Corporate Governance, Investor Protection, and Performance

in Emerging Markets

Leora F. Klapper
The World Bank

Inessa Love
The World Bank

Abstract:

Recent research studying the link between law and finance has concentrated on country-level
investor protection measures and focused on differences in legal systems across countries and legal
families. We use recent data on firm-level corporate governance rankings across 14 emerging
markets and find that there is wide variation in firm-level governance across countries in our
sample and that the average firm-level governance is lower in countries with weaker legal systems.
We explore the determinants of firm-level governance and find that governance is correlated with
the extent of the asymmetric information and contracting imperfections that firms face. We also
find that better corporate governance is highly correlated with better operating performance and
market valuation. Finally, we provide evidence that firm-level corporate governance provisions
matter more in countries with weak legal environments. These results suggest that firms can
partially compensate for ineffective laws and enforcement by establishing good corporate
governance and providing credible investor protection.

JEL Classifications: G3, F3

Keywords: Corporate Governance, International Finance, Law and Finance

World Bank Policy Research Working Paper 2818, April 2002

The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange
of ideas about development issues. An objective of the series is to get the findings out quickly, even if the
presentations are less than fully polished. The papers carry the names of the authors and should be cited
accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors.
They do not necessarily represent the view of the World Bank, its Executive Directors, or the countries they
represent. Policy Research Working Papers are available online at http://econ.worldbank.org.

Corresponding author: Leora Klapper, The World Bank, 1818 H Street, NW, Washington, DC, 20433,
tel: 202-473-8738, fax: 202-522-1155, email:[email protected].

We thank Geert Bekaert, Stijn Claessens, Asli Demirguc-Kunt, Simeon Djankov, Olivier Fremond,
Mariassunta Giannetti, Christian Harm, Charles Himmelberg, Simon Johnson, and Rohan Williamson for
useful discussions. We also thank Victor Sulla for providing excellent research assistance.
1. Introduction

Previous research studying the link between law and finance has concentrated on

corporate governance around the world and focused on differences in legal systems across

countries and legal families. This rapidly developing body of literature began with the finding

that the laws that protect investors differ significantly across countries, in part because of

differences in legal origins (see La Porta, Lopez-de-Silanes, Shleifer, and Vishny, 1998). Recent

literature finds that cross-country differences in laws and their enforcement affect ownership

structure, dividend payout, availability and cost of external finance, and market valuations. 1

However, many provisions in country-level investor protection laws may not be binding

since firms have the flexibility in their corporate charters and bylaws to either choose to “opt-

out” and decline specific provisions or adopt additional provisions not listed in their legal code

(see Easterbrook and Fische l, 1991; Black and Gilson, 1998). For example, firms could improve

investor protection rights by increasing disclosure, selecting well- functioning and independent

boards, imposing disciplinary mechanisms to prevent management and controlling shareholders

from engaging in expropriation of minority shareholders, etc. Therefore, it is likely that firms

within the same country will offer varying degrees of protection to their investors.

A number of recent papers have studied firm- level corporate governance mechanisms,

but most of these studies have concentrated almost exclusively on OECD and US countries (see

Shleifer and Vishny, 1997, and Maher and Andersson, 2000, for comprehensive surveys). For

example, a recent paper by Gompers, Ishi, and Metrick (2001) used differences in takeover

defense provisions to create a corporate governance index of US firms and found that firms with

stronger shareholder rights have better operating performance, higher market valuation, and are

1
more likely to make acquisitions. However, until recently there was no empirical evidence on

the differences in firm- level governance mechanisms across firms in emerging markets. An

exception is Black (2000), which found that the governance practices of Russian corporations are

strongly related to implied value ratios.

In addition to the recent attention given to differences in legal systems across countries,

another interesting empirical question is whether there is variation in firm- level governance

standards within countries and the relationship between firm-specific governance mechanisms

and country-level laws governing investor protection. The relationship between the country-

level legal infrastructure and firm- level corporate governance mechanisms is far from obvious.

One supposition is that firms in countries with weak laws would want to adopt better firm- level

governance to counterbalance the weaknesses in their country’s laws and their enforcement and

signal their intentions to offer greater investor rights. This would suggest a negative correlation

between the strength of firm- level governance and country- level laws. A second possibility is

that in countries with weak laws the degree of flexibility of firms to affect their own governance

is likely to be smaller (i.e. the firm is likely be constrained by the country- level legal provisions),

which would imply a positive correlation. This question has not previously been empirically

studied.

The second question that we address is which firms within countries have relatively better

governance? LLSV (1998) argued that greater investor protection increases investors’

willingness to provide financing and should be reflected in lower costs and greater availability of

external financing. This suggests that we should find that firms with the greatest needs for

financing in the future will find it the most beneficial to adopt better governance mechanisms

1
For example, La Porta, Lopez-de-Silanes, Shleifer, and Vishny (1999a, 1999b, and 2000,
further referred to as LLSV), Claessens, Djankov, and Lang (2000), Berkowitz, Pistor, and

2
today. Finally, we address the most important – and difficult – question, which is whether or not

firm- level differences in corporate governance matter for future performance, market valuation,

and access to external finance. This paper is a first attempt to address some of these questions

and suggest avenues for future research.

The surge of interest in the topic of corporate governance among investment banks, rating

agencies, and other specialized financial institutions has made it possible to address these

questions empirically as a number of private firms have started to collect firm- level data on

differences in corporate governance across firms in different countries. 2 In a recent report,

Credit Lyonnais Securities Asia (further referred as CLSA) calculated an index with corporate

governance rankings for 495 firms across 25 emerging markets and 18 sectors. 3 The descriptive

statistics presented in the CLSA report show that companies ranked high on the governance

index have better operating performance and higher stock returns. We use the governance

rankings produced by CLSA to further investigate the relationship between firm- level

governance, other firm- level characteristics, and the country- level legal environment.

Our study proceeds in two parts. In the first part we investigate the determinants of firm-

level governance. First, we observe that there is wide variation in firm- level governance,

although the degree of variation is not systematically related to countries’ legal environments. In

other words, we find that there are well- governed firms in countries with weak legal systems and

badly governed firms in countries with strong legal systems. However, we find that the overall

level of firm- level governance is strongly positively related to country- level measures of investor

Richard (2002), Lombardo and Pagano (2000), and Beck, Demirguc-Kunt, and Levine (2001).
2
In addition to the data used for this paper, described below, similar governance rankings were
produced by Deutche Bank for Latin American countries, Deminor Rating for Western European
countries, and more recently S&P 500 has entered the governance ranking market and is
producing rankings for both developed and emerging markets.

3
protection, i.e. average governance is higher in countries with stronger legal protection. For

example, increasing the country- level index of efficiency of the legal system from the low to the

median value increases the average firm- level governance by about half a standard deviation, a

large and economically significant effect. This supports the argument that firms have limited

flexibility to affect their governance, which implies that improving the country- level efficiency

of the legal system is likely to lead to an increase in the average firm- level governance.

Next, we investigate the firm- level determinants of governance and find support for the

hypothesis that a growing firm with large needs for outside financing has more incentive to adopt

better governance practices in order to lower its cost of capital. We also investigate whether

differences in firm- level contracting environments affect a firm’s choice of governance

mechanisms, in line with arguments put forth in Himmelberg, Hubbard, and Palia (1999). They

argued that some firms would find it easier to expropriate from minority shareholders due to the

nature of their operations; therefore, these firms would find it optimal to impose ax-ante stricter

governance mechanisms to prevent ex-post expropriation. For example, the composition of the

assets of a firm will affect its contracting environment because it is easier to monitor and harder

to steal fixed assets (i.e. machinery and equipment) then “soft” capital (i.e. intangibles, R&D

capital, and short-term assets, such as inventories.) Therefore, firms operating with higher

proportions of intangible assets may find it optimal to adopt stricter governance mechanisms to

signal to investors that they intend to prevent the future misuse of these assets. We find support

for this hypothesis using a capital intensity measure, which is significantly negatively correlated

with governance.

3
Credit Lyonnais Securities Asia report entitled “Saints and Sinners: Who’s got religion”, April
2001.

4
 In the second part of our paper we investigate the relationship between governance and

performance. We find that better corporate governance is associated with higher operating

performance (ROA) and higher Tobin’s-Q. After including country fixed effects we find that the

correlation of performance measures with governance becomes twice as large and statistically

more significant. This suggests that improvements in governance relative to the country-average

are more important than the absolute value of the index.

Finally, we explore the cross-country nature of our sample and look at the interaction of

firm- level governance and country-level investor protection. We test whether good corporate

governance matters more or less in countries with weak shareholder protection and judicial

efficiency. One hypothesis is that in countries with weak judicial efficiency, additional charter

provisions would not be enforced and therefore firms would be powerless to independently

improve their investor protection. In this case we should find that firm- level governance matters

less in countries with weak legal systems. An alternative hypothesis is that in countries with

weak legal systems, investors would welcome even small improvements in governance relative

to other firms, in which case we should find that good governance matters more in bad legal

environments. This is consistent with the assertion in Doidge, Karolyi, and Stulz (2001) that by

establishing good governance mechanisms the controlling shareholders give up more of their

benefits of control in countries where such benefits are high (i.e. investor protection is low),

which will be reflected in measures of performance and market valuation. Our results suggest

that good go vernance practices are more important in countries with weak shareholder rights and

inefficient enforcement. This finding has strong policy implications and suggests that

recommending to firms to adopt good governance practices is even more important in countries

with weak legal systems.

5
 An important caveat on our results showing the link between governance and

performance is the likely endogeneity of corporate governance practices. For example, as we

have already argued, a growing firm with large needs for outside financing has more incentive to

adopt better governance practices in order to lower its cost of capital. These growth

opportunities would also be reflected in the market valuation of the firm, thus inducing a positive

correlation between governance and Tobin’s-Q.4 Since our governance data have no time-

variation we cannot address the issue of causality directly and leave this issue for future research.

However, we attempt to mitigate this problem by adding several control variables that could

proxy for growth opportunities such as size, average growth in sales, and the rate of investment

and find that our governance results are not spuriously caused by these omitted variables. We

also control for capital intensity in our performance regressions and the governance results

remain significant. Thus, although the concern of reverse causality is a clear drawback of our

governance-performance results, at a minimum we confirm that our results are not caused by

omitted variable bias and leave establishment of causality for further research.

The paper proceeds as follows: Section 2 describes the CLSA corporate governance

survey and summarizes our firm- and country-level data. Section 3 discusses the potential

endogeneity of governance and reports results on the determinants of corporate governance

behavior. Section 4 reports our results, which include correlation tests between measures of

corporate governance and legality, Tobin’s-Q, and return on assets (ROA.) Section 5 concludes.

4 Similar endogeneity problem arises in the studies of ownership and performance, as argued
by Himmelberg, Hubbard, and Palia (1999) who propose panel data techniques and instrumental

6
2. Data

The CLSA report includes corporate governance (CG) rankings on 495 companies in 25

countries. The CLSA sample is selected based on two criteria – firm size and investor interest.5

The CG ranking compiled by CLSA is a composite of 57 qualitative, binary (yes/no) questions,

designed to avoid subjectivity. Appendix 1 reports an abbreviated version of the questionnaire.

Each question is constructed such that answer ‘Yes’ adds one point to the governance score. The

analysts were given strict instructions to answer negatively if they had any doubts or if there

were any unresolved controversies. According to CLSA, about 70% of the questions are based

on objective facts and the remaining questions represent analysts’ opinions. Unfortunately,

reliance on analysts' opinions worsens the endogeneity problem in the governance-performance

regressions, as it is possible that analysts could rely on past performance to form their opinions.6

However, most of the subjective questions refer to the historical experience of the firm and

specifically ask whether there have been any violations or controversial disputes over

shareholder rights in the past, usually within the past five years (for example, questions 6, 20, 36-

38, 40, 42, 47). These questions are likely to reflect the reputation of the company, i.e. whether

in past years the company had established a reputation for respecting minority shareholder rights.

Such reputational characteristics can be considered a part of corporate governance practices and,

if past behavior is any indication of future behavior, could serve as a useful measure of corporate

governance behavior.

variables to address this question.

5
A recent paper by Khanna, Korgan, and Palepu (2001) uses this data to study convergence of
corporate governance practices across countries and confirms the sample selection criteria based
on their detailed study of India.
6
This bias is likely to be worse for regressions including past returns since analysis answering
the questions are most concerned with return performance, as it appears that the main purpose of
the CLSA original report was to give buy/sell recommendations based on the governance
rankings. This is the main reason why we do not study returns as a measure of performance.

7
 The questions in the CLSA report cover seven broad categories: management discipline,

transparency, independence, accountability, responsibility, fairness, and social awareness. A

representative questions for each category is listed below:

1. Discipline: Is expected remuneration for executive(s) tied to the value of the shares?
2. Transparency: Does the company publish its Annual Report within four months of the
end of the financial year?
3. Independence: Is the Chairman an independent, non-executive director?7
4. Accountability: Are the board members and members of the executive/management
committee substantially different?
5. Responsibility: Does the company have a known record of taking effective measures
in the event of mismanagement? Are there mechanisms to allow punishment of the
executive/management committee in the event of mismanagement?
6. Fairness: Are voting methods easily accessible (i.e. proxy voting)? Do all equity
holders have the right to call General Meetings?
7. Social Awareness: Is the company explicitly environmentally conscious?

Our main governance index, further referred to as GOV, is the sum of first six categories

and excludes the social awareness category, which is not relevant for corporate governance

(although our results are robust to the inclusion of this category). Furthermore, we do not study

the disaggregated indices, since the categories seem to overlap and are categorized with some

subjectivity. (For example, question 28 could easily belong to the Independence section;

questions 37 and 39 could belong to the Discipline section; question 45 to Transparency section;

etc.) Also, the distinction between the Independence and Accountability sections is imprecise

and the Responsibility and Fairness sections both reflect minority shareholder rights (as are the

questions 20 and 22 from the Independence section).

In order to include firm- level accounting data, we merged the CLSA data with

Worldscope data (June 2001). To avoid the anomalous period of the Asian Crisis we included

7
Independence of directors must be demonstrated by either being appointed through nomination
of non-major shareholders or having on record voted on certain issues against the rest of the
Board. If no evidence of independence, other than being stated to be so by the company and the
director(s), then answer "No".

8
only firms that had available accounting data beginning in at least 1998. 8 We began with 451

firms with non- missing accounting data. After excluding 50 banks, 20 firms in Eastern Europe

and China (excluded because of unavailable legal indices), and seven firms in countries with less

than three firms each (Argentina, Columbia, Greece and Mexico), our sample was reduced to

374 firms in 14 countries – Brazil, Chile, Ho ng Kong, India, Indonesia, Korea, Malaysia,

Pakistan, Philippines, Singapore, South Africa, Taiwan, Thailand, and Turkey. We also

excluded some outlier observations in the individual regressions, which we describe in the

relevant sections. 9 We also added a dummy indicating if a firm trades American Depository

Receipts (ADRs) in the United States. 10

The distribution of our GOV index across countries is shown in Table 1, Panel A. As

shown, our sample is not equally distributed across countries - 68% of firms are in East Asia,

19% of firms are in South Asia, and 11% of firms are in Latin America. Mean GOV rankings

overall are 54.16 and vary from a country average of 31.85 in Pakistan to 66.53 in South Korea.

There is also great variation within countries – for example, the corporate governance ranking of

firms in Pakistan varies from 17.25 to 66.68. These summary statistics highlight the firm- level

variations in corporate governance practices even within countries and families of legal origins.

8
Our sample contains 29 firms with 1998 as the last available data. We include in all regressions
a dummy for year 1998 to control for time effects. The dummy is always negative and
significant (it is not reported) but does not affect the significance of our governance results.
9
As is commonly found, the distribution of Tobin's-Q is significantly skewed to the right with
some firms having extremely high values, which could significantly influence our results without
representing the common patterns. Therefore, we excluded firms with Tobin’s Q above 10
(which excludes 17 firms - slightly less than 5% of the samp le) and firms with ROA above the
99th and below the 1st percentiles (6 firms). To use as many observations as possible we exclude
only Q outliers in Q regressions and ROA outliers in ROA regressions – therefore the sample in
ROA regressions is slightly different from the sample in Q regressions. In addition, some of the
control variables are missing for some observations, which further causes slight variation in the
sample size across regressions.
10
We identify ADRs traded on the NYSE, AMEX, and NASDAQ us ing the JP Morgan website:
www.adr.com.

9
 We use three country-level measures of legal efficacy. The first is Judicial Efficiency,

which is an index constructed by the International Country Risk Guide (2000). The second is

Shareholder Rights, which is the sum of dummies identifying one-share/one-vote, proxy by mail,

unblocked shares, cumulative vote/proportional representation, preemptive rights, oppressed

minority, and percentage of shares needed to call a shareholders meeting (LLSV, 1998.) The

third is Legality, which is an index of the strength of the legal system and institutional

environment constructed as a weighted average of Judicial Efficiency (identical to our first

index), Rule of Law, Corruption, Risk of Expropriation, and Risk of Contract Repudiation (this

index is constructed using principal components analysis by Berkowitz, Pistor, and Richard,

2002.) We use three different measures in order to cover separately the existence of laws

(Shareholder Rights) and the effectiveness of their implementation (Judicial Efficiency), as well

as the overall legal environment (Legality). Summary statistics and sample distributions for the

legal indicators are given in Table 1, Panel A.

We use two main performance measures: Tobin’s-Q as a measure of market valuation of

the firm and return on assets (ROA) as a measure of operating performance. 11 Summary statistics

and sample distributions for Tobin’s-Q and ROA are given in Table 1, Panel B. For 1999, the

average Tobin’s-Q is 2.09 and varies from country-average 1.16 in Turkey to 3.67 in Taiwan.

The median Q (1.39) is slightly higher then the median reported in other studies (for example

LLSV, 1999) reflecting the overall good performance of the global economy in 1999. The

standard deviation is 1.68, reflecting the significant variation in performance across firms. The

11
Tobin’s-Q is defined as the market value of assets (calculated as book value of assets minus
book value of equity plus market value of equity) over book value of assets, and return on assets
(ROA) is defined as net income over total assets. Other measures of operating performance –
such as gross margin and return on equity – give similar results.

10
country-average ROA is 0.08, with the highest average performance of 0.11 in India and the

lowest performance of 0.01 in Brazil.

3. Determinants of Governance

3.1. Hypotheses

As discussed in the introduction, corporate governance is likely to be endogenously

determined. In this section we discuss variables that in theory could be associated with firms

adopting better governance mechanisms and present empirical results in support of these

theories. We deliberately do not include any performance-related measures as governance

determinants as we will study governance-performance relationships in the next section.

Recognizing the endogeneity of the governance, we can only interpret all our results as partial

correlations. However, the exe rcise in this section helps us better understand the potential

sources of this endogeneity.

Our discussion closely follows Himmelberg, Hubbard, and Palia (1999), who argued that

the degree of managerial ownership is endogenously determined by a firm’s contracting

environment and therefore ownership-performance regressions could spuriously pick up the

effect of this unobserved heterogeneity. As managerial ownership is only one of many

governance mechanisms, this argument could be easily transferred to othe r mechanisms such as

managerial compensation, board structure, disclosure, and other minority shareholder protections

that are included in our governance index.

We consider several causes for the variation in contracting environments. The most

obvious is the overall country- level measure of shareholder rights and their enforcement. For

example, if a country's laws offer weak shareholder protection it might be costly for firms to

adopt different provisions in their corporate charters because it will be difficult for investors and

11
judges to understand non-standard contracts, as argued by LLSV (1998). Therefore, firms in

countries with overall weak legal environments may not have much flexibility to improve their

own investor protection and may consequently have lower corporate governance indices, on

average. In this case we should expect a positive relationship between the quality of country-

level legal systems and the average of firm- level governance indices within each country. In the

extreme case, for example, firms would be completely powerless to change the overall legal

environment with internal governance mechanisms. In another extreme, if firms could

completely “overwrite” the legal code in their own contracts, we would observe better

governance in countries with bad legal systems as these firms would be more in “need” of good

governance mechanisms to compensate for their bad legal systems. In this case we would

observe a negative relationship between country- level legal systems and firm- level governance

mechanisms. However, this case is very unlikely as a large body of evidence shows that the

legal system does matter, most likely because if the enforcement of contracts is weak, then firms

will be unable to “overwrite” their country’s legal system and their flexibility to improve their

corporate governance would be limited. 12

In addition to country- level differences in legal efficiency, it is likely that there will be

variations across firms within contracting environments, as initially proposed by Himmelberg, et

al. (1999) and further developed by Himmelberg, Hubbard, and Love (2001). These papers

suggest that firm- level characteristics affect the level of investor protection. For example, the

composition of a firm’s assets will affect its contracting environment because it is easier to

monitor and harder to steal fixed assets (i.e. machinery and equipment) then “soft” capital

(intangibles, R&D capital, and some short-term assets, such as inventories). Therefore, a firm

12
In a related study Dimitrov (2001) shows that firms in countries with weak legal environments
are less likely to have independent directors, since firms are unable to show credible

12
operating with a higher proportion of intangible assets may find it optimal to adopt stricter

governance mechanisms to prevent misuse of these assets, i.e. we should observe negative

correlation between the proportion of fixed assets and governance. It is important to keep this

relationship in mind when we later estimate the effect of governance on performance, since the

level of intangibles may also result in higher Tobin’s-Q since, in general, the market values

intangibles higher than their book values. Similarly, operating performance should be higher

since the denominator (for example, total assets) does not fully account for all intangibles. In our

performance regressions we control for asset composition and find that the effect of governance

on performance is not driven by this source of heterogeneity. We use fixed capital (i.e. property

plant and equipment) to total sales ratio, denoted K/S, as a measure of the relative importance of

fixed capital in the firm’s output. 13

Another source of endogeneity could arise because of differences in unobserved growth

opportunities. Firms with good growth opportunities will need to raise external financing in

order to expand and may therefore find it optimal to improve their governance mechanisms as

better governance and better minority shareholder protection will be likely to lower their costs of

capital. 14 For example, if Tobin’s-Q is higher for firms with good growth opportunities, this

could also be a cause of endogeneity of governance in the performance regressions and result in

positive spurious correlations with governance. Unfortunately there is no good measure of the

growth opportunities besides Tobin’s-Q. As an arguably imperfect measure, we use the average

commitments not to expropriate.

13
Himmelberg, et al. (1999) also used research and development intensity (R&D) and
advertising expenses as additional measures of the “intangibility” of the assets. Unfortunately,
the Worldscope database does not include advertising expenses as an independent variable and
R&D amounts are reported as missing for most firms in our sample.
14
See La Porta, et al. (1999a), Lombardo and Pagano (2000) and Himmelberg, Hubbard, and
Love (2001) among others on the relationship between investor protection and the cost of capital.

13
real growth rate in sales for the last three years, denoted SalesGR, as a proxy for future growth

(and growth opportunities). 15

We also explore the effects of differences in firm size on governance. The effect of size

is ambiguous as large firms may have greater agency problems (because it is harder to monitor

them or because of the “free cash flows” argument of Jensen, 1986) and therefore need to

compensate with stricter governance mechanisms. Alternatively, small firms may have better

growth opportunities and, as implied by the argument above, greater need for external finance

and better governance mechanisms. For this reason we also use size as a control variable in the

Tobin's-Q regressions. We use the natural log of sales (in US$), denoted Log(SalesUS) as a

measure of firm size.

We also include a dummy indicating if a firm trades American Depository Receipts

(ADRs). There are several reasons to expect that firms that trade in the US should have better

corporate governance rankings. First, firms listed on a US exchange are required to comply with

US GAAP accounting standards, which might improve their transparency. 16 Second, firms that

list shares on a US exchange are subject to many SEC laws and regulations that protect minority

shareholders. Cofee (1999), Stulz (1999) and Reese and Wesibach (2001) find that firms in

countries with weak minority shareholder rights list in the US in order to better protect foreign

investors. We expect that since reporting standards and investor protection in the US are much

15
Past growth rates will be correlated with future growth if there are investment adjustment
costs, “time to build” (i.e. it takes several periods to make new investment fully operational), or
if the shocks to productivity are serially correlated.
16
For example, recent papers by Cantale (1996), Fuerst (1998) and Moel (1999) argue that firms
list in the US in order to signal their higher quality to investors through greater disclosure and
transparency and higher accounting standards.

14
higher than in most other countries, firms in emerging markets would be required to improve

their corporate governance provisions in order to list overseas. 17

Finally we test whether the relationship between ADR issuance and governance varies

with differences in country- level investor protection and enforcement. For example, in a country

with a weak legal system, a firm issuing ADRs would need to make more changes to its

governance to be able to meet the more stringent disclosure and investor protection requirements.

This suggests that the increase in the governance index for ADR firms will be larger in countries

with weaker legal system, i.e. the interaction of ADR and Efficiency is expected to be negative.

To summarize, our model to study governance determinants is given by:

GOV f = β1 Log(Sales)f +β 2 SalesGrf +β3 K/Sf +β 4 LegalSystem c +β5 ADRsf

+β6 ADRsf* LegalSystem +γ. (1)

As previously implied, we expect β2 >0, β3 <0, β 4>0, β5 >0, β 6<0 and β 1 is ambiguous.

We have three different country- level indicators for the Legal System: the laws on the book

(Shareholder Rights) and the effectiveness of their implementation (Judicial Efficiency), as well

as the overall legal environment (Legality). Since Efficiency and Legality measures are both

indicators of the quality of legal enforcement, we do not include them together, but we include

shareholder rights in combinations with either Efficiency or Legality.

3.2 Evidence

We begin our exploration of the relationship between the distribution of firm- level

governance and country-level indicators with graphs and descriptive statistics. Figure 1 presents

17
Unfortunately, we are unable to establish causality – it is possible that only the firms with
good governance issue ADR’s, so our results should be interpreted as a partial correlation and
not a causal relationship.

15
a Box-and-Whisker plot of governance by country and shows that there is wide variation in the

governance rankings in most countries in our sample. The countries are sorted based on their

index of Legality and the plot shows that the degree of the variation in governance within

countries does not appear to be systematically related to the country- level measures of legal

effectiveness. Figure 2 presents a plot of country-average values of the firm- level governance

index plotted against the legality index. The relationship is strongly positive, indicating that

countries with better legal systems have on average higher firm- level governance. The

correlations presented in Panel C confirm the evidence in the figures. The correlations of

country-average governance with any of the legal indicators are significantly positive, while the

correlations of standard deviations and the interquartile range with legal indicators are

(marginally) insignificantly negative, indicating that, if anything, there is more variation in firm-

level governance in countries with weaker legal systems.

We next begin a more formal investigation. Table 2 reports our estimates of the

governance determinants and finds all signs consistent with our hypotheses. In Column 1 we

include only our three firm- level variables and find that size is positive and marginally

significant, while average growth rate and capital intensity have predicted signs and are

significant at 1%. However, the explanatory power of this regression is low, with an adjusted R2

of only 0.06. In Column 2 we add an ADR dummy, which is significant, confirming our prior

that firms that list in the US have relatively higher corporate governance standards. The size

variable is no longer significant, but this is not surprising since larger firms are more likely to list

abroad.

In Column 3 we include country- level measures of the legal environment – shareholder

rights and efficiency of the legal system – and the coefficients of these variables are positive and

16
significant, although shareholder rights is weakly significant at 10% (using the Legality index

instead of Efficiency produces very similar results). Adding these country-level variables

doubles the explanatory power of the regression (R2 raises to 0.15). Thus, we can conclude that

firms in countries with weak overall legal system have on average lower governance rankings.

The coefficients from the regression of governance on the efficiency index alone (not reported)

imply that an increase in efficiency from the country with the lowest efficiency (Indonesia) to the

country with the median efficiency (South Africa) results in an improvement in average

governance ranking of 7.7, or about half a standard deviation, which is an economically

significant effect. These results support the theoretical arguments in Shleifer and Wolfenson

(2002) that firms are unable to completely replicate a good legal environment on their own, but

must depend on a supporting efficient judicial system.

In Column 4 we add the ADR dummy and the interaction of the ADR dummy and

Efficiency and find that the interaction term is significantly negative (at 5%). This confirms our

intuition that firms that issue ADRs have better governance, and the effect is stronger in

countries with weak legal systems. 18 In Column 5 we include both firm- level and country- level

variables together and find that sales growth, capital intensity, and both measures of enforcement

remain significant while size and shareholder rights become insignificant at conventional levels

(size becomes insignificant in Column 4 and shareholder rights in Column 5). Finally, instead of

country- level legal indicators we run a regression with country dummies in Column 7 and find

that capital intensity and growth rates are still significant while size is not. The R2 in this model

is much larger, implying that unobserved country effects account for large differences in the

18
These results are larger and more significant for firms that trade on a US exchange (NYSE,
AMEX or NASDAQ) than for firms that trade 144a or OTC. This is not surprising as exchange
listed firms are likely to have greater disclosure and shareholder protection requirements and is
consistent with findings in Doidge, et al., 2001.

17
variation in governance rank ings. However, over 60% of this variation is not explained by

country effects, suggesting that firms have enough flexibility to affect their corporate governance

and investor protection.

To summarize, we find that (1) firms in countries with weak overall legal systems on

average have lower governance rankings, however there is no systematic relationship between

the variation in firm- level rankings and country- level legal efficiency; (2) past growth rates are

positively associated with good governance; (3) firms with higher proportions of fixed assets

have lower governance; and (4) firms that trade shares in the US have higher governance

rankings, especially so in countries with weak legal systems. These results confirm our intuition

of the endogeneity of governance and emphasize that it is important to control for these factors in

our performance regressions to ensure that the governance effect on performance is not

spuriously caused by any of these omitted factors.

4. Governance and Performance

4.1. Hypotheses

In this section we study the relationship between firm- level governance, the country- level

legal environment, and firm performance. Our first hypothesis tests the correlation between firm

governance and equity valuation:

Qf = α + β1 (Gov f) + γ, (2)

Where Qf equals Tobin’s-Q and Gov f equals the firm- level corporate governance ranking. To

test the robustness of this relationship we sequentially add 1-digit industry dummies, country

dummies and other firm- level control variables discussed in Section 3. We then repeat the same

exercise using ROA as an alternative performance measure.

18
 Second, we test whether the country- level legal environment is correlated with market

and operating performance and whether firm- level governance is robust to controlling for

country- level legal efficiency. Regardless of individual firm- level investor protection, firms are

also subject to country- level regulatory and legal environments that differ in laws and

enforcement. Previous literature has shown the legal environment to be related to firm

performance on the international, country, and state levels (Lombardo and Pagano, 2000, La

Porta et al., 1999, and Daines, 2001, respectively). This relationship can be explained for a

number of reasons: First, block shareholders are more likely to exploit minority shareholders in

countries that have weaker protection of minority shareholders. Second, firm- level protection of

minority rights is less likely to be effective if legal enforcement and judicial efficiency is weak.

We address this concern by first, replacing the firm- level governance index in our model with

country- level legal indicators and second, including the two measures simultaneously to test for

their relative importance.

Our third hypothesis tests the effect of the effect of the interaction of corporate

governance and judicial efficiency on firm valuation. We include in this regression:

Qf = α + β1 (Gov f) + β 2 (Effc) + β 3 (Gov f*Eff c)f + γ, (3)

where Gov equals the firm- level corporate governance ranking; Effc equals the country- level

judicial efficiency; and (Gov*Eff)f equals the interaction. We interpret the interaction term to

identify whether corporate governance matters more or less in countries with weak legal

enforcement (an alternative interpretation is explored below). There are two alternative

hypotheses relating to the differential effect of governance in countries with different levels of

investor protection. One hypothesis is that in countries with weak law enforcement the adoption

of firm- specific governance-related provisions could be less effective then in countries with good

19
enforcement (because the provisions are not enforceable and additional mechanisms such as

independent board of directors or audit committees will be powerless to discipline the insiders).

Therefore, we could find that governance matters less in countries with weak legal system (i.e. a

positive interaction). An alternative hypothesis is that governance matters more in countries with

weak legal systems. One reason for this relationship could be that investors in countries with

weak legal systems will reward more a firm that establishes a good corporate governance

framework – as even a little bit of improvement relative to other firms in a country will make a

big difference for investors – which will improve market valuation and decrease the cost of

capital (and subsequently improve operating performance).

Another reason for a negative interaction effect is suggested in a recent paper by Doidge,

Karolyi, and Stulz (2001). Their argument is based on the mounting evidence that controlling

shareholders have more incentives to expropriate from minority shareholders in countries with

less investor protection and therefore the benefits of control are greater in countries with weaker

rule of law, as recently evidenced by Nenova (2002). By establishing good governance

mechanisms, the controlling shareholders give up more of their benefits of control in countries

where such benefits are high (i.e. investor protection is low), which will be reflected in the firm’s

performance and market valuation.

Because our empirical exploration is not based on a structural model and the interaction

term treats governance and legality as symmetric variables, an alternative interpretation of the

interaction term is to test for which firms the legal environment matters more. For example, the

negative interaction term would suggest that a good legal environment matters more for firms

with weak firm- level governance. In other words, investors in firms with weak governance must

rely more on the enforcement of country-level laws and the efficiency of the courts to uphold

20
their rights, while investors in well- governed firms are less dependent on their countries’ legal

systems.

4.2 Evidence

Table 3 reports our first set of results. Panel A uses Tobin’s-Q as the dependent variable

and finds a significant positive correlation with our governance indicator. This supports our

hypothesis that firms with better corporate governance have higher market valuation. Columns 2

and 3 show that this result becomes stronger with the inclusion of country dummies (Column 2)

and does not change much with addition of 1-digit SIC code dummies (Column 3). It is

interesting to note that the governance coefficient doubles in magnitude after including country

dummies, which suggests that relative governance (i.e. relative to the country average) is more

important than the absolute value of the index. The magnitude of this effect is large, as one

standard deviation change in governance results in about a 23% increase in the value of

Tobin’s-Q.

To test whether this relationship could be spuriously caused by some omitted variables,

we add the variables that we found in Section 3 to be associated with higher governance

rankings. Column 4 shows that our results are robust to the inclusion of log sales, which

suggests that the relationship between good governance and market valuation holds regardless of

firm size. Column 5 adds the average growth rate of real sales, SalesGR, which is significantly

positive, in line with our intuition discussed in Section 3 that past sales growth is likely to be

correlated with future growth opportunities and therefore increases market valuation. Column 6

adds our measure of capital intensity, K/S, which is significantly negative, suggesting that firms

with more intangible assets (i.e. lower capital intensity) have higher Tobin’s-Q. However,

21
governance is robust to the inclusion of these additional controls, which makes us confident that

the relationship is not spuriously caused by any of the omitted variables. 19

Table 3, Panel B also shows a similar positive correlation between corporate governance

behavior and firm performance, as estimated by ROA. These results are consistent with results

found in Gompers et al. (2001), which find that firms with weaker corporate governance have

relatively lower profits in the US. However, ours is the first evidence that a relationship between

corporate governance and firm performance holds across several emerging markets, which unlike

the US are categorized by more concentrated ownership and weaker legal environments.

Table 4 extends the previous regressions to include the effect of legal indicators, with

Tobin’s-Q as the dependent variable. We include three measures of legal performance: Panel A

includes Judicial Efficiency, which indicates the implementation of laws; Panel B includes

Shareholders Rights, which ind icates the existence of laws protecting investors; and Panel C

includes Legality, which is an overall measure of the legal environment.

Table 4, Column 1 shows tests of Tobin’s-Q and legal indicators (all regressions include

1-digit SIC code and year1998 dummies.) We find only a significant correlation with Legality,

which is the broadest measure of a good legal environment. 20 Column 2 shows that good

corporate governance continues to be significantly related to valuation, even after correcting for

the legal environment. This suggests that even though our governance measure is significantly

19
The results are unchanged when we add all three additional controls simultaneously (not
reported, available on request)– governance and all control variables remain significant at 1%
level. We have also experimented with Investment to Assets ratio as another proxy for the future
growth opportunities. It was not significant and did not affect our governance results. Finally,
we included a dummy indicating firms that trade ADRs in all performance regressions, which
was insignificant and did not affect our other results.
20
However, when we add the interaction with governance in column 4 we find that all three legal
indicators are significant at least at 10% or better. This is consistent with La Porta, et al (2002)
that finds evidence in 27 high- income countries that firms with better protection of minority
shareholders have higher market valuation.

22
correlated with country- level legal indicators (as shown in Section 3), firm-specific governance

measures are of greater importance than the constraints of country- level laws in determining

market valuation.

In Table 4, Column 3 we include the interaction of firm- level governance and a measure

of country- level legal efficiency. 21 We find that this interaction is negative, which indicates that

governance is more important in countries with overall weak legal systems. 22 As discussed

above, the alternative interpretation of this interaction term is that the legal system matters less

for the well- governed firms, which is plausible because firms with better governance will have

less need to rely on the legal system to resolve governance conflicts. In Column 3 the interaction

is only significant for the measure of Judicial Efficiency. However, when we include the Latin

dummy in Column 4 the interaction becomes significant for all three legal indicators (significant

at 1% for Efficiency and 10% for Shareholder Rights and Legality). We also obtain similar

results (i.e. all legal indicators are significant at least at 10%) when we exclude these two Latin

American countries from our regressions. We argue that the two Latin American countries in

our sample (Chile and Brazil) are different from the rest of the sample because they have the

lowest mean and median Q in the sample (except for Turkey) and relatively low ROA, however

they have relatively strong governance indicators and country- level indices of investor

protection. One explanation may be that the governments of Brazil and Chile both enacted in

21
An important caveat for our interaction results is the small sample – 14 countries may not be a
large enough sample to properly test this hypothesis. We therefore should be cautious in
attaching strong interpretations to these results and encourage further research.
22
Our finding of the negative interaction effect parallels that of Doidge, Karolyi and Stulz (2001)
who find that the premium associated with ADR issuance is larger for firms in countries with
weaker investor protection.

23
2000 new laws offering greater protection to minority shareholders and these recent

improvements in governance would not be correlated with 1999 data. 23

Table 5 reproduces these results with ROA as the dependent variable. With the exception

of the results including Shareholder Rights, we find a strong correlation between governance and

ROA, and a significantly negative correlation with the interaction of judicial effectiveness. The

weaker results of Shareholder Rights could reflect the smaller variation within this variables as

well as the possible weaker importance of actual laws relative to their implementation.

The results in Tables 4 and 5 suggest that firm- level investor protection is more important

for firm valuation in countries with weaker investor protection from the courts. In terms of

magnitude, a one standard deviation improvement in governance increases Tobin’s-Q by 33% of

its standard deviation if the Efficiency score is five, and improves Tobin’s-Q by 18% of its

standard deviation if the Efficiency score is eight. Although an improvement in firm- level

governance always improves performance and market valuation, the improvements are higher in

countries with weaker legal and judicial infrastructures.

5. Conclusion

Although it is well established that country- level shareholder rights and judicial

efficiency affects firm value, in this paper we explore the differences in firm- level governance

mechanisms, their relationship with the country- level legal environment, and the correlations

between governance and performance. We use data from a recent report by Credit Lyonnais

Securities Asia (CLSA) that constructed corporate governance rankings for 495 firms across 25

emerging markets and 18 sectors. We find that (1) firms in countries with weak overall legal

23
This includes “Rule 345” in Brazil and the OPA law in Chile, which protect minority
shareholders during acquisitions, and the Corporate Law bill in Brazil, which provide broad

24
systems have on average lower governance rankings, (2) firm- level governance is correlated with

variables related to the extent of the asymmetric information and contracting imperfections that

firms face, which we proxy with firm size, sales growth (proxy for the growth opportunities) and

intangibility of assets, (3) firms that trade shares in the US have higher governance rankings,

especially so in countries with weak legal systems, (4) good governance is positively correlated

with market valuation and operating performance and (5) this relationship is stronger in countries

with weaker legal systems.

One interpretation of the last result is that firm- level corporate governance matters more

in countries with weak shareholder protection and poor judicial efficiency. An alternative

interpretation is that the legal system matters less for the well- governed firms, which is plausible

because firms with better governance will have less need to rely on the legal system to resolve

governance conflicts.

Our results suggest that firms in countries with poor investor protection can improve their

corporate governance, which may improve their performance and valuation. Our results do not

attempt to imply that firm- level corporate governance is a replacement for country- level judicial

reform. For example, we also find that firms have on average significantly lower governance

rankings in countries with weak legal systems, which suggests that firms cannot completely

compensate for the absence of strong laws and good enforcement. Although we do find that

firms can independently improve their investor protection and minority shareholder rights to a

certain degree, this adjustment mechanism is a second best solution and does not fully substitute

for the absence of a good legal infrastructure.

Our results also have important policy implications. Although the task of reforming

investor protection laws and improving judicial quality is difficult, lengthy, and requires the

protections to minority shareholders.

25
support of politicians and other interest groups, improving corporate governance on a firm- level

is a feasible goal. Our results suggest that even prior to legal and judicial reform, firms can still

reduce their cost of capital by establishing credible investor protection provisions. Our paper

proposes that firms in countries with poor investor protection can use provisions in their charters

to improve their corporate governance, which may improve their performance and valuation.

However, the task of reforming the legal systems should remain a priority on the policymaker’s

agenda.

26
Appendix 1: Abbreviated CLSA questionnaire

Discipline (15%)24
1. Has the company issued a "mission statement" that explicitly places a priority on good corporate governance?
<…> 25 ?
2. Is senior management incentivised to work towards a higher share price for the company eg, <…> expected
remuneration for the top executive(s) is tied to the value of the shares?
3. Does management stick to clearly defined core businesses? (Any diversification into an unrelated area in last 3
years would count as "No".)
4. <…> Is management's view of its cost of equity within 10% of a CAPM derived estimate?
5. <…> Is management's estimate of its cost of capital within 10% of our estimate based on its capital structure?
6. Over the past 5 years, is it true that the Company has not issued equity, or warrants for new equity, for
acquisitions and/or financing new projects where there was any controversy over whether the
acquisition/project was financially sound? <…>
7. Does senior management use debt for investments/capex only where ROA (or average ROI) is clearly higher
than cost of debt and where interest cover is no less than 2.5x? <…>
8. Over the past 5 years, is it true that the company has not built up cash levels <…>?
9. Does the company's Annual Report include a section devoted to the company's performance in implementing
corporate governance principles?

Transparency (15%)
10. Has management disclosed three- or five-year ROA or ROE targets? <…>
11. Does the company publish its Annual Report within four months of the end of the financial year?
12. Does the company publish/announce semiannual reports within two months of the end of the half-year?
13. Does the company publish/announce quarterly reports within two months of the end of the quarter?
14. Has the public announcement of results been no longer than two working days of the Board meeting? <…>
15. Are the reports clear and informative? (Based on perception of analyst.) <…>
16. Are accounts presented according to IGAAP? <…>
17. Does the company consistently disclose major and market sensitive information punctually? <…>
18. Do analysts have good access to senior management? Good access implies accessibility soon after results are
announced and timely meetings where analysts are given all relevant information and are not misled.
19. Does the Company have an English language web-site where results and other announcements are updated
promptly (no later than one business day)?

Independence (15%)
20. Is it true that there has been no controversy or questions raised over whether the board and senior management
have made decisions in the past five years that benefit them, at the expense of shareholders? (Any loans to
group companies/Vs, non-core/non-controlled group-investments, would mean "No").
21. Is the Chairman an independent, non-executive director?
22. Does the company have an executive or management committee <…> which is substantially different from
members of the Board and not believed to be dominated by major shareholders? (ie, no more than half are also
Board members and major shareholder not perceived as dominating executive decision making.)
23. Does the company have an audit committee? Is it chaired by a perceived genuine independent director?
24. Does the company have a remuneration committee? Is it chaired by a perceived genuine independent director?
25. Does the company have a nominating committee? Is it chaired by a perceived genuine independent director?
26. Are the external auditors of the company in other respects seen to be completely unrelated to the company?
27. Does the board include no direct representatives of banks and other large creditors of the company? (Having
any representatives is a negative.)

24
Percents reflect the weight in the CLSA weighted average index.
25
We kept the wording of the questions exactly as specified in the CLSA report, however to save the space and
without loss of contents we cut out portions of the questions, these cuts are marked with <…> . For example we
removed all clarifications as to how the analysts should answer the questions and endings such as “as far as the
analyst can tell”.

27
Accountability (15%)
28. Are the board members and members of the executive/management committee substantially different <…>?
(ie, no more than half of one committee sits on the other?)
29. Does the company have non-executive directors who are demonstrably and unquestionably independent?
(Independence of directors must be demonstrated by either being appointed through nomination of non-major
shareholders or having on record voted on certain issues against the rest of the Board. <…>)
30. Do independent, non-executive directors account for more than 50% of the Board?
31. Are there any foreign nationals on the Board <…>?
32. Are full Board meetings held at least once a quarter?
33. Are Board members well briefed before Board meetings? <…>(Answers 33-35 must be based on direct contact
with an independent Board member. If no access is provided <…> answer "No" to each question.)
34. Does the audit committee nominate and conduct a proper review the work of external auditors <…>?
35. Does the audit committee supervise internal audit and accounting procedures <…>?

Responsibility (15%)
36. If the Board/senior management have made decisions in recent years seen to benefit them at the expense of
shareholders (cf Q20 above), has the Company been seen as acting effectively against individuals responsible
and corrected such behavior promptly, ie, within 6 months? (If no such case, answer this question as "Yes".)
37. <…> Over the past five years, if there were flagrant business failures or misdemeanors, were the persons
responsible appropriately and voluntarily punished? (If no cases <…> then answer "No.")
38. Is there any controversy or questions over whether the Board and/or senior management take measures to
safeguard the interests of all and not just the dominant shareholders? <…>
39. Are there mechanisms to allow punishment of the executive/management committee in the event of
mismanagement <…>?
40. Is it true that there have been no controversies/ questions over whether the share trading by Board members
have been fair, fully transparent, and well intentioned? <…>
41. <…> Is the board small enough to be efficient and effective? (If more than 12, answer "No".)

Fairness (15%)
42. Is it true that there have not been any controversy or questions raised over any decisions by senior management
in the past 5 years where majority shareholders are believed to have gained at the expense of minority
shareholders?
43. Do all equity holders have the right to call General Meetings? <…>
44. Are voting methods easily accessible (eg proxy voting)?
45. Are all necessary <…> information for General Meetings made available prior to General Meeting?
46. Is senior management unquestionably seen as trying to ensure fair value is reflected in the market price of the
stock <…>?
47. Is it true that there has been no questions or perceived controversy over whether the Company has issued
depositary receipts that benefited primarily major shareholders <…>?
48. Does the majority shareholder group own less than 40% of the company?
49. Do foreign portfolio managers, and/or domestic portfolio investors who have a track record in engaging
management on CG issues, own at least 20% of the total shares with voting rights?
50. Does the head of Investor Relations report to either the CEO or a Board member?
51. <…> Over the past five years, is it true that total directors remuneration has not increased faster than net profit
after exceptionals? <…>

Social awareness (10%)

52. Does the company have an explicit (clearly worded) public policy statements that emphasize strict ethical
behavior: ie, one that looks at the spirit and not just the letter of the law?
53. Does the company have a policy/culture that prohibits the employment of the under-aged <…>?
54. Does the company have an explicit equal employment policy <…>?
55. Does the Company adhere to specified industry guidelines on sourcing of materials <…>?
56. Is the company explicitly environmentally conscious? <…>
57. Is it true that the company has no investments operations in Myanmar?

28
References

Beck, T, A. Demirguc-Kunt, and R. Levine, 2001. Law, politics and finance. Unpublished
working paper. World Bank no. 2585.

Berkowitz, D., K. Pistor, and J. Richard, 2002. Economic development, legality, and the
transplant effect. Forthcoming in the European Economic Review.

Black, B., 2001, The corporate governance behavior and market value of Russian firms.
Forthcoming as Does corporate governance matter? A crude test using Russian data. University
of Pennsylvania Law Review, 149.

Black, B. and R. Gilson, 1998. Venture capital and the structure of capital markets: Banks
versus stock markets. Journal of Financial Economics 47, 243-77.

Cantale, S., 1996. The choice of a foreign market as a signal. Unpublished working paper.
Tulane University.

Claessens, S., S. Djankov, and L. Larry, 2000. The separation of ownership and control in East
Asian corporations. Journal of Financial Economics v58, 81-112.

Coffee, J., 1999. The future as history: The prospects for global convergence in corporate
governance and its implications. Northwestern University Law Review 93, 641-708.

Daines, R., 2001. Does Delaware law improve firm value? Forthcoming in the Journal of
Financial Economics.

Dimitrov, O., 2001. Controlling shareholders on the board of directors: A cross-country

comparison. Unpublished working paper. Purdue University.

Doidge, C., A. Karolyi, and R. Stulz, 2001. Why are foreign firms listed in the US worth more?
Unpublished working paper. NBER no. 8538.

Easterbrook, F. and D. Fischel, 1991. The Economic Structure of Corporate Law. Harvard
University Press, Cambridge, Massachusetts.

Fuerst, O., 1998. A theoretical analysis of the investor protection regulations argument for
global listing of stocks. Unpublished working paper. Yale University.

Gompers, P., J. Ishi, and A. Metrick, 2001. Corporate governance and equity prices.
Unpublished working paper. NBER no. 8449.

Himmelberg, C., R. G. Hubbard, and I. Love, 2001. Investor protection, ownership and the cost
of capital. Unpublished working paper. Columbia University.

29
Himmelberg, C., R. G. Hubbard, and D. Palia, 1999. Understanding the determinants of
managerial ownership and the link between ownership and performance. Journal of Financial
Economics 53, 353-384.

Jensen, M., 1986. Agency costs of free cash flows, corporate finance and takeovers. American
Economic Review 76, 323-329.

Jensen, M., and W. Meckling, 1976. Theory of the firm: Managerial behavior, agency costs, and
ownership structure. Journal of Financial Economics 3, 305-360.

International Country Risk Guide, 2000.

Khanna, T., J. Korgan, and K. Palepu, 2001. Globalization and corporate governance
convergence? A cross-country analysis. Unpublished working paper. Harvard University.

La Porta, R., F. Lopez-de-Silanes, A. Shleifer, and R. Vishny, 1997. Legal determinants of

external finance. Journal of Finance 52, 1131-1150.

La Porta, R., F. Lopez-de-Silanes, A. Shleifer, and R. Vishny, 1998. Law and finance. Journal
of Political Economy 106, 1113-1155.

La Porta, R., F. Lopez-de-Silanes, A. Shleifer, and R. Vishny, 1999. Corporate ownership

around the world. Journal of Finance 54, 471-517.

La Porta, R., F. Lopez-de-Silanes, A. Shleifer, and R. Vishny, 2000. Investor protection and
corporate governance. Journal of Financial Economics 58, 3-27.

La Porta, R., F. Lopez-de-Silanes, A. Shleifer, and R. Vishny, 2002. Investor protection and
corporate valuation. Forthcoming in Journal of Finance.

Lombardo, D. and M. Pagano, 2000. Legal determinants of the return on equity. Unpublished
manuscript.

Maher, M. and T. Andersson, 2000. Convergence and Diversity of Corporate Governance

Regimes and Capital Markets, Oxford University Press, Oxford.

Moel, A., 1999. The role of information disclosure on stock market listing decisions: The case
of foreign firms listing in the US. Unp ublished working paper. Harvard Business School.

Nenova, T., 2002. The value of corporate voting rights and control: a cross-country analysis.
Forthcoming in Journal of Financial Economics.

Reese, W. and M. Weisbach, 2001. Protection of minority shareholder interests, cross- listings in
the United States and subsequent equity offerings, Unpublished working paper. NBER.

30
Shleifer, A. and R. Vishny, 1997. A survey of corporate governance. Journal of Finance 52,
737-783.

Shleifer, A. and D. Wolfenson, 2002. Investor protection and equity markets. Forthcoming in
Journal of Financial Economics.

Stulz, R., 1999. Globalization, corporate finance, and the cost of capital. Journal of Applied
Corporate Finance 12, 8-25.

31
Table 1: Summary Statistics
Governance is CLSA’s firm-level corporate governance rankings. Tobin’s-Q is defined as the market value of
equity plus total liabilities divided by total assets. ROA is a firm-level measure of firm performance and is defined
as net income plus interest expense divided by total assets. Legality is an index of legal and economic development
constructed as a weighted average of Efficiency of the Judiciary, Rule of Law, Corruption, Risk of Expropriation,
and Risk of Contract Repudiation. (Berkowitz, et al., 2002). Shareholder Rights is the sum of dummies identifying
one-share/one-vote, proxy by mail, unblocked shares, cumulative vote/proportional representation, preemptive
rights, oppressed minority, and percentage of shares needed to call an ESM (Shleifer, et al., 1999). Judicial
Efficiency is from the International Country Risk Guide (2000). The Mean, Standard deviation, and Range of firm-
level governance are calculated separately for each country. All correlations are based on 14 country-level
observations. P-values are in parenthesis.

Panel A: Firm-Level Corporate Governance Index and Country-Level Legal Indicators

Firm-Level Governance Index Country-Level Indices

Shareholder Judicial
Obs Mean Median Min. Max. Std Dev Legality
Rights Efficiency

All Sample 374 54.11 54.97 11.77 92.77 14.00 13.88 3.57 6.30
Brazil 24 57.26 59.87 43.08 68.22 7.99 14.07 3.00 5.75
Chile 13 61.63 60.62 48.22 69.25 5.18 14.68 5.00 7.25
Hong Kong 35 58.27 59.73 30.90 92.77 14.80 19.09 5.00 10.00
India 68 52.78 51.07 32.33 92.52 10.76 12.78 5.00 8.00
Indonesia 16 37.81 38.52 11.77 62.85 12.91 9.14 2.00 2.50
Malaysia 40 54.44 58.64 21.63 78.30 14.40 16.65 4.00 9.00
Pakistan 9 31.85 26.83 17.25 66.68 15.56 8.96 5.00 5.00
Philippines 17 40.72 34.08 19.40 64.35 13.66 8.50 3.00 4.75
Singapore 38 65.34 66.10 45.37 85.97 9.82 19.51 4.00 10.00
South Africa 32 66.53 67.16 42.62 80.38 8.55 14.32 5.00 6.00
South Korea 18 40.66 39.73 33.00 55.82 5.73 14.21 2.00 6.00
Taiwan 37 53.45 53.13 38.95 74.52 8.39 17.60 3.00 6.75
Thailand 18 53.54 49.69 28.33 79.02 14.53 12.92 2.00 3.25
Turkey 9 43.04 46.58 23.43 56.77 12.90 11.82 2.00 4.00

32
Table 1: Summary Statistics (cont.)
Panel B: Firm-Level Performance Variables, 1999
Tobin’s-Q: ROA:
Obs Mean Median Std Dev Obs Mean Median Std Dev
All Sample 336 2.09 1.39 1.68 357 0.08 0.06 0.16
Brazil 24 1.27 1.10 0.67 24 0.01 0.02 0.05
Chile 12 1.38 1.20 0.65 12 0.04 0.05 0.04
Hong Kong 29 1.95 1.37 1.67 33 0.09 0.06 0.09
India 55 2.82 1.66 2.26 66 0.11 0.09 0.08
Indonesia 16 2.23 1.89 1.28 16 0.10 0.08 0.10
Malaysia 39 1.84 1.45 1.06 39 0.10 0.08 0.08
Pakistan 9 1.49 1.46 0.47 9 0.06 0.09 0.11
Philippines 17 1.46 1.36 0.65 17 0.04 0.04 0.05
Singapore 35 1.73 1.19 1.37 37 0.05 0.04 0.05
South Africa 32 1.90 1.37 1.40 29 0.09 0.07 0.07
South Korea 16 1.58 1.09 1.32 16 0.03 0.02 0.04
Taiwan 30 3.67 3.12 2.31 31 0.10 0.07 0.07
Thailand 16 2.07 1.50 1.50 15 0.05 0.04 0.09
Turkey 6 1.16 1.16 0.53 7 0.05 0.00 0.08

Panel C. Correlations of Country-level governance statistics and Legal Indicators

Interquartile
Mean of St. Dev. of
Range of Legality Efficiency
Governance Governance
Governance

St. Dev. Governance -0.42

(0.14)

Interquartile Range
-0.39 0.74
Governance
(0.16) (0.00)

Legality 0.77 -0.30 -0.43

(0.00) (0.29) (0.13)

Efficiency 0.64 -0.14 -0.45 0.81

(0.01) (0.64) (0.11) (0.00)

Shareholder Rights 0.47 0.00 -0.39 0.28 0.64

(0.09) (1.00) (0.17) (0.33) (0.01)

33
Table 2: Governance Determinants

The dependent variable is Governance, which is CLSA’s firm-level corporate governance ranking.
Log(SalesUS) is Log of Sales in US$. SalesGR is 3-year average growth rate of sales (in US$). K/S is the ratio
of the 3-year average of fixed capital (property plant and equipment) to sales. ADR Dummy equals one if the
firm lists in the US, 0 otherwise. Judicial Efficiency is from the International Country Risk Guide (2000).
Shareholder Rights is the sum of dummies identifying one-share/one-vote, proxy by mail, unblocked shares,
cumulative vote/proportional representation, preemptive rights, oppressed minority, and percentage of shares
needed to call an ESM (La Porta et al., 1999). Firm-level data is for 1999 except for 29 firms whose last
available data is 1998. All regressions include a year dummy for firms whose last available data is 1998 (not
shown). T-statistics are in parenthesis, *, **, and *** indicate significance at 10%, 5%, and 1% respectively.

(1) (2) (3) (4) (5) (6) (7)

0.98* 0.39 0.52 0.47 0.38
Log(SalesUS)
(1.84) (0.69) (0.90) (0.82) (0.81)
9.98*** 9.96*** 8.61*** 8.62*** 6.86**
SalesGR
(2.6) (2.63) (2.38) (2.38) (2.11)
-1.48*** -0.99** -1.05*** -0.96*** -0.89**
K/S
(-2.97) (-2.09) (-2.64) (-2.51) (-2.12)
5.18*** 4.49*** 4.47*** 5.27*** 2.39*
ADR Dummy
(3.23) (2.68) (3.03) (2.79) (1.68)
1.65*** 2.11*** 1.77*** 2.22***
Judicial Efficiency
(3.56) (4.03) (3.71) (4.16)
1.67** 1.56* 0.63 0.94
Shareholder Rights
(1.96) (1.85) (0.69) (1.04)
Judicial Efficiency -1.45** -1.52**
*ADR Dummy (-2.04) (-2.14)

42.5*** 47.99*** 34.76*** 31.48*** 31.44*** 27.68***

Intercept ___
(6.1) (6.59) (12.82) (9.79) (3.94) (3.33)
Country Dummies No No No No No No Yes

Adjusted R^2 0.06 0.07 0.13 0.16 0.17 0.18 0.39

# of Firms 335 335 374 374 335 335 335

34
Table 3: Corporate Governance and Firm Valuation

The dependent variable in Panel A, Tobin’s-Q, measures expected market performance and is defined as the market
value of equity plus total liabilities divided by total assets. The dependent variable in Panel B, ROA, measures the
return on assets and is calculated as net income plus interest expense divided by total assets. Governance is
CLSA’s firm-level corporate governance rankings. Log(SalesUS) is the natural log of total sales. SalesGR is 3-
year average growth rate of sales (in US$). K/S is the ratio of the 3-year average of fixed capital (property plant and
equipment) to sales. Firm-level data is for 1999 except for 29 firms whose last available data is 1998. All
regressions include a year dummy for firms whose last available data is 1998 (not shown.) T-statistics are in
parenthesis, *, **, and *** indicate significance at 10%, 5%, and 1% respectively.

Panel A: Dependent Variable = Tobin’s-Q

(1) (2) (3) (4) (5) (6)
1.56*** _____ _____ ____ _____ ____
Intercept
(4.99)
0.011** 0.023*** 0.023*** 0.025*** 0.021*** 0.0196***
Governance
(1.94) (3.12) (3.36) (3.54) (2.97) (2.76)
-0.25***
Log(SalesUS)
(-4.24)
1.58***
SalesGR
(3.91)
-0.24***
K/S
(-4.49)
Country dummies No Yes Yes Yes Yes Yes
1-Digit SIC dummies No No Yes Yes Yes Yes

Adjusted R^2 0.03 0.21 0.30 0.34 0.30 0.33

# of Firms 336 336 336 334 323 333

Panel B: Dependent Variable = ROA

(1) (2) (3) (4) (5) (6)
2.3 _____ _____ ____ _____ ____
Intercept
(1.59)
0.08*** 0.14*** 0.13*** 0.13*** 0.12*** 0.11***
Governance
(2.83) (3.58) (3.48) (3.52) (3.17) (2.84)
-0.26
Log(SalesUS)
(-0.85)
0.05**
SalesGR
(2.13)
-0.01***
K/S No
(-4.48)
Country dummies No Yes Yes Yes Yes Yes
1-Digit SIC dummies No No Yes Yes Yes Yes

Adjusted R^2 0.03 0.19 0.25 0.25 0.26 0.28

# of Firms 351 351 351 347 335 346

35
Table 4: Corporate Governance, Legality and Firm Valuation

The dependent variable, Tobin’s-Q is defined as the market value of equity plus total liabilities divided by total
assets. Governance is CLSA’s firm-level corporate governance rankings. Legality is an index of legal and
economic development constructed as a weighted average of Efficiency of the Judiciary, Rule of Law, Corruption,
Risk of Expropriation, and Risk of Contract Repudiation (Berkowitz, et al., 2002). Judicial Efficiency is from the
International Country Risk Guide (2000). Shareholder Rights is the sum of dummies identifying one-share/one-
vote, proxy by mail, unblocked shares, cumulative vote/proportional representation, preemptive rights, oppressed
minority, and % of shares needed to call an ESM (Shleifer, et al., 1999). Latin Dummy is equal to 1 if the country
is in Latin America, 0 otherwise. Firm-level data is for 1999 except for 29 firms whose last available data is 1998.
All regressions include 336 firms, 1-digit SIC dummies and a year dummy for firms whose last available data is
1998 (not shown.) T-statistics are in parenthesis, *, **, and *** indicate significance at 10%, 5%, and 1%
respectively.

Panel A: (1) (2) (3) (4)

0.02 -0.02 0.22** 0.27**
Judicial Efficiency
(0.62) (-0.53) (2.14) (2.61)
0.019*** 0.049*** 0.066***
Governance
(3.20) (3.44) (4.31)
-0.004** -0.006***
Judicial Efficiency* Governance
(-2.25) (-2.98)
-1.05
Latin Dummy
(-5.62)
Adjusted R^2 0.11 0.13 0.14 0.19

Panel B: (1) (2) (3) (4)

0.44 -0.02 0.39 0.43*
Shareholder Rights
(0.58) (-0.30) (1.49) (1.68)
0.02*** 0.05*** 0.06***
Governance
(3.19) (2.33) (2.82)
-0.01 -0.01*
Shareholder Rights* Governance
(-1.52) (-1.84)
-1.15***
Latin Dummy
(-6.26)
Adjusted R^2 0.11 0.13 0.14 0.18

Panel C: (1) (2) (3) (4)

0.47** 0.17 0.08 0.14*
Legality
(2.05) (0.67) (1.18) (1.87)
0.017*** 0.03* 0.06***
Governance
(2.71) (1.74) (2.74)
-0.001 -0.003*
Legality*Governance
(-0.92) (-1.79)
-1.18***
Latin Dummy
(-6.16)
Adjusted R^2 0.12 0.13 0.13 0.18

36
Table 5: Corporate Governance, Legality and Firm Performance

The dependent variable, ROA, measures the return on assets and is calculated as net income plus interest expense
divided by total assets. Governance is CLSA’s firm-level corporate governance rankings. Legality is an index of
legal and economic development constructed as a weighted average of Efficiency of the Judiciary, Rule of Law,
Corruption, Risk of Expropriation, and Risk of Contract Repudiation (Berkowitz, et al., 2002). Judicial Efficiency is
from the International Country Risk Guide (2000). Shareholder Rights is the sum of dummies identifying one-
share/one-vote, proxy by mail, unblocked shares, cumulative vote/proportional representation, preemptive rights,
oppressed minority, and % of shares needed to call an ESM (Shleifer, et al., 1999). Latin Dummy is equal to 1 if the
country is in Latin America, 0 otherwise. Firm-level data is for 1999 except for 29 firms whose last available data is
1998. All regressions include 348 firms, 1-digit SIC dummies and a year dummy for firms whose last available data
is 1998 (not shown.) T-statistics are in parenthesis, *, **, and *** indicate significance at 10%, 5%, and 1%
respectively.

Panel A: (1) (2) (3) (4)

0.39** 0.20 1.2** 0.01**

Judicial Efficiency
(1.95) (0.96) (1.97) (2.38)
0.09*** 0.22*** 0.31***
Governance
(2.87) (2.67) (3.56)
-0.019 -0.028**
Judicial Efficiency* Governance
(-1.38) (-2.44)
-6.1***
Latin Dummy
(-5.69)
Adjusted R^2 0.12 0.14 0.15 0.20

Panel B: (1) (2) (3) (4)

1.2*** 0.95** 1.7 1.8

Shareholder Rights
(3.15) (2.27) (1.31) (1.34)
0.08*** 0.14 0.18*
Governance
(2.55) (1.31) (1.79)
-0.01 -0.02
Shareholder Rights* Governance
(-0.53) (-0.81)
-5.4***
Latin Dummy
(-5.50)
Adjusted R^2 0.14 0.16 0.16 0.20

Panel C: (1) (2) (3) (4)

0.15 -0.03 0.71* 0.98**

Legality
(1.22) (-0.25) (1.62) (2.20)
0.10*** 0.31*** 0.43***
Governance
(3.14) (2.47) (3.31)
-0.014* -0.021***
Legality*Governance
(-1.79) (-2.55)
-6.4***
Latin Dummy
(-6.09)
Adjusted R^2 0.11 0.14 0.15 0.20

37
 Figure 1: Box –and-Whisker Plot of the Distribution of Governance Rankings, by Country

gov
92.7667

11.7667
.1PHL .3IDN .5IND .7BRA .9ZAF 1.1MY 1.3HK
.2PAK .4TUR .6THA .8KOR 1.0CHL 1.2TW 1.4SG

Notes: Countries are sorted in the order of increasing index of Legality.

38
 Figure 2: Scatterplot of Average Governance and Index of Legality, by Country

average governance Fitted values

67.6088 ZAF
ARG
SGP

ISR HKG
CHL

MEX
average governance

BRA
THA MYS
COL IND
TWN

TUR KOR
PHL
IDN

33.9602 PAK

8 14 20
legality - composite

Notes: Plotted are country-means of firm- level governance rankings against index of Legality.

39