[1]
T H E I N F L U E N C E O F L E G A L S Y S T E M S O N
F I R M P E R F O R M A N C E D U R I N G F I N A N C I A L
T U R M O I L
Amadée Nicolaas 1729624 Rijksuniversiteit Groningen Faculty of Economics and Business Master of Science: Business AdministrationSpecialization: Finance Supervisor: T.H.N. Vu Second evaluator: N. Brunia
August, 2012
Acknowledgements
[2]
T H E I N F L U E N C E O F L E G A L S Y S T E M S O N
F I R M P E R F O R M A N C E D U R I N G F I N A N C I A L
T U R M O I L
A BST R A C TPrior studies support the notion that corporate law has an influence on firm value. This paper examines the relationship between firm performance and legal systems during the late 2000s global financial crisis. The focus is on the influence of legal systems on firm performance of firms from 39 countries, namely countries with French civil law, German civil law, Scandinavian civil law and common law. We investigate this by using random effects regression model on balanced panel data from January 2007 to September 2008 and find that firms operating in common law countries perform better than firms in civil law countries during the global financial crisis. This study can help making better recommendations for firms that want to invest abroad during the ongoing financial crisis.
JE L Classification: C23, G01, G30, K40
[3] 1. IN T R O DU C T I O N
The year 2008 was the year that the world experienced the biggest economic crisis since the Great Depression. Major financial institutions were either bailed out or went bankrupt (Tarraf, 2011). Remarkably the financial crisis of 2008 occurred despite the strengthening of the U.S. corporate governance over those past few years. Due to corporate scandals in the early 2000s, the Sarbanes-Oxley Act and new exchange listing requirements at the NYSE and NASDAQ were implemented. These new regulations served as models for corporate governance around the world. However, current corporate governance systems fail to protect against too much risk taking and to provide the control needed so that companies can promote safe business practices (Tarraf, 2011). According to Harrison and Sepúlveda (2011), there are three factors that have made this crisis unusual in comparison to previous crises. Firstly, the crisis began in a developed country. Secondly, many developing countries were able to cushion the effects of the crisis. Lastly, developing countries recovered much faster than the developed countries.
[4]
/HJDO HQYLURQPHQW XQGRXEWHGO\KDV DVWURQJHIIHFW RQFRPSDQLHV¶EHKDYLRr regulating various aspects of market activity, influencing firm valuation through regulations concerning corporate governance. Extensive numbers of antecedent studies have found a strong relationship between WKH RULJLQ RI D FRXQWU\¶V OHJDO WUDGLWLRQ DQG the operation of its financial system (Beck, Demirguc-Kunt, and Levine, 2003).
Despite a number of both theoretical and empirical researches concerning effects of corporate governance on firm performance, there is a lack of papers describing efficiency of governance measures during the late 2000s global financial crisis. This issue is especially relevant due to the ongoing financial turmoil affecting all aspects of real economy. Therefore, this paper will partially fill the gap in the literature by dealing with corporate governance during unusual times and address the topic by analyzing the influence of different legal systems on firm valuation during late 2000s global financial crisis. Such a study can help making better recommendations for firms that want to invest or operate abroad during the ongoing financial crisis.
Comparing different legal systems inevitably leads to certain questions. Therefore, the main contribution to the existing literature is investigating which legal system was best able to cope with the late 2000s global financial crisis, which is an aspect that has not been discussed in previous literature.
Hence, in this paper the main question that will be tested and answered is:
Which legal system provides better firm performance during global financial crisis?
[5]
2008 is used to measure the effects of legal systems on firm performance during the global financial crisis.
The remainder of this paper is organized in the following way. Section 2 presents the overview of the literature relevant for this study. Section 3 describes the applied methodology, whereas description of the data and characteristics used in this study are described in section 4. Section 5 presents the estimation results and discusses the main findings followed by the conclusion and limitations of this research in section 6.
2. L I T E R A T UR E O V E R V I E W 2.1 Theoretical background
[6]
imitation. It was a priority for Napoleon1 to secure the adoption of the French civil code in all
conquered territories. Furthermore, the French civil law had a major influence on the legal systems of Portugal and Spain. The German civil code was not imposed but exerted a significant influence on China, Greece, and Japan. As a result, the laws reflect both the influence of their traditional family and the revisions specific to individual countries. Due to the spread of legal families and the subsequent evolution of the laws, both the individual rules and whole families across a large number of countries can be compared (Watson, 1993).
The different types of laws and the quality of their enforcement are crucial determinants of what rights investors have and how well these rights are protected. Legal systems differ around the world and in most countries they give investors only a limited bundle of rights. Differences between legal protections may help in explaining the reason behind differences in financial and ownership structure of firms in various countries. In general, common law countries tend to protect investors better than civil law countries, especially in comparison to the French civil law countries. German civil law countries and Scandinavian civil law countries take an intermediate position towards investor protection (La Porta et al., 1998). Furthermore, the differences between legal systems could explain why companies are financed so different across countries. Demirgüç-Kunt and Maksimovic (1998) state that, countries whose financial system grants more investor protection have more developed capital markets. The higher the level of investor protection the more external financing is accessible for firms. Hence, being a shareholder in France does not give the investor the same privileges as being a shareholder in the United States.
1 Napoleon Bonaparte (15 August 1769 ± 5 May 1821) was a French military and a political leader during the final
[7]
In addition, common law countries tend to have market-based financial systems and civil law
countries have bank-based financial systems (Demirgüç-Kunt and Maksimovic, 2002)2. During
the financial crisis banks scaled down their balance sheets and tightened lending conditions, as a consequence of fear of default due to exposition to overly high risk. The deleveraging process reduced the risk exposition but led to substantial decrease in the amount of loans granted, making
sources of financing market activity unattainable for firms (ECB, 2009)3. Thus, firms under civil
law perform worse than their counterparts under common law, as they have fewer possibilities to undertake investment opportunities due to the lack of liquidity. Moreover, it is well known that in countries with civil law, which has more lenient anti-trust regulations, cross-ownership is wider spread (Tyrell and Schmidt, 2001). )LUPV¶ LQWHUGHSHQGHQFH PD\ LQFUHDVH WKHLU performance in normal times but during crises it may increase the probability of failure, as default of one firm may cause difficulties to several others or even trigger a chain of bankruptcy. 7KHVHHIIHFWVPLJKWKDYHEHHQHPSKDVL]HGE\LQYHVWRUV¶EHKDYLRUGXULQJWKHFULVLV. Hence, due to the problematical situation of firms under civil law concerning financing, there will be lower valuation of firms in the civil law countries during the crisis.
2.2 Empirical literature
Law and Finance
Daines (2001) declares that corporate law influences firm value and he presents evidence that is in accordance with the theory, which states that Delaware corporate law increases firm value. 'HODZDUH¶VUHJXODWLRQVDUHPRUHEHQHILFLDOWRPLQRULW\LQYHVWRUVWKDQUHJXODWLRQVLQRWher states
2 In bank-based systems, banks are the most important players. In other words, banks mobilize savings, allocate
capital, oversee decisions of corporate managers, and provide management risk instruments. In contrast, in market-based system, securities markets share the leading role with banks.
[8]
LQWKH8QLWHG6WDWHVGXHWR³'HODZDUH¶VXQLTXHDWWUDFWLRQVIRUSXEOLFILUPV>ZKLFK@LQFOXGHLWV UXOHVFRXUWVSUHFHGHQWDQGSROLWLFDOHFRQRP\´+HXVHV7RELQ¶V4DVDQHVWLPDWHIRUILUPYDOXH and finds that Delaware firms have higher value than identical firms incorporated elsewhere in 12 of 16 years between 1981 and 1996. Furthermore, according to Daines (2001), the difference LQ 7RELQ¶V 4 EHWZHHQ 'HODZDUH DQG QRQ-Delaware firms presents, that Delaware law clearly improves firm value. Hence, law can be seen as an intangible asset that can have a positive or negative value.
[9]
countries have lower protection of creditor and equity shareholder rights and thus lower firm performance compared to common law countries. Furthermore, Beck, Demirguc-Kunt, and Levine (2004) find that firms under common law face smaller obstacles to accessing external finance than firms from civil law countries.
$OOLQDOOWKH³WKHRU\RIODZDQGILQDQFH´VKRZVWKDWWKHFRPPRQODZV\VWHPSURYLGHVDEHWWHU framework for financial development and economic growth than the civil law system. On the contrary, Graff (2008) states that there is not much evidence to conclude that common law countries protect financial investor better than civil law countries.
Financial crisis
A financial crisis is an unexpected and unpredictable event, thus it is very difficult for firms to adjust their optimal corporate governance structures in response to a future financial crisis. Hence, we can treat corporate governance structures as exogenous variables that explain the performance of a company during a financial crisis (Liu, Uchida, and Yang, 2012).
[10]
Besides, the East Asian crisis provides an interesting opportunity to investigate the valuation effects of ownership structure (Lemmon and Lins, 2003). They investigate the effect of ownership structure on changes in shareholder value during the East Asian financial crisis. Their main hypothesis is that, during the crisis, values should decline the most in countries with weak legal protection. In other words, during the crisis firm values are expected to decline the most in civil law countries. Using data from 800 firms they find evidence consistent with their main hypothesis. Both the univariate and multivariate tests are consistent with the main hypothesis. In JHQHUDOWKH\ILQGDSRVLWLYHUHODWLRQVKLSEHWZHHQILUPYDOXHDQGWKHFRXQWU\¶VOHJDOSURWHFWLRQ which is consistent with the findings presented in La Porta et al. (2002). Furthermore, La Porta et al. (1998) discuss that the absence of strong legal protection in many emerging economies will
result in more agency problems4 between controlling insiders and outside investors.
Tarraf (2011) explains how failures in corporate governance contributed to the late 2000s financial crisis. In fact, he shows how the current corporate governance systems failed to protect against aggressive risk taking and how it also failed to provide the control needed, so that the companies can promote sound business practices. The major cause of the 2007-2008 financial crisis is aggressive risk taking, which is a corporate governance aspect. Inadequate risk management by executives and board of directors is to be blamed for the credit market collapse and resulting financial crisis.
In addition, Erkens, Hung, and Matos (2012) investigate the influence of corporate governance on the performance of financial firms during the 2007-2008 financial crisis. The relation between firm performance and corporate governance is being examined by regressing stock returns during
4 Due to conflicts of interest between the principal (shareholders) and the agent (managers), agency costs arise.
[11]
the crisis on measures of corporate governance and control variables. Three factors of corporate governance are used, namely, board independence, institutional ownership, and the presence of large shareholders. They find worse stock returns during the global crisis period for firms with more independent boards and higher institutional ownership. An explanation for their finding is that independent directors and institutional shareholders stimulate managers to increase shareholder returns through more risk-taking prior to the crisis. The study by Erkens, Hung, and Matos (2012) is in contrast with prior studies on the 1997-1998 East Asian financial crisis, which find that more external monitoring, leads to better performance during crisis. Erkens, Hung, and Matos (2012) argue that firms with greater external monitoring perform worse during the late 2000s financial crisis. Besides, their study suggests that the impact of corporate governance on firm performance during the crisis in developed markets differs from that in emerging markets.
Altogether, the findings of the studies discussed above support the notion that legal systems influence firm performance. Moreover, it can be assumed that firms operating under common law system, which is known to have higher investor protection regulations, attract investors and positively affect firm valuation. This intuition is reflected in the differences between financial
systems of the common law countries and the civil law countries. In common law countries the
[12]
worse than their counterparts under common law, as they have fewer possibilities to undertake investment opportunities due to the lack of liquidity.
This study will investigate which legal system was most capable of coping with the late 2000s financial crisis. Giving the fact that the British common law system provides investors with better legal protection and better law enforcement than the civil law systems do, one would expect firm performance to differ between these legal systems also during a global financial crisis. According to prior research and their findings the main hypothesis has been formulated:
Firms in common law countries have significantly higher firm values than firms in civil law countries during the global financial crisis.
According to the abovementioned intuition, the differences in firm valuations under different legal systems during the late 2000s financial crisis will be analyzed.
[13] 3. M E T H O D O L O G Y
3.1 Model specification
This study is aimed at finding the relationship between legal systems and firm performance during the global financial crisis. In order to find the relationship between the two variables the following regression model will be estimated:
FPit Įit ȕ1FRENCHit ȕ2GERMANit ȕ3SCANDINAVIANit ȕ4SIZEit +
ȕ5CAPEXit ȕ6LEVit + uit
[14] 3.2 Measurements of variables
Dependent Variables
In most corporate governance studies financial ratios have been used to measure firm performance. In this paper three proxies for measuring firm performance will be used, namely, 7RELQ¶V4 74 5HWXUQRQ$VVHWV 52$ DQG5HWXUQRQ(TXLW\ 52( 7KHYDULDbles used in previous empirical studies in corporate governance are summarized in Table 1. ROA and ROE are accounting-based performance measures and are seen as backward looking. Due to this fact, 52$ DQG 52( DUH QRW DIIHFWHG E\ LQYHVWRUV¶ H[SHFWDWLRQV DQG the fluctuations of the equity market. However, with respect to a correct valuation of the core business activities, there may be a disadvantage when using accounting-EDVHG SHUIRUPDQFH PHDVXUHV ,Q FRQWUDVW 7RELQ¶V 4 LV seen as forward looking. Hence, it takes market effects into account and is expected to capture WKHPDUNHW¶VH[SHFWDWLRQVDQGPRYHPHQWV(Demsetz and Villalonga, 2001). However, according to Morck, Shleifer, and Vishny (1988) 7RELQ¶V 4 DVVXPHV WR EH D QRLV\ PHDVXUHPHQW IRU management or board performance. Furthermore, neither ROA nor ROE are considered to be RSWLPDO SHUIRUPDQFH PHDVXUHPHQWV $V D UHVXOW 7RELQ¶V 4 DV ZHOO DV 52$ DQG 52( ZLOO EH included because they have been applied in corporate governance studies and they are expected to capture limitations of one another.
[15]
(2001), this simple measure produces coefficient estimates that have unbiased signs and are conservative in that they are less likely to produce significant results. Even though there are more VRSKLVWLFDWHGPHDVXUHVRI7RELQ¶V4WKLVDSSUR[LPDWLRQLVFKRVHQWRHQVXUHWKDWWKHUHDUHGDWD available for most of the sample (Bauer, Guenster, and Otten, 2004). ROA is calculated as the ratio of operating income to total assets and ROE is taken directly from Thomson Financial DataStream. In the regression analysis the dependent variables are labeled as TQ, ROA, and ROE. Analyses are conducted with all performance measures, thus possible differences might occur in the findings.
Table 1: Variables used in previous empirical studies in corporate governance Study Sample
Collection Performance measure Dependent Variable: Control Variables Regression results Beiner, Drobetz, Schmid, and Zimmermann (2006) Switzerland
109 Firms 7RELQ¶V4 Growth Firm Size ROA
Growth(+) Firm Size(-) ROA(+) Bhagat and
Black (1999) 205 large U.S. public companies 1988-1991 7RELQ¶V4 ROA Turnover ratio Operating margin Board Size CEO Ownership Firm size Board Size(-) CEO Ownership(+) Firm size (+/-) Bhagat and Bolton (2008) U.S. 1990-2004 7RELQ¶V4 ROA Stock returns Leverage Assets R&D Expenses Board Size Risk CEO age CEO tenure Leverage(-) Asset (-) R&D Expenses(+) Board Size(-) Risk(-) CEO age(+) CEO tenure(-) Bhattacharya and Graham (2007) Finland 7RELQ¶V4 ROE Leverage Capital Expenditure Market Risk Firm size Leverage(-) Capital Expenditure(-) Market Risk(+/-) Firm size(-) Black, Love, and Rachinsky (2006)
Russia 7RELQ¶V4 Firm Size Liquidity Growth Leverage ROA Firm Size(-) Liquidity(+) Growth(+) Leverage(+) ROA(+) Demsetz and
Villalonga U.S. 7RELQ¶V4 R&D to sales Fixed
[16] (2001) sales Leverage sales(+) Leverage(-) Durnev and
Kim (2005) Cross country (27 countries)
7RELQ¶V4 Firm Size R&D/assets Export Firm Size(-) R&D/assets(+) Export(+) Gompers, Ishii, and Metrick (2003) U.S. 1.500 large companies 1990s 7RELQ¶V4 - - Gedajlovic and Shapiro (1998) Cross country (5 countries: Canada, France, Germany, U.K., and U.S.) ROA Growth Firm size Geographic scope Industry effects Growth(+) Firm size(+/-) Geographic scope(+/-) Industry effects(+/-) Henry (2008) Australia 1992-2002 7RELQ¶V4 Adjusted 7RELQ¶V4 Firm Risk Firm Size Leverage Profitability Dividend yield Firm Risk(-) Firm Size(-) Leverage(+) Profitability(+) Dividend Yield(-) Klapper and
Love (2004) 14 emerging markets 7RELQ¶V4 ROA
Firm Size Fixed capital/sales Firm Size(-) Fixed capital/sales(-) Lins (2003) 18 emerging
markets 7RELQ¶V4 Firms Size Capital expenditure Leverage Firms Size(-) Capital expenditure(+) Leverage(-) Morck, Shleifer, and Vishny (1988) U.S. 7RELQ¶V4 Profit rate R&D/assets Leverage R&D/assets(+) Leverage(-) Ntim, Opong, and Danbolt (2011)
South Africa 7RELQ¶V4 ROA Total share returns Capital Expenditures Firm size Growth Cross-listing Audit firm size
Capital
Expenditures(+/-) Firm size(-) Growth(+) Cross-listing(+) Audit firm size(+) Shakir (2008) Malaysia 567 firm-year observations 1999-2005 7RELQ¶V4 - - Weir, Laing, and McKnight (2002)
[17] Independent Variables
The main independent variables are dummies to distinguish between common law firms and civil law firms. There are four dummy variables, namely, common law, French civil law, German civil law, and Scandinavian civil law dummies. In accordance with the main hypothesis, the
coefficients on FrenchCivDUM, GermanCivDUM, and ScandinavianCivDUM are expected to have
smaller values than the coefficient on CommonDUM. However, in the regression analysis we will
include three dummies and an intercept term. Including all four dummies and the intercept coefficient will lead to the problem of perfect multicollinearity (Brooks, 2008).
The French Commercial Code was written under Napoleon in 1807. Napoleon brought the French code to Belgium, The Netherlands, Italy, and Western regions of Germany. In the colonial era, France extended her legal influence to the Near East and Northern and Sub Saharan Africa, Indochina, Oceania, and French Caribbean islands. Furthermore, the French legal influence has been significant in Luxembourg, Portugal, Spain, and Italy. When the Spanish and Portuguese empires separated Latin America in the 19th century, the lawmakers looked for inspiration from the French civil law. The sample in this paper contains 18 countries with laws in the French civil law tradition.
7KH*HUPDQ&RPPHUFLDO&RGHZDVZULWWHQLQ DIWHU%LVPDUFN¶V5 unification of Germany.
The German code had an important influence on the legal theory and doctrine in Austria, Greece, Hungary, Switzerland, Japan, and Korea. There are 5 countries from this family in the dataset.
5 Otto von Bismarck (1 April 1815- 30 July 1898) was a conservative German statesman. After a series of short
[18]
The Scandinavian code is viewed as part of the civil law tradition even though its law is less derivative of Roman law than the French and German families. The sample consists of 4 Scandinavian countries.
The Common law tradition includes the law of England and those laws modeled on English law. Common law has spread to British colonies, including the United States, Canada, Australia, India, and some other countries. In this sample there are 12 common law countries (La Porta et al., 1998).
Control Variables
To minimize potential omitted variables, control variables will be introduced. The model will control for a number of firm-specific factors that previous studies have shown to be correlated with firm performance. These are variables such as firm size, capital expenditure, and leverage.
[19]
Capital expenditure is seen as a proxy for investment opportunity. Firms that have more investment in innovation and technology may gain competitive advantages by releasing new products and/or services. Gaining competitive advantage may help firms receive premium prices and thus a higher firm value. On the other hand, innovation is capital intensive and may have a negative effect on the value of the firm. Hence, in line with previous literature capital expenditure is expected to correlate significantly with firm performance, without specifying if the coefficient is positive or negative (Ntim, Opong, and Danbolt, 2011).
Leverage is the potential value of risk caused by borrowing. Depending on their capital structure, firms might take investment decisions that could influence the overall firm performance negatively. However, the sign for leverage is ambiguous according to Henry (2008). A positive relationship between leverage and firm performance can be associated with the effective usage of debt. In contrast, a negative sign would be in line with increasing cost of capital and financial distress. Moreover, firms that have high level of debt during economic downturns will experience poor stock price performance, thus the effect of leverage on firm performance is expected to be negative (Liu, Uchida, and Yang, 2012). The expected regression results, excluding those for the legal system dummies, are summarized in Table 2. In addition, the definition of the variables and their description are given in Table 3.
Table 2: Summary of the expected regression results
Control Variables Expected Sign
SIZE +/-
CAPEX +/-
[20]
Table 3: Definition and description of the variables Regression Variables
Abbreviation Variable Formula/Description Source
Dependent Variables
T Q 7RELQ¶V4 (MV+PS+ND)/TA DataStream
R O A Return on Assets OPI (t)/TA (t-1) !100 DataStream R O E Return on Equity (Net Income before Preferred Dividends - Preferred
Dividend Requirement) / Last Year's Common Equity * 100
DataStream
Independent Variables Legal systems
dummies CommonFrenchCivDUM DUM
GermanCivDUM ScandinavianCivDUM Intercept FRENCH GERMAN SCANDINAVIAN La Porta et al. (1998) Choi and Meek (2008)
SI Z E Firm Size Log(NS) DataStream
C APE X Investment
Opportunities CAP/TA DataStream
L E V Leverage TL/TA DataStream
Computational Variables
M V Market Value Share price multiplied by the number of ordinary
shares in issue DataStream
PS Preferred Stock A claim prior to the common shareholders on the earnings of a company and on the assets in the event of liquidation
DataStream
ND Net Debt Total debt minus cash DataStream
T A Total Assets The sum of total current assets, long-term receivables, investment in unconsolidated
subsidiaries, other investments, net property plant and equipment and other assets
DataStream
NS Net Sales Gross sales and other operating revenue less discounts, returns and allowances
DataStream OPI Operating Income The difference between sales and total operating
expenses DataStream
T L Total Liabilities Short and long term obligations expected to be satisfied by the company
DataStream C AP Capital Expenditure Funds used to acquire fixed assets other than those
associated with acquisitions
[21] 3.3 Estimation method
Panel data
The results of the regression model will decide whether the main hypothesis will be accepted or rejected. This study will follow Claessens and Djankov (1999) and Anderson and Reeb (2003) and use the panel techniques to investigate the main hypothesis. According to Brooks (2008), a panel of data will include information across both time and space. Hence, it differs from normal time series or cross sectional analyses because it blends both characteristics together. In order to estimate the regression model panel data will be used.
Panel data has some important advantages compared to cross-sectional or time series data. Firstly, panel data sets are usually larger than cross-sectional or time series data sets, and the independent variables vary over two dimensions, hence estimators based on panel data are often more accurate. Secondly, panel data reduces identification problems (Verbeek, 2000). Lastly, panel data also has the advantage that it gives more insight into the development of variables and their relationship over time and it helps to mitigate multicollinearity problems among variables that may arise when time series are modeled individually (Brooks, 2008).
[22] Estimation methods
Pooled regression model
With the pooled regression model, we obtain time-series of data of a firm just as different firms at a point in time. This is the simplest way to deal with a panel data; it involves estimating a single equation on the whole data together by using Ordinary Least Squares (OLS). This method does not use information on who is who regarding company identity. However, the pooled least squares method effectively maximizes the sample size.
Pooling the data has some severe limitations due to the fact that the pooled regression model implicitly assumes homoscedasticity between the variables over time and across all cross-sectional units in the sample. However, the random effects model does not make this assumption, and hence overcomes its limitations. In addition, several other assumptions are made in order to apply the OLS estimator. One of the assumptions is that the error terms are assumed to have constant variance, the assumption of homoscedasticity. Due to the presence of heteroscedasticity the OLS estimators will still be unbiased, but they will be inefficient, and the p-values might be XQUHOLDEOH7RWHVWIRUKHWHURVFHGDVWLFLW\WKH:KLWH¶VWHVWZLOOEHXVHG If the null hypothesis that the errors are homoscedastic is rejected, there will be evidence of heteroscedasticity and the random effects model will be the superior estimation method (Brooks, 2008).
Random effects model
[23]
cross-sectionally but it is constant over time and it measures the random deviation of each FRPSDQ\¶VLQWHUFHSW WHUP IURP WKHFRPPRQLQWHUFHSW WHUP ,QWKHIL[HGHIIHFWV PRGHOGXPP\ variables are used to capture the heterogeneity in the cross-sectional dimension. Instead, this occurs via the error terms in the random effects model. Furthermore, the random effects model
requires some restrictive assumptions for the new error term. Ithas zero mean, is independent of
the individual observation error term, has constant variance, and is independent of the explanatory variables.
To test which estimation method is superior in this study the Breusch-Pagan test (Lagrange multiplier (LM) test) on the OLS residuals will be performed. This LM test will help in deciding between a random effects regression and a simple pooled OLS regression (Greene and Zhang, 2003). The null hypothesis in the LM test is that the variances across entities are zero. In other words, there is no panel effect. If the null hypothesis that there are no individual specific effects is rejected, Breusch-Pagan test lets one prefer the random effects model.
Additionally, in order to apply the random effects model, the dummy variables for legal system must be exogenous. In fact, dummy variables for legal origin, such as English, French, German, or Scandinavian are less prone to endogeneity problems (Levine, 1998). Thus, the general regression model can be estimated by using the random effects model.
Summarizing the steps for estimating the regression model:
Step 1: Panel techniques will be used to investigate the main hypothesis.
[24]
Step 3: To help in deciding between a random effects model and a simple pooled OLS model, :KLWH¶VWHVWDQG%UHXVFK-Pagan test will be performed.
4. D A T A
The crisis period in this paper is defined as January 2007 to September 2008, therefore the empirical analysis is conducted by using monthly data from January 2007 to September 2008. This crisis period definition corresponds with the one used by Erkens, Hung, and Matos (2012). +HQFHILUP¶VYDULDEOHVDUHPHDVXUHGRQDPRQWKO\EDVLVGXULQJWKHSHULRGIURP-DQXDU\WR September 2008. The investigation is initiated at the beginning of 2007 because this is regarded as the period when the market first realized the severity of the losses related to subprime mortgages. There are three reasons why the investigation ends in September 2008. Firstly, the huge amounts of government bailouts were initiated from October 2008 onwards. Secondly, at the end of the third quarter, regulators of various countries imposed short-selling bans on the stocks of many financial institutions to restrain the steep decline of their share prices. Thirdly, modifications in the International Financial Reporting Standards (IFRS) permitted financial institutions to avoid reorganizing asset write-downs in October 2008. Thus, by using the data from January 2007 to September 2008 we could test which legal system has the highest firm value during this global financial crisis.
4.1 Data collection
[25]
countries that have stock exchanges are chosen (Choi and Meek, 2008). The sample is collected from Thomson Financial DataStream and contains firm-level data in US dollars from both developed and developing countries as listed in Table 4. Initially, the total sample contained 6211 firms. From this total sample, financial institutions, including investment trusts and real estate companies, are excluded due to their great debt structure. Including these financial institutions would bias the results. Also firms with missing observations are excluded from the sample. Hence, for a firm to be included in the sample it must have data on market value, preferred stock, net debt, total assets, net sales, operating income, capital expenditure, return on equity, and total liabilities. In total, 4472 firms are available resulting in a total of 93912 firm-month observations. A summary of the firm collection is given in Table 5. This dataset can be defined as a balanced panel data, because it has the same number of cross-sectional units for each month.
Table 4: List of countries
Country Stock Index Legal System
1. Argentina MERVAL French code law
2. Belgium BEL All share French code law
3. Brazil Brazil IBX Index French code law
4. Chile IPSA/IGPA French code law
5. Colombia IGBC French code law
6. France CAC All share index French code law 7. Greece Athex All share French code law 8. Indonesia Jakarta Islamic index French code law 9. Italy FTSE Italian all share French code law 10. Jordan ASE weighted index French code law
11. Luxembourg LuxX index French code law
[26]
13. Netherlands AEX Index French code law
14. Peru IGBVC French code law
15. Philippines PSE all share index French code law
16. Portugal PSI general French code law
17. Spain IBEX-35 French code law
18. Turkey ISE all share French code law
19. Austria ATX prime German code law
20. Czech Republic PX Index German code law 21. Germany Prime All share index German code law
22. Japan NIKEI 225 German code law
23. Switzerland SMI expanded index German code law
24. Denmark KAX Index Scandinavian code law
25. Finland OMXH Scandinavian code law
26. Norway Oslo all share index Scandinavian code law 27. Sweden OMX all share Scandinavian code law 28. Australia S&P/ASX 300 Common law
29. Canada S&P/TSX Composite Common law 30. Hong Kong Hang Seng Index Common law
31. India BSE 500 Common law
32. Israel TA 100 index Common law
33. Malaysia FTSE Bursa Malaysia Common law
34. New Zealand NZSX all Common law
35. Pakistan KSE 100 Common law
36. South Africa FTSE/JSE all share Common law
37. Thailand SET 100 Common law
38. U.K. FTSE all share index Common law
[27]
Table 5: Collection of firms
Country Initial Firms Exclusion financials and Missing variables Firms Final Final Obs.
Argentina 70 14 56 1176 Belgium 123 32 91 1911 Brazil 100 31 69 1449 Chile 107 5 102 2142 Colombia 17 5 12 252 France 517 76 441 9261 Greece 209 30 178 3738 Indonesia 30 9 21 441 Italy 234 60 174 3654 Jordan 100 49 51 1071 Luxembourg 11 2 9 189 Mexico 35 7 28 588 Netherlands 114 27 87 1827 Peru 12 0 12 252 Philippines 267 97 170 3570 Portugal 53 20 33 693 Spain 113 34 79 1659 Turkey 102 33 69 1449
Subtotal French Civil
law countries 2214 1682 35322 Austria 40 13 27 567 Czech Republic 25 11 14 294 Germany 360 99 261 5481 Japan 225 36 189 3969 Switzerland 50 13 37 777 Subtotal German
Civil law countries 700 528 11088
Denmark 173 66 107 2247 Finland 129 23 106 2226 Norway 176 53 123 2583 Sweden 280 99 181 3801 Subtotal Scandinavian Civil law 758 517 10857
Subtotal Civil law
[28] 4.2 Descriptive statistics
[29]
Table 6: Descriptive statistics
Panel A : Civil and Common law
Variable M ean M edian M in M ax StDev Obs.*
T Q 1.45 1.00 -0.79 497.17 4.79 93912 R O E 10.90 12.61 -1165.23 1798.23 40.67 93912 R O A 4.98 4.69 -1219.07 81.94 16.26 93912 SI Z E** 5.62 5.79 0.00 8.70 1.45 93912 C APE X 0.06 0.04 -0.01 5.88 0.08 93912 L E V 0.51 0.53 -1.64 1.40 0.22 93912
Panel B: F rench Civil law
T Q 1.35 0.86 -0.79 497.17 7.43 35322 R O E 8.59 10.14 -833.56 525.93 37.73 35322 R O A 5.32 4.73 -148.54 81.94 11.50 35322 SI Z E** 5.41 5.48 0.00 8.69 1.31 35322 C APE X 0.05 0.04 -0.01 0.90 0.07 35322 L E V 0.52 0.54 -1.64 1.23 0.23 35322
Panel C : German Civil law
T Q 1.16 0.91 -0.18 11.29 1.03 11088 R O E 9.02 10.92 -189.69 1798.23 36.62 11088 R O A 5.08 5.26 -160.52 49.81 11.49 11088 SI Z E** 6.24 6.40 0.00 8.41 1.04 11088 C APE X 0.06 0.05 0.00 0.42 0.05 11088 L E V 0.54 0.57 -0.10 0.97 0.20 11088
Panel D: Scandinavian Civil law
T Q 1.44 1.00 0.70 26.03 1.54 10857 R O E 6.21 12.38 -11165.23 215.55 52.19 10857 R O A 2.43 2.93 -292.99 45.00 19.31 10857 SI Z E** 5.35 5.37 0.00 8.11 1.13 10857 C APE X 0.07 0.04 0.00 0.74 0.08 10857 L E V 0.52 0.56 -0.11 1.40 0.21 10857
Panel E : Common law
T Q 1.65 1.21 -0.73 149.03 2.10 36645 R O E 15.09 15.83 -823.81 833.33 40.30 36645 R O A 5.38 4.85 -1219.07 75.35 20.22 36645 SI Z E** 5.73 6.05 0.00 8.70 1.70 36645 C APE X 0.08 0.05 0.00 5.88 0.10 36645 L E V 0.49 0.52 -0.07 1.08 0.23 36645
*Observations are equal to number of companies times 21 months (e.g. for Common law, N = 36645 = 1745 companies times 21 months)
[30]
Table 7: Mean of the law systems during the global financial crisis Variable Common law F rench civil law German civil law Scandinavian civil law
T Q 1.65 1.35 1.16 1.44 (2.10) (7.43) (1.03) (1.54) R O E 15.09 8.59 9.02 6.21 (40.30) (37.73) (36.62) (52.19) R O A 5.38 5.32 5.08 2.43 (20.22) (11.50) (11.49) (19.31) SI Z E** 5.73 5.41 6.24 5.35 (1.70) (1.31) (1.04) (1.13) C AP E X 0.08 0.05 0.06 0.07 (0.10) (0.07) (0.05) (0.08) L E V 0.49 0.52 0.54 0.52 (0.23) (0.23) (0.20) (0.21)
**SIZE is displayed in thousands of units of dollars (logarithm) Standard deviations are reported in parentheses.
4.3 Correlation diagrams
Table 8 presents the correlation between the variables for the entire dataset. The most important and interesting correlations will be discussed. Almost all the variables correlate strongly significantly with each other.
Table 8: The correlation matrix
[31]
The correlation matrix indicates a strong negative correlation between firm performance (TQ and
ROE) and FrenchCivDUM,GermanCivDUM, and ScandinavianCivDUM. On the other hand, there is
a strong positive correlation between firm performance (TQ, ROA, and ROE) and CommonDUM,
which is also consistent with the main hypothesis, but of course it does not control for the other effects and it is not a sufficient condition to establish a causal relationship in either direction. Moreover, firm size and leverage correlate positively with ROA and ROE, but negatively with TQ. Conversely, CAPEX has a positive correlation with TQ and negative correlations for both ROA and ROE. This is also in line with my ambiguous expectations regarding the coefficients of these control variables. In addition, to avoid the phenomenon of multicollinearity between the independent variables Table 8 can be used to look for signs of highly correlated variables. The correlations are all considerably less than 0.7, suggesting that multicollinearity is not likely to be a significant issue in the regression estimations (Brooks, 2008).
5. R ESU L TS
The econometric model building applied in this paper, is that of starting with the simplest model and adding to it sequentially so that it becomes more complex. This approach is known as a µVSHFLILF-to-JHQHUDO¶RUµERWWRPV-XS¶PRGHOLQJDSSURDFK(Brooks, 2008).
[32]
variance, the assumption of homoscedasticity. To test for heteroscedasticity we will use the :KLWH¶VWHVWTables 9, 10, and 11 suggest that there is evidence of heteroscedasticity because the p-values for every single test are below 1%. Due to the problem of heteroscedasticity, the random effects model will be preferred. To confirm the superiority of the random effects model the Breusch-Pagan test on the OLS residuals will be performed. When testing the regressions in this study, the Breusch-Pagan test provides evidence that the p-values are lower than 1% as can be seen in Tables 9,10, and 11. Therefore, the null hypothesis that there are no individual specific effects is rejected. As a result, the pooled OLS model is not appropriate and the random effects model is preferred. Hence, the regression model will be estimated using the random effects model for all three dependent variables. There are four dummy variables, namely, common law, French civil law, German civil law, and Scandinavian civil law dummies. In accordance with the
main hypothesis, we expect the coefficients on FrenchCivDUM, GermanCivDUM, and
ScandinavianCivDUM to have smaller values than the coefficient on CommonDUM. However, in
the regression analysis we will include only three dummies and an intercept term. If we include all four dummies and the intercept coefficient the problem of perfect multicollinearity will arise. Hence, the dummy variables operate by changing the intercept (Brooks, 2008).
5HVXOWVEDVHGRQ7RELQ¶V4
[33]
are valuated higher than their counterparties from civil law countries during the global financial crisis when TQ is used to measure firm performance. In column (3), CAPEX is included, which is positive and also strongly significant. Finally, in column (4), LEV is added which has the predicted negative sign but the effect of leverage on TQ is not statistically significant. Similarly, in the columns (3) and (4) the coefficients of the civil law firms are in the expected directions but the effects of German and Scandinavian civil law firms on TQ are not statistically significant. Furthermore, adding control variables increases the explanatory power of the regression
(Adjusted R2 raises to 0.002, 0.021, and 0.021 in columns (2), (3), and (4) respectively). All in
all, from Table 9 can be concluded that firms in common law countries have significantly higher firm values than firms in civil law countries during the global financial crisis, which is consistent with the main hypothesis. However, the results are only stable for firms in French civil law countries. The effects of German and Scandinavian civil law firms on TQ are not stable; adding more control variables to the model make the effects of German and Scandinavian civil law firms on TQ statistically insignificant.
[34]
7DEOH 5DQGRPHIIHFWPRGHOUHVXOWVRI7RELQ¶V4
Explanatory variables Expected (1) (2) (3) (4)
sign Intercept 1.650*** 3.942*** 3.034*** 3.024*** (20.312) (21.478) (16.479) (16.276) F renchCivDU M - -.305*** -0.432*** -0.193* -0.196* (-2.631) (-3.740) (-1.667) (-1.690) GermanC ivDU M - -0.494*** -.286* -0.104 -0.106 (-2.928) (-1.700) (-0.619) (-0.629) ScandinavianCivDU M - -0.211 -0.361** -0.257 -0.260 (-1.244) (-2.132) (-1.520) (-1.538) SI Z E +/- -0.400*** -0.377*** -0.381*** (-13.904) (-13.161) (-12.519) C APE X +/- 10.417*** 10.411*** (43.124) (43.045) L E V - -0.064 (-0.370) No. of observations 93912 93912 93912 93912 No. of firms 4472 4472 4472 4472 :KLWH¶V7HVW 8.883 13.484 17.550 15.014 p-value 0.000 0.000 0.000 0.000 BP Test 8.883 13.908 14.458 13.115 p-value 0.000 0.000 0.000 0.000 R-squared 0.000 0.002 0.021 0.021 Adjusted R-squared 0.000 0.002 0.021 0.021 F-statistics 3.875*** 51.25*** 412.637*** 343.814*** *, **, and *** indicate significant at 10%, 5%, and 1%, respectively
T-statistics are in parentheses
Results based on Return on Assets
[35]
column (4) LEV is included which has the predicted negative sign and is also strongly significant, consistent with Liu, Uchida, and Yang (2012). Furthermore, when adding control
variables the explanatory power of the regression increases (Adjusted R2 raises to 0.016, 0.224,
and 0.234 in columns (2), (3), and (4) respectively) but the effect of French civil law firms on ROA become statistically insignificant.
Table 10: Random effect model results of R O A
Explanatory variables Expected (1) (2) (3) (4)
sign Intercept 0.054*** -0.157*** -0.084*** -0.045*** (16.861) (-25.602) (-14.316) (-7.633) F renchCivDU M - -0.001 -0.011*** -0.011*** -0.003 (-0.124) (-2.616) (-2.475) (-0.788) GermanC ivDU M - -0.003 -0.022*** -0.040*** -0.035*** (-0.453) (-3.565) (-6.462) (-5.765) ScandinavianCivDU M - -0.029*** -0.016*** -0.025*** -0.017*** (-4.417) (-2.524) (-3.957) (-2.714) SI Z E +/- 0.037*** 0.037*** 0.045*** (39.327) (41.968) (50.167) C APE X +/- -0.970*** -0.957*** (-160.138) (-158.840) L E V - -0.176*** (-36.847) No. of observations 93912 93912 93912 93912 No. of firms 4472 4472 4472 4472 :KLWH¶V7HVW 6.501 26.318 24115.410 16392.770 p-value 0.000 0.000 0.000 0.000 BP Test 6.501 39.900 2725.433 2284.154 p-value 0.000 0.000 0.000 0.000 R-squared 0.000 0.016 0.224 0.234 Adjusted R-squared 0.000 0.016 0.224 0.234 F-statistics 7.202*** 392.836*** 5411.675*** 4794.175*** *, **, and *** indicate significant at 10%, 5%, and 1%, respectively
[36]
On the whole, Table 10 shows that firms in common law countries have significantly higher firm values than firms in civil law countries during the global financial crisis, which is consistent with the main hypothesis. However, the results are only stable for firms in Scandinavian and German civil law countries.
Results based on Return on Equity
Results of the random effects regression model based on ROE are presented in Table 11. In column (1), we find that the civil law firms have the predicted negative signs just as in Table 10. Column (2) shows the same results as in Table 10; the coefficients of the civil law firms are negative and significant. Similarly, column (3) shows the same results as in Table 10. Finally, in column (4), LEV is included, which has the predicted negative sign and is also strongly significant. Overall, Table 11 also shows that firms in common law countries have significantly higher firm values than firms in civil law countries during the global financial crisis, which is consistent with the main hypothesis. However, the results are only stable for firms in German civil law countries. When adding control variables the explanatory power of the regression
increases (Adjusted R2 raises to 0.004, 0.005, and 0.069 in columns (2), (3), and (4) respectively)
but the effects of French and Scandinavian civil law firms on ROE become statistically insignificant.
In general, the choice of dependent variable has little effect on the relationship between legal systems and firm performance.
[37]
significant negative relationship between TQ and the French civil law firms, between ROA and the German and Scandinavian civil law firms, and between ROE and the German civil law firms.
Table 11: Random effect model results of R O E
Explanatory variables Expected (1) (2) (3) (4)
sign Intercept 0.151*** -0.098*** -0.083*** 0.137*** (16.861) (-6.261) (-5.278) (8.769) F renchCivDU M - -0.001 -0.051*** -0.055*** -0.010 (-0.124) (-4.785) (-5.167) (-0.893) GermanC ivDU M - -0.003 -0.083*** -0.086*** -0.059*** (-0.453) (-5.351) (-5.558) (-3.817) ScandinavianCivDU M - -0.029*** -0.073*** -0.074*** -0.024 (-4.417) (-4.638) (-4.749) (-1.538) SI Z E +/- 0.043*** 0.043*** 0.098*** (18.076) (17.970) (39.967) C APE X +/- -0.018*** -0.104*** (-10.079) (-6.014) L E V - -1.090*** (-82.078) No. of observations 93912 93912 93912 93912 No. of firms 4472 4472 4472 4472 :KLWH¶V7HVW 6.724 18.129 12.685 182.147 p-value 0.000 0.000 0.000 0.000 BP Test 6.724 7.886 7.508 152.512 p-value 0.000 0.000 0.000 0.000 R-squared 0.001 0.004 0.005 0.069 Adjusted R-squared 0.001 0.004 0.005 0.069 F-statistics 16.773*** 95.186*** 96.433*** 1167.354*** *, **, and *** indicate significant at 10%, 5%, and 1%, respectively
T-statistics are in parentheses
[38]
Moreover, the results are also consistent with the study of Lemmon and Lins (2003). They investigate the effects of corporate governance on firm performance during the East Asian financial crisis and find that firm values decline more in civil law countries, thus in countries with weak legal protection. For instance, the fact that common law countries have market-based financial systems and civil law countries have bank-based financial systems can explain why civil law firms have lower firm values during financial crises. During the global financial crisis banks tightened their lending conditions, which reduced the risk exposition but led to substantial decrease in the amount of loans granted. As a result, civil law firms perform worse than their counterparts under common law, because they have fewer possibilities to undertake investment opportunities due to lack of liquidity (Demirgüç-Kunt and Maksimovic, 2002).
6. C O N C L USI O N
[39]
law countries. In total, there are 4472 firms available resulting in 93912 firm-month observations.
Due to ³the WKHRU\RIODZDQGILQDQFH´, that shows that the common law system provides a better framework for financial development and economic growth than the civil law system, and the fact that the British common law system provides investors with better legal protection and better law enforcement than the civil law systems do, one would expect firm performance to differ between these legal systems also during a global financial crisis. According to prior research and their findings we have formulated the main hypothesis that during the global financial crisis, firm values are expected to be significantly higher for firms in common law countries. To test the main hypothesis regression analyses are conducted for all dependent variables: TQ, ROA, and ROE that will be regressed on dummy variables and several control variables.
[40]
another reason why civil law firms have lower values is because in civil law countries cross-ownership is wider spread. In normal times this may increase firm performance but during crises it may increase the probability of failure, as default of one firm may cause problems to several others firms (Tyrell and Schmidt, 2001).
All in all, from the results can be concluded that the common law system was the superior law system during this turmoil.
Limitations
[41]
[42] R E F E R E N C ES
Anderson, R.C. and Reeb, D.M., 2003, FoundingǦfamily ownership and firm performance: evidence from the S&P 500, The Journal of Finance, vol. 58, no. 3, pp. 1301-1327. Bauer, R., Guenster, N. and Otten, R., 2004, Empirical evidence on corporate governance in
Europe: The effect on stock returns, firm value and performance, Journal of Asset Management, vol. 5, no. 2, pp. 91-104.
Beck, T., Demirguc-Kunt, A. and Levine, R., 2004, Law and firms' access to finance, American Law and Economics Review, vol. 7, no. 1, pp. 211-252.
Beck, T., Demirguc-Kunt, A. and Levine, R., 2003, Law and finance: why does legal origin matter?, Journal of Comparative Economics, vol. 31, no. 4, pp. 653-675.
Beiner, S., Drobetz, W., Schmid, M.M., and Zimmermann, H., 2006, An integrated framework of corporate governance and firm valuation, European Financial Management, vol. 12, no. 2, pp. 249-283.
Bhagat, S. and Black, B., 1999, The Uncertain Relationship between Board Composition and Firm Performance, Business Lawyer, vol. 54, pp. 921-963.
Bhagat, S. and Bolton, B., 2008, Corporate governance and firm performance, Journal of Corporate Finance, vol. 14, no. 3, pp. 257-273.
Bhattacharya, P. and Graham, M., 2007, Institutional ownership and firm performance: evidence from Finland, Working paper, Deakin University.
Black, B.S., Love, I., and Rachinsky, A., 2006, Corporate governance indices and firms' market values: Time series evidence from Russia, Emerging Markets Review, vol. 7, no. 4, pp. 361-379.
Brooks, C., 2008, Introductory econometrics for finance, Cambridge Univ Pr.
Choi, F.D.S. and Meek, G.K., 2008, International accounting, Pearson: Prentice Hall: Pearson Education International, Upper Saddle River (New Jersey).
Claessens, S. and Djankov, S., 1999, Ownership concentration and corporate performance in the Czech Republic, Journal of Comparative Economics, vol. 27, no. 3, pp. 498-513.
Daines, R., 2001, Does Delaware law improve firm value?, Journal of Financial Economics, vol. 62, no. 3, pp. 525-558.
[43]
Demirgüç-Kunt, A. and Maksimovic, V., 2002, Funding growth in bank-based and market-based financial systems: evidence from firm-level data, Journal of Financial Economics, vol. 65, no. 3, pp. 337-363.
Demsetz, H. and Villalonga, B., 2001, Ownership structure and corporate performance, Journal of Corporate Finance, vol. 7, no. 3, pp. 209-233.
Durnev, A. and Kim, E., 2005, To steal or not to steal: Firm attributes, legal environment, and valuation, The Journal of Finance, vol. 60, no. 3, pp. 1461-1493.
Erkens, D.H., Hung, M., and Matos, P., 2012, Corporate governance in the 2007±2008 financial crisis: Evidence from financial institutions worldwide, Journal of Corporate Finance, vol. 18, no. 2, pp. 389-411.
Gedajlovic, E.R. and Shapiro, D.M., 1998, Management and ownership effects: Evidence from five countries, Strategic Management Journal, vol. 19, no. 6, pp. 533-553.
Gompers, P., Ishii, J., and Metrick, A., 2003, Corporate governance and equity prices, The Quarterly Journal of Economics, vol. 118, no. 1, pp. 107-156.
Graff, M., 2008, Law and Finance: Common Law and Civil Law Countries Compared²An Empirical Critique, Economica, vol. 75, no. 297, pp. 60-83.
Greene, W.H. and Zhang, C., 2003, Econometric analysis, Prentice hall Upper Saddle River, NJ.
Harrison, A. and Sepúlveda, C., 2011, Learning from developing country experience: growth and economic thought before and after the 2008-09 crisis, World Bank Policy Research
Working Paper No. 5257.
Henry, D., 2008, Corporate Governance Structure and the Valuation of Australian Firms: Is There Value in Ticking the Boxes?, Journal of Business Finance & Accounting, vol. 35, no. 7Ǧ8, pp. 912-942.
Klapper, L.F. and Love, I., 2004, Corporate governance, investor protection, and performance in emerging markets, Journal of Corporate Finance, vol. 10, no. 5, pp. 703-728.
La Porta, R., Lopez-de-Silanes, F., Shleifer, A., and Vishny, R., 2000, Investor protection and corporate governance, Journal of Financial Economics, vol. 58, no. 1, pp. 3-27.
La Porta, R., Lopez-de-Silanes, F., Shleifer, A., and Vishny, R.W., 2002, Investor Protection and Corporate Valuation, Journal of Finance, vol. 57, no. 3, pp. 1147-1170.
[44]
Lemmon, M.L. and Lins, K.V., 2003, Ownership Structure, Corporate Governance, and Firm Value: Evidence from the East Asian Financial Crisis, Journal of Finance, vol. 58, no. 4, pp. 1445-1468.
Levine, R., 1998, The Legal Environment, Banks, and Long-Run Economic Growth, Journal of Money, Credit and Banking, vol. 30, no. 3, Part 2: Comparative Financial Systems, pp. 596-613.
Lins, K.V., 2003, Equity Ownership and Firm Value in Emerging Markets, Journal of Financial & Quantitative Analysis, vol. 38, no. 1, pp. 159-184.
Liu, C., Uchida, K., and Yang, Y., 2012, Corporate governance and firm value during the global financial crisis: Evidence from China, International Review of Financial Analysis, vol. 21, pp. 70-80.
Mitton, T., 2002, A cross-firm analysis of the impact of corporate governance on the East Asian financial crisis, Journal of Financial Economics, vol. 64, no. 2, pp. 215-241.
Morck, R., Shleifer, A., and Vishny, R.W., 1988, Management ownership and market valuation: An empirical analysis, Journal of Financial Economics, vol. 20, pp. 293-315.
Ntim, C., Opong, K., and Danbolt, J., 2011, Board Size and Firm Value: Evidence from South Africa, Working paper, University of Glasgow
Shakir, R., 2008, Board size, executive directors and property firm performance in Malaysia, Pacific Rim Property Research Journal, vol. 14, no. 1, pp. 66-80.
Tarraf, H., 2011, The Role of Corporate Governance in the Events Leading up to the Global Financial Crisis: Analysis of Aggressive Risk-Taking, Global Journal of Business Research, vol. 5, no. 4, pp. 93-105.
Tirole, J., 2006, The theory of corporate finance, Princeton University Press, Princeton, N.J. Tyrell, M. and Schmidt, R.H., 2001, Pension systems and financial systems in Europe: a
comparison from the point of view of complementarity, ifo studien, vol. 47, no. 4, pp. 469-503.
Verbeek, M., 2000, A Guide to Modern Econometrics, John Wiley & Sons.
Watson, A., 1993, Legal transplants: An approach to comparative law, Univ of Georgia Pr. Weir, C., Laing, D., and McKnight, P.J., 2002, Internal and external governance mechanisms:
