Download - ESSAYS ON CORPORATE GOVERNANCE - t U
ESSAYS ON CORPORATE GOVERNANCE
BY
MS. JUTAMAS WONGKANTARAKORN
A DISSERTATION SUBMITTED IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY (BUSINESS ADMINSTRATION)
FACALTY OF COMMERCE AND ACCOUNTANCY
THAMMASAT UNIVERSITY
ACADEMIC YEAR 2017
COPYRIGHT OF THAMMASAT UNIVERSITY
Ref. code: 25605502310021IBC
ESSAYS ON CORPORATE GOVERNANCE
BY
MS. JUTAMAS WONGKANTARAKORN
A DISSERTATION SUBMITTED IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY (BUSINESS ADMINSTRATION)
FACULTY OF COMMERCE AND ACCOUNTANCY
THAMMASAT UNIVERSITY
ACADEMIC YEAR 2017
COPYRIGHT OF THAMMASAT UNIVERSITY
Ref. code: 25605502310021IBC
(1)
Dissertation Title ESSAYS ON CORPORATE GOVERNANCE
Author Ms. Jutamas Wongkantarakorn
Degree Doctor of Philosophy (Business Administration)
Faculty Faculty of Commerce and Accountancy
University Thammasat University
Dissertation Advisor Assistant Professor Chaiyuth Padungsaksawasdi, Ph.D.
Academic Years 2017
ABSTRACT
These essays investigate the role of corporate governance from international-
evidence to firm-level. The first essay provides international corporate governance
spillover among country groups, namely G7, BRICS, and PIGS in six dimensions of
corporate governance of the Worldwide Governance Indicators (WGI) developed by
the World Bank. Empirical evidence shows that there exists corporate governance
spillover across these groups. The second essay examines international agency cost in
state-owned enterprise (SOE) that determines the effect of ownership structure of SOE
on firm performance. Empirically, SOE has positive impact on firm performance,
suggesting that SOE firm perform better than non-SOE firm. The third essay explores
the relationship between corporate governance and stock liquidity in Thailand using a
treatment effects. Consistent with prior literature, corporate governance has positive
effect on stock liquidity and its impact is more pronounced for listed firms with good
corporate governance. Firms with good corporate governance are viewed to have lower
information asymmetry that investors are more confident to trade more stocks, leading
to higher stock liquidity.
Keywords: corporate governance, index, state-owned enterprise, liquidity
Ref. code: 25605502310021IBC
(2)
ACKNOWLEDGEMENTS
First and foremost, I would like to express my sincere gratitude to my
advisor, Assistant Professor Dr. Chaiyuth Padungsaksawasdi for tremendous and
continuous support of my Ph.D. study. I highly appreciate all his contributions of time,
motivation, advice, and idea. His guidance helps me in all the time of research and
writing this dissertation.
Besides my advisor, I would like to thank the rest of my dissertation
committees, and Associate Professor Dr. Seksak Jumreornvong and Professor Dr.
Pornchai Chunhachinda for insightful comments and encouragement. My special
appreciation goes to Associate Professor Dr. Tatre Jantarakolica and Dr. Thanomsak
Suwannoi for their brilliant suggestions that incent me to widen my dissertation from
various perspectives.
My time at Thammasat University was made enjoyable due to many friends.
Special mention goes to Massaporn Cheuathonghua and Phasin Wanidwaranan for time
spent throughout my study and their kind assistance.
Last but not least, I would like to thank my family for encouragement and
support that motivate me to complete my Ph.D. study.
Ms. Jutamas Wongkantarakorn
Ref. code: 25605502310021IBC
(3)
TABLE OF CONTENTS
Page
ABSTRACT (1)
ACKNOWLEDGEMENTS (2)
LIST OF TABLES (6)
LIST OF FIGURES (8)
LIST OF ABBREVIATIONS (9)
CHAPTER 1 INTRODUCTION 1
1.1 Motivation 1
1.2 Background 3
1.2.1 Corporate governance 3
1.2.2 Corporate governance index 4
1.2.3 History of State-Owned Enterprise (SOE) 5
1.2.4 Liquidity and corporate governance 5
1.3 Objective and contribution 5
1.4 Structure of dissertation 6
CHAPTER 2 INTERNATIONAL CORPORATE GOVERNANCE
SPILLOVER; EVIDENCE FROM PANEL DATA
8
2.1 Introduction 8
2.2 Literature Review 9
2.2.1 Governance indicators as a proxy for corporate governance
practices
9
Ref. code: 25605502310021IBC
(4)
Page
2.2.2 Corporate governance spillovers 10
2.2.3 Hypothesis development 12
2.3 Data description and research methodology 12
2.3.1 Data 12
2.3.2 Methodology 13
2.4 Results and discussion 18
2.4.1 Data and descriptive statistics 18
2.4.2 Testing results of international corporate governance spillover 22
2.4.3 Testing results of relationship between financial integration
and corporate governance spillover 23
2.4.4 Testing results of panel cointegration 23
2.4.5 Testing results of panel VARs 30
2.5 Conclusion 35
CHAPTER 3 AGENCY COST IN STATE-OWNED ENTERPRISES:
INTERNATIONAL EVIDENCE 36
3.1 Introduction 36
3.2 Theoretical framework 38
3.2.1 Theory of SOE 38
3.3 Literature review 39
3.3.1 SOE and performance 39
3.3.2 SOE by Industry 39
3.3.3 Hypothesis development 40
3.4 Data description and research methodology 41
3.4.1 Data 41
3.4.2 Research methodology 41
3.4.2.1 Difference-in-differences (DiD) 41
3.4.2.2 Matching firm method 43
3.5 Empirical results 44
3.6 Conclusion 60
Ref. code: 25605502310021IBC
(5)
Page
CHAPTER 4 CORPORATE GOVERNANCE AND LIQUIDITY 61
4.1 Introduction 61
4.2 Literature review 62
4.2.1 Trading under information asymmetry and adverse selection 62
4.2.2 Corporate governance and liquidity 63
4.2.3 Hypothesis development 65
4.3 Data and data description 65
4.3.1 Data 65
4.3.1.1 Corporate governance of the firms 66
4.3.1.2 Liquidity measure 66
4.3.1.3 Control variables 66
4.3.2 Methodology 66
4.3.2.1 Panel Random-effects Tobit Model 66
4.3.2.2 Panel Random-effects Tobit Model 68
4.3.2.3 Panel Fixed-effects Quantile Regression Model 68
4.3.2.4 Robustness Check 68
4.4 Empirical results 69
4.4.1 Descriptive statistics 69
4.4.2 Estimated results of econometric models 71
4.4.3 Robustness tests 74
4.5 Discussion & Conclusion 78
REFERENCES 80
APPENDIX
APPENDIX A: WGI DATA SOURCES 93
BIOGRAPHY 94
Ref. code: 25605502310021IBC
(6)
LIST OF TABLES
Tables Page
2.1 Countries by group of G7, PIGS, and BRICS 13
2.2 Descriptive statistics 19
2.3 Panel Unitroot test 22
2.4 Panel-data cointegration, KAO test of BRICS country 24
2.5 Panel-data cointegration, Pedoni test of BRICS country 25
2.6 Panel-data cointegration, Pedroni test of BRICS country 25
2.7 Panel-data cointegration, Westerlund test of BRICS country 26
2.8 Panel-data cointegration, KAO test of PIGS country 26
2.9 Panel-data cointegration, Pedroni test of PIGS country 27
2.10 Panel-data cointegration, KAO test of G7 country 28
2.11 Panel-data cointegration, Pedroni test of G7 country 28
2.12 Panel-data cointegration, Westerlund test of G7 country 29
2.13 Multivariate Panel vector autoregressive models (XTVAR) 32
3.1 Number of firms in each country that have SOE 45
3.2 Number of observations by industry 46
3.3 Descriptive statistics of variables 46
3.4 Correlation of variables 47
3.5 Panel-data regression of all sample 49
3.6 Panel regression random effects of China 50
3.7 Panel regression random effects by industry 52
3.8 Panel-data regression by group of law system 57
3.9 Effect of SOE on performance of firm by using the treatment effects and
propensity-score matching method 59
4.1 Descriptive statistics of stock return and liquidity measures. 70
4.2 Estimated Results of Random-effects Linear Model, Random-effects Tobit
Model, and Fixed-effects Quantile Regression Model using Annual Data 72
4.3 Estimated Results of Random-effects Linear Model, Random-effects Tobit
Model, and Fixed-effects Quantile Regression Model using Monthly Data 73
Ref. code: 25605502310021IBC
(7)
Tables Page
4.4 Frequency of Firm-year Categorized by IOD’s Corporate Governance
Index (GovIndex) and Level of Illiquidity (ILLIQ_Level) 74
4.5 Frequency of Firm-month Categorized by IOD’s Corporate Governance
Index (GovIndex) and Level of Illiquidity (ILLIQ_Level) 75
4.6 Estimated Results of Random-effects Ordered Probit Model using
Annually Data and Monthly Data 76
4.7 Descriptive Statistical Indices of Change of Amihud’s Illiquidity (ILLIQ)
After Change in IOD’s Corporate Governance Index (GovIndex) during
2007-2011 and 2012-2017 77
Ref. code: 25605502310021IBC
(8)
LIST OF FIGURES
Figures Page
2.1 Average of six dimensions of corporate governance index of 16
countries, 1998-2014
22
4.1 Histogram of Amihud’s Illiquidity (ILLIQit) 67
Ref. code: 25605502310021IBC
(9)
LIST OF ABBREVIATIONS
Symbols/Abbreviations Terms
SOE State-owned enterprise
CG Corporate governance
CSS Country share SOE share
OECD The Organization for Economic and
Cooperation Development
WGI The World Governance indicators
ISS Institutional Shareholder Services
ESG The Environmental, Social, and Governance
of corporate
G7 Group of Seven
BRICS Brazil, Russia, India, China, and South Africa
PIGS Portugal, Italy, Greece, and Spain
M&A Mergers and acquisitions
UCM The unobserved components model
XTVAR The multivariate panel vector autoregressive
models
Ref. code: 25605502310021IBC
1
CHAPTER 1
INTRODUCTION
1.1 Motivation
Corporate governance has been an important issue in modern finance for
more than 20 years (Cheffins, 2012). It was blamed as one of the major causes that
ignited Asian financial crisis in 1997. During that period, corporate governance did not
only cause financial crisis in the Asian countries, but it also affected financial condition
in the United States and European countries on both macro-level and micro-level. At
the macro-level, the changes of country’s corporate governance are in rules and
regulations that impact social and financial environments. On the other hand, corporate
governance also influences investors and managers decision. Additionally, publics are
not only interested in country-level governance indicators, but they also consider firm-
level corporate governance measures such as Dow Jones Sustainability Index (DJSI).
Therefore, companies, especially the listed ones, are encouraged to adopt good
governance practices which mainly raise business profit by reducing various types of
cost, such as agency, monitoring, and bonding costs.
Since corporate governance is the system of guideline to control and direct
an organization, the framework is related to rules and procedures regarding company’s
decisions. Most importantly, the mechanism is designed for the balance of
stakeholders’ interest. According to the agency theory, there are two principal-agent
problems. First, the conflict between manager and shareholders is the circumstance
that manager does not perform for the best interest of shareholders. Second, the conflict
between majority and minority shareholders is largely driven by the one-share one-vote
principle (Jensen & Meckling, 1976; La Porta, Lopez-De-Silanes, & Shleifer, 1999; La
Porta, Lopez-de-Silanes, Shleifer, & Vishny, 2000). With an imbalance of interest of
stakeholders, this is an evidence of agency problem. As integrity and reliability of the
firm are in doubt, bad governance is an important source of business difficulty which
easily spreads to entire industry, country, and global economy. The development of
corporate governance is normally stimulated by financial crises. Stakeholders, who
Ref. code: 25605502310021IBC
2
suffer from the cost of bad governance, try to manage their future. As the loss is much
higher for developed countries, they are the leader of corporate governance
improvement. Furthermore, they have better resources than developing economies.
Developed countries also motivate others for an adoption of their framework by
providing incentive and punishment. Besides, the corporate governance spillover,
which is a transfer of corporate governance practices between countries, is also fueled
by market integration, the advancement of information technology, merger, and
acquisitions. Stulz (1999) suggests that good governance reduces cost through
globalization. Moreover, everyone is better-off from the acceptance of corporate
governance (Bris, Brisley, & Cabolis, 2008). For these reasons, corporate governance
spillover is an important issue. However, prior studies primarily investigate the firm-
level spillover of corporate governance, especially through merger deals (Martynova &
Renneboog, 2008), and cross-border trading (Abdallah & Goergen, 2008). In terms of
country-level, Marshall, Nguyen, Nguyen, & Visaltanachoti (2016) examine the
relationship between country governance and international equity returns prediction.
On the contrary, this study focuses on country-level corporate governance spillover by
using country corporate governance index with panel regression.
State-owned enterprise (SOE) is an economic entity. It is a company which
is significantly controlled by government through an ownership. SOE is established by
government in order to conduct commercial activities on behalf of state. SOE is
important for world economy, especially in developing countries, because it accounts
for approximately 10 percent of the world’s GDP (Peng, Bruton, Stan, & Huang, 2016).
For example, China has the Country SOE Share (CSS) index more than 90 percent. The
Country SOE Share (CSS) index represents the share of state ownership which accounts
for all sales and market value of each country. The change in ownership structure and
corporate behavior can be driven by corporate governance framework, especially in
SOE (Grosman, Okhmatovskiy, & Wright, 2016). This ownership structure has some
benefits and drawbacks because government is expected to act for the best interest of
public which is in this case, the company’s stakeholders. Unfortunately, the
effectiveness of federal institution is still in doubt. As a result, corporate governance of
SOE is unclear. The effect of this ownership structure on corporate governance should
be investigated. Previous studies regarding corporate governance of SOE are limited.
Ref. code: 25605502310021IBC
3
Research in SOE mainly focuses on Chinese firms which mostly studies the relationship
between privatization companies and their performances. Moreover, they also
emphasize on the association in both firm and industry levels. This study fulfills the
gap by exploring the firm-level corporate governance of SOE in global context.
Lastly, the micro-level issue of corporate governance, which is stock
liquidity, is examined. Liquidity is crucial for listed company because it affects market
value of firm (Amihud & Mendelson, 2008). Since illiquidity is caused by adverse
selection, the information asymmetry links to stock liquidity and pricing. As corporate
governance reduces asymmetric information, it influences the stock liquidity (Glosten
& Milgrom, 1985). Prior studies in asset pricing have some arguments about an impact
of liquidity to the firm. They consider various kinds of control variables such as market
condition and rule and regulation which affect liquidity and the relationship between
liquidity and corporate governance. Literature regarding the association between
corporate governance and liquidity generally employs specific characteristics as a
proxy of corporate governance which cannot represent every perspective of firm’s
governance. Besides, most of them focus on developed markets. This study aims to
explore the relationship between corporate governance and liquidity in firm-level.
1.2 Background
1.2.1 Corporate governance
The history of corporate governance occurred since corporation was
formed in the 16th century and officially stated in the Federal Registrar in 1976 in US
(Cheffins, 2011). The issue that is emphasized is about management accountability.
There is no consensus of definition of corporate governance. There are two dimensions
of the definitions which are corporate activity, and finance and law environment
dimensions (Claessens & Yurtoglu, 2013). According to the difference in economic and
financial system between U.S. (the earliest country that uses the term corporate
governance) and the rest of the world, corporate governance in each country is different
(Morck & Steier, 2005). However, the term reached the world interest by the 1997
Asian financial crisis. Until 1976, Jensen & Meckling (1976) reveal that agency cost
is the important issue of corporate governance.
Ref. code: 25605502310021IBC
4
In capitalism, both developed and developing countries are affected by
problems of corporate governance. Economy, market structure, politics, laws, culture,
and financial system make corporate governance of each country distinctive. Therefore,
mutual understanding among countries is very important. In 1999, Asian roundtable on
corporate governance was organized by the Organization for Economic and
Cooperation Development (OECD) (OECD, 2014) to establish comprehensive
guideline for all countries. However, the role model country of corporate governance
also faces the scandals, for example Enron and WorldCom (Cheffins, 2012; Ke,
Huddart, & Petroni, 2003).
1.2.2 Corporate governance index
Corporate governance cannot be quantified directly and therefore
researchers have to measure corporate governance via proxy (Black, Gledson De
Carvalho, Khanna, Kim, & Yurtoglu, 2017). Prior studies use two types of proxies of
corporate governance. The first type is a secondary data proxy which is derived by
institutions or data providers that construct corporate governance according to their
objectives of study. Data consist of corporate governance and relate to indices such as
world governance index (WGI) by World Bank, and Institutional Shareholder Services
(ISS), the Environmental, Social, and Governance of corporate (ESG) by Thomson
Reuters Corporate Responsibility Ratings. The second type is to construct corporate
governance index. Researchers primarily construct index from check-list of each
dimension of corporate governance or from survey data (K. H. Chung, Elder, & Kim,
2010; K. H. Chung, Kim, Park, & Sung, 2012; Lei, Lin, & Wei, 2013; Prommin,
Jumreornvong, & Jiraporn, 2014; Tang & Wang, 2011). Recently, study on corporate
governance construction is G index (Gompers, Ishii, & Metrick, 2003) which construct
index by using equally weight of 24 check-lists in firm level. The benefit of corporate
governance index is to reduce information asymmetry by signaling investors about the
corporate governance status of country or company. However, the limitation of
corporate governance index is a comparison across countries due to different
measurements and availability of data.
Ref. code: 25605502310021IBC
5
1.2.3 History of State-Owned Enterprise (SOE)
SOE is an important type of ownership structure. Definition of SOE
varies according to law of each country and area of study. In general, SOE is usually
under government ownership and control, as PwC (2015) stated “enterprises where the
state has significant control through full, majority, or significant minority ownership”.
SOE has been used as government tools for social and economic objectives since
ancient Greece and ancient China. However, SOE has problems in management,
control, and allocation that affect efficiency in competitive economy. Problem of
inefficiency stems from performance of the firms that leads to privatization. According
to property rights theory, private ownership structure is better than SOE in terms of
performance (Boardman & Vining, 1989). This study defines SOE with more than 50%
shareholder of government ownership.
1.2.4 Liquidity and corporate governance
Stock liquidity is defined as “the ease with which it is traded”
(Brunnermeier & Pedersen, 2009). Liquidity of stock affects value of the firm (Amihud
& Mendelson, 2008). There are many studies about stock liquidity. Moreover, there are
issues of corporate governance relationship to stock liquidity. Trading against adverse
selection from information asymmetry affects stock liquidity (Glosten & Milgrom,
1985). Illiquidity implies to trading cost in study on asset pricing which affects expected
returns. The information asymmetry between manager and shareholders leads to
implicit transaction cost like illiquidity. The well-known liquidity measure is Illiquidity
ratio (Amihud, 2002).
1.3 Objective and contribution
The objective of this study is to investigate corporate governance in both
macro and micro levels which separate into three parts. The first part is to examine
international corporate governance spillover. This study focuses on corporate
governance transmission among countries by using panel data which is corporate
governance data in country level overtime, and also uses corporate governance index
which is measured by six dimensions of corporate governance.
Ref. code: 25605502310021IBC
6
The second part is to investigate corporate governance and ownership
structure focusing on listed SOE of separate industries. This part uses constructed
corporate governance index of listed SOE around the world by using panel data in firm
level with multi-level which consists of country, industry, and firm level, and employs
matching firm and difference-in-differences methods to compare effect of corporate
governance in ownership structure to stock returns and performances.
The third part is to determine the relationship between corporate
governance and stock liquidity in Thailand. This part uses the same constructed
corporate governance index method to determine effect of corporate governance on
stock liquidity in multi-level effect in firm, industry, and corporate governance levels.
The contributions of this study are as follows. The first one is
comprehensive corporate governance spillover which covers a number of countries
around the world. This study highlights the dimension of corporate governance that has
the most impact on corporate governance transmission. Second, this study is the first
one that investigates corporate governance in all industry of listed companies. The last
contribution is to examine the relationship between corporate governance and liquidity
in microstructure level under emerging market condition.
1.4 Structure of dissertation
Chapter 2 examines corporate governance spillover among a group of
countries based on economic relevance (G7 and BRICS) and regional union (PIGS).
This chapter employs the Worldwide Governance Indicators (WGI) developed by the
World Bank to test corporate governance spillovers in six dimensions among the group
of these countries.
Chapter 3 investigates corporate governance and ownership structure
focusing on state-owned enterprise (SOE). This chapter uses difference-in-difference
(DiD) to determine effects of SOE on the independent variable and matching firm
methods to determine effects of SOE on non-experimental causal study.
Chapter 4 examines the relationship between corporate governance and
stock liquidity of listed companies in Thailand by using a panel analysis. This analysis
employs Random-effects Tobit Model and Fixed-effects Quantile Regression Model to
Ref. code: 25605502310021IBC
7
assess corporate governance index of Thai Institute of Director (Thai IOD) and
illiquidity measure of Amihud (2002).
Ref. code: 25605502310021IBC
8
CHAPTER 2
INTERNATIONAL CORPORATE GOVERNANCE SPILLOVER;
EVIDENCE FROM PANEL DATA
2.1 Introduction
Corporate governance has been used to rationalize the Asian financial crisis
in 1997, subprime crisis of mortgage-backed securities in the United States in 2008 and
European sovereign debt crisis in 2012 (See Aebi, Sabato, & Schmid (2012) and Van
Essen, Engelen, & Carney (2013) for a discussion of the role of good corporate
governance in the crisis). When the crisis occurs, a negative market disturbance usually
spreads from one country to the others. The process of co-movements in the financial
asset prices (e.g., exchange rats and stock prices) and international capital flows is
referred in the literature as financial contagion (See Bekaert, Ehrmansn, Fratzscher, &
Mehl (2014) and Popov & Van Horen (2015) for a literature of the contagion effect
during the crisis). When a country attempts to integrate its financial system into the
international financial markets and institutions through globalization, financial
contagion can be a potential risk.
However, there is a hero in every villain. A good corporate governance
paradigm is considered an effective means to minimize, if not avert, a possibility of a
crisis and eventually financial contagion (See Beltratti & Stulz (2012) and Van Essen
et al. (2013) for a discussion of the importance of good corporate governance to avert
the crisis). This paper aims to examine whether the good governance practices can
extend across the economically-relevant countries (e.g., G7 and BRICS) or regions
(e.g., PIGS). Studies of the spillovers of corporate governance practices can be at the
firm and industry level (e.g., Bris, Brisley, & Cabolis (2008), Cheng (2011), and
Holmstrom (1982)) or at the cross-countries level (e.g., Bris et al. (2008), Martynova
& Renneboog (2008), Bhagat, Malhotra, & Zhu (2011), and Martynova & Renneboog,
(2011)). This study is interested to understand if the governance practices can extend
across countries that are economically cohesive (e.g., G7 and BRICS) or regionally
unified (e.g., PIGS).
Ref. code: 25605502310021IBC
9
Using the Worldwide Governance Indicators (WGI) developed by the
World Bank, the results show that there are governance spillovers among the G7,
BRICS and PIGS countries. The panel unit root tests suggest stationary panel time
series properties of the WGIs among the G7, BRICS and PIGS countries. The panel
cointegration tests confirm the spillover effect among the examining countries.
However, the panel VARs tests partially supports the governance spillover hypothesis.
Specifically, the results from the VARs suggest that the US has a governance-influence
over the PIGS and BRICS countries but has no impact on the G7 countries.
The paper is organized as follows. The next section reviews the literature
of corporate governance spillovers. Section 3 describes the data and methodology.
Section 4 discusses the empirical results and Section 5 contains summary and
conclusion.
2.2 Literature review
This section first explains a proxy for corporate governance practices used in this
study. It then discusses a literature on corporate governance spillovers and proposes a
hypothesis for the unit root, cointegration, and panel VAR tests.
2.2.1 Governance indicators as a proxy for corporate governance practices
Kaufmann, Kraay, & Zoido-Lobatón (1999b) provide evidence of a
positive relationship between governance and economic development outcomes in 150
countries. They use the Worldwide Governance Indicators (WGI) which covers over
200 countries and territories and measures six dimensions of governance: (1) Voice and
Accountability, (2) Political Stability and Absence of Violence/Terrorism, (3)
Government Effectiveness, (4) Regulatory Quality, (5) Rule of Law, and (6) Control of
Corruption. The data start in 1996 and reflect the perceptions of interested parties –
residents of a country, entrepreneurs, foreign investors, and civil society at large
regarding the quality of governance in a country. The aggregate indicators are based on
several underlying variables, taken from a wide variety of existing data sources (See
Kaufmann, Kraay, & Zoido (1999a, 1999b) for the governance dimensions measured
and aggregation methodology).
Ref. code: 25605502310021IBC
10
This paper follows Kaufmann, Kraay, & Zoido (1999a, 1999b) and
Kaufmann, Kraay and Mastruzzi (2009) and use the WGI as a proxy for governance
practices in a country. The WGI captures the perceptions of 6 dimensions of
governance. First, Voice and Accountability captures the degree to which a country's
citizens are able to participate in selecting their government, as well as freedom of
expression, freedom of association, and a free media. Second, Political Stability and
Absence of Violence captures the likelihood that the government will be destabilized or
overthrown by unconstitutional or violent means (e.g., politically-motivated violence
and terrorism). Third, Government Effectiveness measures the quality of public and
civil services, degree of independence from political pressures, quality of policy
formulation and implementation, and credibility of the government's commitment to
such policies. Fourth, Regulatory Quality measures the ability of the government to
formulate and implement sound policies and regulations promoting private sector
development. Fifth, Rule of Law measures the extent to which agents have confidence
in and abide by the rules of society, and the quality of contract enforcement, property
rights, the police, and the courts, as well as the likelihood of crime and violence. Lastly,
Control of Corruption indicates the perceptions of the extent to which public power is
exercised for private gain.
The WGI is not without a limitation. Critics have focused on problems
of bias or lack of comparability of these indicators. For example, Thomas (2010)
focuses on whether the indicators have a construction validity and whether they
measure what they are supposed to measure. He argue that the WGI is based on personal
and untested notions of governance and that the WGI claim to measure governance; as
yet no evidence has been offered that this is true. Langbein and Knack (2010) also argue
that the WGI is tautological and is not measuring what claims to measure. However,
this study believes that the WGI is a reasonable proxy for the country’s governance
practices as mentioned in (Kaufmann, Kraay, & Mastruzzi, 2010).
2.2.2 Corporate governance spillovers
Corporate governance cross-firm spillovers can be explained by the
model of career concerns (Cheng, 2011; Holmstrom, 1982; Meyer & Vickers, 1997).
According to the career concerns model suggested by Cheng (2011), managers of the
Ref. code: 25605502310021IBC
11
two firms compare each other’s performance and decide how much to inflate earnings
based on their performance differential. The competing managers, who are motivated
by their own career concerns, are more likely to inflate their own earnings to boost up
the stock prices1.
Based on the micro foundation of spillovers, the corporate governance
practices can extend across the borders due to, for instance, the economic integration,
cross-border mergers and acquisitions and financial market integration. For the past
decades, free trade agreements among countries in the same region and across regions
have been widely established and implemented (e.g., the European Union (EU),
ASEAN Economic Community (AEC) and The North American Free Trade Agreement
(NAFTA), resulting in an increased level of economic integration among the groups of
countries under the agreements have rapidly increased2. In addition, firms in the
countries under the free trade agreement have been integrated through the mergers and
acquisitions (M&A). Corporate governance of the firms can then be spillover between
the two merged firms3. With the higher level of international trade and economic
integration among countries, the cross-country M&As have enforced the notion of the
corporate governance spillovers. That is, corporate governance spillovers can spread
between the two countries through the spillover of the two mergers firms.
Financial market integration among countries have also largely
increased through the world financial market during the past decades. The financial
integration allows cross-border capital flows and cross-border listings in foreign stock
1 The inflating earnings minimize bad perceptions on their performance (Holmstrom, 1982; Meyer &
Vickers, 1997; Stein, 1989). Graham, Harvey, & Rajgopal (2005) found evidence from survey study
supporting that career concerns and external labor market reputations are the first concern for managers.
Gibbons & Murphy (1990) and Jenter & Kanaan (2015) revealed the significant relationship between
relative performance evaluation in stock price performance and observably poor labor market outcomes such as being fired. 2 Petri, Plummer, & Zhai (2012) found that ASEAN market, especially ASEAN-5, had been integrated
and converged in term of economic growth, both productivity and unemployment. Trade among ASEAN
has increased from about 18% in 1985 to more than 30% in 2015. 3 Mostly, the acquirer firm with high level of corporate governance transfers its governance practice to
the target firm with lower level of corporate governance. Bris et al. (2008) revealed that the spillover of
corporate governance had been implemented via transfer of accounting standards and shareholder
protection improve Tobin's Q. Based on the law hypothesis, positive spillover states that spillover of
corporate governance caused by M&A spreads from high level bidder firm to improve the low level
corporate governance of target firm (Goergen & Renne, Lim, Brooks, & Hinich, 2008; Martynova &
Renneboog, 2008)
Ref. code: 25605502310021IBC
12
markets. For example, Liao & Ferris (2015) examined the intra-industry spillover from
the cross-listing firms by explaining that foreign cross-listing firm must comply with
SEC and exchange regulation indicating higher level of corporate governance. All in
all, there is strong evidence that the level of governance practices can extend across the
borders from one country to the other.
2.2.3 Hypothesis development
This paper hypothesizes that the cross-country corporate governance
spillover – caused by the economic integration, financial market integration, cross-
border M&As, and cross-listing of the foreign companies – exists. This study focuses
on three major economic group countries, i.e., the G7, BRICS and PIGS countries and
tests whether there exist corporate governance spillovers among these groups of
countries.
Hypothesis: There exists corporate governance spillover among countries.
This study focuses on three major economic group countries, including
G7, BRICS, and PIGS countries.
2.3 Data description and research methodology
2.3.1 Data
2.3.1.1 Governance indices
While other studies focus on corporate governance at the firm
level based on several aspects, such as, the sustainability index, individual governance
index, Environmental, Social, and Governance of corporate (ESG) by Thomson Reuters
Corporate Responsibility Rating, this study focuses on the country level using the
Worldwide Governance Indicators (WGI) of the World Bank.4 Kaufmann, Kraay, &
Mastruzzi (2011) employed the unobserved components model (UCM) to create WGI.5
They grouped 31 sources of survey data into six dimensions of governance. The initial
data consists of annual data of 215 countries with six dimensions of governance: Voice
4 For the data sources of WGI, see Appendix A. 5 For methodology that constructed WGI, see Kaufmann et al. (2011)
Ref. code: 25605502310021IBC
13
and Accountability, Political Stability and Absence of Violence/Terrorism, Government
Effectiveness, Regulatory Quality, Rule of Law, and Control of Corruption.
The WGI data are based on the survey of participant perceptions
of governance. The WGI is constructed from many sources of expert survey, for
example, the Global Competitiveness survey, BERI survey, and Economic Freedom
Index Poll. The sample excludes missing data and countries that do not have the stock
exchanges. The aggregate governance indicators are available from 1996-2014. The
remaining sample is 182 countries and the period of study is 1996-2014. To provide an
economic justification on the governance spillovers of the economic/financial
integration, this study uses 3 groups of sample countries including the G7, BRICS and
PIGS countries as follows.
Table 2.1 Countries by group of G7, PIGS, and BRICS
G7 PIGS BRICS
Canada Portugal Brazil
France Italy Russia
Germany Ireland India
Italy Greece China
Japan Spain South Africa
United Kingdom
United States
2.3.1.2 Stock market indices
The annual stock market index returns are calculated from
daily closing stock index price. The stock market indices are collected from Thomson
Reuters Eikon.
2.3.2 Methodology
Methods of the study in this study consist of (i) testing methods of
international corporate governance spillover; and (ii) testing method of the relationship
corporate governance spillover.
Ref. code: 25605502310021IBC
14
2.3.2.1 Testing methods of international corporate governance
spillover
This study employs a panel unit root test. This study
hypothesizes that a cross-country corporate governance spillover exists among
countries with the same economic groups. If so, the panel time series using the annual
WGI of these groups (G7, BRICS and PIGS) should be stationary. These tests assume
that cross-sectional (countries) WGIs in the panel are all independent or there is no
relationship among the cross-sectional (countries) WGIs. The tests include the LLC
test (Levin, Lin, & Chu (2002).
The LLC test computes the test statistic by averaging single
time-series Augmented Dickey Fuller (ADF) t-tests of all cross-sectional units
(countries) assuming homogenous across cross-sectional units (countries). The null
hypothesis is that each individual country CG time series contains a unit root against
the alternative that each country CG time series is stationary. The testing model is
, 1
1
ip
it i i t iL it L im imt it
L
CG CG CG d − −
=
= + + + (1)
where: i represents countries which 1,2,3, ,i k= , mtd is vector of deterministic
variables, 1,2,3m = , 1 { }td empty set= , 2 {1}td = , 3 {1, }td t= , and mi is vector of
coefficients.
The test procedures can be divided into three steps. The first
step begins by performing separate augmented Dickey-Fuller (ADF) (equation (1)) for
each i cross-sectional country currency where lag order ip can be varied across i. For
given time T, optimal lag ip can be determined. The two regression models are then
estimated (i) using itCG as dependent variable and it LCG − (for all L=1,…, pi) and
imtd as independent variables to obtain residual ite and (ii) using 1itCG − as dependent
variable and it LCG − (for all L=1,…, pi) and imtd as independent variables to get
residual 1itv − . Then, standardized values of the two residuals should be computed as
ˆ ˆit it ie e = and , 1
ˆ ˆi t it iv v − = , where ˆ
i is standard error from each ADF test.
Ref. code: 25605502310021IBC
15
The second step is to compute the ratio of long-run to short-run
standard deviations. The long-run variance can be computed as:
2 2
, 1
2 1 2
1 1ˆ 2
1 1i
T K T
CG it it i tKLt L t L
CG w CG CGT T
−
= = = +
= + − −
(2)
where: K is optimal truncated lag and ( )( )1 1KL
w L K= − + . Then, ratio of long- run
standard deviation to innovation standard deviation is computed as ˆ ˆ ˆi ii Rs = and
average standard deviation can also be computed as 1
1ˆ
N
i
i
S sN =
= which N=k countries.
The last step is to compute the panel unit root test statistics by
estimating pooled regression based on NT observations of
, 1it i t ite v −= + (3)
where: 1T T p= − − and 1
N
i
i
p p N=
= .
Then, the panel unit root t-test for 0 : 0H = can be computed:
ˆ
ˆ
ˆt
= (4)
where: 2
, 1 , 1
1 2 1 2
ˆ
i i
N T N T
i t it i t
i t p i t p
v e v − −
= = + = = +
= and
1 2
2
ˆ , 1
1 2
ˆ ˆi
i
TN
i t
i t p
v −
= = +
=
and ( )22
, 1
1 2
1ˆˆ
i
N T
it i t
i t p
e vNT
−
= = +
= −
Finally, to obtain asymptotic property, the adjusted t-statistic
can be computed:
2 *
ˆ*
*
ˆ ˆ ˆ(0,1)
N mT
mT
t NTSt N
−−= (5)
Ref. code: 25605502310021IBC
16
where: *
mT and
*
mT are the mean and standard deviation adjustments obtained from
LLC computations. As a result, the *t is asymptotically distribution as (0,1)N .
Note that the LLC test has also been claimed that its limitations
are caused by cross-sectional independent assumption and test only no unit-root of all
cross-sectional units.
2.3.2.2 Panel-data cointegration tests
This study also employs a panel cointegration test and the
multivariate panel vector autoregressive models (XTVAR). Based on the economic and
financial market integration, this study hypothesizes that a cross-countries corporate
governance spillover exists among these groups of countries. If so, cointegrating
relationship between the domestic governance and foreign (US in this study)
governance should be found. Thus, a test in this section include a panel cointegration
test using the panel annual WGIs against the US’s WGI.
To test the governance spillover from the foreign (US) country
to other countries, a panel cointegration test between the panel of WGIs of other
countries and WGI of the US is applied to test the existence of long-run relationship
between the two series. The Pedroni (2004) test (based on Engle-Granger) is employed
to test the long-run cointegrated relationship between the two WGI panel series by
assuming the asymptotic and finite sample properties of the panel data. Consider the
long-run relationship.
To test the governance spillover from the foreign (US) country
to other countries, a panel cointegration test between the panel of WGIs of other
countries and WGI of the US is applied to test the existence of long-run relationship
between the two series. The Pedroni (2004) test (based on Engle-Granger) is employed
to test the long-run cointegrated relationship between the two WGI panel series by
assuming the asymptotic and finite sample properties of the panel data. Consider the
long-run relationship.
it i i USt itCG CG = + + (6)
Ref. code: 25605502310021IBC
17
where: i and i are cointegrating equation parameters, which may or may not be
homogeneous across i.
Based on Pedroni (1996), the between-dimension, group-mean
panel Fully Modified Ordinary Least Squares (FMOLS) can be estimated as
1
1
ˆ ˆi
N
GFM FMOLS
i
N −
=
= (7)
where: ˆiFMOLS is the conventional time series FMOLS estimator (Phillips & Hansen,
1990) for country i. Then, t-statistic for the between-dimension estimator can be
computed as
1 2
ˆ ˆ
1GFM FMOLSi
N
i
t N t
−
=
= (8)
where: ˆFMOLSi
t
is t-statistic of FMOLS estimator ˆiFMOLS .
In order to reconfirm the results, this study also employs panel
VARs to determine interdependence and dynamic relationship between economic and
financial market integration and CG index of among the countries within the same
group.
Currently, there are two methods of estimation for panel VARs,
including, Least Squares Dummy Variable (LSDV) procedure known as XTVAR
(Cagala & Glogowsky, 2014), and panel Generalized Method of Moments (PGMM)
known as PVAR (Abrigo & Love, 2016).
PGMM requires asymptotic properties and suitable for large
size sample with more cross-sectional units (i) and long length period of time (t) (Bun
& Kiviet, 2006) while XTVAR employs least squares dummy variable technique, which
is suitable for panel data with less number of cross-sectional units. Since this study
emphasized on G7, BRICS, and PIGS, which consists of only 5-7 countries (cross-
sectional units), XTVAR is applied since it is more appropriate for finite data than
PGMM. Shank & Vianna (2016) uses XTVAR to investigate investors’ behavior in ETFs
that their data are similar to this study. Their data have the fixed number of ETFs and
Ref. code: 25605502310021IBC
18
time tends to be increasing. Therefore, this study uses XTVAR, the dynamic spillover
is analyzed by Panel Vector Autoregressions based on Cagala & Glogowsky (2014).
The panel VAR model can be stated as:
, , 1 , 211 11 12 12
, , 1 , 221 21 22 22
1 1 , , 1 ,1
2 2 , , 2 ,2
...
i t i t i t
j t j t j t
p p i t p i t i ti
p p j t p j t i ti
CG CG CG
CG CG CG
CG Dumf
CG Dumf
− −
− −
−
−
= +
+ + + +
, (9)
where ,i tCG is corporate governance index of country i at period t. The panel index i
represents k countries of the three group countries. The optimal lag length p is obtained
from LLC unitroot test. fi is unobserved fixed effect of the panel model, which
represents the effects of each country.
2.4 Results and discussion
2.4.1 Data and descriptive statistics
Table 2.2 shows descriptive statistics of all six aspects (including,
Voice & Accountability, Political Stability, Government Effectiveness, Regulatory
Quality, Rule of Law, and Control of Corruption) of CG index of G7, BRICS, and PIGS
countries. Panel A illustrates Voice and Accountability aspect, Panel B reveals Political
Stability and Absence of Violence/Terrorism aspects, Panel C shows Government
Effectiveness aspects, Panel D states Regulatory Quality aspect, Panel E illustrates Rule
of Law aspects, and Panel F shows Control of Corruption aspect. Based on group
evaluation, mean values of each indicates the same direction that mean of CG index of
G7 and PIGS are mostly at the same level for all six aspects while mean of CG index
of all aspects of BRICS are deviated from each other. G7 countries have average CG
index higher than those of the two groups countries. BRICS countries which are
differences in terms of location, economic background, and culture show mean
differences among CG index of all aspects. And figure 2.1 shows graph comparision of
average of corporate goverance index of all six aspects.
Ref. code: 25605502310021IBC
19
Table 2.2 Descriptive statistics
Panel A Voice and Accountability
Country Mean Median SD Min Max
Brazil 1.4924 1.4511 0.0910 1.3757 1.6752
Canada 1.2411 1.2203 0.1076 1.0885 1.4745
China 1.3757 1.3619 0.0556 1.2911 1.4728
France 1.0128 1.0235 0.0766 0.8913 1.1555
Germany 1.0045 1.0222 0.0636 0.8888 1.0999
Greece 1.3356 1.3144 0.0897 1.1976 1.6109
India 1.2571 1.2287 0.1729 0.9967 1.5407
Ireland 1.3818 1.3721 0.0939 1.2247 1.6146
Italy 0.9053 0.9463 0.1625 0.5649 1.1355
Japan 1.1616 1.1457 0.1191 0.9657 1.3275
Portugal 0.3869 0.4212 0.1328 0.0924 0.5297
Russia -0.7420 -0.8581 0.2350 -1.0423 -0.2984
South Africa 0.3913 0.4076 0.0544 0.2574 0.4502
Spain -1.5359 -1.5743 0.1262 -1.6816 -1.2854
United Kingdom 0.6516 0.6384 0.1015 0.5516 0.8740
United States 1.1988 1.1275 0.1237 1.0503 1.3659
Panel B Political Stability and Absence of Violence/Terrorism
Country Mean Median SD Min Max
Brazil 1.0179 1.0300 0.1136 0.7906 1.1762
Canada 0.5439 0.5523 0.1820 0.1751 0.8502
China 0.9235 0.9279 0.2023 0.5454 1.3246
France 0.5766 0.5009 0.2407 0.2750 1.1261
Germany 1.0063 0.9902 0.0977 0.8365 1.1895
Greece 0.4832 0.4480 0.2738 0.0947 0.9834
India 0.9756 0.9382 0.2460 0.7019 1.3586
Ireland 1.1662 1.1518 0.1943 0.8772 1.4961
Italy 0.2924 0.4565 0.3665 -0.2240 0.7943
Japan -0.0064 0.0100 0.2697 -0.4656 0.4193
Portugal -0.1142 -0.1840 0.2108 -0.3779 0.2860
Russia -1.0146 -0.9320 0.2344 -1.4622 -0.7360
South Africa -1.1617 -1.1647 0.1589 -1.5269 -0.9124
Spain -0.4635 -0.4787 0.1260 -0.6572 -0.1665
United Kingdom -0.1332 -0.0944 0.2025 -0.5774 0.1986
United States 0.4653 0.5244 0.3362 -0.1960 1.0132
Ref. code: 25605502310021IBC
20
Table 2.2 Continued
Panel C Government Effectiveness
Country Mean Median SD Min Max
Brazil 1.8553 1.8339 0.0962 1.7523 2.0131
Canada 1.5451 1.5355 0.1426 1.3293 1.8147
China 1.6367 1.5809 0.1508 1.4013 1.9312
France 0.5430 0.4489 0.2136 0.2136 0.8687
Germany 1.3762 1.4376 0.2231 0.9566 1.8190
Greece 1.7025 1.6902 0.1552 1.4708 1.9165
India 1.0683 1.0710 0.1006 0.8826 1.2273
Ireland 1.5643 1.5717 0.1278 1.3363 1.7772
Italy 0.6221 0.6312 0.1543 0.3075 0.8344
Japan 1.3037 1.1501 0.3626 0.8907 1.8988
Portugal -0.0725 -0.0975 0.1072 -0.2299 0.1797
Russia -0.4349 -0.4114 0.1604 -0.7660 -0.0785
South Africa -0.0772 -0.0829 0.0838 -0.2044 0.1110
Spain 0.0269 0.0032 0.1405 -0.2483 0.3393
United Kingdom 0.5395 0.5107 0.1547 0.3252 0.8765
United States 1.6198 1.6028 0.1194 1.4575 1.8426
Panel D Regulatory Quality
Country Mean Median SD Min Max
Brazil 1.6183 1.6222 0.1014 1.4258 1.8309
Canada 1.1330 1.1843 0.1497 0.8075 1.3109
China 1.5127 1.5262 0.1042 1.2184 1.6951
France 0.8714 0.9053 0.1259 0.6614 1.0925
Germany 1.0081 1.1023 0.2381 0.4841 1.2597
Greece 1.7718 1.7608 0.1263 1.5931 2.0229
India 1.0206 1.0741 0.2162 0.6337 1.2896
Ireland 1.6887 1.6536 0.1241 1.5351 1.9169
Italy 0.7408 0.8082 0.1911 0.3450 0.9980
Japan 1.1601 1.1903 0.1603 0.7772 1.3535
Portugal 0.1472 0.0957 0.1614 -0.0726 0.4120
Russia -0.3331 -0.3587 0.1126 -0.5640 -0.1130
South Africa -0.3560 -0.3680 0.0926 -0.4807 -0.1584
Spain -0.2448 -0.2351 0.1020 -0.5306 -0.1294
United Kingdom 0.4840 0.4060 0.1585 0.2672 0.7784
United States 1.5070 1.5596 0.1460 1.2524 1.7394
Ref. code: 25605502310021IBC
21
Table 2.2 Continued
Panel E Rule of Law
Country Mean Median SD Min Max
Brazil 1.7430 1.7415 0.0691 1.6328 1.8925
Canada 1.4141 1.4302 0.0718 1.1967 1.5115
China 1.6547 1.6272 0.0757 1.5654 1.8522
France 0.5225 0.4279 0.1998 0.3370 0.9822
Germany 1.3129 1.3188 0.1006 1.1359 1.5987
Greece 1.6835 1.6690 0.0789 1.5468 1.8870
India 1.1176 1.0885 0.1158 0.9529 1.2941
Ireland 1.6484 1.6931 0.1109 1.4666 1.8010
Italy 0.7000 0.7450 0.1906 0.3448 0.9762
Japan 1.1638 1.1456 0.1266 0.9370 1.3918
Portugal -0.2680 -0.3069 0.1577 -0.4924 -0.0037
Russia -0.8709 -0.8687 0.1058 -1.1265 -0.7114
South Africa 0.0652 0.0625 0.1404 -0.1119 0.2896
Spain -0.4202 -0.4328 0.0691 -0.5473 -0.3219
United Kingdom 0.0951 0.0910 0.0584 -0.0126 0.2372
United States 1.5539 1.5628 0.0581 1.4307 1.6298
Panel F Control of Corruption
Country Mean Median SD Min Max
Brazil 2.0161 1.9946 0.1345 1.8178 2.2386
Canada 1.3713 1.3678 0.0783 1.2391 1.5221
China 1.8447 1.8072 0.1312 1.6976 2.1648
France 0.2802 0.3326 0.2500 -0.1100 0.7208
Germany 1.3138 1.2712 0.2570 0.8569 1.7303
Greece 1.8507 1.7571 0.2412 1.5605 2.2409
India 1.0885 1.0479 0.1710 0.8846 1.5229
Ireland 1.5885 1.5783 0.1491 1.2967 1.7933
Italy 0.2293 0.2946 0.3718 -0.2546 1.0572
Japan 1.1081 1.0799 0.2292 0.5260 1.3734
Portugal -0.0541 -0.0463 0.1243 -0.3783 0.1457
Russia -0.9339 -0.9432 0.1164 -1.0876 -0.7105
South Africa -0.4393 -0.4303 0.0891 -0.5728 -0.2830
Spain -0.4692 -0.5259 0.1424 -0.6537 -0.2405
United Kingdom 0.2803 0.2778 0.2945 -0.1653 0.7609
United States 1.4853 1.3935 0.2336 1.2597 2.0114
Table 2.2 shows descriptive statistics of six dimensions of corporate
governance index of 16 countries of G7, BRICS, and PIGS from 1998 to 2014.
Ref. code: 25605502310021IBC
22
Figure 2.1 Average of six dimensions of corporate governance index of 16 countries,
1998-2014
2.4.2 Testing results of international corporate governance spillover
Panel unit root tests of CG index of the countries are performed by
grouping countries into three groups, G7, BRICS, and PIGS.
Table 2.3 Panel Unitroot test
Group Variables Adjusted t delta star
G7 Voice & Accountability -2.6109 ***
G7 Political Stability -8.3776 ***
G7 Government Effectiveness -1.6778 **
G7 Regulatory Quality -0.1972
G7 Rule of Law -1.6103 *
G7 Control of Corruption -2.6279 ***
PIGS Voice & Accountability -0.4565
PIGS Political Stability -2.5185 ***
PIGS Government Effectiveness -1.3146 *
PIGS Regulatory Quality 1.8646
PIGS Rule of Law -1.5527 *
PIGS Control of Corruption -0.5782
BRICS Voice & Accountability -6.0127 ***
BRICS Political Stability -3.2020 ***
BRICS Government Effectiveness -0.8396
BRICS Regulatory Quality -1.5006 *
BRICS Rule of Law -2.2306 **
BRICS Control of Corruption -1.5306 *
-10
-5
0
5
10
15
Average of six dimensions of corporate governance
index of 16 countries, 1998-2014
Voice and Accountability Political Stability Government Effectiveness
Regulatory Quality Rule of Law Control of Corruption
Ref. code: 25605502310021IBC
23
Table 2.3 presents the panel unitroot test (LLC) of governance indexes
of six dimensions in the G7, BRICS and PIGS countries. The Adjusted t delta star
provides tests for no unit-root of all cross-sectional units. The The Adjusted t delta star
equation is
2 *
ˆ*
*
ˆ ˆ ˆ(0,1)
N mT
mT
t NTSt N
−−= . The ***, **, and * indicate statistical
significance at the 1%, 5%, and 10% level, respectively.
Table 2.3 shows panel unit root tests of all six aspects (including,
Voice & Accountability, Political Stability, Government Effectiveness, Regulatory
Quality, Rule of Law, and Control of Corruption) of the CG index of G7, BRICS, and
PIGS countries. The results reveal that panel unit root tests of four aspects including
Voice & Accountability, Political Stability, Government Effectiveness, and Control of
Corruption, of CG index of G7 countries group are stationary, thus, the four aspects
have corporate governance spillover among G7 countries, except Rule of Law and
Regulatory Quality, which are nonstationary, indicating no CG spillover of these two
aspects. The test results of CG index of BRICS countries show stationary properties of
only three aspects including Voice & Accountability, Political Stability, and Rule of
Law indicating the CG spillover on these three aspects are still valid while the other
two aspects are not. However, the tests of stationarity of CG index of PIGS show less
significantly level of the panel unit roots. Only one aspect, Political Stability, shows
significant result indicating stationary of this aspect. Political Stability aspect of CG
index has spread and spillover thought out the peer firm.
2.4.3 Testing results of relationship between financial integration and
corporate governance spillover
The tests are divided into two testing methods, including, panel
cointegration tests and panel VARs.
2.4.4 Testing results of panel cointegration
Test results of panel cointegration are divided into three panel
cointegration testing methods, including Kao test, Pedroni test, and Westerlund test.
Ref. code: 25605502310021IBC
24
Table 2.4 Panel-data cointegration, KAO test of BRICS country
Corporate Governance
dimensions
Modified
Dickey-Fuller t Dickey-Fuller t
Augmented
Dickey-Fuller t
Unadjusted
modified
Dickey-Fuller t
Unadjusted
Dickey-Fuller t
Voice and Accountability 0.1305 -2.5980 *** 0.4072 -1.7262 ** -3.8482 ***
Political Stability -4.5599 *** -4.5358 *** -3.4739 *** -4.5599 *** -4.5358 ***
Government Effectiveness -1.6547 ** -1.6166 * 0.2405 -2.4159 *** -1.9263 **
Regulatory Quality -2.3058 ** -3.2272 *** -2.5552 *** -2.9382 *** -3.4308 ***
Rule of Law -2.1242 ** -2.1902 ** 0.3534 -1.9952 ** -2.1454 **
Control of Corruption -2.5200 *** -2.0565 ** 0.1159 -2.5200 *** -2.0565 **
Table 2.4 presents panel-data cointegration test null hypothesis of non-
stationarity of corporate governance index of six dimensions in BRICS country,
including Brazil, Russia, India, China, and South Africa. The KAO test provides
Modified Dickey-Fuller t, Dickey-Fuller t, Augmented Dickey-Fuller t, Unadjusted
modified Dickey-Fuller t, Unadjusted Dickey-Fuller t, which are tests for stationary of
ite in cointegration equation it it i it i itCG UScg z e = + + . ***, **, and * indicate
statistical significance at the 1%, 5%, and 10% level, respectively.
Table 2.4 shows Panel-data cointegration using KAO testing method
of CG index of BRICS country. Table 2.5 illustrates Panel-data cointegration using
Pedroni testing method of CG index of BRICS country. Table 2.6 reveals Panel-data
cointegration using Westerlund testing method of CG index of BRICS country. The test
results of KAO testing method and Pedroni testing method confirm panel cointegrating
equations of all aspects of CG index of BRICS countries and those of US country exist.
However, Westerlund testing method detect only two aspects, including, Voice and
Accountability aspect and Government Effectiveness aspect.
Ref. code: 25605502310021IBC
25
Table 2.5 Panel-data cointegration, Pedoni test of BRICS country
Corporate Governance
dimensions
Modified variance
ratio
Modified Phillips-Perron t
Phillips-Perron t Augmented
Dickey-Fuller t
Voice and Accountability -5.9377 *** -2.1261 ** 10.5771 *** 15.5371 ***
Political Stability 1.6516 ** -1.1040 -2.9380 *** -2.8384 ***
Government Effectiveness 0.7887 -1.7058 ** -2.4352 *** -2.1608 **
Regulatory Quality 0.1054 -1.0454 -2.6633 *** -2.4490 ***
Rule of Law -0.3240 -1.2109 -2.4549 *** -3.6267 ***
Control of Corruption -0.5046 -0.8209 -1.7587 ** -1.6814 **
Table 2.5 presents the panel-data cointegration test of the null
hypothesis of non-stationarity of governance indexes of six dimensions in the G7,
BRICS and PIGS countries. The Pedroni test provides the Phillips-Perron t-test, which
are tests for a stationary of itein the cointegration equation it it i it i itCG UScg z e = + +
.
The ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level,
respectively.
Table 2.6 Panel-data cointegration, Pedroni test of BRICS country
Corporate Governance
dimensions
Modified
variance ratio
Modified
Phillips-Perron t
Phillips-
Perron t
Augmented
Dickey-Fuller t
Voice and Accountability -5.9377 *** -2.1261 ** 10.5771 *** 15.5371 ***
Political Stability 1.6516 ** -1.1040 -2.9380 *** -2.8384 ***
Government Effectiveness 0.7887 -1.7058 ** -2.4352 *** -2.1608 **
Regulatory Quality 0.1054 -1.0454 -2.6633 *** -2.4490 ***
Rule of Law -0.3240 -1.2109 -2.4549 *** -3.6267 ***
Control of Corruption -0.5046 -0.8209 -1.7587 ** -1.6814 **
Table 2.6 presents panel-data cointegration test null hypothesis of non-
stationarity of corporate governance index of six dimensions in BRICS country,
including Brazil, Russia, India, China, and South Africa. The Pedroni test provides
Modified variance ratio, Modified Phillips-Perron t, Phillips-Perron t, Augmented
Dickey-Fuller t, which are tests for stationary of ite in cointegration equation
Ref. code: 25605502310021IBC
26
it it i it i itCG UScg z e = + + . ***, **, and * indicate statistical significance at the 1%,
5%, and 10% level, respectively.
Table 2.7 Panel-data cointegration, Westerlund test of BRICS country
Corporate Governance dimensions Variance ratio
Voice and Accountability -1.8642 **
Political Stability -0.2314
Government Effectiveness -1.7373 **
Regulatory Quality -1.4862 *
Rule of Law -0.5463
Control of Corruption -0.9509
Table 2.7 presents panel-data cointegration test null hypothesis of non-
stationarity of corporate governance index of six dimensions in BRICS country,
including Brazil, Russia, India, China, and South Africa. The Westerlund test provides,
which is test for stationary of ite in cointegration equation it it i it i itCG UScg z e = + + .
***, **, and * indicate statistical significance at the 1%, 5%, and 10% level, respectively.
Table 2.8 Panel-data cointegration, KAO test of PIGS country
Corporate Governance
dimensions
Modified
Dickey-
Fuller t
Dickey-
Fuller t
Augmented
Dickey-
Fuller t
Unadjusted
modified
Dickey-
Fuller t
Unadjusted
Dickey-
Fuller t
Voice and Accountability 0.3836 0.2959 -0.4320 -0.8053 -0.5828
Political Stability -0.7471 -1.7377 ** -2.3834 *** -0.7471 -1.7377 **
Government Effectiveness -1.5885 * -2.1292 ** -1.4239 * -2.6676 *** -2.5654 ***
Regulatory Quality -3.6059 *** -3.1889 *** -2.4040 *** -4.0619 *** -3.3042 ***
Rule of Law -0.7368 -0.9813 0.4972 -0.7368 -0.9813
Control of Corruption -1.1606 -1.2113 0.6837 -1.1606 -1.2113
Table 2.8 presents panel-data cointegration test null hypothesis of non-
stationarity of corporate governance index of six dimensions in PIGS country, including
Portugal, Italy, Greece, and Spain. The KAO test provides Modified Dickey-Fuller t,
Dickey-Fuller t, Augmented Dickey-Fuller t, Unadjusted modified Dickey-Fuller t,
Ref. code: 25605502310021IBC
27
Unadjusted Dickey-Fuller t, which are tests for stationary of ite in cointegration
equation it it i it i itCG UScg z e = + + . ***, **, and * indicate statistical significance at
the 1%, 5%, and 10% level, respectively.
Table 2.9 Panel-data cointegration, Pedroni test of PIGS country
Corporate Governance
dimensions
Modified
variance ratio
Modified Phillips-
Perron t
Phillips-
Perron t
Augmented Dickey-
Fuller t
Voice and Accountability -0.9702 -0.2139 -0.7228 -0.6850
Political Stability -1.1141 1.1298 -0.2138 -0.8856
Government Effectiveness -0.9735 -1.2089 -2.5823 *** -2.8050 ***
Regulatory Quality -0.7684 -2.0953 ** -3.0954 *** -3.8420 ***
Rule of Law -0.6857 -0.2819 -1.7074 ** -3.5510 ***
Control of Corruption -1.0484 -0.8957 -2.6119 *** -1.8599 **
Table 2.9 presents panel-data cointegration test null hypothesis of non-
stationarity of corporate governance index of six dimensions in PIGS country, including
Portugal, Italy, Greece, and Spain. The Pedroni test provides Modified variance ratio,
Modified Phillips- Perron t, Phillips- Perron t, Augmented Dickey- Fuller t, which are
tests for stationary of ite in cointegration equation it it i it i itCG UScg z e = + + . ***, **,
and * indicate statistical significance at the 1%, 5%, and 10% level, respectively.
Table 2.7 shows Panel-data cointegration using KAO testing method
of CG index of PIGS country. Table 2.8 illustrates Panel-data cointegration using
Pedroni testing method of CG index of PIGS country. Table 2.9 reveals Panel-data
cointegration using Westerlund testing method of CG index of PIGS country.The test
results of KAO testing method, Pedroni testing method, and Westerlund testing method
confirm panel cointegrating equations of three aspects, including, Political Stability,
Government Effectiveness, and Regulatory Quality of CG index of PIGS countries and
those of US country exist.
Ref. code: 25605502310021IBC
28
Table 2.10 Panel-data cointegration, KAO test of G7 country
Corporate
Governance
dimensions
Modified
Dickey-
Fuller t
Dickey-
Fuller t
Augmented
Dickey-
Fuller t
Unadjusted
modified
Dickey-
Fuller t
Unadjusted
Dickey-
Fuller t
Voice and Accountability -2.7407 *** -3.0543 *** -2.8143 *** -3.6090 *** -3.3333 ***
Political Stability -5.3351 *** -5.1080 *** -6.2044 *** -5.3351 *** -5.1080 ***
Government Effectiveness -0.4666 -0.4601 -1.6159 * -0.8206 -0.6727
Regulatory Quality -1.1564 -1.7153 ** -3.5965 *** -2.6872 *** -2.4284 ***
Rule of Law -1.2679 -1.0664 -0.9111 -1.6617 ** -1.2542
Control of Corruption -0.2069 -0.6686 -0.4395 -0.2069 -0.6686
Table 2.10 presents panel-data cointegration test null hypothesis of
non-stationarity of corporate governance index of six dimensions in G7 country,
including Canada, France, Germany, Italy, Japan, the United Kingdom and the United
States. The KAO test provides Modified Dickey-Fuller t, Dickey-Fuller t, Augmented
Dickey-Fuller t, Unadjusted modified Dickey-Fuller t, Unadjusted Dickey-Fuller t,
which are tests for stationary of ite in cointegration equation it it i it i itCG UScg z e = + +
. ***, **, and * indicate statistical significance at the 1%, 5%, and 10% level,
respectively.
Table 2.11 Panel-data cointegration, Pedroni test of G7 country
Corporate Governance
dimensions
Modified
variance ratio
Modified Phillips-
Perron t
Phillips-
Perron t
Augmented Dickey-
Fuller t
Voice and Accountability 0.2893 -1.2914 * -2.5963 *** -2.4112 ***
Political Stability 1.8060 ** -1.1595 -3.2217 *** -2.6931 ***
Government Effectiveness -3.7686 *** -4.8416 *** 15.8355 *** 14.8547 ***
Regulatory Quality 0.7077 -1.2603 -3.8583 *** -2.5581 ***
Rule of Law 0.7689 -1.0146 -1.9406 ** -3.0982 ***
Control of Corruption -1.0823 -1.1351 -2.9233 *** -3.1471 ***
Table 2.11 presents panel-data cointegration test null hypothesis of
non-stationarity of corporate governance index of six dimensions in G7 country,
including Canada, France, Germany, Italy, Japan, the United Kingdom and the United
Ref. code: 25605502310021IBC
29
States. The Pedroni test provides Modified variance ratio, Modified Phillips-Perron t,
Phillips-Perron t, Augmented Dickey-Fuller t, which are tests for stationary of ite in
cointegration equation it it i it i itCG UScg z e = + + . ***, **, and * indicate statistical
significance at the 1%, 5%, and 10% level, respectively.
Table 2.12 Panel-data cointegration, Westerlund test of G7 country
Corporate Governance dimensions Variance ratio
Voice and Accountability -1.8680 **
Political Stability 1.1168
Government Effectiveness -1.4844 *
Regulatory Quality -1.7975 **
Rule of Law -1.2858 *
Control of Corruption -0.4763
Table 2.12 presents panel-data cointegration test null hypothesis of
non-stationarity of corporate governance index of six dimensions in G7 country,
including Canada, France, Germany, Italy, Japan, the United Kingdom and the United
States. The Westerlund test provides, which is test for stationary of ite in cointegration
equation it it i it i itCG UScg z e = + + . ***, **, and * indicate statistical significance at
the 1%, 5%, and 10% level, respectively.
Table 2.10 shows Panel-data cointegration using KAO testing
method of CG index of G7 country. Table 2.11 illustrates Panel-data cointegration
using Pedroni test of CG index of G7 country. Table 2.12 shows Panel-data
cointegration using Westerlund test of CG index of G7 country. The test results of KAO
testing method, Pedroni testing method, and Westerlund testing method confirm panel
cointegrating equations of three aspects, including, Political Stability, Government
Effectiveness, and Regulatory Quality of CG index of G7 countries and those of US
country exist.
The test results of Pedroni testing method confirm panel cointegrating
equations of all aspects of CG index of G7 countries and those of US country exist.
However, KAO testing method and Westerlund testing method detect only two aspects,
including, Voice and Accountability aspect and Regulatory Quality aspect.
Ref. code: 25605502310021IBC
30
For regulatory quality in BRICS and PIGS, government effectiveness,
and rule of law are highly driven by individual countries characteristics. As a result,
they are insignificant and unlikely to spillover. Regulatory quality, which are
nonstationary, indicating there is no corporate governance spillover of two aspects. The
test results of corporate governance index of BRICS countries show stationary
properties of only three aspects including Voice & Accountability, Political Stability,
and Rule of Law indicating the CG spillover on these three aspects are still valid while
the other two aspects are not.
The test results of Pedroni testing method confirm panel cointegrating
equations of all aspects of CG index of G7, PIGS, and BRICS countries and those of
US country exist. G7 and BRICS are well cointegrated in all dimension, according to
the same level of economy, developed country and emerging market.
However, in PIGS country, Voice and Accountability and Political
Stability and Absence of Violence/Terrorism dimension are not cointegrating within
the country group. According to the different within country group such as economics
level, the group consists of developed and developing countries. According to panel
unitroot test and panel cointegration test in table 2.3 to table 2.12, the results of mean
reversion testing show that corporate governance index of group of country G7, PIGS,
and BRICS are stationary, except regulatory quality, government effectiveness, and
rule of law of G7 and PIGS in panel cointegration using Pedoni test. Which mean that
the three significant dimensions of corporate governance can be spillover within group
of country. The implication of this study is corporate governance practices can be
improved by voice of accountability, control of corruption, and political stability. There
exists spillover in term of Political Stability, the group of counties that share the similar
economy and political conditions. According to Panel cointegration and unitroot test,
the results partially confirm that there exists corporate governance spillover.
2.4.5 Testing results of panel VARs
In order to confirm the results of CG spillover among G7, BRICS, and
PIGS countries, this study employs Panel VARs in order to determine the interdepence
and dynamic relationship of CG index among countries within the same group and CG
index of foreign (US) countries. Table 2.13 shows the estimated results of Multivariate
Ref. code: 25605502310021IBC
31
Panel vector autoregressive models (XTVAR) between CG index among countries
within G7, BRICS, and PIGS group and CG index of foreign (US) countries.
For G7 country group, corporate government dimension of Voice and
Accountability, Political Stability, and Absence of Violence/Terrorism dimension of
US statistically significant positive effect to those corporate government of G7. The
spillover support by that US is part a of G7 country which normally utilize the same
economic policy and join in many military operations, US has leading country in G7
both economy and military policy, and US is a country who set standard and policy of
corporate governance, financial standard, and military movement. While Control of
Corruption of G7 has positive effect US., the effect of developed country group will
spillover via accounting standard, and Voice and Accountability of G7have negative
effect to US. The negative effect can be explained by the Career Concerns Model. The
bad governance in the peer group will induced a bad effect to the high governance.
Panel B: PIGS, Voice and Accountability, Government Effectiveness,
Regulatory Quality, and Control of Corruption of US has positive effect to PIGS.
However, The European debt crisis of PIGS spillover to other countries including US.
PIGS has positive effect of and Absence of Violence/Terrorism, Government
Effectiveness, Regulatory Quality, and Control of Corruption to US. While Political
Stability, and Rule of Law has negative effect to US. The negative effect can be
explained by the Career Concerns Model. The bad governance in the peer group will
induced a bad effect to the high governance.
Panel C: BRICS, Political Stability and Absence of Violence/Terrorism,
Regulatory Quality, Control of Corruption of US has positive effect to PIGS. While
BRICS has positive effect of Control of Corruption to US.
Ref. code: 25605502310021IBC
32
Table 2.13 Multivariate Panel vector autoregressive models (XTVAR)
Panel A: G7
Voice and
Accountability Political Stability
Government
Effectiveness
G7 US G7 US G7 US
L1_G7 0.428*** -0.174** 0.260** -0.224 0.768*** 0.01
L1_US 0.121* 0.864*** 0.161*** 0.623*** -0.06 0.540***
N 84 84 84 84 84 84
R-Squared 0.31 0.780 0.319 0.342 0.617 0.399
Regulatory Quality Rule of Law Control of Corruption
G7 US G7 US G7 US
L1_G7 0.506*** -0.06 0.701*** -0.017 0.8096*** 0.2584**
L1_US -0.057 0.859*** -0.053 0.400*** 0.017 0.711***
N 84 84 84 84 84 84
R-Squared 0.382 0.663 0.428 0.145 0.735 0.591
Panel A presents the multivariate panel vector autoregressive models
results of corporate governance spillovers between the G7 countries and the US. There
are six dimensions of corporate governance spillovers. There are the coefficients of lag
term ( ) of equation , 1 ,
1
p
it i t it ij i t j it
j
CG CG CG u − −
=
= + + + in the table 2.13. There are two
equations of the spillover tests in XTVAR. The first equation has the WGIs of the G7
countries as a dependent variable and a lagged dependent variable and the US’s WGI
as the independent variables. The second equation uses the US’s WGI as a dependent
variable and the same set of independent variables. The ***, **, and * indicate
statistical significance at the 1%, 5%, and 10% level, respectively.
Ref. code: 25605502310021IBC
33
Table 2.13 Continued
Panel B: PIGS
Voice and Accountability Political Stability Government Effectiveness
PIGS US PIGS US PIGS US
PIGS
L.PIGS 0.6839*** 0.0071 0.7509*** -0.2830** 0.6170*** 0.1648***
L.US 0.1774* 0.8418*** 0.0584 0.5660*** 0.4715*** 0.4472***
N 70 70 70 70 70 70
R-Squared 0.6225 0.7642 0.6726 0.3881 0.6856 0.4567
Regulatory Quality Rule of Law Control of Corruption
PIGS US PIGS US PIGS US
PIGS
L.PIGS 0.5559*** 0.1860** 0.8564*** -0.1122** 0.7922*** 0.2453**
L.US 0.5196*** 0.7503*** 0.0059 0.3199** 0.1314** 0.6495***
N 70 70 70 70 70 70
R-Squared 0.6851 0.6843 0.699 0.2124 0.8052 0.5963
Panel B presents the multivariate panel vector autoregressive models
results of corporate governance spillovers between the PIGS countries and the US.
There are six dimensions of corporate governance spillovers. There are the coefficients
of lag term ( ) of equation , 1 ,
1
p
it i t it ij i t j it
j
CG CG CG u − −
=
= + + + in the table. There are
two equations of the spillover tests in XTVAR. The first equation has the WGIs of the
PIGS countries as a dependent variable and a lagged dependent variable and the US’s
WGI as the independent variables. The second equation uses the US’s WGI as a
dependent variable and the same set of independent variables. The ***, **, and *
indicate statistical significance at the 1%, 5%, and 10% level, respectively.
Ref. code: 25605502310021IBC
34
Table 2.13 Continued
Panel C: BRICS
Voice and Accountability Political Stability Government Effectiveness
BRICS US BRICS US BRICS US
L.BRICS 0.7263*** 0.0158 0.3024*** -0.0487 0.6456*** -0.0183
L.US 0.0567 0.8398*** 0.1089* 0.5547*** 0.0601 0.6015***
N 70 70 70 70 70 70
R-Squared 0.6723 0.7644 0.2301 0.332 0.3799 0.384
Regulatory Quality Rule of Law Control of Corruption
BRICS US BRICS US BRICS US
L.BRICS 0.5411*** 0.0512 0.6503*** 0.0591 0.6510*** 0.3429**
L.US 0.1858** 0.8562*** 0.1948 0.3728*** 0.1326** 0.6722***
N 70 70 70 70 70 70
R-Squared 0.4776 0.6612 0.4702 0.1583 0.5801 0.5932
Panel C presents the multivariate panel vector autoregressive models
results of corporate governance spillovers between the BRICS countries and the US.
There are six dimensions of corporate governance spillovers. There are the coefficients
of lag term ( ) of equation , 1 ,
1
p
it i t it ij i t j it
j
CG CG CG u − −
=
= + + + in the table 2.13. There
are two equations of the spillover tests in XTVAR. The first equation has the WGIs of
the BRICS countries as a dependent variable and a lagged dependent variable and the
US’s WGI as the independent variables. The second equation uses the US’s WGI as a
dependent variable and the same set of independent variables. The ***, **, and *
indicate statistical significance at the 1%, 5%, and 10% level, respectively.
The estimated results of Panel VARs between CG index of PIGS
countries and CG index of US and the estimated results of Panel VARs between CG
index of BRICS countries and CG index of US indicated the dependency of CG index
of PIGS and BRICS countries on the aspects of Government Effectiveness, Regulatory
Quality, and Control of Corruption, on the CG index of the US. The results of Panel
VARs between CG index of US and those of G7 should insignificant interdependent
and dynamic relationship between the two variables. These results can be implied that
Ref. code: 25605502310021IBC
35
US has dominated over PIGS and BRICS, but less impact on G7 countries. As a result,
corporate governance of US has spread to PIGS and BRICS countries on the aspects of
Government Effectiveness, Regulatory Quality, and Control of Corruption. According
to both Panel cointegration test and Panel VARs, the results partially confirm that
financial integration leads to corporate governance spillover.
2.5 Conclusion
This study intends to reveal cross-country corporate governance spillover.
The worldwide corporate governance indicators (WGI) is a proxy for corporate
governance practices spillover. This There are six dimensions of governance: Voice and
Accountability, Political Stability and Absence of Violence/Terrorism, Government
Effectiveness, Regulatory Quality, Rule of Law, and Control of Corruption. According
to fundamental economic and financial integration, cross-border M&A, and cross-
listing among countries, a positive and negative effect of corporate governance
spillover cause an increasing and decreeing in corporate governance (Martynova &
Renneboog, 2008). There are empirical evidences in prior study positive spillover by
law hypothesis and the bootstrapping hypothesis (Lim, Brooks, & Hinich, 2008;
Martynova & Renneboog, 2008). The study employs panel data technique in determine
corporate governance spillover among countries within economic group. Based on
panel unit root tests, the testing results show stationary panel time series of CG index
among US paring with G7, BRICS, and PIGS, thus, the hypothesis that “there exist
corporate governance spillover among economic countries,” is confirmed. However,
the results of panel cointegration tests and panel VARs partially (not all aspects)
confirms the hypothesis that “there exist corporate governance spillover among
economic countries.
Ref. code: 25605502310021IBC
36
CHAPTER 3
AGENCY COST IN STATE-OWNED ENTERPRISES :
INTERNATIONAL EVIDENCE
3.1 Introduction
The state-owned enterprise (SOE) has the important role in society and
economy. SOE is used as political tools or political ideology. The objectives of SOE
are political control, society and culture, public goods, government financing, etc.
However, the performance and efficiency of SOE have the arguments in empirical
evidence. Corporate governance in SOE is doubtful. Researches in SOE are mainly in
industry level, region, group of countries such as transitions economy, especially in
China. The issues mostly study on the relationship between privatization firms and
performance.
The state-owned enterprise is one type of ownership structure of the firm.
The definitions of SOE depend on law of the country and objective of study. Consistent
with Megginson & Netter (2001), this study follows the World Bank (1995) that defines
SOE as "government-owned or government-controlled economic entities that generate
the bulk of their revenues from selling goods and services. Furthermore, OECD defines
SOE as “enterprises where the state has significant control through full, majority, or
significant minority ownership” (PwC, 2015). Moreover, Robinett (2006) indicates that
SOE is parastatals, public enterprises, or public sector enterprises. SOE occur because
of social and economic factors.
If there are Pareto optimum and no externality, government activity is not
necessary. When there exist public goods or natural monopoly, there are free riders
which cause inefficiency for private sector to produce public goods. Therefore,
government intervention or state-ownership will solve public goods problem (Pindyck
& Rubinfeld, 2005).
SOE exists in both western and eastern economies. In ancient Greece 200-
300 B.C., for example, the deforestation by government for fuel purposes such as coal
and petroleum is present (Hughes & Thirgood, 1982). In ancient China 221 to 206 B.C.,
Ref. code: 25605502310021IBC
37
the absolute monarchy, Qin dynasty government provides public goods which are the
national defense by building the Great Wall of China (Lattimore, 1937). Under the
political system in ancient era, the feudalism is the reason of doing government activity.
SOE is important for economy as well as political and social benefits that government
has to produce public goods to solve market failure (Megginson & Netter, 2001). Until
the Industrial Revolution (Jensen, 1993), there is an increase in competitive market.
Then, the concept of competitive economy without social or political goals influences
SOE to become inefficient allocations. The performance of SOE is less than private
firm under property rights theory of the firm (Boardman & Vining, 1989). Importance
of SOE is measured by Country SOE Share, combining total assets (Kowalski, Buge,
Sztajerowska, & Egeland, 2013).
Literature about SOE is close to privatization studies. SOE has unique
board structure, ownership structure, board composition. Moreover, SOE is subject to
different rules and regulations. Mostly, studies on SOE are about assessment of
efficiency of SOE or performance, stock pricing, performance, the impact of
privatization (change in ownership structure) and compare to private ownership. The
summary of empirical studies in group of countries, industry, on comparing firms
before and after privatization are in Megginson, (2005a); Megginson & Netter (2001).
Literature of SOE on firm performance in natural monopoly industry such as utilities,
energy and regulated market in banks, airline (Belloc, 2014). SOE in transition
economy is also important for developing countries such as China (Zhong, 2015) and
India (SOM, 2013).
Prior literature uses the elements of corporate governance as a corporate
governance proxy, but this study uses constructed corporate governance index. The
objectives of this study are to test whether SOE structure affects corporate governance
index, SOE structure affects returns, and SOE structure affects firm performance. The
main contribution is that this study is the first one in corporate governance centric in
SOE.
This study is organized as follows; section 2 theoretical framework about
SOE, section 3 literature review of SOE and corporate governance. Section 4 describes
the data and methodology. Section 5 provides empirical results, and section 6 contains
summary and conclusion.
Ref. code: 25605502310021IBC
38
3.2 Theoretical framework
3.2.1 Theory of SOE
The difference of SOE and private firms is ownership structure. Peng,
Bruton, Stan, & Huang (2016) summarize the theories that explain SOE which consists
of property right theory, transaction cost theory of the firm, agency theory, and
resource-based theory. These theories support and against the existence of SOE.
Property right theory involves the right of owner in term of the claim of firm’s income,
control and transfer or sale of property. Transaction cost theory supports SOE according
to economy of scales under good governance condition. Agency cost occurs when
principal and agent have conflicts of interests and this problem is important in SOE.
Ownership structure is also important in previous studies on property rights and
transaction cost theory (Williamson, 1985).
The property right theory is based on the theory of production and
exchange. The traditional analysis of decision making by manager regarding to
maximize shareholder profit must change according to a new utility function of
manager or decision maker agent. Moreover, the more efficient resources allocation are
in the private-ownership (Furubotn & Pejovich, 1972). Literature on property right
theory is to assess the efficiency of resources allocation by measuring cost efficiency.
The studies compare production efficiency in public and private firms in many
industries. The study in water utility industry in US uses Cobb-Douglas production
function to estimate marginal productivity of firm and modifies into cost function to
compare operating of private and public firms (Crain & Zardkoohi, 1978). Meanwhile,
in line with the property theory, public firm performs less efficiently than private firm
(Frech III, 1976). Transaction cost and agency theories focus on unit analysis of
individuals while property right theory mainly focuses on institution of utilization and
social welfare. The comparison of these theories are in an example of oil field utilization
that is comprehensive in contracts and analysis of ownership (J. Kim & Mahoney,
2005).
Ref. code: 25605502310021IBC
39
3.3 Literature review
3.3.1 SOE and performance
Literature of privatization until 2000 focuses on privatization effect to
performance of firm. The studies investigate in industry across countries such as utility,
airline, telecommunication, financial, especially in comparing pre and post
privatization. There are SOE studies in developed country such as OECD counties, US,
UK, Mexico, and developing countries such as China and South America. These studies
mostly focus on region or industry because of data limitation and comparable reason.
The objective of SOE also concerns about social and political issues. Therefore, the
comparison among firms has many concerns about social benefit more than economic
benefit. For example, banking industry, La Porta, Lopez-de-Silanes, Shleifer, &
Vishny (2000) study performance of SOE banks in 92 countries. The board structure of
SOE affects SOE performance in Lithuania (Jurkonis & Petrusauskaitė, 2014; Jurkonis
& Aničas, 2015). Moreover, SOE structure affects growth of the firms (Khuong, 2015).
According to, Megginson & Netter (2001), survey on privatization literature suggests
that recent studies mostly are in developed country, OECD, and some in transitory
economy, but little in ASEAN country.
The ownership structure is used as corporate governance proxy. The
ownership structures are categorized into block holders’ activity (Becht, 1999; Cueto
& Switzer, 2013; Edmans, Fang, & Zur, 2013; Sakawa, Ubukata, & Watanabel, 2014),
institutional investor (Cheung, Chung, & Fung, 2015), foreign (Jackson, 2013; Sakawa
et al., 2014), and family firm (Fu, Lu-Andrews, & Yu-Thompson, 2015; Jackson, 2013)
while others studies use transparency as a corporate governance (W. P. Chen, Chung,
Lee, & Liao, 2007; Li, Chen, & French, 2012).
3.3.2 SOE by industry
Prior literature focuses on country level or across country in the same
industry. The main industries of SOE are bank, utility, and airline have empirical
evidence as follows.
Banking sector is not likely to be a natural monopoly, but it is an
oligopoly or nearly competitive market in some countries. The purposes of government
Ref. code: 25605502310021IBC
40
ownership in banks are to facilitate country development and special purpose for special
unit of economy. Country development and economy stability are usually the main
objectives of SOE. Therefore, performance of banks has to measure not only as private
banks, but also measure other output such as country development and growth of
economy.
Empirical evidence about bank in country level shows that the more
government ownership leads to lower country development in 92 countries (La Porta,
Lopez de Silanes, & Shleifer, 2002) and finds the same effect in the former Soviet
Union and in Russia (Berkowitz, Hoekstra, & Schoors, 2014).
The survey of empirical literature examining bank privatization before
and after privatized finds that privatized banks outperform after privatization
(Megginson, 2005). Mohsni & Otchere (2014) find a decrease in risk taking behavior
after privatization and the U-shaped relationship between private ownership and risk
taking. Stated-owned bank has lower performance than domestic and foreign banks
(Berger, Clarke, Cull, Klapper, & Udell, 2005).
Literature that focuses on the relationship of corporate governance of
SOE in banking industry to many measures of outputs of bank. Governance in country
level affects performance of banks and privatized banks outperform normal banks,
especially in developing countries and good national governance (Ho, Lin, & Tsai,
2016). Other important factors for banks are risk taking behavior, country governance
effect and risk taking of privatized banks. There is the U-shaped relationship between
risk taking and ownership concentration (Williams, 2014). Other issues in SOE are
ownership structure effect, effective tax rate (Zhang, M, Zhang, & Yi, 2016), and
corporate financial fraud (G. Chen, Firth, & Rui, 2006).
3.3.3 Hypothesis development
According to property rights theory, ownership power is how
economic agent has rights to control firm (Peng et al., 2016). The studies of ownership
structure show that agency cost is cost of separation of ownership and control (Fama &
Jensen, 1983; Jensen & Meckling, 1976) and this problem also occurs in SOE.
However, SOE is special case of separation of ownership and control. SOE firms are
object to different market conditions, monopoly market, and board composition. SOE
Ref. code: 25605502310021IBC
41
has both positive and negative effects to performance of firms. The negative effect of
SOE is caused by inefficiency in assets allocation (Boardman & Vining, 1989). There
is positive effect of SOE to performance, political, connections and economy of scale.
Therefore, ownership structure affects behavior of manager, especially on SOE
(Grosman et al., 2016). Hypothesis is defined as follows;
H1: Performance of SOE firm is lower than that of private firm.
3.4 Data description and research methodology
3.4.1 Data
This study uses listed SOE which have more than 25% of government
ownership classified by Orbis Bureau van Dijk ownership. Financial data are retrieved
from Thomson Reuters Eikon, DataStream and Orbis Bureau van Dijk. The criterion is
an ultimate owner of SOE obtained from Orbis Bureau van Dijk ownership data with a
total number of 134,463 companies (both listed and non-listed companies) between
2000 and 2017. The listed companies consist of 39 countries. The performance
efficiency proxies are ROA, ROE, TobinQ, and stock index return (Boardman &
Vining, 1989; Megginson, Nash, & Randenborgh, 1994). Control variables in this study
include stock market capitalization, long-term debt, short-term debt, cash holdings,
profit, book-to-market (Lins, Servaes, & Tamayo, 2017) and price inverse.
3.4.2 Research methodology
3.4.2.1 Difference-in-differences (DiD)
This study uses difference-in-difference (DiD) method
(Jiraporn, Jumreornvong, Jiraporn, & Singh, 2015) to determine effects of SOE on
independent variables. The difference-in-difference effect is determined by coefficient
of dummy variables in regression analysis. Following, Boardman & Vining (1989),
Megginson et al. (1994), and Lins, Servaes, & Tamayo (2017) this study employs
regression analysis for SOE performance. Moreover, this study includes financial crisis
dummy variable to capture performance of SOE in crisis period. The full equation as
follows;
, , , , , ,i t i t i t SOE it D i t i t SOE i t XY SOE D SOE D X = + + + + (10)
Ref. code: 25605502310021IBC
42
where
, 1 , 2 ,it i t i t i tD Crisis Law Law = (11)
and
, 1 , 2 , 3 , 4 , 5 , 6 , 7 , , , ,i t i t i t i t i t i t i t i t i t i t i tX X X X X X X X Ind Country Region = (12)
Yi,t refers to matrix of firm performances that are returns, ROA,
ROE, return, and TobinQ. i represents an individual firm and t is monthly and annually
data. SOE refers to dummy variable of state-owned enterprise of firms that its value
equals to 1 if firm has government ownership greater than 25 percent, and 0 otherwise.
Di,t is matrix of dummy variables of interest which consists of crisis, civil law, and .
common law. Crisis refers to time dummy variable that its value equals to 1 if year
equals to 2008 - 2009, and 0 otherwise. Law dummy variables consist of civil law,
common law, and mixed law system of country, this study take mixed law system as a
based dummy variable according to the least number of country and to focusing on civil
and common law. Moreover, this study investigates the difference-in-difference (DiD)
method which employ interaction terms. , ,i t i tSOE D , is the interaction terms between
SOE dummy variables and crisis and law system. These variables are classified as the
treatment effects.
For control variables, 1 , 2 , 3 , 4 , 5 , 6 , 7 ,i t i t i t i t i t i t i tX X X X X X X is
matrix of control variables that effect liquidity and represent firm financial health (Lins
et al., 2017; Prommin et al., 2014; Prommin, Jumreornvong, Jiraporn, & Tong, 2016),
the control variables as follow. MarketCap is the logarithm of stock market
capitalization of firms. LongDebt and ShortDebt are long-term debt and short-term debt
divided by total assets. CashHoldings is cash holdings divided by total assets. Profit is
operating income divided by total assets. Book2Market is book value of equity divided
by market value of equity. 1/P is one divided by price of stock. This study also includes
dummy variables for control. IndustryDummy, CountryDummy, RegionDummy, and
YearDummy refer to the dummy variables of industry, country, region, and year,
respectively.
Ref. code: 25605502310021IBC
43
3.4.2.2 Matching firm method
According to unique characteristics of SOE in each industry
and country, there are limitations of data, SOE consist of very unique characteristics of
the firm. The estimations of effects of SOE have to estimate treatment effect (SOE
firms) and control effect (private firms). Matching method is frequency used in labor
economics and clinical experiment studies (Heinrich, Maffioli, & Vázquez, 2010). This
study uses matching methods to determine effects of SOE in nonexperimental causal
study. This study considers many dimensions of observable characteristics. There are
many matching methods that this study uses propensity score-matching method because
it is appropriate to estimate treatment effect in nonexperimental study (Dehejia &
Wahba, 2002). This study follows propensity score matching in Ho, Lin, & Tsai (2016).
The matching methods compare groups of the firms by matching the set of
characteristics of the firm. For example, treatment (SOE firm) and control (private firm)
have the similar set of characteristics which are considered random sampling (Ho et al.,
2016). The treatment effect is the difference between two groups.
The propensity score-matching reduce the dimension of
characteristics of firm into one probability to reduce bias in matching method
(Rosenbaum & Rubin, 1983). The score helps in matching the same firm to compare
the effect of treatment. The probability score as follows;
( ) ( )1SOEP D F X= = , (13)
where SOE refers state-owned enterprises dummy variable that its value equals to 1 if
firm has government ownership greater than 2007, and 0 otherwise. The ( )F X is
the cumulative probability density function of logistic or probit distribution of X set of
characteristics of firm. The set of characteristics which important in this study are
government ownership, stock market capitalization, profit, region, law system, and
industry. The outcome of treatment in this study are firm performance which consists
of ROA, ROE, TobinQ, and return. The sample are both government and non-
government ownership. The model selects non-government ownership firm that have
the same score as government firm to compare effect of government ownership to firm
performance.
Ref. code: 25605502310021IBC
44
The perfectly matched firm to compare performance of SOE
and non-SOE cannot be performed in real world. Therefore, theoretical comparing the
effect of SOE is possible by assuming non-SOE characteristics that likely to be SOE.
ATT is the average effect of treatment on nonSOE that match characteristics with SOE.
The probability of set of characteristics are used in matching firm. Then the estimates
of mean difference of firm’s performance between government ownership and
nongovernment ownership firms calculate as follow;
ATT = E[YSOE - Y NonSOE] (14)
where Y(1) is performance of SOE firm, Y(0) is performance of non-SOE firm.
3.5 Empirical results
Table 3.1 shows descriptive statistics of number of firms in each country
that have SOE. There are 32 countries in this study during 2000-2017. The annually
data consists of 18 years of total 95,976 observations. The highest number of SOE is
China, which accounts more than 50 percent of all SOE, and account for about 1/3 of
number of listed firms in their country. The listed SOE are mostly in developing
country, for example, India, Bermuda, and Malaysia. The detail of firm by industry are
in table 3.2. There are eighth industries in this study; agriculture, mining, construction,
manufacturing, transportation, wholesale trade, retail trade, and services. This study
excludes finance industry. Manufacturing industry is the highest number of firms in this
sample.
Table 3.3 show descriptive statistics of variables of all sample, non-SOE,
and SOE samples in panel A, B, and C respectively. Panel C shows mean different of
variables in this study between non-SOE and SOE. The results form panel C indicate
that ROA, ROE, TOBINQ, return, and stock market capitalization of non-SOE are
statistically significant less than SOE. Table 3.4 shows correlation relationship of all
variables in this study.
Ref. code: 25605502310021IBC
45
Table 3.1 Number of firms in each country that have SOE
No. Country Non-SOE SOE Total
1 CHINA 11,913 5,700 17,613
2 INDIA 4,503 494 4,997
3 BERMUDA 4,180 418 4,598
4 MALAYSIA 4,769 342 5,111
5 HONG KONG 1,273 342 1,615
6 BRAZIL 1,672 285 1,957
7 SINGAPORE 2,280 209 2,489
8 GERMANY 4,655 171 4,826
9 ITALY 1,634 133 1,767
10 NEW ZEALAND 475 133 608
11 THAILAND 3,230 114 3,344
12 INDONESIA 1,881 114 1,995
13 RUSSIAN FEDERATION 152 114 266
14 KOREA, REPUBLIC OF 8,626 95 8,721
15 FRANCE 5,073 95 5,168
16 SWITZERLAND 1,691 95 1,786
17 PAKISTAN 1,026 95 1,121
18 SOUTH AFRICA 1,805 76 1,881
19 GREECE 1,444 76 1,520
20 NORWAY 684 76 760
21 AUSTRIA 418 76 494
22 EGYPT 95 76 171
23 BELGIUM 931 57 988
24 ARGENTINA 684 57 741
25 CAYMAN ISLANDS 684 57 741
26 POLAND 627 57 684
27 HUNGARY 95 57 152
28 JAPAN 39,596 38 39,634
29 SWEDEN 1,729 38 1,767
30 CHILE 1,330 38 1,368
31 FINLAND 1,216 38 1,254
32 CZECH REPUBLIC 38 38 76
33 UNITED KINGDOM 6,631 19 6,650
34 AUSTRALIA 4,180 19 4,199
35 TURKEY 1,596 19 1,615
36 NETHERLANDS 1,140 19 1,159
37 IRELAND 551 19 570
38 VENEZUELA 19 19 38
39 MONACO 0 19 19 Total 124,526 9,937 134,463
Ref. code: 25605502310021IBC
46
Table 3.2 Number of observations by industry
Industry Freq. Percent
Agriculture 1,672 1.24
Mining 4,788 3.56
Construction 6,346 4.72
Manufacturing 78,926 58.7
Transportation 13,433 9.99
Wholesale Trade 7,600 5.65
Retail Trade 7,505 5.58
Services 14,193 10.56
Total 134,463 100
Table 3.3 Descriptive statistics of variables
Panel A: All sample
Variable N Mean S.D. Min Max
ROA 131,905 0.0131 3.6393 -1319.9000 8.2564
ROE 120,870 4.9307 31.6785 -3043.6800 456.1500
TOBINQ 130,661 1.5217 3.8539 -4.5683 1090.7330
RETURN 125,629 0.1702 0.6254 -1.0000 29.6484
MARKET CAP 133,428 5.2691 1.9657 -4.6052 12.2822
LONG-TERM DEBT 131,217 0.2962 13.9615 0.0000 1226.1130
SHORT-TERM DEBT 130,296 0.3865 20.6741 0.0000 1825.8590
CASH HOLDDINGS 131,854 0.2125 2.8238 0.0000 433.8000
PROFIT 131,786 0.0433 7.1199 -2561.6000 62.8929
BOOK TO MARKET 132,257 1.0405 1.5807 -100.0000 100.0000
PRICE INVERSE 133,274 3.4023 17.0933 0.00003 1428.5710
Panel B: Non-SOE
Variable Obs Mean Std. Dev. Min Max
ROA 122,317 0.0115 3.7791 -1319.9000 8.2564
ROE 112,113 4.8094 32.0371 -3043.6800 456.1500
TOBINQ 121,434 1.5065 3.9805 -4.5683 1090.7330
RETURN 116,357 0.1651 0.6134 -1.0000 29.6484
MARKET CAP 123,621 5.1832 1.9556 -4.6052 12.2822
LONG-TERM DEBT 121,643 0.3094 14.4997 0.0000 1226.1130
SHORT-TERM DEBT 120,767 0.4042 21.4737 0.0000 1825.8590
CASH HOLDDINGS 122,271 0.2162 2.9320 0.0000 433.8000
PROFIT 122,215 0.0430 7.3933 -2561.6000 62.8929
BOOK TO MARKET 122,580 1.0650 1.6151 -100.0000 100.0000
PRICE INVERSE 123,470 3.4480 17.5888 0.0000 1428.5710
Ref. code: 25605502310021IBC
47
Table 3.3 Continued
Panel C: SOE
Variable N Mean S.D. Min Max
ROA 9,588 0.0341 0.0977 -2.4552 0.6659
ROE 8,757 6.4845 26.6170 -1212.7700 271.8200
TOBINQ 9,227 1.7223 1.3284 -2.7358 21.6776
RETURN 9,272 0.2340 0.7575 -1.0000 6.5244
MARKET CAP 9,807 6.3511 1.7607 -2.8134 11.8653
LONG-TERM DEBT 9,574 0.1286 0.5256 0.0000 49.4175
SHORT-TERM DEBT 9,529 0.1622 0.4881 0.0000 45.0472
CASH HOLDDINGS 9,583 0.1645 0.1762 0.0000 2.4132
PROFIT 9,571 0.0465 0.0926 -0.8883 1.2758
BOOK TO MARKET 9,677 0.7298 1.0017 -8.3333 33.3333
PRICE INVERSE 9,804 2.8269 8.6837 0.0002 166.6667
Panel D: Mean different of variables between Non-SOE and SOE
Variables Mean for Non-SOE Mean for SOE Diff. (NonSOE-SOE)
ROA 0.0115 0.0341 -0.0226 **
ROE 4.8094 6.4845 -1.6751 ***
TOBINQ 1.5065 1.7223 -0.2158 ***
RETURN 0.1651 0.2340 -0.0689 ***
MARKET CAP 5.1832 6.3511 -1.1678 ***
LONG-TERM DEBT 0.3094 0.1286 0.1808
SHORT-TERM DEBT 0.4042 0.1622 0.2420
CASH HOLDDINGS 0.2162 0.1645 0.0518
PROFIT 0.0430 0.0465 -0.0035
BOOK TO MARKET 1.0650 0.7298 0.3352
PRICE INVERSE 3.4480 2.8269 0.6212
Table 3.4 Correlation of variables
ROA ROE TobinQ Return Market
Cap
Long-
term
debt
Short-
term
debt
Cash
holdings Profit
Book-
to-
market
1/P
ROA 1
ROE 0.57 1
TobinQ -0.06 0.00 1
Return 0.11 0.14 0.12 1
Market Cap 0.19 0.22 0.12 0.12 1
Long-term debt 0.01 0.00 0.00 0.01 -0.01 1
Short-term debt 0.00 0.00 0.00 0.00 -0.01 0.97 1
Cash holdings 0.03 0.00 0.01 0.01 -0.02 0.90 0.89 1
Profit 0.13 0.09 0.00 0.02 0.03 0.82 0.85 0.81 1
Book-to-market -0.06 -0.05 -0.19 -0.13 -0.39 0.00 0.00 0.00 -0.02 1
1/P -0.10 -0.08 -0.01 -0.04 -0.19 0.00 0.00 0.00 -0.02 0.12 1
Ref. code: 25605502310021IBC
48
This study focuses on state-own enterprises as a proxy of agency cost in
corporate governance. Therefore, the methodology of treatment effects is employed to
test effects of SOE to performance of firm. The results of panel regression of firm
performance and government ownership firm are in table 3.5 The treatment effects in
this model are interaction term of government ownership (SOE) and crisis, and
interaction term of government ownership (SOE) and law system.
The estimated results of panel-data regression of all sample are in table 3.5.
The result shows that there are significantly and negatively of SOE dummy variable on
ROE and return. State-owned enterprise significantly less performs than private firms.
This confirms with hypothesis that government ownership firm is less perform than
private firm, and support property right theory. However, dummy variable of
SOExT2008-9, SOE during crisis period 2008 to 2009, are positively significant to
TobinQ and return. According to the results, SOE is better than private firm in crisis
period. This support the prior literature that SOE can access more funding form
government, and nature of SOE that not sensitive to change in financial crisis. For law
system in SOE, SOExCivil and SOExCommon, state-owned firm with civil law system
are positively significant to ROE and return. While state-owned firm with common law
system positive significantly to ROA, which inconclusive with prior literature that civil
law system is better than common law system in term of minority protection and
process (La Porta et al., 2000).
Ref. code: 25605502310021IBC
49
Table 3.5 Panel-data regression of all sample
ROA ROE TobinQ Return
SOE -0.0233 -6.2546** -0.1963 -0.0710***
T2008-9 -0.0070*** -1.0770*** -0.2098*** -0.2357***
SOExT2008-9 0.0075 1.5813 0.2368*** 0.1782***
Civil -0.0231*** -3.2342*** 0.1772*** -0.0565***
Common -0.0284*** -0.1129 0.4360*** 0.0278***
SOExCivil 0.0205 4.6138* 0.3507 0.0970***
SOExCommon 0.0349* 2.6231 -0.2935 0.0442
Asia & Pacific -0.0127** 0.2926 -0.3371*** -0.0796***
Europe -0.0035 0.9009 -0.3072*** -0.1154***
Market Cap 0.0204*** 5.1339*** 0.1450*** 0.0331***
Long-term debt -0.0002 0.1192*** -0.0050* 0.0006
Short-term debt -0.0029*** -0.2320*** 0.0077*** -0.0016***
Cash holdings 0.0084*** -0.0033 0.0134* 0.0016
Profit 0.0748*** 5.6615*** -0.1993*** 0.0401***
Book-to-market 0.0042*** 2.0389*** -0.1097*** -0.0312***
1/P -0.0003*** -0.0349*** -0.0018*** -0.0015***
Agriculture 0.0000 0.0000 0.0000 0.0000
Mining -0.0869*** -10.6648*** -0.1378 -0.0148
Construction -0.0025 -0.5204 -0.7675*** -0.0275
Manufacturing 0.0074 0.8256 -0.4150** -0.0188
Transportation -0.0122 -2.2029 -0.6165*** -0.0836***
Wholesale Trade 0.0105 1.7468 -0.5711*** -0.0277
Retail Trade 0.0108 3.0982 -0.4150** -0.0674***
Services -0.0093 -1.0435 -0.0999 -0.0355*
Constant -0.0624*** -22.9592*** 1.4102*** 0.1954***
N 128,412 118,144 126,859 122,352
No. group 7,077 7,077 7,077 7,075
Degree of freedom 23 23 23 23
RMSE 0.1598 27.2152 1.996 0.6103
R2_overall group 0.0613 0.0616 0.0314 0.0403
Chi-Square 8,753.64 5,467.34 2,065.29 5,140.94
p 0 0 0 0
* p < 0.05, ** p < 0.01, *** p < 0.001
Table 3.5 shows panel-data regression of all sample, SOE is dummy variable
of state-owned enterprise of firms that its value equals to 1 if firm has government
Ref. code: 25605502310021IBC
50
ownership greater than 25 percent, and 0 otherwise. T2008-9 is time dummy variable
that its value equals to 1 if year equals to 2008 - 2009, and 0 otherwise. SOExT2008-9
is an interaction term between SOE and time dummy of crisis. Civil and Common are
civil and common law systems. SOExCivil and SOExCommon are intraction terms
between SOE and law system. Asia & Pacific and Europe are region dummy variables.
Market Cap is the logarithm of stock market capitalization of firms. Long-term debt
and Short-term debt are long-term debt and short-term debt divided by total assets. Cash
holdings is cash holdings divided by total assets. Profit is operating income divided by
total assets. Book-to-market is book value of equity divided by market value of equity.
1/P is one divided by price of stock. This study also includes dummy variables for
control. * p < 0.05, ** p < 0.01, *** p < 0.001.
Table 3.6 shows significant result of SOE effect on performance in China.
There exists negative significant effect of SOE on performance on TobinQ. However,
dummy variable of SOExT2008-9, SOE during crisis period 2008 to 2009, are
positively significant to TobinQ. These results is confirmed with all sample data that at
normal period, SOE less performs than private firm. However, in crisis period, SOE
performs better than private firm.
Table 3.6 Panel regression random effects of China
ROA ROE TobinQ Return
SOE -0.0008 -0.0766 -0.2637*** -0.0122
T2008-9 -0.0005 0.0606 -0.1813*** -0.1689***
SOExT2008-9 0.0018 0.3975 0.0397 0.1144**
Market Cap 0.0012** 0.6282*** 0.1960*** 0.1029***
Long-term debt -0.0280*** -3.0749** -1.9093*** -0.0635
Short-term debt -0.0454*** -11.8781*** -1.4929*** 0.0922*
Cash holdings 0.0555*** -2.7715*** 1.1431*** 0.1081**
Profit 0.7303*** 167.6518*** -0.0752 0.6988***
Book-to-market 0.0078*** 5.5577*** -2.0701*** -0.2473***
1/P -0.0009*** -0.5961*** 0.0027 -0.0305***
Constant -0.0074* -4.1533*** 2.4118*** -0.2636***
N 16509 15193 15168 16228
No. group 927 927 927 927
Degree of freedom 10 10 10 10
RMSE 0.0435 12.9327 1.2783 0.7827
R2_overall group 0.6054 0.4215 0.2926 0.0708
Chi-Square 24129.8796 10852.883 4480.7687 1235.3358
p 0 0 0 0
Ref. code: 25605502310021IBC
51
Table 3.6 shows panel-data regression of China. SOE is dummy variable of
state-owned enterprise of firms that its value equals to 1 if firm has government
ownership greater than 25 percent, and 0 otherwise. T2008-9 is time dummy variable
that its value equals to 1 if year equals to 2008 - 2009, and 0 otherwise. SOExT2008-9
is an interaction term between SOE and time dummy of crisis. Civil and Common are
civil and common law systems. Market Cap is the logarithm of stock market
capitalization of firms. Long-term debt and Short-term debt are long-term debt and
short-term debt divided by total assets. Cash holdings is cash holdings divided by total
assets. Profit is operating income divided by total assets. Book-to-market is book value
of equity divided by market value of equity. 1/P is one divided by price of stock. This
study also includes dummy variables for control. * p < 0.05, ** p < 0.01, *** p < 0.001
Subsample panel-data regression by industry are in table 3.7, there
subsample of eighth industries. The result in Panel B: Mining industry shows that there
are significantly and positively of SOExT2008-9 on return. The result in Panel D:
Manufacturing industry, shows that there are significantly and negatively of SOE
dummy variable on ROA and ROE, except, negatively on TobinQ. The result in Panel
E: Transportation & Communications industry shows that there are significantly and
positively of SOE dummy variable on return. The result in Panel F: Wholesale Trade
industry shows that there are significantly and negatively of SOE dummy variable on
ROE. The result in Panel H: Services industry shows that there are significantly and
positively of SOExT2008-9 on return. While the results in Panel A: Agriculture, Panel
C: Construction, and Panel G: Retail Trade are not significantly SOE dummy variable
on performance.
Ref. code: 25605502310021IBC
52
Table 3.7 Panel regression random effects by industry
Panel A: Agriculture ROA ROE TobinQ Return
SOE 0.0297 -3.9599 -1.7488 -0.0844
T2008-9 -0.0137 3.2180 0.3693 -0.1988*
SOExT2008-9 -0.0033 -1.9950 1.3257 0.2486
Civil 0.0081 5.8998 -0.0360 -0.0405
Common -0.0115 14.5679 -0.7853 0.0004
Asia & Pacific 0.0011 -6.4378 -0.7030 -0.0895
Europe -0.0041 -21.3384* 0.3020 0.0284
Market Cap -0.0076** 3.7846* 0.4276** 0.0444**
Long-term debt 0.0339*** -8.1416 -1.2774* 0.0397
Short-term debt -0.0757*** -16.5967 -2.4335 0.2308
Cash holdings 0.0608*** 0.7478 -0.4670*** 0.0019
Profit 1.5408*** 65.3700*** -40.6720*** -0.0150
Book-to-market -0.0012* 0.5139 -0.0001 -0.0047
1/P -0.0003* -0.1251 -0.0050 -0.0018
Constant -0.0078 -12.3205 3.0463** 0.0504
N 1574 1431 1540 1499
No. group 88 88 88 88
Degree of freedom 14 14 14 14
RMSE 0.0936 82.9248 6.8874 0.9646
R2_overall group 0.9723 0.0293 0.3336 0.0149
Chi-Square 73670.1728 42.0803 891.7833 22.4319
p 0 0.0001 0 0.0702
Panel B: Mining ROA ROE TobinQ Return
SOE -0.0151 -0.6878 -0.4888 -0.0394
T2008-9 -0.0223 -2.7887 -0.0861 -0.2437***
SOExT2008-9 0.0260 5.5131 0.2326 0.2661*
Civil 0.0401 1.9171 0.7459* 0.0467
Common -0.0739* -7.0429* 0.0655 0.0470
Asia & Pacific -0.0110 -0.6650 0.0366 0.0335
Europe -0.0001 -2.0428 -0.0415 -0.0862
Market Cap 0.0185*** 4.1545*** 0.0681* 0.0254***
Long-term debt -0.1952*** -16.4323*** -0.6253 -0.0217
Short-term debt -0.4779*** -16.4070* -1.5541*** -0.2387*
Cash holdings -0.2047*** 36.7685*** -1.2394*** 0.3627***
Profit 0.4773*** 91.6216*** -4.8959*** 0.1884***
Book-to-market 0.0108* 4.3101*** -0.4910*** -0.0844***
1/P -0.0008*** -0.0616** -0.0011 -0.0017***
Constant -0.0201 -31.4560*** 2.2497*** 0.1102
N 4403 4059 4348 4250
No. group 252 252 252 252
Degree of freedom 14 14 14 14
RMSE 0.3406 37.8613 3.0125 0.8435
R2_overall group 0.4222 0.4259 0.2995 0.065
Chi-Square 2405.4467 2261.8522 1642.6311 294.5831
p 0 0 0 0
Ref. code: 25605502310021IBC
53
Table 3.7 Continued
Panel C: Construction ROA ROE TobinQ Return
SOE 0.0113 2.6300 0.1157 -0.0110
T2008-9 -0.0043 -0.1374 -0.0738* -0.2495***
SOExT2008-9 0.0061 1.5464 0.0480 0.1911
Civil -0.0066 0.2920 -0.0594 -0.0345
Common 0.0096 3.8652* -0.0732 0.0502
Asia & Pacific -0.0249*** -4.8348* -0.0686 -0.1567***
Europe -0.0164* -0.9472 0.0876 -0.1621***
Market Cap 0.0002 2.2802*** 0.1051*** 0.0376***
Long-term debt 0.0222* -8.4168** -0.5931*** 0.1015
Short-term debt -0.0336** -32.8546*** 0.3570** 0.1065
Cash holdings 0.0516*** 26.0092*** 1.0798*** 0.3878***
Profit 1.0657*** 188.3224*** -3.0757*** 1.1672***
Book-to-market 0.0034*** 2.4311*** -0.1056*** -0.0093*
1/P -0.0012*** 0.1042** 0.0181*** -0.0018**
Constant -0.0035 -15.1661*** 0.6928*** 0.0433
N 6138 5620 6090 5815
No. group 334 334 334 334
Degree of freedom 14 14 14 14
RMSE 0.0962 20.7305 0.8097 0.5959
R2_overall group 0.4286 0.2969 0.2045 0.07
Chi-Square 4286.3613 2146.1525 1408.4146 436.7634
p 0 0 0 0
Panel D: Manufacturing ROA ROE TobinQ Return
SOE -0.0157*** -3.7193*** 0.2152** 0.0054
T2008-9 -0.0101*** -2.0433*** -0.1933*** -0.2357***
SOExT2008-9 0.0079* 1.9125* 0.2785*** 0.1925***
Civil -0.0218*** -3.6855*** 0.2268*** -0.0732***
Common -0.0068* 1.2187* 0.4172*** 0.0241**
Asia & Pacific 0.0027 2.2030** -0.3144*** -0.0685***
Europe 0.0089* 3.2293*** -0.3068*** -0.1092***
Market Cap 0.0182*** 4.3578*** 0.1716*** 0.0331***
Long-term debt 0.0005*** 0.0525* -0.0025 0.0015*
Short-term debt -0.0007*** -0.0945*** -0.0004 -0.0012**
Cash holdings -0.0009 -0.0252 0.0141 -0.0072*
Profit 0.0172*** 2.4217*** 0.0206 0.0397***
Book-to-market 0.0045*** 1.4458*** -0.1116*** -0.0465***
1/P 0.0000 0.0188** -0.0049*** -0.0010***
Constant -0.0575*** -18.9838*** 0.7946*** 0.1951***
N 75550 69429 74486 71922
No. group 4154 4154 4154 4152
Degree of freedom 14 14 14 14
RMSE 0.0832 18.9973 1.5967 0.6041
R2_overall group 0.0619 0.0659 0.0492 0.0438
Chi-Square 3966.7952 3622.0142 1873.8353 3294.3971
p 0 0 0 0
Ref. code: 25605502310021IBC
54
Table 3.7 Continued
Panel E: Transportation
& Communications ROA ROE TobinQ Return
SOE 0.0063 -0.5590 -0.1341 0.0309*
T2008-9 -0.0013 -0.3556 -0.1841*** -0.2440***
SOExT2008-9 0.0030 -0.2894 0.0520 0.0813
Civil -0.0048 0.7208 -0.2103 -0.0160
Common -0.0064 3.5632 0.0530 0.0292
Asia & Pacific -0.0022 4.0701* -0.1030 -0.1091***
Europe -0.0108* 2.5223 -0.0964 -0.1301***
Market Cap 0.0060*** 3.2539*** 0.0959*** 0.0102***
Long-term debt -0.0021 1.5220** -0.0917*** -0.0113
Short-term debt -0.0983*** -26.9705*** 0.3356* 0.0334
Cash holdings -0.0632*** -8.1287*** 0.0879** -0.0463***
Profit 0.3934*** 54.2034*** -0.3094** 0.2749***
Book-to-market 0.0021 1.1841** -0.2267*** -0.0404***
1/P -0.0010*** -0.0508 0.0012 -0.0032***
Constant 0.0009 -17.9328*** 1.3514*** 0.2545***
N 12826 11809 12697 12276
No. group 707 707 707 707
Degree of freedom 14 14 14 14
RMSE 0.1285 36.6506 1.5114 0.5716
R2_overall group 0.1837 0.0689 0.0404 0.0395
Chi-Square 2346.1841 676.9807 339.7406 504.7295
p 0 0 0 0
Panel F: Wholesale Trade ROA ROE TobinQ Return
SOE -0.0174 -12.4644*** 0.1635 0.0102
T2008-9 -0.0085** -2.1246* -0.1328* -0.2067***
SOExT2008-9 0.0091 11.8616 -0.0208 0.1820
Civil -0.0180*** 1.0020 -0.0361 -0.0905***
Common -0.0145* 5.8445** 0.4952* 0.0149
Asia & Pacific -0.0011 2.6995 -0.0378 -0.0137
Europe 0.0091 1.9591 0.2790 -0.0145
Market Cap 0.0124*** 4.7794*** 0.0808*** 0.0368***
Long-term debt -0.0239*** -7.6613*** -0.0049 -0.0400
Short-term debt 0.0017 -0.1902 0.0166 -0.0084
Cash holdings 0.0148*** 4.2656*** 0.0224 0.0220**
Profit 0.0747*** 21.4162*** 0.0569 0.0799**
Book-to-market 0.0035*** 3.7320*** -0.1733*** -0.0394***
1/P -0.0001 0.0370 -0.0091** -0.0027***
Constant -0.0293** -28.4884*** 1.0509*** 0.1191**
N 7301 6769 7234 6960
No. group 400 400 400 400
Degree of freedom 14 14 14 14
RMSE 0.0769 26.7813 1.4297 0.5218
R2_overall group 0.0936 0.0705 0.0687 0.0523
Chi-Square 527.5153 457.0076 173.8674 383.3871
p 0 0 0 0
Ref. code: 25605502310021IBC
55
Table 3.7 Continued
Panel G: Retail Trade ROA ROE TobinQ Return
SOE 0.0076 2.4476 0.1083 0.0021
T2008-9 -0.0007 0.5873 -0.1494 -0.1655***
SOExT2008-9 -0.0143 -1.3738 0.1865 0.1513
Civil -0.0256*** -6.2216* -0.2099 -0.0419
Common -0.0174*** -0.7056 0.3400 0.0260
Asia & Pacific -0.0074 -0.1569 -0.3231 -0.1027***
Europe -0.0050 1.6012 -0.1778 -0.1490***
Market Cap 0.0028*** 1.9502*** 0.0506 0.0280***
Long-term debt -0.0316*** 6.7808** 1.1128*** 0.0164
Short-term debt -0.0687*** -14.2988*** 1.4854*** 0.1709**
Cash holdings 0.0027 5.7279** 1.3994*** 0.2382***
Profit 0.7258*** 164.7810*** 3.3705*** 0.5147***
Book-to-market 0.0032*** 2.2102*** -0.1720*** -0.0249***
1/P 0.0000 -0.0372 -0.0047 -0.0020**
Constant 0.0129 -10.2232** 1.0909*** 0.0667
N 7262 6768 7195 6916
No. group 395 395 395 395
Degree of freedom 14 14 14 14
RMSE 0.0523 15.7406 2.3225 0.5205
R2_overall group 0.5452 0.3663 0.0523 0.0511
Chi-Square 6146.5848 2699.0364 184.6681 371.4004
p 0 0 0 0
Panel H: Services ROA ROE TobinQ Return
SOE 0.0232 2.5544 -0.3606 0.0038
T2008-9 -0.0110 1.4311 -0.3913*** -0.2798***
SOExT2008-9 0.0053 1.5750 0.3029 0.3353***
Civil -0.0040 -1.6967 0.6805* -0.0412*
Common -0.0300* -4.2478 0.8267** -0.0042
Asia & Pacific 0.0066 3.1605 -1.6057*** -0.1142***
Europe 0.0072 2.0779 -1.5142*** -0.1328***
Market Cap 0.0118*** 5.6224*** 0.3517*** 0.0422***
Long-term debt 0.0133*** 1.4679*** -0.0514*** -0.0027
Short-term debt -0.1268*** -38.5138*** -0.2340*** -0.0774***
Cash holdings -0.0450*** 5.8907*** 0.2955*** 0.0632***
Profit 0.7463*** 72.5378*** -1.1362*** -0.0005
Book-to-market 0.0014 5.1060*** -0.0551*** -0.0143***
1/P 0.0005** -0.1931*** -0.0028 -0.0022***
Constant -0.0590** -32.0055*** 1.0350** 0.1270***
N 13358 12259 13269 12714
No. group 747 747 747 747
Degree of freedom 14 14 14 14
RMSE 0.2134 37.2275 2.0695 0.6128
R2_overall group 0.5153 0.2398 0.0459 0.0514
Chi-Square 12097.3244 2709.9422 622.0588 688.7519
p 0 0 0 0
Ref. code: 25605502310021IBC
56
Table 3.7 shows panel-data regression by industry. SOE is dummy variable
of state-owned enterprise of firms that its value equals to 1 if firm has government
ownership greater than 25 percent, and 0 otherwise. T2008-9 is time dummy variable
that its value equals to 1 if year equals to 2008 - 2009, and 0 otherwise. SOExT2008-9
is an interaction term between SOE and time dummy of crisis. Civil and Common are
civil and common law systems. Asia & Pacific and Europe are region dummy variables.
Market Cap is the logarithm of stock market capitalization of firms. Long-term debt
and Short-term debt are long-term debt and short-term debt divided by total assets. Cash
holdings is cash holdings divided by total assets. Profit is operating income divided by
total assets. Book-to-market is book value of equity divided by market value of equity.
1/P is one divided by price of stock. This study also includes dummy variables for
control. * p < 0.05, ** p < 0.01, *** p < 0.001
The estimated results of panel-data regression by group of law system,
civil, common, and mixed law systems, are in table 3.8. Panel A: ROA and Panel B:
ROE, the result shows that there are significantly and negatively of SOE dummy
variable on ROA, but positively significant of SOExT2008-9 in civil law system
sample.
Panel C: TobinQ, the result shows that there are significantly and
negatively of SOE dummy variable on TobinQ in common law system sample.
Moreover, there is a significantly and positively of SOExT2008-9 on TobinQ in civil
law system sample.
Panel D: Return, the result shows that there are significantly and positively
of SOE dummy variable on return, and SOExT2008-9 in civil law system sample.
While, there are significantly and negatively of SOE dummy variable on return, but
positively significant of SOExT2008-9 in mixed law system sample.
Ref. code: 25605502310021IBC
57
Table 3.8 Panel-data regression by group of law system
Panel A: ROA Civil Common Mixed
SOE -0.0059** -0.0161 -0.0058
T2008-9 -0.0072*** -0.0135* 0.0011
SOExT2008-9 0.0083** 0.0211 -0.0134
Market Cap 0.0132*** 0.0309*** 0.0117***
Long-term debt -0.0122*** 0.0049*** -0.1900***
Short-term debt -0.0306*** -0.0055*** -0.1176***
Cash holdings 0.0162*** -0.0245*** 0.0410***
Profit 0.0669*** 0.1780*** 0.1853***
Book-to-market 0.0014*** 0.0180*** 0.0042***
1/P 0.0001 -0.0005*** -0.0005***
Constant -0.0460*** -0.1645*** -0.0052
N 83271 27665 17476
No. group 4573 1543 961
Degree of freedom 10 10 10
RMSE 0.0709 0.3272 0.0861
R2_overall group 0.1025 0.1579 0.4142
Chi-Square 7091.2428 4878.6313 12041.9006
p 0 0 0
Panel B: ROE Civil Common Mixed
SOE -1.6686** -7.0870*** -3.4809
T2008-9 -1.7931*** -0.2842 1.0445
SOExT2008-9 2.2417** -0.3988 0.7699
Market Cap 3.5880*** 6.8123*** 3.4149***
Long-term debt -1.3374*** 0.1786*** -25.1891***
Short-term debt -8.7993*** -0.2543*** -27.0415***
Cash holding 3.4214*** -0.5742* 0.1596
Profit 17.9745*** 7.5181*** 54.5686***
Book-to-market 0.6500*** 4.0731*** 2.6643***
1/P 0.0479*** -0.0590*** -0.0587
Constant -15.4909*** -33.7630*** -11.7393***
N 77080 25207 15857
No. group 4573 1543 961
Degree of freedom 10 10 10
RMSE 18.2696 38.3446 40.816
R2_overall group 0.0939 0.1171 0.072
Chi-Square 6315.1928 2124.968 1070.5215
p 0 0 0
Ref. code: 25605502310021IBC
58
Table 3.8 Continued
Panel C: TobinQ Civil Common Mixed
SOE 0.1372 -0.4897*** -0.1740
T2008-9 -0.2054*** -0.2531*** -0.1199*
SOExT2008-9 0.2795*** 0.0922 0.1517
Market Cap 0.1456*** 0.1541*** 0.0942***
Long-term debt -0.0208 -0.0077*** 1.2225***
Short-term debt 0.1683*** 0.0064*** 0.0637
Cash holdings 0.0256 0.0413*** -0.0154
Profit -0.3644*** -0.1955*** -2.5107***
Book-to-market -0.0789*** -0.2245*** -0.1795***
1/P -0.0035* -0.0006 -0.0045
Constant 0.8071*** 1.2559*** 1.0366***
N 81868 27547 17444
No. group 4573 1543 961
Degree of freedom 10 10 10
RMSE 1.9789 2.0198 2.0163
R2_overall group 0.0183 0.0526 0.0369
Chi-Square 872.0866 941.7966 761.3795
p 0 0 0
Panel D: Return Civil Common Mixed
SOE 0.0382*** -0.0088 -0.0806***
T2008-9 -0.1957*** -0.3191*** -0.2837***
SOExT2008-9 0.1440*** 0.2296*** 0.2360***
Market Cap 0.0320*** 0.0212*** 0.0265***
Long-term debt -0.0336*** 0.0006 -0.1515***
Short-term debt -0.0739*** -0.0008 0.0106
Cash holdings 0.0420*** 0.0007 0.0013
Profit 0.2789*** 0.0167* 0.2671***
Book-to-market -0.0220*** -0.0520*** -0.0538***
1/P -0.0012*** -0.0013*** -0.0025***
Constant 0.0090 0.2084*** 0.1715***
N 79392 26493 16467
No. group 4573 1543 959
Degree of freedom 10 10 10
RMSE 0.5694 0.7247 0.5949
R2_overall group 0.0366 0.0411 0.0567
Chi-Square 3018.1641 1134.9412 989.8379
p 0 0 0
Ref. code: 25605502310021IBC
59
Table 3.8 shows panel-data regression by group of law system, civil, common,
and mixed law systems. SOE is dummy variable of state-owned enterprise of firms that its
value equals to 1 if firm has government ownership greater than 25 percent, and 0 otherwise.
T2008-9 is time dummy variable that its value equals to 1 if year equals to 2008 - 2009, and 0
otherwise. SOExT2008-9 is an interaction term between SOE and time dummy of crisis.
Market Cap is the logarithm of stock market capitalization of firms. Long-term debt and Short-
term debt are long-term debt and short-term debt divided by total assets. Cash holdings is cash
holdings divided by total assets. Profit is operating income divided by total assets. Book-to-
market is book value of equity divided by market value of equity. 1/P is one divided by price
of stock. This study also includes dummy variables for control. * p < 0.05, ** p < 0.01, *** p <
0.001
For the propensity score-matching, Table 3.9 shows that effect of SOE on
performance of firm by using the treatment effects and propensity-score matching method. The
set of control variables for matching are stock market capitalization, operating profit, region,
industry, and law system. ATT is average treatment effects of treatment on treated (SOE firm)
on performance. The ATE are significantly positively on performance variables except TobinQ.
Table 3.9 Effect of SOE on performance of firm by using the treatment effects and
propensity-score matching method
Variable Sample Treated Controls Difference S.E. t-stat
ROA Unmatched 0.0355 0.0253 0.0101 0.0018 5.6
ATT 0.0355 0.0241 0.0114 0.0024 4.68
ROE Unmatched 6.4260 4.8899 1.5360 0.3594 4.27
ATT 6.4260 6.3073 0.1186 0.4171 0.28
TobinQ Unmatched 1.6946 1.4541 0.2405 0.0439 5.48
ATT 1.6946 1.5696 0.1250 0.0261 4.8
Return Unmatched 0.2408 0.1641 0.0767 0.0071 10.82
ATT 0.2408 0.1911 0.0497 0.0114 4.35
SOE is dummy variable. SOE eqauls to 1 when firm is state-own enterprise.
Average Treatment Effects on the Treated (ATT) of SOE on performance of firms by using
psmatch2 model. The set of control variables for matching are stock market capitalization,
operating profit, region, industry, and law system. Number of observations equals to 96,338.
Number of SOE observation equals to 3,438.
Ref. code: 25605502310021IBC
60
3.6 Conclusion
This study focuses on state-own enterprises as a proxy of agency cost in
governance in the world context. The study uses DiD and treatment effect technique
with panel data and propensity score matching to determine effect of ownership
structure of SOE to firm performance. The treatment effects in this model are
interaction term of government ownership (SOE) and crisis, and interaction term of
government ownership (SOE) and law system. Based on DiD test, the result shows that
there are significantly and negatively SOE dummy variable on firm performance.
However, interaction term SOE during crisis period 2008 to 2009, are positively
significant to firm performance. The results support property right theory, theory of
firm, and hypothesis of this study that state-owned enterprise significantly less performs
than private firms.
Another treatment effect of SOE is law system in SOE, civil and common
law system are positively significant to ROE and return. While state-owned firm with
common law system positive significantly to ROA, which inconclusive with prior
literature that civil law system is better than common law system in term of minority
protection and process (La Porta et al., 2000). The results about law system results is
confirmed with by subsample panel-data regression by law system, that negatively
significant in SOE dummy variable but positively significant in interaction term.
In addition to treatment effect by panel-data regression, propensity score
matching is employed to investigate effect of SOE on firm performance. For the
propensity score-matching, shows that effect of SOE on performance of firm by using
the treatment effects and propensity-score matching method. After control matching for
all aspect of charateristics of firm like firm financial health, stock market capitalization,
operating profit, region, industry, law system, and governament ownership, the results
show that the ATE are significantly positively on performance variables except
TobinQ. However, in treatment effect with propensity score matching, SOE has
positive effect on firm performance. Therefore, the hypothesis that SOE firm less
perform than non-SOE firm is contrary to the results.
Ref. code: 25605502310021IBC
61
CHAPTER 4
CORPORATE GOVERNANCE AND LIQUIDITY
4.1 Introduction
According to the separation of ownership and control (Fama & Jensen,
1983) and information asymmetry, stock liquidity becomes important because it affects
value of the firm (Amihud & Mendelson, 2008). Though researchers categorize
meaning of liquidity into different terms based on areas of study such as micro and
macro levels of liquidity, its concept is not clear (Benson, Faff, & Smith, 2015). In asset
pricing, a number of previous studies document the positive relationship between
liquidity and stock return. Specifically, low information asymmetry improves liquidity
in the stock market because informed and uninformed traders are likely to know the
same set of information. Thus, high liquidity is associated with high stock return.
Moreover, corporate governance affects stock liquidity by the flow of information
asymmetry which causes adverse selection and hence affects liquidity (Glosten & Milgrom,
1985). Prior literature examines corporate governance and stock liquidity and demonstrates that
high corporate governance is associated with high liquidity (Chung et al., 2010&2012; and
Prommin et al., 2014). In particular, high corporate governance suggests high public
information that leads to low information asymmetry. Meanwhile, Lei et al. (2013) study the
relationship among stock liquidity, corporate governance, family firm, and state-owned
enterprise in China and find the consistent evidence with recent literature. However, those
studies typically use specific characteristics of corporate governance that represent all
perspectives of corporate governance rather than a general corporate governance index.
This study concentrates on a panel analysis for corporate governance and
stock liquidity in Thailand. The main objective is to assess whether corporate governance
has an impact on stock liquidity of listed firms that employs data on corporate governance
index of Thai Institute of Director (Thai IOD) and illiquidity measure of Amihud (2002).
This analysis employs Random-effects Tobit Model and Fixed-effects Quantile
Regression Model that consider the positively skew distribution of stock liquidity
measured by Amihud’s illiquidity. The sample period ranges from 2006 to 2017.
Ref. code: 25605502310021IBC
62
The main finding demonstrates the consistent evidence with recent
literature that firms with high corporate governance have high stock liquidity. The
positive impact of corporate governance on stock liquidity is more pronounced for firms
with good corporate governance scores (3-star, 4-star and 5-star). This empirical
evidence exhibits that listed firms with better corporate governance score have lower
information asymmetry that leads to an increase in their stock liquidity because
investors have more confidence in stocks of these firms and subsequently trade more
on their stocks.
This study is organized as follows. In the next section, literature review of
stock liquidity measure and the constructed corporate governance index. Section 3
describes the data and methodology. Section 4 provides empirical results, and section
5 contains summary and conclusion.
4.2 Literature review
4.2.1 Trading under information asymmetry and adverse selection
Information asymmetry is a friction in security markets. Adverse
selection theory describes an influence of asymmetric information on market liquidity
as follows. Since a lack of information transparency weakens trading’s decisions,
informed traders normally get more benefits which is on cost to counterparties. In order
to protect their interest, uninformed traders and market-makers widen bid-ask spread to
insurance the risk of missing information (Copeland & Galai, 1983; Glosten &
Milgrom, 1985). Moreover, Kavajecz (1999) studies trading activities around
information events. With an intension to manage their exposure to information opacity,
market-makers also lower bid and ask quoted sizes. Therefore, information asymmetry
deteriorates both market width and depth which are the key measures of market
liquidity.
Prior studies suggest a strong link between information asymmetry and
corporate governance practice. Leuz & Verrecchia (2000) focus on a change of
financial reporting standard in German companies. Under better disclosure
environment, firms’ information asymmetry is reduced. Diamond (1985) & Verrecchia
(2001) specify that voluntary disclosure enhances firms’ public information which
Ref. code: 25605502310021IBC
63
declines information cost and asymmetric information. Whereas, quality of voluntary
disclosure is influenced by ownership structure and board composition (Eng & Mak,
2003). Ajinkya, Bhojraj, & Sengupta (2005) and Karamanou & Vafeas (2005) examine
an effect of board structure, institutional ownership, and audit committee on quantity
and quality of earnings forecast. They indicate that information fairness can be
improved by corporate governance policy. Also, Kanagaretnam, Lobo, & Whalen
(2007) investigate quarterly announcement periods. Their results indicate that board
independence, board activity, and corporate insiders’ stock holdings lessen bid-ask
spreads which is used as a proxy of information asymmetry. Besides, (Cormier,
Ledoux, Magnan, & Aerts, 2010) and Cai, Liu, Qian, & Yu (2015) show that corporate
governance practices, such as monitoring activities and voluntary disclosures, decrease
information asymmetry. While, Elbadry, Gounopoulos, & Skinner (2015) add that
managerial monitoring is driven by level of board independence, board activeness,
executive compensation and debt financing.
On the other hand, good governance can also escalate information
asymmetry when costs of disclosure are high (Bamber & Cheon, 1998; Verrecchia,
1983). Furthermore, sophisticated investors have better information processing ability
than unsophisticated investors. The announcement, that provides new information to
both types of investors, intensifies information gap (Coller & Yohn, 1997; O. Kim &
Verrecchia, 1994; Lee, Mucklow, & Ready, 1993). Still, (Amiram, Owens, &
Rozenbaum, 2016) study analyst forecast announcements. They propose that
information from analyst forecast does not provide additional information to
sophisticated traders. But it is valuable for ingenuous ones. Consequently, analyst
forecast announcements decrease information asymmetry.
4.2.2 Corporate governance and liquidity
Considering many dimensions of governance, plenty of governance
measures are constructed by institutions and researchers(Jackson, 2013; Lei et al., 2013;
Prommin et al., 2014; Tang & Wang, 2011), for example, the Environmental, Social,
and Governance of corporate (ESG) by Thomson Reuters Corporate Responsibility
Ratings, Transparency index, Corruption index, World Governance Index (WGI) by
World Bank, and International Shareholder Services (ISS). Researchers mostly
Ref. code: 25605502310021IBC
64
constructed governance indices by employing equally weighed technique, however
other methodologies also existed (Jackson, 2013).
There are many literatures regarding an association between corporate
governance, information asymmetry, and liquidity. Diamond & Verrecchia (1991),
Welker (1995), and Healy, Hutton, & Palepu (1999) do not only demonstrate that
information disclosure diminish asymmetric information. But they also conclude that
information disclosure encourages more investors which increase securities’ liquidity.
Some researchers conclude that regulatory environment, which is external corporate
governance factor, enhance market liquidity (Bacidore & Sofianos, 2002; Brockman &
Chung, 2003; H. Chung, 2006).
Chung, Elder, & Kim (2010) is the pioneer paper that emphasis on the
impact of internal corporate governance attributes on stock liquidity. By using US data
during 2001 to 2004, they construct their own index based on 24 governance standards.
Panel regression’s results reveal that stock liquidity is significantly improved by
corporate governance policy. By employing limited corporate governance
characteristics and liquidity proxies with Malaysian listed companies in 2007, Foo &
Zain (2010) support Chung, Elder, & Kim (2010). The same conclusions are found in
155 French stocks during 2008 and 2009 (Karmani & Ajina, 2012). By using survey
data from 25 international markets during 2003 to 2010, Chung, Kim, Park, & Sung
(2012) indicate positive impact of shareholder protection right on stock liquidity. Li et
al. (2012) examine the association in Russian companies. They document that liquidity
enhances corporate governance. Tang & Wang (2011) and Lei, Lin, & Wei (2013) study
Chinese market over the period from 1999 to 2004 and 2006 to 2008. They are not only
backing Chung, Elder, & Kim (2010), but the latter also propose the effect of different
types of agency conflicts on the relationship between corporate governance and stock
liquidity. Edmans et al. (2013) employ Amihud’s liquidity measure with panel
regression. The positive relationship between liquidity and the likelihood of
blockholder formation is documented. While, Cueto & Switzer (2013) analyze the
association by using Brazil and Chile intraday data. As high concentration of ownership
structure is a proxy of weak minority shareholder protection, dominant shareholders do
not decrease market liquidity. Because the dominant owners have to maintain a low-
cost exit strategy. Jackson (2013) shows conflicting findings. 71 Caribbean firms with
Ref. code: 25605502310021IBC
65
concentrated ownership are linked to lower liquidity during 2005 to 2011. By
constructing four-dimensions corporate governance index, (Prommin et al., 2014) and
(Prommin et al., 2016) denote the similar results as Chung, Elder, & Kim (2010). The
positive relationship between governance quality and liquidity in found within firm-
level. However, they investigate the relationship by employing only 100 largest stocks
in Thailand for a small period between 2006 and 2009. On the other hand, Fu et al.
(2015) suggest that family firms have more liquidity than the others due to effective
monitoring activities and lower agency problem. As prior studies are mostly affected
by limited sample and inadequate liquidity proxies, Ali, Liu, & Su (2017) examine the
influence of corporate governance quality (CGQ) index on various types of liquidity
measure by using 1,207 Australian listed firms for the period of 2001 to 2013. Their
results show a positive relationship between corporate governance and stock liquidity.
4.2.3 Hypothesis development
This study considers information asymmetry as corporate governance
proxy that affect stock liquidity. Trading under information asymmetry leads to an
adverse selection problem of investor behavior, uninform traders , dealer (Glosten &
Milgrom, 1985).
This study focusses on Thailand which is an emerging market and
transition in corporate governance since Asian financial crisis in 1997. This study
categorizes firm by level of corporate governance ranking into groups to investigate
effect of corporate governance to stock liquidity.
H3: High corporate governance firm has more stock liquidity than low
corporate governance firm.
4.3 Data and data description
4.3.1 Data
This study uses listed company in the stock exchange of Thailand. Data
are retrieved from Thomson Reuters Eikon. The data consists of 395 companies, during
2006 – 2017. Total observation is 4,740 firm-year.
Ref. code: 25605502310021IBC
66
4.3.1.1 Corporate governance of the firms
This study uses corporate governance index by Thai Institute
of Director (Thai IOD). Data collected from Thai IOD website. This data is publicly
available and generally used by financial institution, for example mutual fund that has
objective of investment on good corporate governance.
4.3.1.2 Liquidity measure
This study uses liquidity measure follow (Prommin et al.,
2014), which is includes Illiquidity ratio (Amihud, 2002). Illiquidity ratio Amihud's
(2002) the average ratio of the daily absolute stock return to trading volume on one-
time period.
1
1/
iyD
iy iyd iydtiy
ILLIQ R VOLDD =
= (15)
where Riyd represents the return on stock i on day d of year y, VOLDiyd is the respective
daily volumet, and Diy is the number of days when data are available for stock i in year
y (Amihud, 2002).
4.3.1.3 Control variables
This study uses firm characteristic as control variables, total
assets, stock price, return of stock, stock return volatility, firm age, institutional
ownership, industry, country, and year dummy follow Chung et al. (2010).
4.3.2 Methodology
4.3.2.1 Panel Random-effects Tobit Model
Based on panel data, this study employs Panel Random-effects
linear regression model to analyze the impact of level of good corporate governance of
the firm on level of illiquidity of that particular firm (Roberts & Whited, 2011).
Following Chung et al., (2010); Lei et al., (2013), the Panel Random-effects regression
model can be stated as follows.
it it itILLIQ X u= + (16)
and it i itu = + , 1,2, ,i N= , 1,2, ,t T=
where ILLIQ is illiquidity measure that is Amihud’s Illiquidity ratio (Amihud, 2002).
Ref. code: 25605502310021IBC
67
Xit is NTx8 matrix of independent and control variables
1it it it it it it it
it
X GovIndex Return Volatility FirmSize Age TradingVolumeprice
=
(17)
GovIndexit is corporate governance index of Thai Institute of
Directors of stock i in year t. Priceit is price of stock i in year t. Returnit is return of
stock i in year t. Volatilityit is stock return volatility of stock i in year t. FirmSizeit
represents firm size which is measured by total assets of stock i in year t. Ageit is firm
age of stock i in year t. TradingVolumeit is trading volume of stock i in year t. i is
cross-sectional Random-effects of stock i. it is stochastic random error term of stock i
in year t.
However, since the dependent variable, ILLIQit (Illiquidity
ratio), has positively skew distribution with very high extreme values, random-effects
upper bound Tobit model is also applied in order to avoid biased result which is caused
by the high extreme values. Figure illustrates distribution of Amihud’s illiquidity
(ILLIQit) of the data in this study, which show extremely positively skew distribution.
Figure 4.1 Histogram of Amihud’s Illiquidity (ILLIQit)
0
.00
2.0
04
.00
6
Den
sity
0 1000 2000 3000 4000 5000Amihud
Ref. code: 25605502310021IBC
68
4.3.2.2 Panel Random-effects Tobit Model
The random-effects upper bound Tobit model can be stated as:
it it it
it
it
X u if ILLIQILLIQ
if ILLIQ
+ =
(18)
and it i itu = +
where represents the upper limit censored point.
This Random-effects upper bound Tobit model is estimated by
Maximum Likelihood estimation using Guass-Hermite Quadrature method.
4.3.2.3 Panel Fixed-effects Quantile Regression Model
Alternative model, panel fixed-effects quantile regression
model, is also applied.
( )it i it itQ ILLIQ X = + + (19)
where ( )itQ ILLIQ represents quantile of ILLIQ of stock i at year t. i is cross-
sectional fixed-effects.
This panel fixed-effects quantile regression model is estimated
by using Markov Chain Monte Carlo (MCMC) methods.
4.3.2.4 Robustness Check
In order to ensure the results of the study, nonparametric test,
Goodman & Kruskal Gamma, is employed. The test attempt to test the rank correlation
between the two ordinal variables. Since IOD’s Corporate Governance Index is
measured as ordinal level measure variable, thus, rank order correlation should also be
measured to reveal the relationship with liquidity level. Hence, to perform the tests,
ILLIQ is transformed from ratio level measure to be ordinal measure based on its
quartile, as ILLIQ_Level, which has value range from 1 to 4. Then, Goodman & Kruskal
Gamma of the rank correlation between GovIndex and ILLIQ_Level can be computed.
Goodman & Kruskal Gamma can be determined by the following formula:
Ref. code: 25605502310021IBC
69
c d
c d
N NGamma
N N
+=
− (20)
where Gamma is rank correlation, value ranges between -1 to 1. Nc is the total number
of concordant pairs. Nd is the total number of discordant pairs.
Additionally, multivariate analysis of the ordinal level measure
of Amihud’s illiquidity, ILLIQ_Level, using Random-effects Ordered Probit model is
also estimated. The model can be stated as:
it it itI X u= + (21)
and it i itu = +
1
1 2 1
1 2 3 1 2
1 2 3
Pr( _ 1) ( )
Pr( _ 2) ( ) ( )
Pr( _ 3) ( ) ( )
Pr( _ 4) 1 ( )
it it
it it it
it it it
it it
ILLIQ Level I
ILLIQ Level I I
ILLIQ Level I I
ILLIQ Level I
= = +
= = + + − +
= = + + + − + +
= = − + + +
(22)
where Iit is unobservable latent variable of ordered probit model. (.) is cumulative
normal probability distribution function. j is threshold value at level j. j = 1, 2, 3.
This Random-effects Ordered Probit model is estimated by
Maximum Likelihood estimation using Guass-Hermite Quadrature method.
4.4 Empirical results
This study collects stock data from Thomson Reuters Eikon, and corporate
governance index from Thai Institute of Directors during 2006-2017. The analysis
includes both annual and monthly data for the result verification. The data are from
total of 364 listed companies in the stock exchange of Thailand (SET).
4.4.1 Descriptive statistics
To determine whether there exists the differences between annual and
monthly data, descriptive statistics of the two data are separately shown in Panel A and
Panel B of table 4.1, respectively. illustrates descriptive statistics of Amihud’s
Ref. code: 25605502310021IBC
70
illiquidity, all data and categorized by corporate governance index of Thai Institute of
Directors (IOD), and the factors determining liquidity, including price inverse, stock
return volatility, firm age, size, turnover by volume.
Table 4.1 Descriptive statistics of stock return and liquidity measures.
Panel A: Annual Data
Variable Obs Mean Median Std. Dev. Min Max
Illiq 2,977 112.7085 3.5022 302.1050 0.0000 4783.2490
CG No-star 811 116.8941 4.1077 280.1586 0.0000 2396.8790
CG 3-star 902 155.8630 6.1194 333.8065 0.0000 4359.7980
CG 4-star 901 92.2577 3.0384 319.4313 0.0002 4783.2490
CG 5-star 363 46.8856 0.2432 183.3089 0.0001 2084.5550
1/Price 2,977 0.4894 0.1980 1.2567 0.0013 33.3333
Return Volatility 2,977 0.0553 0.0287 0.3947 0.0013 12.4128
Firm age 2,977 23.4068 23.8028 8.2900 10.8389 42.9972
Ln(Firm Size) 2,977 15.6822 15.3448 1.6965 11.2037 21.8458
Ln(Turnover by Volume) 2,977 10.8437 10.8766 4.4799 0.5596 23.5943
Panel B: Monthly Data
Variable Obs Mean Median Std. Dev. Min Max
Illiq 53,479 94.6613 0.5273 356.2681 0.0000 9881.1620
CG No-star 18,752 135.9254 0.7901 438.7916 0.0000 9881.1620
CG 3-star 15,189 104.4503 1.2660 361.5682 0.0000 7774.7830
CG 4-star 14,131 59.6603 0.3453 274.0753 0.0000 5905.4650
CG 5-star 5,407 15.5283 0.1035 92.2504 0.0000 2175.2220
1/Price 53,479 0.7366 0.1905 3.8962 0.0003 100.0000
Return Volatility 53,479 0.0956 0.0510 0.2335 0.0000 15.0625
Firm age 53,479 17.1174 17.2868 8.6402 0.0411 42.6585
Ln(Firm Size) 53,479 11.0066 11.3318 2.8721 1.2321 24.0579
Ln(Turnover by Volume) 53,479 8.9640 9.6499 3.4891 -2.3026 19.3114
Table 4.1 shows descriptive statistics of variables in this study during
2006-2017. Illiq is Amihud illiquidity ratio, all data and categorized by corporate
governance index of Thai Institute of Directors (IOD). 1/Price is an inverse of stock
price data. Return Volatility is the standard deviation of stock return. Firm age is age
Ref. code: 25605502310021IBC
71
of firm from IPO date. Ln(Firm Size) is measured by logarithm of total assets.
Ln(Turnover by Volume) is measured by logarithm of turnover by volume.
For both annual and monthly data, mean and median of Amihud’s
illiquidity are differences among different groups of corporate governance. High level
of corporate governance groups, 4-star and 5-star firms, have lower mean and median
of Amihud’s illiquidity, implying that they have higher liquidity compare to those with
lower level of corporate governance. Additionally, with the hugh differences between
mean and median of Amihud’s illiquidity, it indicates that this variable, Amihud’s
illiquidity, has positively skew distribution with very high positive extreme value (the
maximum value is very high). This positively skew distribution suggests that the
econometric model of this dependent variable should be either Panel Random-effects
Tobit Model or Panel Fixed-effects Quantile Regression Model.
Descriptive statistics of the control variables, including price inverse,
stock return volatility, firm age, firm size, and turnover by volume, show relatively
symetric distribution since their mean and median are less different with relatively
moderate standard deviation.
4.4.2 Estimated results of econometric models
The estimated results of Random-effects Linear Model, Random-
effects Tobit Model, and Fixed-effects Quantile Regression Model using annual data
and monthly data are illustrated in and table 4.2, respectively. Based on the estimated
results using annual data in table 4.2, the estimated results of Random-effects Linear
Model reveal positive significant impact of 3-star CG score on Amihud’s illiquidity,
which is opposite direction from that suggested by the theory, and insignificant negative
impacts of 4-star and 5-star. These unfavorable results might be caused by the positively
skew distribution of the dependent variable, Amihud’s illiquidity. To cope with this
problem, Random-effects Tobit Model and Fixed-effects Quantile Regression are
estimated. According to the results of Random-effects Tobit Model, 3-star CG score
indicates insignificant positive coefficient, implied no impact, while 4-star and 5-star
reports significant negative impacts on illiquidity. Negative impacts of corporate
governance are also confirmed by the estimated results of Fixed-effects Quantile
Regression Model. All corporate governance variables, 3-star, 4-star, and 5-star reveal
Ref. code: 25605502310021IBC
72
negative impacts on Amihud’s illiquidity. These results imply that good corporate
governance can lead to higher market liquidity of the listed company. This conclusion
is also confirmed by the estimated results of all three models using monthly data. As
shown in table 4.3, the estimated results of all three models illustrate negative impacts
of corporate governance index, 3-star, 4-star, and 5-star, on Amihud’s illiquidity.
Table 4.2 Estimated Results of Random-effects Linear Model, Random-effects Tobit
Model, and Fixed-effects Quantile Regression Model using Annual Data
RE-Linear RE-Tobit FE-QReg
cg3 24.8099 * 0.2229 -1.0934
cg4 -7.0201 -0.4572 ** -14.0985 ***
cg5 -0.3996 -1.3231 *** -4.9257 ***
priceinverse 17.5850 *** 0.4501 *** 20.9306 ***
volatility 49.9615 *** 0.9726 *** 29.7871 ***
age -0.7653 -0.0728 *** -0.4736 ***
lnta 2.1368 0.0753 * 1.4423 ***
lntv -30.4129 *** -1.1105 *** -11.5906 ***
Constant 410.1833 *** 18.1968 ***
sigma_u 1.7872 ***
sigma_e 3.3717 ***
N 2977 2977 2977
No Group 364 364 364
Chi-square 777.19 *** 2733.61 ***
Overall R2 0.2075
* p < 0.05, ** p < 0.01, *** p < 0.001
RE-Linear is Random-effects Linear Model. RE-Tobit is Random-
effects Tobit Model. FE-QReg is Fixed-effects Quantile Regression Model. cg3 is
dummy variable of IOD corporate governance index, value equals to 1 for 3-star and 0
otherwise. cg4 is dummy variable of IOD corporate governance index, value equals to
1 for 4-star and 0 otherwise. cg5 is dummy variable of IOD corporate governance index,
value equals to 1 for 5-star and 0 otherwise. priceinverse is an inverse of stock price
data. volatility is the standard deviation of stock return. age is age of firm from IPO
date. lnta represents size of the firm measured by logarithm of total assets. lntv
represents turnover of the stock measured by logarithm of turnover by volume.
Ref. code: 25605502310021IBC
73
Based on the estimated results of both annual data and monthly data,
all control variables do have significant impacts on Amihud’s illiquidity as suggested
by the conceptual framework. Inverse of stock price (priceinverse), volatility of stock
return (volatility), and size of the firm (lnta) have significant positive influence on
Amihud’s illiquidity. This implies that when price of stock increase or decrease,
liquidity of that stock will increase and decrease as well. Increase in volatility of stock
return will result in reducing liquidity of that stock. Smaller size firms, in term of total
asset, tend to have higher level of liquidity than those with bigger size.
Age of the firm (age) and turnover of the stock (lntv) all have
significant impacts on Amihud’s illiquidity. The results are according to what are
expected by the conceptual framework. The companies that have been listed for longer
period of time have more liquidity than those with shorter period of listing time. Higher
turnover of the stock leads to lower liquidity of that stock.
Table 4.3 Estimated Results of Random-effects Linear Model, Random-effects Tobit
Model, and Fixed-effects Quantile Regression Model using Monthly Data
RE-Linear RE-Tobit FE-QReg
cg3 -4.0607 ** -0.0453 -2.7144 ***
cg4 -2.2887 -0.2215 *** -3.8422 ***
cg5 -2.7418 -0.1668 ** -4.2833 ***
priceinverse 0.7860 *** 0.0180 *** 0.5381 ***
volatility 149.0707 *** 1.5877 *** 370.1578 ***
age -0.2912 -0.0500 *** -0.3468 ***
lnta 2.8783 *** 0.0410 *** 0.8160 ***
lntv -19.7908 *** -1.0099 *** -5.7240 ***
Constant 201.7362 *** 13.2204 ***
sigma_u 2.1887 ***
sigma_e 2.5556 ***
N 53479 53479 53479
No Group 364 364 364
Chi-square 2527.76 *** 16356.04 ***
Overall R2 0.1965
* p < 0.05, ** p < 0.01, *** p < 0.001
RE-Linear is Random-effects Linear Model. RE-Tobit is Random-
effects Tobit Model. FE-QReg is Fixed-effects Quantile Regression Model. cg3 is
Ref. code: 25605502310021IBC
74
dummy variable of IOD corporate governance index, value equals to 1 for 3-star and 0
otherwise. cg4 is dummy variable of IOD corporate governance index, value equals to
1 for 4-star and 0 otherwise. cg5 is dummy variable of IOD corporate governance index,
value equals to 1 for 5-star and 0 otherwise. priceinverse is an inverse of stock price
data. volatility is the standard deviation of stock return. age is age of firm from IPO
date. lnta represents size of the firm measured by logarithm of total assets. lntv
represents turnover of the stock measured by logarithm of turnover by volume.
4.4.3 Robustness tests
In order to confirm the finding of this study, robustness tests are also
performed. The nonparametric testing results of Gamma rank correlation between
IOD’s Corporate Governance Index (GovIndex) and Level of Illiquidity (ILLIQ_Level)
using annual data and monthly data indicate significant negative rank correlation
between these two ordinal-measured variables. Table 4.6 and table 4.7 illustrate
frequency of observations categorized by IOD’s Corporate Governance Index
(GovIndex) and Level of Illiquidity (ILLIQ_Level) and Gamma rank correlation using
annual data and monthly data, respectively. Gamma values of -0.1964 and -0.2032
imply that firms with higher level of IOD corporate governance score can be expected
to have lower level of Amihud’s illiquidity (or higher liquidity).
Table 4.4 Frequency of Firm-year Categorized by IOD’s Corporate Governance Index
(GovIndex) and Level of Illiquidity (ILLIQ_Level).
ILLIQ_Level
GovIndex 1 2 3 4 Total
No Star 158 220 213 220 811 19.5% 27.1% 26.3% 27.1% 100%
3-Star 160 217 224 301 902 17.7% 24.1% 24.8% 33.4% 100%
4-Star 230 241 247 183 901 25.5% 26.8% 27.4% 20.3% 100%
5-Star 185 73 61 44 363 51.0% 20.1% 16.8% 12.1% 100%
Total 733 751 745 748 2,977 24.7% 25.2% 25.0% 25.1% 100%
Gamma = -0.1964***
Note: *** indicates significant at 0.01.
Ref. code: 25605502310021IBC
75
Table 4.5 Frequency of Firm-month Categorized by IOD’s Corporate Governance
Index (GovIndex) and Level of Illiquidity (ILLIQ_Level).
ILLIQ_Level
GovIndex 1 2 3 4 Total
No Star 3,421 4,655 4,525 6,151 18,752 18.2% 24.8% 24.1% 32.8% 100%
3-Star 2,091 3,670 4,595 4,833 15,189 13.8% 24.2% 30.3% 31.8% 100%
4-Star 3,240 4,182 3,895 2,814 14,131 22.9% 29.6% 27.6% 19.9% 100%
5-Star 2,109 1,681 1,201 416 5,407 39.0% 31.1% 22.2% 7.7% 100%
Total 10,861 14,188 14,216 14,214 53,479 20.3% 26.5% 26.6% 26.6% 100%
Gamma = -0.2032***
Note: *** indicates significant at 0.01.
To reconfirm the robustness test of the rank correlation, Random-
effects Ordered Probit models are estimated using annual data and monthly data.
Table 4.8 reveals the estimated results of Random-effects Ordered
Probit Model using annual data and monthly data. The significant negative estimated
results of coefficients of corporate governance dummy variables (cg3, cg4, and cg5)
confirm that higher level of corporate governance score lead to lower level (rank) of
Amihud’s illiquidity.
Ref. code: 25605502310021IBC
76
Table 4.6 Estimated Results of Random-effects Ordered Probit Model using Annually
Data and Monthly Data.
ILLIQ_Level Annual Data Monthly Data
cg3 0.1079 -0.0499 ***
cg4 -0.0423 ** -0.0533 **
cg5 -0.4179 *** -0.1383 ***
priceinverse 0.1218 *** 0.0521 ***
volatility 0.4347 *** 0.5012 **
age -0.0263 *** -0.0394 ***
lnta -0.1300 *** -0.0405 ***
lntv -0.3884 *** -0.6028 ***
1 -8.1586 *** -7.7730 ***
2 -6.6801 *** -5.9117 ***
3 -5.1983 *** -3.9149 ***
sigma_u 0.5391 *** 0.8298 ***
N 2977 53479
No Group 364 364
Log-likelihood -2584.96 -34739.76
Overall Chi-square Test 1587.47 *** 19072.99 ***
Chi-square-Bar 237.34 *** 8295.53 ***
* p < 0.05, ** p < 0.01, *** p < 0.001
cg3 is dummy variable of IOD corporate governance index, value
equals to 1 for 3-star and 0 otherwise. cg4 is dummy variable of IOD corporate
governance index, value equals to 1 for 4-star and 0 otherwise. cg5 is dummy variable
of IOD corporate governance index, value equals to 1 for 5-star and 0 otherwise.
priceinverse is an inverse of stock price data. volatility is the standard deviation of stock
return. age is age of firm from IPO date. lnta represents size of the firm measured by
logarithm of total assets. lntv represents turnover of the stock measured by logarithm
of turnover by volume.
According to the robustness tests results in table 4.4, table 4.5, and
table 4.6, these ordinal level measure of illiquidity analyses also help confirm impacts
of good corporate governance on liquidity of the stock. In addition, table 4.7 shows
average change of Amihud’s illiquidity (ILLIQ) caused by one level change in IOD’s
corporate governance index (GovIndex) during two sub-periods (2007-2011 and 2012-
2017).
Ref. code: 25605502310021IBC
77
Table 4.7 Descriptive Statistical Indices of Change of Amihud’s Illiquidity (ILLIQ)
After Change in IOD’s Corporate Governance Index (GovIndex) during 2007-2011 and
2012-2017
Period 2007-2011 2012-2017
CG-Change 0 → 3 3 → 4 4 → 5 0 → 3 3 → 4 4 → 5
Firm-year (# obs.) 74 70 32 96 119 71
Mean 102.663 3.180 0.930 -20.514 -35.417 -4.455
Median 67.553 -1.981 -1.563 -0.008 -0.001 -0.001
Std. Dev. 446.413 459.553 373.120 75.107 131.842 22.906
Minimum -972.357 -1723.934 -569.291 -508.400 -702.241 -146.127
Maximum 1187.658 1390.975 1833.472 12.596 13.193 5.623
0 → 3 represents the case that the firm IOD corporate governance score
increases one level from no-star to 3-star. 3 → 4 represents the case that the firm IOD
corporate governance score increases one level from 3-star to 4-star. 4 → 5 represents
the case that the firm IOD corporate governance score increases one level from 4-star
to 5-star.
During the first sub-period 2007-2011, the cases of one level change
in IOD corporate governance score, including no-star to 3-star, 3-star to 4-star, and 4-
star to 5-star, have positive mean (increase) in Amihud’s illiquidity (102.663, 3.180,
and 0.930) with very high standard deviation (446.413, 459.553, and 373.120). Since
the median of the changes are a lot less than the mean, it implies that the magnitudes of
the changes in Amihud’s illiquidity caused by one level change in IOD corporate
governance score are positively skew distributed with high positive extreme value
(maximum).
Based on the second sub-period 2012-2017, all cases of one level
change in IOD corporate governance score, including no-star to 3-star, 3-star to 4-star,
and 4-star to 5-star, have negative mean (decrease) in Amihud’s illiquidity (-20.514, -
35.417, and -4.455) with moderately level of standard deviation (75.107, 131.842, and
22.906). The median of the changes are more than the mean, which indicates negatively
skew distribution of the changes in Amihud’s illiquidity with low negative extreme
value (minimum).
Ref. code: 25605502310021IBC
78
The opposite direction of the impacts of improving corporate
governance during the first sub-period 2007-2011 and the second sub-period 2012-2017
indicate that before 2012, impact of improving corporate governance score does not
help increasing liquidity of the stock but instead reducing liquidity. After 2012,
improving one level of corporate governance score helps increasing liquidity of the
stock. Therefore, these findings also help confirm the relationship between good
corporate governance and liquidity.
4.5 Discussion & Conclusion
This study has provided evidences of relationship between good corporate
governance and stock market liquidity. Based on information asymmetry and adverse
selection concept, listed companies can provide investors their better operating
performance information by sending signal through good corporate governance
practice (Chung, et al., 2010; Prommin et al., 2014; and Prommin, et al., 2016). Better
corporate governance score of the stocks help reducing level of information asymmetry.
Then, investors have more confident in these stocks and trade more on these stocks,
thus, trading volumes of these stocks increase as well as stocks market liquidity. Unlike
other studies, this study takes into account of the positively skew distribution of stock
market liquidity measured by Amihud’s illiquidity by employing Random-effects Tobit
Model, and Fixed-effects Quantile Regression Model. Then, the estimated results of
this study reveal the significant impacts of corporate governance measured by IOD
corporate governance index (no-star, 3-star, 4-star, or 5-star) on stock liquidity
measured by Amihud’s illiquidity, which is in line with previous studies (Chung, et al.,
2010; Prommin et al., 2014; and Prommin, et al., 2016). Thus, hypothesis of the study
is confirmed. Similar to Chung, et al. (2010), the results suggest that listed companies
can alleviate information-based trading and improve stock market liquidity by
improving their corporate governance score, which can help lower information
asymmetry problem.
By changing scale of measurement of liquidity as ordinal level, robustness
test using nonparametric rank correlation and Random-effects Ordered Probit model
also reveal the ordinal-level relationship between IOD corporate governance score and
Ref. code: 25605502310021IBC
79
level of liquidity. The findings of this study are remarkably robust to alternative
statistical tests and different scale of measurement of liquidity. Additionally, corporate
governance score improvement lead to higher level of liquidity. Based on sub-period
analysis, Thai investors seems not to value much on IOD corporate governance index
during 2007-2011 since the result indicate positive average change in Amihud’s
illiquidity. After 2011, Thai IOD reconstructed its corporate governance index, mutual
fund managers then took into account of corporate governance score of the stocks to
help forming their portfolio. As a result, corporate governance score improvement
during 2012-2017 reveals negative average change in Amihud’s illiquidity, which
imply increasing liquidity. Furthermore, similar to previous studies (Foo & Zain, 2010;
Chung, et al., 2010; Karmani & Ajina, 2012; Chung, et al., 2012; Lei, et al., 2013), the
estimated results of all regression models show significant effects of all control
variables on stock liquidity.
Ref. code: 25605502310021IBC
80
REFERENCES
Articles
Abdallah, W., & Goergen, M. (2008). Does corporate control determine the cross-
listing location? Journal of Corporate Finance, 14, 183–199. Retrieved from
https://ac.els-cdn.com/S0929119908000217/1-s2.0-S0929119908000217-
main.pdf?_tid=b22b09b7-3f07-4a5f-b769-
b3f600d1302d&acdnat=1526117115_050ade09ce904e16bc60ce28426f5bc2
Abrigo, M. R. M., & Love, I. (2016). Estimation of panel vector autoregression in Stata.
Stata Journal, 16(3), 778–804. Retrieved from http://www.stata-
journal.com/article.html?article=st0455
Aebi, V., Sabato, G., & Schmid, M. (2012). Risk management, corporate governance,
and bank performance in the financial crisis. Journal of Banking and Finance,
36(12), 3213–3226. https://doi.org/10.1016/j.jbankfin.2011.10.020
Ajinkya, B., Bhojraj, S., & Sengupta, P. (2005). The association between outside
directors, institutional investors and the properties of management earnings
forecasts. Journal of Accounting Research, 43(3), 343–376.
https://doi.org/10.1111/j.1475-679x.2005.00174.x
Ali, S., Liu, B., & Su, J. J. (2017). Corporate governance and stock liquidity
dimensions: Panel evidence from pure order-driven Australian market.
International Review of Economics and Finance, 50(March), 275–304.
https://doi.org/10.1016/j.iref.2017.03.005
Amihud, Y. (2002). Illiquidity and stock returns: cross-section and time-series effects.
Journal of Financial Markets, 5, 31–56.
Amihud, Y., & Mendelson, H. (2008). Liquidity, the Value of the Firm, and Corporate
Finance. Journal of Applied Corporate Finance, 20(2), 32–45.
https://doi.org/10.1111/j.1745-6622.2008.00179.x
Amiram, D., Owens, E., & Rozenbaum, O. (2016). Do information releases increase or
decrease information asymmetry? New evidence from analyst forecast
announcements. Journal of Accounting and Economics, 62(1), 121–138.
https://doi.org/10.1016/j.jacceco.2016.06.001
Ref. code: 25605502310021IBC
81
Bacidore, J. M., & Sofianos, G. (2002). Liquidity Provision and Specialist Trading in
NYSE-Listed Non-U.S. Stocks. Journal of Financial Economics, 63, 133–158.
Retrieved from https://ac.els-cdn.com/S0304405X01000927/1-s2.0-
S0304405X01000927-main.pdf?_tid=11b07d5b-78b0-415d-8bbb-
131a800b757b&acdnat=1525577780_a163916b166467cd421a1b7559f6bbe7
Bamber, L. S., & Cheon, Y. S. (1998). Discretionary Management Earnings Forecast
Disclosures: Antecedents and Outcomes Associated with Forecast Venue and
Forecast Specificity Choices. Journal of Accounting Research, 36(2), 167–190.
https://doi.org/10.2307/2491473
Becht, M. (1999). European corporate governance: Trading off liquidity against
control. European Economic Review, 43(4–6), 1071–1083.
https://doi.org/10.1016/S0014-2921(98)00115-9
Bekaert, G., Ehrmann, M., Fratzscher, M., & Mehl, A. (2014). The Global Crisis and
Equity Market Contagion. Journal of Finance, 69(6), 2597–2649.
https://doi.org/10.1111/jofi.12203
Belloc, F. (2014). Innovation in State-Owned Enterprises: Reconsidering the
Conventional Wisdom. Journal of Economic Issues (M.E. Sharpe Inc.), 48(3),
821–848. https://doi.org/10.2753/JEI0021-3624480311
Beltratti, A., & Stulz, R. M. (2012). The credit crisis around the globe: Why did some
banks perform better? Journal of Financial Economics, 105(1), 1–17.
https://doi.org/10.1016/j.jfineco.2011.12.005
Benson, K., Faff, R., & Smith, T. (2015). Injecting liquidity into liquidity research.
Pacific-Basin Finance Journal, 35, 533–540. https://doi.org/10.1016/j.pacfin.2015.10.001
Berger, A. N., Clarke, G. R. G., Cull, R., Klapper, L., & Udell, G. F. (2005). Corporate
governance and bank performance: A joint analysis of the static, selection, and
dynamic effects of domestic, foreign, and state ownership. Journal of Banking &
Finance, 29(8), 2179–2221. https://doi.org/10.1016/j.jbankfin.2005.03.013
Berkowitz, D., Hoekstra, M., & Schoors, K. (2014). Bank privatization, finance, and
growth. Journal of Development Economics, 110, 93–106.
https://doi.org/10.1016/j.jdeveco.2014.05.005
Ref. code: 25605502310021IBC
82
Bhagat, S., Malhotra, S., & Zhu, P. C. (2011). Emerging country cross-border
acquisitions: Characteristics, acquirer returns and cross-sectional determinants.
Emerging Markets Review, 12(3), 250–271. https://doi.org/10.1016/j.ememar.2011.04.001
Black, B., Gledson De Carvalho, A., Khanna, V., Kim, W., & Yurtoglu, B. (2017).
Corporate Governance Indices and Construct Validity (ECGI Working Paper
Series in Finance No. 483/2016). Retrieved from http://ssrn.com/abstract_id=2838273
Boardman, A. E., & Vining, A. R. (1989). Ownership and Performance in Competitive
Environments: A Comparison of the Performance of Private, Mixed, and State-
Owned Enterprises. The Journal of Law and Economics, 32(1), 1–33.
https://doi.org/10.1086/467167
Bris, A., Brisley, N., & Cabolis, C. (2008). Adopting better corporate governance:
Evidence from cross-border mergers. Journal of Corporate Finance, 14(3), 224–
240. https://doi.org/10.1016/j.jcorpfin.2008.03.005
Brockman, P., & Chung, D. Y. (2003). American Finance Association Investor
Protection and Firm Liquidity. Source: The Journal of Finance, 58(2), 921–937.
Retrieved from http://www.jstor.org/stable/3094564
Brunnermeier, M. K., & Pedersen, L. H. (2009). Market liquidity and funding liquidity.
Review of Financial Studies, 22(6), 2201–2238. https://doi.org/10.1093/rfs/hhn098
Bun, M. J. G., & Kiviet, J. F. (2006). The effects of dynamic feedbacks on LS and MM
estimator accuracy in panel data models. Journal of Econometrics, 132, 409–444.
https://doi.org/10.1016/j.jeconom.2005.02.006
Cagala, T., & Glogowsky, U. (2014). Panel Vector Autoregressions for Stata (xtvar).
Cai, J., Liu, Y., Qian, Y., & Yu, M. (2015). Information Asymmetry and Corporate
Governance. Quarterly Journal of Finance, 05(03), 1550014.
https://doi.org/10.1142/S2010139215500147
Cheffins, B. R. (2011). The History of Corporate Governance. SSRN Electronic
Journal. https://doi.org/10.2139/ssrn.1975404
Cheffins, B. R. (2012). The History of Corporate Governance. European Corporate
Governance Institute, (January). Retrieved from http://ssrn.com/abstract=1975404
Chen, G., Firth, M., & Rui, O. (2006). Have China’s enterprise reforms led to improved
efficiency and profitability? Emerging Markets Review, 7(1), 82–109.
https://doi.org/10.1016/j.ememar.2005.05.003
Ref. code: 25605502310021IBC
83
Chen, W. P., Chung, H., Lee, C., & Liao, W. L. (2007). Corporate governance and
equity liquidity: Analysis of S&P transparency and disclosure rankings. Corporate
Governance, 15(4), 644–660. https://doi.org/10.1111/j.1467-8683.2007.00594.x
Cheng, I.-H. (2011). Corporate Governance Spillovers. SSRN Electronic Journal,
(April). https://doi.org/10.2139/ssrn.1299652
Cheung, W. M., Chung, R., & Fung, S. (2015). The effects of stock liquidity on firm
value and corporate governance: Endogeneity and the REIT experiment. Journal
of Corporate Finance, 35, 211–231. https://doi.org/10.1016/j.jcorpfin.2015.09.001
Chung, H. (2006). Investor protection and the liquidity of cross-listed securities:
Evidence from the ADR market. Journal of Banking & Finance, 30, 1485–1505.
https://doi.org/10.1016/j.jbankfin.2005.03.021
Chung, K. H., Elder, J., & Kim, J.-C. (2010). Corporate Governance and Liquidity.
Journal of Financial and Quantitative Analysis, 45(02), 265–291.
https://doi.org/10.1017/S0022109010000104
Chung, K. H., Kim, J. S., Park, K., & Sung, T. (2012). Corporate governance, legal
system, and stock market liquidity: Evidence around the world. Asia-Pacific
Journal of Financial Studies, 41(6), 686–703. https://doi.org/10.1111/ajfs.12002
Claessens, S., & Yurtoglu, B. B. (2013). Corporate governance in emerging markets: A
survey. Emerging Markets Review, 15, 1–33. https://doi.org/10.1016/j.ememar.2012.03.002
Coller, M., & Yohn, T. L. (1997). Management Forecasts and Information Asymmetry:
An Examination of Bid-Ask Spreads. Journal of Accounting Research, 35(2), 181.
https://doi.org/10.2307/2491359
Copeland, T. E., & Galai, D. (1983). American Finance Association Information
Effects on the Bid-Ask Spread. Source: The Journal of Finance, 38(5), 1457–
1469. Retrieved from http://www.jstor.org/stable/2327580
Cormier, D., Ledoux, M., Magnan, M., & Aerts, W. (2010). Corporate governance and
information asymmetry between managers and investors. Corporate Governance:
The International Journal of Business in Society, 10(5), 574–589.
https://doi.org/10.1108/14720701011085553
Crain, W. M., & Zardkoohi, A. (1978). A Test of the Property-Rights Theory of the
Firm: Water Utilities in the United States. The Journal of Law and Economics,
21(2), 395–408. https://doi.org/10.1086/466927
Ref. code: 25605502310021IBC
84
Cueto, D. C., & Switzer, L. N. (2013). Intraday market liquidity, corporate governance,
and ownership structure in markets with weak shareholder protection: evidence
from Brazil and Chile. Journal of Management & Governance, 19(2), 395–419.
https://doi.org/10.1007/s10997-013-9263-8
Dehejia, R. H., & Wahba, S. (2002). Propensity Score-Matching Methods for
Nonexperimental Causal Studies. Review of Economics and Statistics, 84(1), 151–
161. https://doi.org/10.1162/003465302317331982
Diamond, D. W. (1985). Optimal Release of Information By Firms. The Journal of
Finance, 40(4), 1071–1094. https://doi.org/10.1111/j.1540-6261.1985.tb02364.x
Diamond, D. W., & Verrecchia, R. E. (1991). Disclosure , Liquidity , and the Cost of
Capital. The Journal of Finance, XLVI(4), 1325–1359.
Edmans, A., Fang, V. W., & Zur, E. (2013). The effect of liquidity on governance.
Review of Financial Studies, 26(6), 1443–1482. https://doi.org/10.1093/rfs/hht012
Elbadry, A., Gounopoulos, D., & Skinner, F. (2015). Governance Quality and
Information Asymmetry. Financial Markets, Institutions & Instruments, 24(2–3),
127–157. https://doi.org/10.1111/fmii.12026
Eng, L. L., & Mak, Y. T. (2003). Corporate governance and voluntary disclosure.
Journal of Accounting and Public Policy, 22(4), 325–345.
https://doi.org/10.1016/S0278-4254(03)00037-1
Fama, E. F., & Jensen, M. C. (1983). Separation of Ownership and Control. The Journal
of Law & Economics, 26(2), 301–325. Retrieved from http://www.jstor.org/stable/725104
Foo, Y. B., & Zain, M. M. (2010). Board independence, board diligence and liquidity
in Malaysia: A research note. Journal of Contemporary Accounting and
Economics, 6(2), 92–100. https://doi.org/10.1016/j.jcae.2010.10.001
Frech III, H. E. (1976). The Property Rights Theory of the Firm: Empirical Results
from a Natural Experiment. Journal of Political Economy, 84(1), 143--152.
https://doi.org/10.1086/260416
Fu, L., Lu-Andrews, R., & Yu-Thompson, Y. (2015). Liquidity and Corporate
Governance: Evidence from Family Firms. SSRN Electronic Journal. Retrieved
from http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2433764
Ref. code: 25605502310021IBC
85
Furubotn, E. G., & Pejovich, S. (1972). Property Rights and Economic Theory: A
Survey of Recent Literature. Journal of Economic Literature, 19(4), 1137–1162.
Retrieved from http://www.jstor.org/stable/pdf/2721541.pdf?refreqid=excelsior%
3A3f5fbd0cb5d2d4ab9a874af849cfde9d
Gibbons, R., & Murphy, K. J. (1990). Relative Performance Evaluation for Chief
Executive Officers. Industrial and Labor Relations Review, 43(3), 30S.
https://doi.org/10.2307/2523570
Glosten, L. R., & Milgrom, P. R. (1985). Bid, ask and transaction prices in a specialist
market with heterogeneously informed traders. Journal of Financial Economics,
14(1), 71–100. https://doi.org/10.1016/0304-405X(85)90044-3
Gompers, P. A., Ishii, J., & Metrick, A. (2003). Corporate Governance and Equity
Prices. The Quarterly Journal of Economics, 118(1), 107–155.
https://doi.org/10.2307/25053900
Graham, J. R., Harvey, C. R., & Rajgopal, S. (2005). The economic implications of
corporate financial reporting. Journal of Accounting and Economics.
https://doi.org/10.1016/j.jacceco.2005.01.002
Grosman, A., Okhmatovskiy, I., & Wright, M. (2016). State Control and Corporate
Governance in Transition Economies: 25 Years on from 1989. Corporate
Governance (Oxford), 24(3), 200–221. https://doi.org/10.1111/corg.12145
Healy, P. M., Hutton, A. P., & Palepu, K. G. (1999). Stock Performance and
Intermediation Changes Surrounding Sustained Increases in Disclosure.
Contemporary Accounting Research, 16(3), 485–520.
https://doi.org/10.1111/j.1911-3846.1999.tb00592.x
Heinrich, C., Maffioli, A., & Vázquez, G. (2010). A Primer for Applying Propensity-
Score Matching:Impact-Evaluation Guidelines. Technical Notes, No. IDB-TN-
161, (August), 1–56. https://doi.org/NEP-ECM-2010-10-23
Ho, P. H., Lin, C. Y., & Tsai, W. C. (2016). Effect of country governance on bank
privatization performance. International Review of Economics and Finance, 43,
3–18. https://doi.org/10.1016/j.iref.2015.10.028
Holmstrom, B. (1982). Moral Hazard in Teams. The Bell Journal of Economics, 13(2),
324. https://doi.org/10.2307/3003457
Ref. code: 25605502310021IBC
86
Hughes, J. D., & Thirgood, V. (1982). Deforestation, Erosion, and Forest Management
in Ancient Greece and Rome. Journal of Forest History, 26(2), 60–75.
https://doi.org/10.2307/4004530
Jackson, M. K. (2013). Ownership, Corporate Governance and Liquidity in Caribbean
Firms.
Jensen, M. C. (1993). The Modern Industrial Revolution , Exit , and the Failure of
Internal Control Systems the Failure of Internal Control Systems. Journal of
Finance, 48(3), 831–880. https://doi.org/10.1111/j.1540-6261.1993.tb04022.x
Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behavior,
agency costs and ownership structure. Journal of Financial Economics, 3(4), 305–
360. https://doi.org/10.1016/0304-405X(76)90026-X
Jenter, D., & Kanaan, F. (2015). CEO turnover and relative performance evaluation.
The Journal of Finance, 70(5), 2155–2184. https://doi.org/10.1111/jofi.12282
Jiraporn, P., Jumreornvong, S., Jiraporn, N., & Singh, S. (2015). How do independent
directors view powerful CEOs? Evidence from a quasi-natural experiment.
Finance Research Letters, 16, 268–274. https://doi.org/10.1016/j.frl.2015.12.008
Jurkonis, L., & Aničas, I. (2015). Impact of The Board On Management of Lithuanian
State-Owned Enterprises. EKONOMIKA, 94(3), 139–151.
Jurkonis, L., & Petrusauskaitė, D. (2014). Effects of Corporate Governance State-
Owned Enterprises. Ekonomika, 93(2), 77–97.
Kanagaretnam, K., Lobo, G. J., & Whalen, D. J. (2007). Does good corporate
governance reduce information asymmetry around quarterly earnings
announcements? Journal of Accounting and Public Policy, 26(4), 497–522.
https://doi.org/10.1016/j.jaccpubpol.2007.05.003
Karamanou, I., & Vafeas, N. (2005). The association between corporate boards, audit
committees, and management earnings forecasts: An empirical analysis. Journal
of Accounting Research, 43(3), 453–486. https://doi.org/10.1111/j.1475-
679X.2005.00177.x
Karmani, M., & Ajina, A. (2012). Market Stock Liquidity and Corporate Governance.
SSRN Electronic Journal. https://doi.org/10.2139/ssrn.2084707
Ref. code: 25605502310021IBC
87
Kaufmann, D., Kraay, A., & Mastruzzi, M. (2009). Governance Matters VIII:
Aggregate and Individual Governance Indicators, 1996-2008. World Bank Policy
Research Working Paper No. 4978. Retrieved from
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1424591
Kaufmann, D., Kraay, A., & Mastruzzi, M. (2010). Response to: “The Worldwide
Governance Indicators: Six, One, or None.” European Journal of Development
Research, 22(1), 55–58. https://doi.org/10.1057/ejdr.2009.49
Kaufmann, D., Kraay, A., & Mastruzzi, M. (2011). The Worldwide Governance
Indicators: Methodology and Analytical Issues. Hague Journal on the Rule of
Law, 3(02), 220–246. https://doi.org/10.1017/S1876404511200046
Kaufmann, D., Kraay, A., Mastruzzi, M., & Thomas, M. A. (2010). What Do the
Worldwide Governance Indicators Measure? European Journal of Development
Research, 22(1), 31–54. https://doi.org/10.1057/ejdr.2009.32
Kaufmann, D., Kraay, A., & Zoido-Lobatón, P. (1999). Governance Matters. SSRN
Electronic Journal. Retrieved from https://papers.ssrn.com/sol3/papers.cfm?
abstract_id=188568
Kaufmann, D., Kraay, A., & Zoido, P. (1999). Aggregating Governance Indicators.
SSRN Electronic Journal. https://doi.org/10.2139/ssrn.188548
Kavajecz, K. A. (1999). American Finance Association A Specialist ’ s Quoted Depth
and the Limit Order Book Published by : Wiley for the American Finance
Association Stable URL : http://www.jstor.org/stable/2697726 A Specialist ’ s
Quoted Depth and the Limit Order Book. The Journal of Finance, 54(2), 747–771.
Ke, B., Huddart, S., & Petroni, K. (2003). What insiders know about future earnings
and how they use it: Evidence from insider trades. Journal of Accounting and
Economics, 35(3), 315–346. https://doi.org/10.1016/S0165-4101(03)00036-3
Khuong, V. M. (2015). Can Vietnam Achieve More Robust Economic Growth?
Insights from a Comparative Analysis of Economic Reforms in Vietnam and
China. Southeast Asian Economies, 32(1), 52. https://doi.org/10.1355/ae32-1d
Kim, J., & Mahoney, J. T. (2005). Property rights theory, transaction costs theory, and
agency theory: An organizational economics approach to strategic management.
Managerial and Decision Economics, 26(4), 223–242.
https://doi.org/10.1002/mde.1218
Ref. code: 25605502310021IBC
88
Kim, O., & Verrecchia, R. E. (1994). Market liquidity and volume around earnings
announcements. Journal of Accounting and Economics, 17, 41–67. Retrieved from
http://cyber.sci-
hub.tw/MTAuMTAxNi8wMTY1LTQxMDEoOTQpOTAwMDQtMw==/10.101
6%400165-4101%2894%2990004-3.pdf
Kowalski, P., Buge, M., Sztajerowska, M., & Egeland, M. (2013). State-Owned
Enterprises: Trade Effects and Policy Implications. Publisher Information:OECD
Publishing, (147).
La Porta, R., Lopez-De-Silanes, F., & Shleifer, A. (1999). Corporate Ownership around
the World Corporate Ownership Around the World. Journal of Finance, 54(2),
471–517. https://doi.org/10.1111/0022-1082.00115
La Porta, R., Lopez-de-Silanes, F., Shleifer, A., & Vishny, R. (2000). Investor
protection and corporate governance. Journal of Financial Economics, 58(1–2),
3–27. https://doi.org/10.1016/S0304-405X(00)00065-9
La Porta, R., Lopez-de-Silanes, F., Shleifer, A., & Vishny, R. W. (2000). Agency
Problems and Dividend Policies around the World. The Journal of Finance, 55(1),
1–33. https://doi.org/10.1111/0022-1082.00199
La Porta, R., Lopez‐de‐Silanes, F., & Shleifer, A. (2002). Government ownership of
banks. THE JOURNAL OF FINANCE, 57(1), 265–301.
Langbein, L., & Knack, S. (2010). The worldwide governance indicators: Six, one, or
none? Journal of Development Studies, 46(2), 350–370.
https://doi.org/10.1080/00220380902952399
Lattimore, O. (1937). Origins of the Great Wall of China: A Frontier Concept in Theory
and Practice. American Geographical Society, 27(4), 529–549.
Lee, C. M. C., Mucklow, B., & Ready, M. J. (1993). Spreads, depths, and the impact
of earnings information: An intraday analysis. The Review of Financial Studies,
6(2), 345–374. https://doi.org/10.1017/CBO9781107415324.004
Lei, Q., Lin, B., & Wei, M. (2013). Types of agency cost, corporate governance and
liquidity. Journal of Accounting and Public Policy, 32(3), 147–172.
https://doi.org/10.1016/j.jaccpubpol.2013.02.008
Ref. code: 25605502310021IBC
89
Leuz, C., & Verrecchia, R. E. (2000). The Economic Consequences of Increased
Disclosure. Journal of Accounting Research, 38(May), 91–124. Retrieved from
http://www.jstor.org/stable/2672910
Levin, A., Lin, C.-F., & Chu., C.-S. J. (2002). Unit root tests in panel data: Asymptotic
and finite-sample properties. Journal of Econometrics, 108, 1–24.
Li, W.-X., Chen, C. C.-S., & French, J. J. (2012). The relationship between liquidity,
corporate governance, and firm valuation: Evidence from Russia. Emerging
Markets Review, 13(4), 465–477. https://doi.org/10.1016/j.ememar.2012.07.004
Lim, K. P., Brooks, R. D., & Hinich, M. J. (2008). Nonlinear serial dependence and the
weak-form efficiency of Asian emerging stock markets. Journal of International
Financial Markets, Institutions and Money, 18(5), 527–544.
https://doi.org/10.1016/j.intfin.2007.08.001
Lins, K. V., Servaes, H., & Tamayo, A. (2017). Social Capital, Trust, and Firm
Performance: The Value of Corporate Social Responsibility during the Financial
Crisis. Journal of Finance, 72(4), 1785–1824. https://doi.org/10.1111/jofi.12505
Marshall, B. R., Nguyen, N. H., Nguyen, T. H., & Visaltanachoti, N. (2016). Country
Governance and International Equity Returns. In Financial Markets and
Corporate Governance; Asian Finance Association (AsianFA) 2016 Conference.
https://doi.org/10.2139/ssrn.2701603
Martynova, M., & Renneboog, L. (2008). Spillover of corporate governance standards
in cross-border mergers and acquisitions. Journal of Corporate Finance, 14(3),
200–223. https://doi.org/10.1016/j.jcorpfin.2008.03.004
Martynova, M., & Renneboog, L. (2011). Evidence on the international evolution and
convergence of corporate governance regulations. Journal of Corporate Finance,
17(5), 1531–1557. https://doi.org/10.1016/j.jcorpfin.2011.08.006
Megginson, W. L. (2005). The economics of bank privatization. Journal of Banking
and Finance, 29(8–9 SPEC. ISS.), 1931–1980.
https://doi.org/10.1016/j.jbankfin.2005.03.005
Megginson, W. L., Nash, R. C., & Randenborgh, M. Van. (1994). The Financial and
Operating Performance of Newly Privatized Firms: An International Empirical
Analysis. The Journal of Finance, 49(2), 403–452. https://doi.org/10.2307/2329158
Ref. code: 25605502310021IBC
90
Megginson, W. L., & Netter, J. M. (2001). From State to Market : A Survey of Empirical
Studies on Privatization. Journal of Economic Literature, 39(2), 321–389.
Meyer, M. A., & Vickers, J. (1997). Performance Comparisons and Dynamic Incentives.
Journal of Political Economy, 105(3), 547–581. https://doi.org/10.1086/262082
Mohsni, S., & Otchere, I. (2014). Risk taking behavior of privatized banks. Journal of
Corporate Finance, 29, 122–142. https://doi.org/10.1016/j.jcorpfin.2014.07.007
Morck, R. K., & Steier, L. (2005). The Global History of Corporate Governance : An
Introduction. National Bureau of Economics Research (Vol. A History).
University of Chicago Press. https://doi.org/10.3386/w11062
OECD. (2014). Corporate governance in Asia, 30–41.
https://doi.org/10.1108/13581980710726778
Pedroni, P. (1996). Fully modified OLS for heterogenous cointegrated panels and the
case of purchasing power parity. Indiana University Working Papers in
Economics, No. 96-020. Retrieved from https://pdfs.semanticscholar.org/f99a/
f7633ca11e91c6ba51446dae06f45591a573.pdf
Pedroni, P. (2004). Panel cointegration: Asymptotic and finite sample properties of
pooled time series tests with an application to the PPP hypothesis. Econometric
Theory, 20(3), 597–625. https://doi.org/10.1017/S0266466604203073
Peng, M. W., Bruton, G. D., Stan, C. V., & Huang, Y. (2016). Theories of the (state-
owned) firm. Asia Pacific Journal of Management, 33(2), 293–317.
https://doi.org/10.1007/s10490-016-9462-3
Petri, P. A., Plummer, M. G., & Zhai, F. (2012). ASEAN Economic Community: A
General Equilibrium Analysis. Asian Economic Journal, 26(2), 93–118.
https://doi.org/10.1111/j.1467-8381.2012.02079.x
Phillips, P. C. B., & Hansen, B. E. (1990). Statistical Inference in Instrumental
Variables Regression with I(1) Processes Instrumental Variables Regression with
(1 ) Processes. Source: The Review of Economic Studies, 57131220(1), 99–125.
Retrieved from http://www.jstor.org/stable/2297545
Pindyck, R. S., & Rubinfeld, D. L. (2005). Microeconomics (6th ed.). Prentice-Hall.
Popov, A., & Van Horen, N. (2015). Exporting Sovereign Stress: Evidence from
Syndicated Bank Lending during the Euro Area Sovereign Debt Crisis. Review of
Finance, 19(5), 1825–1866. https://doi.org/10.1093/rof/rfu046
Ref. code: 25605502310021IBC
91
Prommin, P., Jumreornvong, S., & Jiraporn, P. (2014). The effect of corporate
governance on stock liquidity: The case of Thailand. International Review of
Economics & Finance, 32, 132–142. https://doi.org/10.1016/j.iref.2014.01.011
Prommin, P., Jumreornvong, S., Jiraporn, P., & Tong, S. (2016). Liquidity, ownership
concentration, corporate governance, and firm value: Evidence from Thailand ☆.
Global Finance Journal, 31, 73–87. https://doi.org/10.1016/j.gfj.2016.06.006
PwC. (2015). State-Owned Enterprises Catalysts for public value creation ?, (April).
Retrieved from https://www.pwc.com/gx/en/psrc/publications/assets/pwc-state-
owned-enterprise-psrc.pdf
Roberts, M. R., & Whited, T. M. (2011). Endogeneity in Empirical Corporate Finance.
SSRN Electronic Journal. https://doi.org/10.2139/ssrn.1748604
Robinett, D. (2006). Held by the visible hand : the challenge of state-owned enterprise
corporate governance for emerging markets. Washington, DC.
Rosenbaum, P. R., & Rubin, D. B. (1983). The central role of the propensity score in
observational studies for causal effects. Biometrika, 70(1), 41–55.
https://doi.org/10.1093/biomet/70.1.41
Sakawa, H., Ubukata, M., & Watanabel, N. (2014). Market liquidity and bank-
dominated corporate governance: Evidence from Japan. International Review of
Economics & Finance, 31, 1–11. https://doi.org/10.1016/j.iref.2013.11.005
Shank, C. A., & Vianna, A. C. (2016). Are US-Dollar-Hedged-ETF investors
aggressive on exchange rates? A panel VAR approach. Research in International
Business and Finance, 38, 430–438. https://doi.org/10.1016/j.ribaf.2016.05.002
SOM, L. (2013). Corporate Governance of Public Sector Enterprises in India. Money
& Finance, 53–75.
Stein, J. C. (1989). Efficient Capital Markets, Inefficient Firms: A Model of Myopic
Corporate Behavior. The Quarterly Journal of Economics. Retrieved from
https://scholar.harvard.edu/files/stein/files/qje-1989.pdf
Stulz, R. M. (1999). Golbalization, Corporate Finance, and the Cost of Capital. Journal
of Applied Corporate Finance, 12(3), 8–25. https://doi.org/10.1111/j.1745-
6622.1999.tb00027.x
Ref. code: 25605502310021IBC
92
Tang, K., & Wang, C. (2011). Corporate Governance and Firm Liquidity: Evidence
from the Chinese Stock Market. Emerging Markets Finance & Trade,
47(February), 47–60. https://doi.org/10.2753/REE1540-496X4701S105
Van Essen, M., Engelen, P. J., & Carney, M. (2013). Does “good” corporate governance
help in a crisis? The impact of country- and firm-level governance mechanisms in
the European financial crisis. Corporate Governance: An International Review,
21(3), 201–224. https://doi.org/10.1111/corg.12010
Verrecchia, R. E. (1983). Discretionary disclosure. Journal of Accounting and
Economics, 5, 179–194. Retrieved from
http://www.roselink.com/references/verrecchia_1983.pdf
Verrecchia, R. E. (2001). Essays on disclosure. Journal of Accounting and Economics,
32(1–3), 97–180. https://doi.org/10.1016/S0165-4101(01)00025-8
Welker, M. (1995). Disclosure policy , information asymmetry , and liquidity in e.
Contemporary Accounting Research, 11(2), 801–827.
Williams, B. (2014). Bank risk and national governance in Asia. Journal of Banking
and Finance, 49, 10–26. https://doi.org/10.1016/j.jbankfin.2014.08.014
Williamson, O. E. (1985). The Economic Institutions of Capitalism: Firms, Markets,
Relational Contracting. The Free Press, a Division of Macmillan, Inc.
World Bank. (1995). Bureaucrats in Business: The Economics and Politics of
Government Ownership. World Bank, (5). https://doi.org/10.1016/0024-
6301(96)90336-2
Zhang, M., M, L., Zhang, B., & Yi, Z. (2016). Pyramidal structure, political
intervention and firms’ tax burden: Evidence from China’s local SOEs. Journal of
Corporate Finance, 36, 15–25. https://doi.org/10.1016/j.jcorpfin.2015.10.004
Zhong, N. (2015). Corporate governance of Chinese privatized firms: Evidence from a
survey of non-listed enterprises. Journal of Comparative Economics, 43(4), 1101–
1121. https://doi.org/10.1016/j.jce.2015.05.003
Ref. code: 25605502310021IBC
APPENDIX
Ref. code: 25605502310021IBC
93
APPENDIX A
WGI DATA SOURCES
1 ADB African Development Bank Country Policy and Institutional Assessments
2 IRP African Electoral Index
3 AFR Afrobarometer
4 ASD Asian Development Bank Country Policy and Institutional Assessments
5 BPS Business Enterprise Environment Survey
6 BTI Bertelsmann Transformation Index
7 HUM Cingranelli Richards Human Rights Database
8 EBR European Bank for Reconstruction and Development Transition Report
9 EIU Economist Intelligence Unit
10 FRH Freedom House
11 CCR Freedom House -- Countries at the Crossroads
12 GCB Global Corruption Barometer Survey
13 GCS Global Competitiveness Report
14 WMO Global Insight Business Condition and Risk Indicators
15 GII Global Integrity Index
16 GWP Gallup World Poll
17 HER Heritage Foundation Index of Economic Freedom
18 IFD IFAD Rural Sector Performance Assessments
19 IJT iJET Country Security Risk Ratings
20 WCY Institute for Management & Development World Competitiveness Yearbook
21 IPD Institutional Profiles Database
22 MSI International Research & Exchanges Board Media Sustainability Index
23 OBI International Budget Project Open Budget Index
24 LBO Latinobarometro
25 PRC Political Economic Risk Consultancy
26 PRS Political Risk Services International Country Risk Guide
27 PTS Political Terror Scale
28 RSF Reporters Without Borders Press Freedom Index
29 TPR US State Department Trafficking in People report
30 VAB Vanderbilt University's AmericasBarometer
31 VDM Varieties of Democracy Project
32 PIA World Bank Country Policy and Institutional Assessments
33 WJP World Justice Project Rule of Law Index
Ref. code: 25605502310021IBC
94
BIOGRAPHY
Name Ms. Jutamas Wongkantarakorn
Date of Birth November 3, 1984
Educational Attainment 2007: Bachelor of Economics,
Thammasat University
2009: Master of Science (Finance),
Thammasat University
Work Position Lecturer
Rajamangala University of Technology
Rattanakosin
Scholarship Year 2012: Ph.D., Thammasat University
Scholarship
Ref. code: 25605502310021IBC