. These factors can serve as a reference for competent
authorities in governmental sectors and financial institutions that are promoting corporate
governance. Thus, the essence of corporate governance can be implemented to maintain
stakeholder rights and interests and corporate organizations and operations can develop
steadily.
This study was conducted from the corporate governance mechanism perspective for
investigating idiosyncratic risk in financial institutions. By using dynamic panel data
modeling and by using listed and OTC firms from 2006 to 2012 as the study sample, we
explored vital corporate governance mechanisms, such as ownership structure, board
composition, managerial incentive systems, and information transparency and their
relationship with idiosyncratic risk in financial institutions.
First, the empirical results showed that regarding ownership structure, the institutional
ownership was positively correlated to firm idiosyncratic risk. After further analysis, the
results showed that the foreign investor shareholding ratio and idiosyncratic risk were
positively correlated. The main reason for this result could be that foreign investment in
Taiwan’s financial institution is primarily short term. Thus, foreign investment did not
achieve the effect of an institutional legal personality on monitoring corporate operations.
Based on board composition, more independent directors and supervisors on the board are
correlated to low idiosyncratic risk. This result indicated that when boards of directors of
financial institutions in Taiwan possessed high independence, the firm idiosyncratic risk
decreased. Finally, high information transparency in a financial institution was correlated to
low idiosyncratic risk. Thus, we recommend that the financial industry increase the board of
director independence and information transparency to reduce idiosyncratic risk.
Proceedings of the Second Asia-Pacific Conference on Global Business, Economics, Finance
and Social Sciences (AP15Vietnam Conference) ISBN: 978-1-63415-833-6
Danang-Vietnam, 10-12 July, 2015 Paper ID: V535
17
www.globalbizresearch.org
Table 4: The empirical results from dynamic panel data regression: market model
Explanatory variables
(expected sign)
Dependent variable
, 1iIV
Model 1 Model 2
Intercept 3.2832
(0.9217)
3.5423
(0.9387)
,iIV
( )
2.3341
(0.0000) ***
2.5454
(0.0000) ***
,iBOR
(-)
0.1121
(0.5431)
0.1221
(0.5521)
,iIOR
( )
1.2434
(0.0011) ***
і,τFOREIGN
0.9987
(0.0000) ***
і,τITCS
-1.2563
(0.7676)
і,τDEALERS
-0.8876
(0.5521)
,iTIMELINESS
( )
-0.0451
(0.0185) **
-0.0444
(0.0178) **
,iINF
( )
-0.3321
(0.0703) *
-0.3561
(0.0773) *
,iINDR
( )
-0.2122
(0.0340) **
-0.2139
(0.0355) **
,iMOR
( )
-4.2541
(0.6676)
-4.6657
(0.6709)
,iLNSIZE
( )
-2.1231
(0.0000) ***
-2.1333
(0.0000) ***
,iMTB
( )
-3.2122
(0.5143)
-3.3455
(0.5298)
,iLEV
( )
1.1121
(0.7671)
1.1222
(0.7688)
,iTURN
( )
0.2212
(0.0796) *
0.2393
(0.0788) *
,iCE
( )
0.8878
(0.8522)
0.8999
(0.8437)
,iROA
( ) 0.3122 0.3102
Proceedings of the Second Asia-Pacific Conference on Global Business, Economics, Finance
and Social Sciences (AP15Vietnam Conference) ISBN: 978-1-63415-833-6
Danang-Vietnam, 10-12 July, 2015 Paper ID: V535
18
www.globalbizresearch.org
(0.4771) (0.4777)
Time dummy variables Yes Yes
Industry dummy variables Yes Yes
0.2208 0.2332
18.2119
(0.3229)
19.4531
(0.3131)
Note: ,iIV
is represented the idiosyncratic risk of market model for firm i at quarter . The
definitions of other variables are defined in section 3. The p-value is in bracket. ***, **, and *
significant at the 1%, 5%, and 10% level, respectively.
Table 5: The empirical results from dynamic panel data regression: three-factor model
Explanatory variables
(expected sign)
Dependent variable
Model 1 Model 2
Intercept 2.9877
(0.3565)
2.9889
(0.3566)
,iFIV
( )
3.8901
(0.0000) ***
3.8999
(0.0000) ***
,iBOR
(-)
0.2221
(0.3331)
0.2233
(0.3333)
,iIOR
( )
1.9871
(0.0031) ***
і,τFOREIGN
0.9987
(0.0000) ***
і,τITCS
-1.2563
(0.7676)
і,τDEALERS
-0.8876
(0.5521)
,iTIMELINESS
( )
-0.1111
(0.0085) ***
-0.1231
(0.0088) ***
,iINF
( )
-0.5432
(0.0431) **
-0.5569
(0.0448) **
,iINDR
( )
-0.3321
(0.0255) **
-0.3354
(0.0255) **
Proceedings of the Second Asia-Pacific Conference on Global Business, Economics, Finance
and Social Sciences (AP15Vietnam Conference) ISBN: 978-1-63415-833-6
Danang-Vietnam, 10-12 July, 2015 Paper ID: V535
19
www.globalbizresearch.org
,iMOR
( )
-3.9908
(0.4444)
-3.9967
(0.4434)
,iLNSIZE
( )
-2.2221
(0.0000) ***
-2.2891
(0.0000) ***
,iMTB
( )
-3.4535
(0.4989)
-3.5643
(0.5001)
,iLEV
( )
1.1321
(0.5998)
1.1443
(0.5988)
,iTURN
( )
0.4509
(0.0888) *
0.4565
(0.0889) *
,iCE
( )
0.8779
(0.7677)
0.8760
(0.7543)
,iROA
( )
0.4454
(0.3339)
0.4631
(0.3341)
Time dummy variables Yes Yes
Industry dummy variables Yes Yes
0.3001 0.3021
18.9978
(0.3209)
19.5678
(0.3087)
Note: ,i
FIVis represented the idiosyncratic risk of three factors model for firm i at quarter .
The definitions of other variables are defined in section 3. The p-value is in bracket. ***, **,
and * significant at the 1%, 5%, and 10% level, respectively.
References
Ashbaugh, H., D.W. Collins, and R. LaFond, 2006. Corporate gorvernace and the cost of
Equity Capital. Working paper, University of Wisconsin.
Ashbaugh, H., D.W. Collins, and R. LaFond, 2006. The effects of corporate governance on
firms’ credit ratings. Journal of Accounting and Economics, 42(1-2), 203-243.
Bae, K.H., C. Lim, and K.C. Wei, 2006. Corporate governance and conditional skewness in
the world’s stock markets. Journal of Business, 79(6), 2999-3028.
Bai, C.E., Q. Liu, J. Lu, F.M. Song, and J. Zhang, 2004. Corporate governance and market
valuation in China. Journal of Comparative Economics, 32(4), 599-616.
Bloomfield, R.J., and T.J. Wilks, 2000. Disclosure effects in the laboratory: Liquidity, depth,
and the cost of capital. Accounting Review, 75(1), 13-41.
Proceedings of the Second Asia-Pacific Conference on Global Business, Economics, Finance
and Social Sciences (AP15Vietnam Conference) ISBN: 978-1-63415-833-6
Danang-Vietnam, 10-12 July, 2015 Paper ID: V535
20
www.globalbizresearch.org
Bushman and Smith ,2001. Financial accounting information and corporate governance.
Journal of Accounting and Economics, 32(3:)237-333.
Cebenoyan, A. S., Cooperman, E. S., and Register, C. A., 1995. Deregulation, reregulation,
equity ownership, and S&L risk-taking. Financial Management, 24(3), 63-76.
De Bettignies, J.-E., Baggs, J., 2005. Product Market Competition and Agency Cost, Working
Paper. University of British Columbia.
Fama, E. F,1980. Agency problems and the theory of the firm. Journal of Political Economy.
Vol.88: pp.288-307.
Fama, E. F., and Jensen, M. C., 1983. Separation of Ownership and Control. The Journal of
Law & Economics, 26, pp.301-325.
Ferreira, M.A. and P.A. Laxu, 2007. Corporate Governance, Idiosyncratic Risk, and
Information Flow. Journal of Finance, 62: 951-990.
Flannery, M. J. and K. W. Hankins, 2013.Estimating dynamic panel models in corporate
finance, Journal of Corporate Finance, 19, 1-19.
Gaspar, J. M., and M. Massa, 2006. Idiosyncratic risk and Product Market Competition,
Journal of Business, 79(6), 3125-3152.
Gorton, G., and Kahl, M. Block holder Identity, Equity Ownership Structures and Hostile
Takeovers. NBER working paper number 7123, 1999.
Goyal, V. K. and C. W. Park,2002. Board leadership structure and CEO turnover. Journal
of Corporate Finance, Vol.8,PP.49-66.
Hausman, J. A., 1978. Specification tests in Econometrics, Econometrica, 46(6), 1251-1271.
Himmelberg, C.P., R.G. Hubbard, and, D. Palia, 1999. Understanding the determinants of
managerial ownership and the link between ownership and performance, Journal of Financial
Economics, 53(3), 353-384.
Jagannathan, R., and S. B. Srinivasan., 2000. Does Product Market Competition Reduce
Agency Costs? Working paper 7480, National Bureau of Economic Research, Cambridge,
MA.
Jensen, M. C., and Ruback, R. S., 1983. The Market for Corporate Control: The Scientific
Evidence. Journal of Financial Economics, 11, pp.5-50.
Jensen, M.C. and W.H. Meckling, 1976. Theory of the Firms: Managerial Behavior, Agency
Cost, and Ownership Structure. Journal of Financial Economics, 3(4), 305-360.
Proceedings of the Second Asia-Pacific Conference on Global Business, Economics, Finance
and Social Sciences (AP15Vietnam Conference) ISBN: 978-1-63415-833-6
Danang-Vietnam, 10-12 July, 2015 Paper ID: V535
21
www.globalbizresearch.org
Jiambalvo, J., S. Rajgopal, and M. Venkatachalam. 2002. Institutional ownership and the
extent to which stock prices reflect future earnings. Contemporary Accounting Research
19(1): 117-145.
La Porta, Rafael, Lopez de Silanes, Florencio, A. Shleifer, and R.W. Vishny, 2002. Investor
protection and corporate valuation. Journal of Finance, 57(3), 1147-1170.
Lin, C.H, H.M. Chen, L.H. Wang, 2010. The effects of corporate governance on idiosyncratic
risk. Journal of Management, 27(5), 409-435. (In Chinese)
Morck, R., A. Shleifer, and R.W. Vishny, 1988. Management ownership and market valuation:
An empirical analysis. Journal of Financial Economics, 20(1), 293- 315.
Pound, J., 1988. Proxy Contests and the Efficiency of Shareholders Oversight. Journal of
Financial Economics, 20, pp.237-265.
Saunders, A., Strock, E., & Travlos, N. G. (1990). Ownership structure, deregulation and
bank risk taking. The Jorunal of Finance, 45(2), 643- 654.
Saunders, W., Rowan-Robinson, M., Lawrence, A., Efstathiou, G., Kaiser, N., et al. 1990.
MNRAS 242: 318-37 。
Shleifer, A. and Vishny, R., 1997. A Survey of Corporate Governance. Journal of Finance,
52, pp.737-775.
Shleifer, A., and R.W. Vishny, 1986. Large Shareholders and Corporate Control, Journal of
Political Economy, 94(3), 461- 488.