Journal of Corporate Finance 12 (2006) 560 – 593

www.elsevier.com/locate/jcorpfin

Capital market governance: How do security laws

affect market performance?
Hazem Daouk a,*, Charles M.C. Lee b, David Ng a
a
Department of Applied Economics and Management, Cornell University, Ithaca, NY 14853, United States
b
Johnson Graduate School of Management, Cornell University, Ithaca, NY 14853, United States
Received 10 March 2005; accepted 11 March 2005
Available online 25 July 2005

Abstract

This paper examines the link between capital market governance (CMG) and several key
measures of market performance. Using detailed data from individual stock exchanges, we develop a
composite CMG index that captures three dimensions of security laws: the degree of earnings
opacity, the enforcement of insider laws, and the effect of removing short-selling restrictions. We
find that improvements in the CMG index are associated with decreases in the cost-of-equity capital
(both implied and realized), increases in market liquidity (trading volume, market depth, and U.S.
foreign investments), and increases in market pricing efficiency (reduced price synchronicity and
IPO underpricing). The results are quite consistent across individual components of CMG and over
alternative market performance measures.
D 2005 Elsevier B.V. All rights reserved.

JEL classification: G15; G30

Keywords: Capital market governance; Insider trading; Earnings opacity; Short-selling constraints; Market
performance

1. Introduction

In an increasingly integrated global economy, interest in (and awareness of) good

capital market regulation is on the rise. While presumption of the damaging effects of

* Corresponding author.
E-mail addresses: [email protected] (H. Daouk), [email protected] (C.M.C. Lee), [email protected]
(D. Ng).

0929-1199/$ - see front matter D 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jcorpfin.2005.03.003
 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593 561

bad market governance is widespread, direct evidence on its economic consequences has
been more difficult to document. Contributing to this problem is the elusive nature of
governance. Because the quality of capital market governance can be associated with a
number of other country-level phenomena, its direct impact on the performance of stock
markets may be difficult to isolate.
In this paper, we examine how capital market regulations and their enforcement
might affect a wide range of market performance measures. We focus on exchange-
based (or market-related) regulations, and coin the term capital market governance
(CMG) to describe this aspect of a country’s regulatory environment. Using detailed
data collected from individual exchanges for the period 1969–1998, we construct a
composite CMG index that varies over time, thus reflecting inter-temporal variations in
the quality of market governance in each country. We then examine the relation
between changes in the CMG index and changes in market performance across
countries.
Our study consists of two stages. In the first stage, we use a unique data set gleaned
from market regulators, exchange officials, and industry contacts, to construct a broad
index of capital market governance. Specifically, we exploit innovations developed in
several recent studies to measure three dimensions of capital market governance: (1) a
composite earnings opacity measure, (2) insider trading laws and their enforcement, and
(3) relaxation of short-selling restrictions. Fig. 1 provides an overview of these market
governance variables.

Capital Market Governance

1. Insider 2. Earning 3. Short

Trading Opacity Selling
Constraint

1.1 Insider Trading Dummy 2.1 Earning Aggressiveness 3.1 Short Selling Constraint
Insider trading dummy is zero if The ratio of Accruals to Total A dummy which equals ten after short
insider trading law has never Asset. The higher is this ratio, selling became feasible in a country.
been enforced and ten if it was the higher is earning
enforced. aggressiveness. 3.2 Put Option
A dummy which equals ten after the
2.2 Loss Avoidance date when put option became feasible.
The number of firms with
small positive earnings minus
the number of firms with
small negative earnings, all
divided by their sum. The
higher is this ratio, the higher
is the loss avoidance.

2.3 Earning Smoothing

Correlation between change in
accruals and change in cash
flows. The higher is this ratio,
the higher is earning
smoothing.

Fig. 1. Capital market governance measures.

562 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593

In the second stage, we evaluate the impact of CMG on key dimensions of market
performance. Specifically, we use seven empirical proxies to capture three aspects of
market performance: (1) Cost-of-capital (both implied and realized); (2) Market
liquidity (including trading volume, market depth, and foreign ownership by U.S.
investors); and Pricing efficiency (including price synchronicity and IPO under-
pricing). Fig. 2 offers an overview of these performance measures and our empirical
proxies.
Our goal is to gain an overarching perspective on how changes in security laws can
affect the returns shareholders demand, their willingness to trade, and the efficiency
with which prices incorporate information. Securities laws cover a wide spectrum of
areas, including the distribution of securities, takeovers, stock market manipulations,
insider trading, stock exchanges, and the activities of financial intermediaries. We
focus on three aspects of these regulations: earnings opacity, insider laws, and short-
selling restrictions. We examine how these laws individually affect a wide set of
market performance metrics, and we evaluate their combined effect using a CMG
index.
A distinguishing feature of this study is the range of market performance measures
we examine. Prior studies have typically focused on the effect of individual laws on
one or two aspects of market performance—e.g., the role of accounting disclosure laws
on the cost-of-capital. However, regulatory decisions can hinge on potential trade-offs
across different market performance metrics (e.g., the trade-off between pricing
efficiency and market liquidity in the insider trading law debates). From a policy

Market Performance

1. Cost of 2. Market 3. Pricing

Capital Liquidity Efficiency

1.1 Implied Cost-of-capital 2.1 Trading Volume 3.1 Stock Price Synchronicity
The implied cost of equity The ratio of dollar traded per The average R2 from firm-level
capital, derived from current month to the dollar market regressions of bi-weekly stock returns
market price and a dividend capitalization at the end of the on local and U. S. market indices in
discount model. month. each country (Morck, Yeung and Yu
(2000)). This variable measures the
1.2 Average Realized Returns 2.2 Market Depth ratio of firm-specific to market-level
The average realized returns, The ratio of trading volume to information. Countries with higher
based on an international asset the standard deviation of daily average R2 are deemed to be less
pricing model (Bekaert and returns (or the absolute value efficient.
Harvey (1995)) of monthly return) computed
each month. 3.2 IPO Underpricing
Based on the average initial-day
2.3 U.S. Foreign Investment returns on IPOs. Countries with larger
The over (under) weight in IPO underpricing are deemed to be
stockholdings by U.S. less efficient.
investors, relative to that
country’s weight in a global
index.

Fig. 2. Market performance measures and their empirical proxies.

 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593 563

perspective, it is important to know how regulatory changes might affect multiple

aspects of market performance.1
Our strategy is to examine several key dimensions of market performance, using
multiple empirical proxies for each. For example, in estimating changes in the cost-of-
capital, we use both an implied approach based on discounted cash flow models (Bekaert
and Harvey 2000; Lee et al., 2003), and a more traditional international asset pricing
approach based on realized returns. In measuring market liquidity, we examine changes in
trading volume, market depth (volume scaled by volatility), as well as U.S. foreign
investment. Finally, in examining pricing efficiency, we use both a Morck et al. (2000)
synchronicity measure, and a measure of the degree of underpricing in IPO offerings.
While none of these individual proxies might fully capture the underlying economic
construct, the consistency of our results across multiple methods adds to their
interpretability.
To deal with the empirical challenge posed by country-level correlated omitted
variables, we designed our tests to capture inter-temporal changes in a country’s securities
laws. By estimating most of our models with fixed-country effects, we use each country as
its own control. We are thus able to isolate inter-temporal fluctuations in the CMG index,
and evaluate the effect of changes in a country’s CMG index on changes in its cost-of-
equity, market liquidity and pricing efficiency over time.2
We find that after controlling for other factors, improvements in the CMG index are
associated with economically significant decreases in the cost-of-equity capital. The
estimated economic magnitude of the effect is similar across the two cost-of-capital
measures—using the implied estimation method, the cost of equity decreases by 2.6% per
standard deviation increase in CMG; using the average realized returns from an
international asset pricing model, the effect is 2.9% per standard deviation increase in
CMG. While not all three components of the CMG index are individually significant in
every regression specification, the directional inferences are always the same: improve-
ments in earnings opacity, insider trading, and short-selling laws result in lower costs of
capital.
In addition, we find that improvements in the CMG index are positively correlated with
three measures of market liquidity. Specifically, improved CMG is associated with
increases in trading volume, market depth (i.e., volume divided by volatility), as well as
the level of U.S. stockholdings (suggesting that countries with improving governance laws
attract more U.S. investors). Once again, all three components of the CMG index
contribute to these overall results. The directional inferences are consistent for the earnings
opacity, insider trading, and short-selling variables, even when each is individually
included in the same regression.

1
Our use of a composite CMG measure might also enhance the power of our tests. For example, in assessing
the potential impact of security laws on a country’s cost-of-capital, it seems sensible to use a composite measure
of these laws rather than a variable that just reflects a single regulation.
2
We conduct country fixed-effect regressions whenever data permit. Specifically, all tests involving the cost-of-
capital (both implied and realized) and market liquidity (both trading volume and market depth) are estimated
with country fixed-effect regressions. The pricing efficiency tests (price synchronicity and IPO underpricing) and
the U.S. foreign investment test are based on cross-sectional regressions.
564 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593

Finally, we find that the CMG index is negatively correlated with pricing synchronicity
(Morck et al., 2000) and the amount of IPO underpricing, after controlling for a host of
other factors. These results suggest that improved CMG increases pricing efficiency.
Detailed analysis indicates that the pricing efficiency results are driven primarily by the
insider trading and short-selling variables, with earnings opacity playing a lesser role.
Overall, our findings support the view that improved capital market governance is
associated with lower costs-of-equity capital, increased market liquidity, improved price
efficiency, and increased stock ownership by U.S. investors. We find little evidence that
the improvements in each of the three market performance attributes come at the expense
of the other two. To the extent that market regulators find the directional changes we
document desirable, the prescriptive implications are unambiguous.

2. Capital market governance and equity market characteristics

Capital market governance refers to the set of laws, rules, and regulations that govern
the functioning of capital markets. More importantly, it is the degree of enforcement of
those laws, rules and regulations.
The capital market governance measures we construct are aimed at capturing different
facets of the interaction between insiders and outsiders of the corporations. Corporate
insiders have an informational advantage that they can potentially exploit to the harm of
ordinary investors. A key prediction from the agency theory literature is that in
equilibrium, outsiders will factor in these agency problems and make the insiders bear
the cost. We posit that these costs will translate into higher costs of equity, as well as
possibly lower market liquidity, and lower pricing efficiency.
Each of our capital market governance measures is designed to track a distinct aspect of
regulatory protection of investors from insider activities. Insider trading laws and their
enforcement is the most direct expression of this type of protection. Similarly, our measure
of accounting opacity captures the extent to which insiders might secure an unfair
informational advantage over outsiders. Finally, short-selling prohibitions increase the risk
to outsiders, because they allow corporate values to be potentially overvalued, thus further
increasing informational asymmetry between insiders and outsiders.

2.1. Insider trading enforcement

The first capital market governance measure we use is the enforcement of insider
trading laws. Scores of law, economics and finance papers have argued the pros and cons
of insider trading regulations. Bainbridge (2000), besides providing a comprehensive list
of papers that have discussed insider trading, succinctly summarizes the arguments for and
against allowing insider trading. Briefly stated, in stock markets where insiders trade with
impunity, liquidity providers would protect themselves by increasing their ask price and
decreasing their bid price. This increases transaction costs, leading to lower market
liquidity, and a higher required rate of return on equity. These transaction costs are not
diversified away in equilibrium even in well-integrated global markets (see Lombardo and
Pagano, 2000; Lee and Ng, 2004).
 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593 565

A second reason the cost of equity might be higher in markets with weak insider trading
laws is that in such markets, large controlling shareholders are more easily tempted by
management to profit from stock tips, rather than from effort-intensive monitoring.
Knowing this, ordinary shareholders would again demand a higher return on equity. In
short, the first reason predicts a higher cost of equity because of an implicit transaction tax
inherent in high bid–ask spreads, while the second reason does not depend on an illiquidity
premium argument.3
The directional effect of insider trading enforcement on market efficiency is less clear.
Restricting insider trading may decrease information flow and decrease the ability of
markets to price assets quickly and efficiently. On the other hand, when insider trading
enforcement is more effective, analysts and other investors have greater incentives to
analyze firms, which might lead to better market efficiency.
Our measure of insider trading is based on enforcement of insider trading laws from
Bhattacharya and Daouk (2002). Bhattacharya et al. (2000) had shown that it is the
enforcement rather than the existence of insider trading laws that deter insiders. In our
empirical setup, the insider trading variable is assigned a value of 10 if there had been any
enforcement, and 0 otherwise.

2.2. Earnings opacity

The second measure of governance we use is earnings opacity. Our earnings opacity
measure is based on the methodology in Bhattacharya et al. (BDW, 2003). Briefly, we
define earnings opacity as the extent to which the distribution of reported earnings of firms
in that country fails to provide information about the distribution of the true, but
unobservable, economic earnings (see Appendix A for details).
In their call for more research on the relation between accounting information and
governance, Bushman and Smith (2001) identify three channels by which earnings opacity
may affect financial markets. First, better accounting information helps investors
distinguish between good and bad investments, thus lowering estimation risk and leading
to lower costs of equity and higher pricing efficiency. Second, better accounting
information helps investors distinguish between good and bad managers, thus decreasing
agency costs, and lowering firms’ cost of equity.
Third, earnings opacity, by weakening the link between reported earnings and
unobservable economic earnings, increases asymmetric information. An increase in
asymmetric information leads to an increase in the adverse selection problem a liquidity
provider faces when trading with insiders. Liquidity providers in such a market would seek
to protect themselves by increasing their ask price and decreasing their bid price. The
increased transaction costs in turn result in higher required rates of return on equity, and
lower market liquidity (see Glosten and Milgrom, 1985; Kyle, 1985; Amihud and
Mendelson, 1986; Jacoby et al., 2000; Brennan and Subrahmanyam, 1996).

3
An argument can also be made that the cost of equity is lower in markets where insiders trade freely (e.g.,
Manne, 1966). However, the weight of the evidence appears to suggest otherwise (e.g., Bhattacharya and Daouk,
2002).
566 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593

Following BDW (2003), we infer a country’s general level of earnings opacity from the
distributional properties of its reported earnings. Specifically, our opacity measure is
designed to capture three aspects of reported earnings: earnings aggressiveness, loss
avoidance, and earnings smoothing (see Appendix A). We construct a panel data set for
each of these three measures and then average them to obtain an overall earnings opacity
time-series measure per country. This variable ranges from 0 (most opaque) to 10 (least
opaque).

2.3. Short-selling constraints

Although the directional effect of short-selling constraints on market valuation is less

clear, the preponderance of the theoretical and empirical evidence seems to suggest that
increased short-selling constraints is associated with less efficient price discovery, lower
market liquidity, and higher costs of capital.
The earliest theoretical work on this topic is Miller (1977). Under heterogeneous
expectations, when short-selling constraints are binding, Miller showed that stock prices
will be overvalued because they reflect only the bullish views. Also, short-selling
constraints impedes price discovery and potentially lowers market liquidity.
Short-selling constraints can also affect the cost of equity through its potential effect on
return skewness and volatility. Theoretical models suggest short-selling restrictions will
give rise to greater negative skewness in returns. To the extent that investors are averse to
negative skewness, they will demand a higher rate of return. At the same time, increased
short-selling constraints are associated with higher return volatility. Higher return volatility
could increase cost of equity if international markets are not fully integrated. Empirically,
Charoenrook and Daouk (2003) provide evidence that short-selling restrictions are indeed
associated with lower market liquidity. In addition, Bris et al. (2004) find that short-selling
restriction in a country tends to be associated with lower pricing efficiency. In sum, the
weight of the evidence suggests that increased short-selling constraints will adversely
affect market liquidity and pricing efficiency, and result in higher costs of capital. Thus, we
construct our CMG index such that decreases in short-selling constraints are reflected as
improvements in the index.
Our measure of short-selling restrictions is based on a survey of exchange officials by
Charoenrook and Daouk (2003). They ask in their survey the first date where short-selling
became feasible. Also, they ask about the existence of put options on stocks, because equity
put options offer an alternative method to implement a short position. In our empirical tests,
the insider trading variable is 0 if short-selling was not allowed and put options do not exist
in a given time period, and 10 if short-selling is allowed or put options exist.
After we compute the measures of insider trading, earning opacity, and short-selling
restrictions, we compute the simple average of the three measures to form the CMG index.
This measure ranges from 0 (worst governance) to 10 (best governance).

2.4. Relation to other studies

Some other studies conduct in-depth case study on the issue of governance in a
particular market. For example, Black et al. (2006-this issue) conduct an in-depth country
 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593 567

case study of the factors that predict governance levels for both large and small Korean
firms. On the other hand, our study is in a stream of literature that adopts a multi-country
approach. At least two recent studies have also focused on the relation between market-
based regulations and equity markets. In this section, we discuss the distinctive features of
our study relative to other concurrent projects.
La Porta et al. (2002) conducts a detailed analysis of how securities laws affect the
issue of new equity to the public. Their focus is on the agency problem between
prospective investors in an initial public offering (IPO) and the bpromoterQ who offers
shares for sale. Their data allows them to evaluate the efficacy of specific types of
securities laws, but their focus is primarily on laws governing the IPO process. Using
the disclosure requirement and enforcement indices developed by La Porta et al. (2002),
Hail and Leuz (2003) examine the effect of securities regulation and legal institutions on
firms’ implied cost of capital. Both studies conclude that more extensive securities
regulation yield beneficial results for security markets, such as lower levels of cost of
capital.
Our study is similar in spirit to these two papers. However, we use a unique set of
capital market governance metrics and examine a much wider range of market
performance metrics. Our earnings opacity measure captures three different dimensions
of a country’s accounting system, and allows these dimensions to vary over time.
Similarly our insider-law and short-selling restriction measures identify not only
countries with superior laws, but also the timing of the enforcement of these laws.
These research design innovations allow us to examine the effect of changes in
governance not only on the implied cost of capital, but also the average realized returns.
In addition, our study allows us to evaluate the effect of capital market governance on
multiple dimensions of market performance. Market regulators are understandably
concerned about the effect of securities laws on many different aspects of performance.
For example, regulations that lower the cost of capital might have adverse effects on
market liquidity or pricing efficiency. To our knowledge, we are the first study to
examine how securities laws affect a wide range of market performance metrics. Our
approach allows regulators to evaluate potential trade-offs in performance within a
consistent research framework.

3. Market performance measures

As Fig. 2 shows, the dependent variables for our analysis can be broadly categorized
into three groups: (1) two measures of the cost of equity, (2) market liquidity (including
trading volume, market depth, and foreign U.S. investments), and (3) pricing efficiency
(including price synchronicity and IPO underpricing). In this section, we discuss each of
these measures in detail.

3.1. Cost of equity measures

The cost of equity in a country is defined as the return shareholders require for holding
shares in that country. This is an expectations variable, which must be proxied for using
568 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593

observable data. We employ two approaches: a price implied cost-of-capital, and average
realized returns.

3.1.1. The implied approach

The first approach is to compute the cost of equity by backing it out from the classical
constant growth dividend discount model. Appendix A in Bekaert and Harvey (2000)
explores in great detail the relation between dividend yields and the cost of equity for
many models. Assuming that the best forecast for future growth rates in dividends is the
most current dividend growth rate, which implies that we assume that dividend growth
rates follow a random walk, it follows that the estimated cost of equity = current dividend
yield  (1 + current growth rate of dividends) + growth rate of dividends.
We use dividend yields to measure cost of equity because these yields are observable,
stable, and stationary. In general, a sharp change in cost of equity should lead to a sharp
change in dividend yields. A disadvantage of using dividend yields is that changes in
dividend yields may come about because of repurchases of stock. However, this problem
should be small in emerging markets because repurchases are minor.4

3.1.2. The international asset pricing approach

The second approach to estimating the cost of equity using realized returns while
explicitly accounting for risk. We adopt a simplified version of Bekaert and Harvey (1995)
as our international asset pricing model. Their empirical specification allows for partial
integration of a country to the world equity markets. Their model is appealing because it
permits a country to evolve from a developing segmented market (where risk is measured
by the country’s variance) to a developed country which is integrated to world equity
markets (where risk is measured by the sensitivity of a country’s equity returns to
movements in the world market portfolio). The special case of complete integration, where
the world factor is the only factor, is nested in their model. We describe the details of this
asset pricing model, and our estimation procedure in Appendix B.

3.2. Market liquidity measures

We are also interested in how CMG affects market liquidity. A simple measure of
market liquidity is trading volume. However, volume alone does not fully capture the
concept we have in mind, particularly if increased volume comes at the expense of
increased volatility. We therefore use several different measures, including market depth
(volume divided by volatility) and the amount of U.S. foreign investment.

3.2.1. Trading volume

We use the rate of turnover in a market to measure trading volume. Specifically, our
volume measure is defined as the ratio of volume of dollar trade per month to the dollar

4
Another approach to estimating the implied cost-of-capital is to incorporate analyst forecasts of earnings (e.g.,
Lee et al., 2003; Hail and Leuz, 2003). While the use of forecasted earnings might lead to increased precision, our
technique does not require firms to have analyst coverage. Because our analysis is at the country-level (i.e., not
firm level), these techniques are likely to yield similar results.
 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593 569

market capitalization at the end of the month. To mitigate the effect of outliers, which
occur because the denominator is small in some countries, we take the natural logarithm of
this ratio.

3.2.2. Market depth

While trading volume is the lifeblood of markets, it is not necessarily the best measure
of market quality. This is because, from an investor’s perspective, increased volume does
not necessarily translate into increased market liquidity. The notion of market liquidity is
related to how much trading can be done without adversely affecting prices (i.e., how
much trading can be accomplished with minimal price impact). Unfortunately, increased
volume is generally accompanied by increased volatility, so higher trading volume alone is
insufficient evidence that market depth has increased.
A better measure of market depth, or liquidity, is volume turnover scaled by return
volatility (e.g., Amihud, 2002). This variable captures the quantity of trading per unit of
volatility, and is closer to the notion of market quality investors are most concerned about.
Moreover, Hasbrouck (2003) finds that the Amihud (2002) variable is the best among the
non-intraday measures of price impact costs. We construct two versions of the market
depth variable. In the first, we use the ratio of trading volume (see Section 3.2.1) to the
standard deviation of daily returns computed each month. In the second, we use the ratio
of trading volume to the absolute value of the monthly returns.5

3.2.3. U.S. foreign stockholdings

A liquid market is also characterized by a breadth of ownership, in particular by foreign
investors. We construct a measure that examines U.S. citizens’ stockholding in a country
relative to what the U.S. holdings would have been based on the country’s market
capitalization. If an investor wishes to be fully diversified across different countries, (s)he
should hold a portfolio of country indices whose weights are in proportion to the market
capitalization of those countries. In reality, the actual holdings of U.S. investors can differ
sharply from these theoretic weights.
We compute a measure that captures this difference. Specifically, we use the ratio of the
percentage of U.S. stockholdings in each country to the percentage of the market
capitalization of the country. To mitigate the effect of outliers, we take the natural
logarithm of this ratio. In other words, we use:
0 , 1
X N
B hi hj C
B C
B ,
j¼1 C
U:S: holdi ¼ lnB C
B X N C
@ cap capj A
i
j¼1

where h i is U.S. stockholding in dollars in country i and capi is market capitalization of

country i in dollars where N is the number of countries in the world.

5
We also used the conditional volatility of monthly returns computed from the multivariate ARCH model used
in the International Asset Pricing Model described above. The results are similar and are not reported.
570 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593

This variable captures the over (under) weighting in shareholdings by U.S. investors
for each country, relative to that country’s weight in a global index. We conjecture that
U.S. investors (who are not constrained to buy stocks from any countries) are less
inclined to hold stocks in markets with lower CMG scores.
In constructing this variable, we use data on foreign stockholdings by U.S. citizens
from Bhattacharya and Groznik (2003). The data is based on a survey carried out by
the U.S. Treasury Department and the Board of Governors of the Federal Reserve
System (also see Ahearne et al., 2001). Market capitalization in dollars is from
Datastream.

3.3. Pricing efficiency measures

In addition to the cost-of-capital and market liquidity, market regulators are also
interested in the effect of security laws on markets’ ability to efficiently price stocks and
process new information. We use two empirical proxies suggested by the ex ante
literature.

3.3.1. Price synchronicity

In an interesting study, Morck et al. (2000) argue that in more price efficient
markets, stocks will exhibit higher average idiosyncratic risk. According to this
argument, the ratio of firm-specific information to market-level information should be
higher in informational environments that allow market participants to acquire
information, and act quickly and inexpensively upon it. Therefore, Morck et al.
(2000) nominate stock synchronicity as a measure of the overall price efficiency of the
market.
Following their definition, we compute this variable as the average R 2 from firm-
level regressions of bi-weekly stock returns on local and U.S. market indices in each
country. Countries with higher average R 2 (higher price synchronicity) are deemed to
be less price efficient. Given our overall research design, we examine how the CMG
variables, both collectively and individually, affect a country’s price synchronicity.

3.3.2. IPO underpricing

IPO underpricing refers to the frequent incidence of large initial returns accruing to
investors in IPOs of common stock. We use average IPO first-day returns from
Professor Jay Ritter’s web site (http://bear.cba.ufl.edu/ritter/) for 38 countries. The
range in average first-day returns varies greatly, from China at 257% to Denmark at
only 5%. The prior research has linked IPO underpricing to various agency problems
including information asymmetry and corruption (see Loughran et al., 1994 and Ritter
and Welch, 2002 for a summary of this literature). Generally, greater underpricing is
associated with high pricing inefficiencies. To the extent improved governance will
reduce these types of agency problems, we expect an increased CMG to reduce IPO
underpricing.
Note that although we treat IPO underpricing as a price discovery problem, it can
also be regarded as part of the cost of capital analysis. IPO is a way for entrepreneurs
to raise capital for their enterprise. The larger the underpricing, the less money the
 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593 571

entrepreneur receives for selling claims to outside investors. This is equivalent to an

increased cost of raising capital for existing shareholders.

4. Data description and sample selection

Data on monthly equity indices of 22 developed countries were obtained from

Morgan Stanley Capital International (MSCI). Data on monthly equity indices of 10
emerging markets were obtained from International Financial Corporation (IFC). The
data are from December 1969 to December 1998 (some countries do not have data for
the full time period).
We use the MSCI value-weighted World Index as a proxy for the world market
portfolio. Monthly returns of each country’s stock market and the world market portfolio
are computed from these indices. We obtain monthly data on dividend yields for the 32
countries based on Datastream indices. We obtain monthly data on the volume of trade
and market capitalization from Datastream (this data was available for 29 of the 32
countries in our sample).
Bekaert and Harvey (1997) divide the sum of exports and imports by the country’s
gross domestic product to obtain a variable that captures the level of integration of a
country with the rest of the world. This is because the level of globalization affects the
cost of equity (see Stulz, 1999a). We follow the same procedure. Monthly data on
exports and imports for the 32 countries were obtained from the International Financial
Statistics provided by the International Monetary Fund.
As purchasing power parity is not observed in the data, standard international asset
pricing models like Ferson and Harvey (1993) and Dumas and Solnik (1995) have a
foreign exchange factor (FX factor). We include this control in our international asset
pricing factor model as well. Monthly data on foreign exchange rates are obtained from
the International Financial Statistics.
We also control for the confounding effect of market liberalization. When a country
opens its capital markets to foreigners, its cost of equity is expected to be reduced
through improved risk-sharing and better corporate governance (Stulz, 1999b). Bekaert
and Harvey (2000), and Henry (2000) empirically confirm that market liberalization
activities do reduce a country’s cost of equity. To control for this phenomenon, we use
official liberalization dates from Bekaert and Harvey (2000).
Finally, we use several variables as controls for the level of corporate governance
and protection of minority shareholder rights in different countries. Klapper and Love
(2004) showed that better corporate governance is highly correlated with better
performance in many emerging markets. Following La Porta et al. (1998), we use a
measure of the efficacy of the judicial system (Judsys), ranging from 0 (least efficient)
to 10 (most efficient). Antidir is an aggregate index developed by La Porta et al.
(1998) to capture shareholder rights within a country. Acctstand is a crude measure of
the quality of financial reporting in a country, based on the inclusion or omission of
90 items in seven categories. We also use binary variables that represent the legal
origin of the different countries (English, French, German and Scandinavian legal
regime).
 572
H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593
Table 1
Summary statistics, 1969–1998
Countries Earnings Short- Put Insider Capital Cost of Expected Vol/Cap Market Volatility Syncronicity Log (%U.S. IPO
Opacity Selling Options Trading Market equity return (turnover) depth holdings/% return
Enforcement Governance Mkt cap)
Australia 6.31 1969 1982 1996 5.62 0.112 0.114 0.04 0.74 0.05 0.06  1.47 0.12
Austria 5.31 1969 1991 No 4.77 0.154 0.087 0.04 0.88 0.05 0.09 0.84 0.06
Belgium 6.59 1969 1992 1994 6.22 0.113 0.093 0.01 0.32 0.04 0.15  1.98 0.15
Brazil 6.63 No 1984 1978 8.54 No 0.177 No No 0.08 0.16  1.19 0.79
Canada 6.51 1969 1975 1976 8.50 0.054 0.108 0.03 0.95 0.03 0.06  1.15 0.06
Chile 4.00 No 1994 1996 5.00 0.047 0.077 0.01 0.21 0.05 0.21  2.10 0.09
Denmark 5.64 1969 1990 1996 5.48 0.139 0.094 0.02 0.45 0.04 0.07  1.49 0.05
Finland 5.72 No 1988 1993 6.09 0.094 0.113 No No No 0.14  0.72 0.10
France 6.36 1969 1987 1975 8.45 0.142 0.113 0.03 0.75 0.05 0.07  1.13 0.12
Germany 5.54 1969 1990 1995 5.62 0.101 0.102 0.14 3.57 0.05 0.11  1.77 0.28
Greece 3.11 No No 1996 1.44 0.211 0.085 0.02 0.34 0.08 0.19  2.46 0.49
Hong Kong 5.00 1994 1993 1994 3.83 0.164 0.126 0.04 0.62 0.07 0.15  2.24 0.17
India 3.95 No No No 0.98 0.115 0.076 0.03 0.58 0.06 0.19  2.37 0.35
Ireland 5.16 1986 No No 4.72 0.115 0.109 No No 0.05 0.06  0.29 No
Italy 4.41 1969 1995 1996 4.98 0.108 0.105 0.03 0.45 0.06 0.18  1.22 0.22
Japan 4.05 1969 1989 1990 6.40 0.057 0.120 0.03 0.54 0.06 0.23  2.52 0.28
Malaysia 4.46 1996 No 1996 1.92  0.005 0.115 0.02 0.29 0.07 0.43  1.77 1.04
Mexico 6.52 1969 No No 5.17 0.005 0.106 0.04 0.64 0.08 0.29  0.65 0.33
Netherlands 5.49 1969 1978 1994 5.85 0.135 0.107 0.06 1.63 0.04 No  0.87 0.10
Norway 7.00 No 1990 1990 6.85 0.120 0.111 0.04 0.86 0.06 0.12  0.96 0.13
Pakistan 4.38 No No No 1.13 No 0.056 No No 0.08 0.17 No No
Portugal 7.00 1986 No No 5.33 0.072 0.096 0.03 0.66 0.04 0.07  1.33 0.11
Singapore 4.89 1969 1993 1978 7.96 0.064 0.115 0.02 0.41 0.05 0.19  1.98 0.30
South Africa 4.74 1969 1992 No 4.58 0.091 0.111 0.02 0.37 0.06 0.20  2.20 0.33
South Korea 3.47 No 1997 1988 4.49  0.092 0.089 0.06 0.72 0.12 0.17  1.71 0.74
Spain 5.94 No 1992 No 3.46 0.095 0.103 0.04 0.88 0.05 0.19  1.74 0.11
Sweden 5.59 1993 1987 1990 6.40 0.151 0.105 0.03 0.61 0.06 0.14  0.94 0.31
Switzerland 5.59 1969 1988 1995 5.63 0.150 0.106 0.04 1.07 0.05 No  1.25 0.35

H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593

Taiwan 4.89 1980 No 1989 7.96 0.031 0.093 0.14 1.92 0.08 0.41  3.36 0.31
Thailand 5.29 No No 1993 3.81  0.101 0.082 0.04 0.55 0.09 0.27  2.43 0.47
Turkey 4.78 1995 No 1996 4.04 0.083 0.080 0.07 0.60 0.14 0.39  1.45 0.13
United Kingdom 5.92 1969 1984 1981 8.31 0.132 0.116 0.05 1.19 0.04 0.06  1.97 0.17
United States 7.49 1969 1973 1969 8.83 0.083 0.106 0.06 1.95 0.04 0.02 0.99 0.18
Column 1 presents the countries in our dataset in alphabetical order. For each country, we report the average CMG index value, the average of each of the three
components of the index. The bEarnings OpacityQ score is the average of three earnings opacity decile rankings. To construct this number, we rank each country-month
into deciles on each of three opacity measures (earning aggressiveness, loss aversion, and income smoothing). We then take the average of these three rankings in
computing an opacity index. Table values represent the monthly average index value for each country. Higher scores indicate lower opacity (i.e., greater transparency). The
values associated with bShort-SellingQ and bPut OptionsQ represent the date each was first allowed in a given country. The bInsider Trading EnforcementQ column reports
the date of the first enforcement action against an insider for trading violations. The liquidity variable is the natural logarithm of the ratio of volume to market
capitalization. The equity return for each country is computed from its stock market index. The variable market depth is the ratio of market turnover to return volatility. We
measure volatility as either the standard deviation of daily returns computed each month (Volatility), or the absolute monthly return for that month (|Return|). The variable
log(%U.S. stockholdings/%Market Cap) is the ratio of percentage of U.S. stockholdings in country i to the percentage of market capitalization in country i. To mitigate the
effect of outliers, we take the natural logarithm of this ratio. The variable Synchronicity is measured as the average R 2 of firm-level regressions of bi-weekly stock returns
on local and U.S. market indexes in each country. The variable, IPO returns, is the initial returns accruing to investors in IPOs of common stock. This variable was
downloaded from Jay Ritter’s web site (http://bear.cba.ufl.edu/ritter/).

573
574 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593

5. Empirical analysis

5.1. Descriptive statistics

Table 1 presents the countries in our dataset in alphabetical order. Each country is
associated with values of the components of the CMG index. The values associated with
bShort-SellingQ and bPut OptionsQ represent the date each was first feasible in a given
country. The bInsider Trading EnforcementQ column reports the date of the first
enforcement action against an insider for trading violations. Also, Table 1 shows for
each country, average liquidity, depth, efficiency and cost of equity measures.
The bEarnings OpacityQ score is the average of three earnings opacity decile rankings.
To construct this number, we rank each country-month into deciles on each of three
opacity measures (earning aggressiveness, loss aversion, and income smoothing). We then
take the average of these three rankings in computing an opacity index. Table values
represent the monthly average index value for each country. Higher scores indicate lower
opacity (i.e., greater transparency).
Table 2 ranks countries by the quality of each component of the CMG index, in
descending order. In constructing the short-selling ranking, we took the earlier of either the
date of the relaxation of short-selling constraints, or the date that put options are allowed
into a market. Also reported in this table is the rank of each country according to the CMG
index itself. As can be seen, the top five countries in terms of quality of capital market
governance are United States, Brazil, Canada, France, and the U.K., respectively. The
bottom three countries are India, Greece, and Pakistan, respectively. Most of the countries
that score high on the CMG index are European or American countries. However, Taiwan,
Singapore, and Japan, also count among the top 10. Having the largest market
capitalization in Latin America, Brazil also enjoys a high CMG score due to its long-
standing insider trading law enforcement, feasibility of short-selling and relatively good
earning opacity score.

5.2. Does capital market governance affect stock markets?

Our main empirical tests explore the effect of capital market governance on several
dimensions of market performance—the cost of capital, market liquidity, and pricing
efficiency. This section reports our results.
As a preliminary exploration, we plot the time series of CMG, liquidity, and cost of
equity measures averaged over all countries. This is shown in Fig. 3. We normalize all
three time series to have a value of 1 in 1992. This is arbitrary. However, it allows us to
have all three series with the same scale. As can be seen, CMG and liquidity seem to
exhibit an upward trend. On the other hand, cost of equity seems to show a downward
trend. This is an early indication that CMG has improved over the years, potentially
increasing liquidity and reducing cost of equity.

5.2.1. Cost of equity

We first explore the effect of capital market governance on the cost of equity. Using
the dividend yield-based measure of cost of equity as the dependent variable, we run
 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593 575

Table 2
Rankings of countries
Earnings Opacity Insider Trading Enforcement Short-Selling/Put Option CMG Index
United States United States United States United States
Portugal Brazil Australia Brazil
Norway France Austria Canada
Brazil Canada South Africa France
Belgium Singapore Ireland United Kingdom
Mexico Taiwan Canada Taiwan
Canada United Kingdom Singapore Singapore
France Finland Taiwan Norway
Australia South Korea Germany Japan
Spain Japan Mexico Sweden
United Kingdom Norway Japan Belgium
Finland Sweden Italy Finland
Denmark Thailand Netherlands Netherlands
Sweden Turkey France Switzerland
Switzerland Malaysia Switzerland Germany
Germany Hong Kong United Kingdom Australia
Netherlands Belgium Denmark Denmark
Austria Netherlands Belgium Portugal
Thailand Switzerland Portugal Mexico
Ireland Germany Finland Chile
Hong Kong Chile Brazil Italy
Taiwan Greece Sweden Austria
Singapore Australia Norway Ireland
Turkey Italy Spain South Africa
South Africa Denmark Hong Kong South Korea
Malaysia India Chile Turkey
Italy South Africa Turkey Hong Kong
Pakistan Portugal South Korea Thailand
Japan Spain Malaysia Spain
Chile Ireland Greece Malaysia
India Pakistan Pakistan Greece
South Korea Austria India Pakistan
Greece Mexico Thailand India
Countries are ranked by decreasing quality of different components of the CMG index (columns 1 to 3). In
ranking countries by Short-Selling/Put Option, we take the earlier of the two introduction dates. Also reported in
this table is the rank of each country according to the CMG index itself (column 4).

five panel time-series regressions with country-fixed effects. Model 1 uses CMG as the
independent variable. Models 2 to 4 use each one component of CMG as the
independent variable. Model 5 uses all three components of CMG as the independent
variables. All models control for liberalization, liquidity risk (Brennan and Sub-
rahmanyam, 1996),6 and foreign exchange risk (Ferson and Harvey, 1993; Dumas and

6
The proxy for liquidity risk is turnover. Turnover is the ratio of volume of trade to market capitalization. We
take the natural logarithm of this ratio for reasons mentioned before.
576 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593

1.6

1.5

1.4

1.3

1.2

1.1

0.9

0.8

0.7

0.6
1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998
CMG Cost of equity Liquidity

Fig. 3. Capital market governance, liquidity, and cost of equity. This figure depicts the time-series of CMG,
liquidity, and cost of equity, averaged over all countries. CMG is our measure of capital market governance, based
on three elements: (1) the degree of earning opacity, (2) the enforcement of insider trading laws, and (3) the
relaxation of short-selling restrictions. Liquidity is the ratio of trading volume to market capitalization; Cost of
equity is excess return based on an international asset pricing model. For presentation, we normalize all three time
series to a value of 1 in 1992.

Solnik, 1995).7 The panel regressions use data for the 32 countries for which we have
dividend yield data from January 1986 to December 1998 (some countries do not have
data for the full time period). We correct for country-specific heteroskedasticity and
country-specific autocorrelation.
Table 3 presents the results from these panel time-series regressions. In Model 1, the
coefficient for the CMG variable is negative and statistically significant at the 1% level.
This is consistent with our hypothesis that improved capital market governance lowers the
cost of equity. The association is also economically significant. An increase in CMG of
one standard deviation is associated with a 2.6% decrease in the cost of equity.8 The
coefficient on the currency risk has the right sign, but is not significantly different from 0 at
conventional levels. The liberalization indicator has the correct negative sign, but is not
significantly different from 0 at conventional levels. The liquidity variable has the correct
negative sign, but is also not significantly different from 0. The results from Models 2 to 5 are
broadly consistent with our conclusion. Each of the three components of CMG is significant

7
Because of convergence problems, our estimation is a two-step procedure. In the first step we strip out the effects
of the local variance factor and the world factor, and in the second step, to isolate the effect of earnings opacity, we
strip out the effects of other factors like the FX factor. The FX factor that we use is the conditional covariance of the
return of the stock market index of the country with the return a U.S. investor would receive if she held the foreign
currency. This conditional covariance is obtained by using the multivariate ARCH model we discuss in equation
(B2) in Appendix B—just replace the world portfolio (w) by the foreign exchange portfolio (ifx).
8
Calculated as 0.0025 (per month)  12 months  0.86 (the time-series standard deviation of the CMG index,
averaged across all countries).
 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593 577

Table 3
Effect of capital market governance on the cost of equity (using dividend yield as a proxy)
Dependent variable Cost of equity
Independent variable Model 1 Model 2 Model 3 Model 4 Model 5
CMG 0.0025
(4.23)
Insider Trading 0.0011 0.0005
(2.71) (1.58)
Earning Opacity 0.0029 0.0025
(3.46) (2.98)
Short-Selling 0.0007 0.0005
(1.99) (1.10)
Forex 2.4948 0.1375 0.1006 2.4517 2.4392
(0.98) (0.04) (0.03) (0.91) (0.94)
Liquidity 0.0009 0.0005 0.0022 0.0002 0.0009
(0.82) (0.25) (1.92) (0.12) (0.89)
Liberalization 0.0038 0.0035 0.0029 0.0038 0.0024
(0.65) (0.82) (0.61) (0.77) (0.45)
Country Fixed Effects not included
No. of observations 3414 4479 3498 4375 3414
The panel regressions with country-fixed effects are based on monthly data. They are corrected for country-
specific heteroskedasticity and country-specific autocorrelation. t-statistics are reported in the parentheses. The
dependent variable is k, the cost of equity. It is defined as follows. It is computed as the sum of the dividend yield
forecast and the growth rate of the dividend. The bCMGQ variable is capital market governance and it consists of
three different elements: (1) the level of earning opacity, (2) enforcement of insider trading laws, and (3) short-
selling restrictions. The indicator variable bliberalizationQ changes from 0 to 1 in the month after the official
liberalization date that was obtained from Bekaert and Harvey (2000). The liquidity variable is the natural
logarithm of the ratio of volume to market capitalization. bForexQ is the conditional covariance of the return of the
stock market index with the depreciation of the ith foreign currency with respect to the dollar at time t. It is
estimated using a multivariate ARCH model. The equity data for developed countries are from Morgan Stanley
Capital International, and the equity data for emerging markets are from International Financial Corporation.

and of the correct sign when included on its own. When all components are included
together, they are still of the correct sign, although only earnings opacity remains significant.
We also compute a cost of capital measure based on realized returns using non-linear
least squares. The results are reported in Panel A of Table 4.
Panel A of Table 4 reveals that covariance risk has a positive price (k cov is positive). The
estimates are statistically significant at the 5% level. It also reveals that own country variance
risk has a positive price (k var is positive). The estimates are statistically significant at the 10%
level. Using residuals from the first stage as the dependent variable, we run five panel time-
series regressions with country-fixed effects. Models 1, 2, 3, 4 and 5 are as previously
defined. We control for liberalization, liquidity risk, and foreign exchange risk as before.
Panel B of Table 4 presents the results from this panel time-series regression. The
coefficient of the CMG index (Model 1) is negative and statistically significant at the 1%
level. This is consistent with our hypothesis that capital market governance adversely
affects the cost of equity. An increase in CMG of one standard deviation is associated with
a 2.9% decrease in the cost of equity. At first blush, this result might appear surprisingly
large. However, many of the countries in our sample are emerging markets, with yearly
578 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593

Table 4
Effect of capital market governance on the cost of equity (using an international asset pricing model)
Panel A: Some coefficients of the risk-adjustment model, Model 1
Parameter Coefficient p-value
a0 0.0011 0.5534
a1 15.6094 0.0283
k cov 2.2157 0.0471
k var 2.3984 0.0615

Panel B: Effect on residuals (risk adjusted)

Dependent variable Residual from risk adjustment model
Independent variable Model 1 Model 2 Model 3 Model 4 Model 5
CMG  0.0028
(3.07)
Insider Trading 0.0017 0.0009
(4.30) (1.76)
Earning Opacity  0.0004 0.0002
(0.46) (0.23)
Short-Selling 0.0015 0.0013
(2.57) (1.82)
Forex 7.2954 5.9985 7.1289 6.0448 7.2660
(5.34) (4.26) (5.15) (4.33) (5.32)
Liquidity 0.0054 0.0080 0.0032 0.0075 0.0053
(3.05) (5.30) (1.84) (5.12) (2.94)
Liberalization  0.0076 0.0200  0.0076 0.0205 0.0080
(0.73) (2.72) (0.72) (2.81) (0.76)
Country fixed effects not reported
No. of observations 3200 4214 3272 4121 3200
Model 1: The international asset pricing factor model used for risk-adjusting is
   
ri;t  rf ;t ¼ a0 þ ui;t kcov hi;w;t þ 1 þ ui;t kvar hi;t þ ei;t

where the measure of integration of country i at time t, / i,t , is defined in the text. k cov is the price of the
covariance risk with the world, and k var is the price of own country variance risk. The independent variables are
the conditional covariances and variances, h i,w,t and h i ,t , respectively, and these are obtained from the multivariate
ARCH model as defined in the text.
In Panel B, the dependent variable is the monthly equity return for each country minus the 1-month U.S. T-Bill
return. The equity return for each country is computed from its stock market index. Data on monthly stock market
indices for the 20 developed markets were obtained from Morgan Stanley Capital Market International (MSCI).
Data on monthly stock market indices for the 14 emerging markets were obtained from the International Financial
Corporation (IFC). The data for the 1-month U.S. Treasury bill return was obtained from Datastream. The
measure of a country’s integration with the world, as defined above, is computed from its exports, imports, and
GDP. It is equation (B3) in Appendix B. Data on quarterly/annual GDP, monthly exports and monthly imports
were from the International Financial Statistics of the International Monetary Fund. The statistics for Taiwan
come from Datastream.
The conditional covariance of the return of the stock market index with the depreciation of the ith foreign
currency with respect to the dollar at time t, defined as the foreign exchange risk and denoted as h i,ifx,t , is
estimated from the multivariate ARCH model as defined in the text. t-statistics are reported in the parentheses.
 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593 579

Table 5
Effect of capital market governance on trading volume
Dependent variable Vol/Cap (turnover)
Independent variables Model 1 Model 2 Model 3 Model 4 Model 5
CMG 0.0040
(3.75)
Insider Trading 0.0020 0.0019
(4.91) (3.39)
Earning Opacity 0.0012 0.0004
(1.73) (0.71)
Short-Selling 0.0021 0.0002
(4.93) (0.24)
Liberalization 0.0133 0.0097 0.0161 0.0100 0.0147
(2.38) (1.66) (3.22) (1.71) (2.44)
Country fixed effects
not reported
No. of observations 3469 4720 3553 4615 3469
The panel regressions with country fixed-effects are based on monthly data. They are corrected for country-
specific heteroskedasticity and country-specific autocorrelation. t-statistics are reported in the parentheses. The
dependent variable is Vol/Cap, and it is the ratio of volume to market capitalization. The first independent variable
is bCMGQ. bCMGQ is capital market governance and it consists of three different elements: 1—the level of earning
opacity, 2—enforcement of insider trading laws, and 3—short-selling restrictions. The next three independent
variables are the three components of CMG. The fifth independent variable is the liberalization variable. It is
coded as follows. The indicator variable bliberalizationQ changes from 0 to 1 in the month after the official
liberalization date that was obtained from Bekaert and Harvey (2000). The equity data for developed countries are
from Morgan Stanley Capital International, and the equity data for emerging markets are from International
Financial Corporation.

returns that range from  18% to 28%. In this context, our estimate of the impact of
improved capital market governance on the cost of equity does not seem extreme.
In sum, we derive the cost of equity using two sharply different methods of
estimation—the dividend yield method (implicitly controls for risk, but has less estimation
risk) and an international asset pricing model (explicitly controls for risk, but has more
estimation risk). Yet the results are strikingly similar—a one standard-deviation change in
the CMG index is associated with a 2.6% and 2.9% change in the cost of capital,
respectively. These results suggest our finding is robust to perturbations in the method for
estimating the cost of capital.

5.2.2. Liquidity (trading volume, market depth, and U.S. foreign holdings)
Our first measure of market liquidity is trading volume (or turnover), defined as the ratio
of the dollar volume of trade to the total market capitalization.9 We run five panel time-
series regressions with country-fixed effects in Table 5. The second measure of liquidity is
market depth (Turnover/Volatility), and these regressions are reported in Table 6.
Table 5 presents the results from this panel time-series regression. The coefficient of
the CMG index is positive and statistically significant at the 1% level. An increase in

9
We also use natural log of the ratio as the dependent variable and the results remain the same.
580 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593

Table 6
Effect of capital market governance on market depth
Market depth
Dependent variable Turnover/volatility Turnover/|Return| Volatility
Independent variables Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7
CMG 0.0791 0.1438 0.0024
(2.83) (2.69) (1.93)
Insider Trading 0.0304 0.0205
(1.35) (0.94)
Earning Opacity 0.0419 0.0417
(4.17) (3.16)
Short-Selling 0.0324 0.0090
(3.04) (0.59)
Liberalization 0.1366 0.2004 0.0335 0.0492 0.1368 0.3863  0.0042
(4.08) (8.68) (0.45) (0.54) (3.79) (3.02) (2.84)
Country fixed effects
not reported
No. of observations 3385 3469 4562 4457 3385 3355 3865
The panel regressions with country fixed-effects are based on monthly data. They are corrected for country-
specific heteroskedasticity and country-specific autocorrelation. t-statistics are reported in the parentheses. The
dependent variable in the first six models is market depth, defined as the ratio of market turnover to return
volatility. Market turnover is volume/market cap as defined in Table 5. We measure volatility as either the
standard deviation of daily returns computed each month (Volatility), or the absolute monthly return for that
month (|Return|). In Model 7, the dependent variable is Volatility.
The first independent variable is bCMGQ. bCMGQ is capital market governance and it consists of three different
elements: 1—the level of earning opacity, 2—enforcement of insider trading laws, and 3—short-selling
restrictions. The next three independent variables are the three components of CMG. The fifth independent
variable is the liberalization variable. It is coded as follows. The indicator variable bliberalizationQ changes from 0
to 1 in the month after the official liberalization date that was obtained from Bekaert and Harvey (2000). The
equity data for developed countries are from Morgan Stanley Capital International, and the equity data for
emerging markets are from International Financial Corporation.

CMG of one standard deviation is associated with a 9.5% increase in trading volume
(i.e., the turnover ratio).10 The results from Models 2 to 5 are generally consistent with
our conclusion. The coefficients on liberalization are significant, and have the right
sign.
One concern with the trading volume result is that improved CMG could lead to
greater volatility. In terms of overall market liquidity, we are interested in measuring
how much volume can be traded without excessive price movements. The Market
Depth regressions are aimed at this issue. Specifically, these variables evaluate the
extent to which CMG allows a greater volume to be traded per unit of market
volatility.
Table 6 reports results for the Market Depth regressions. The dependent variable in
Model 7 is Volatility. In Models 1, 2, 3, 4 and 5 we compute market depth

10
This is computed as 0.004  0.86 / 0.036 where 0.86 is average time-series standard deviation of CMG index,
and 0.036 is the overall average of market turnover.
 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593 581

Table 7
Effect of earnings opacity on U.S. foreign stockholdings
Dependent variable %U.S. stockholdings/%Market Cap
Independent variables Model 1 Model 2 Model 3 Model 4 Model 5
CMG 0.1386
(1.72)
Insider Trading 0.3132 0.3122
(2.69) (2.20)
Earning Opacity 0.0813 0.0750
(1.65) (1.35)
Short-Selling 0.0284  0.0183
(0.94) (0.50)
Acctstand 0.0083 0.0054 0.0113 0.0005 0.0072
(0.40) (0.31) (0.63) (0.03) (0.36)
Antidir 0.0431 0.0329 0.0516 0.1208 0.0585
(0.26) (0.24) (0.44) (1.03) (0.39)
Judsys 0.0211 0.0377 0.0202 0.0005 0.0201
(0.25) ( 0.49) (0.27) (0.01) (0.25)
French Law 0.3523 0.0721 0.0005 0.0954 0.2734
(0.58) ( 0.15) (0.0) (0.21) (0.52)
German Law 0.6291 0.1353 0.4166 0.3489 0.2737
(1.24) ( 0.32) (1.01) (0.82) (0.60)
Scandinavian Law 0.2234 0.2333 0.4094 0.4136 0.2150
(0.50) (0.57) (1.01) (0.99) (0.52)
Country fixed effects
not reported
No. of observations 29 29 32 32 29
Foreign stockholdings by U.S. citizens for 1997 are from Bhattacharya and Groznik (2003). The data are from
the survey carried out by the U.S. Treasury Department and the Board of Governors of the Federal Reserve
System. The dependent variable is the ratio of percentage of U.S. stockholdings in country i to the percentage
of market capitalization in country i. To mitigate the effect of outliers, we take the natural logarithm of this
ratio. The first independent variable is bCMGQ. bCMGQ is capital market governance and it consists of three
different elements: 1—the level of earning opacity, 2—enforcement of insider trading laws, and 3—short-
selling restrictions. The next three independent variables are the three components of CMG. We control for
three variables that were featured in La Porta et al. (1998) as measures of the level of corporate governance and
protection of minority shareholder rights. Judsys is a measure of the efficacy of the judicial system, ranging
from 0 (least efficient) to 10 (most efficient). Antidir is an aggregate index developed by La Porta et al. (1998)
to capture shareholder rights within a country. Acctstand is a crude measure of the quality of financial reporting
in a country, based on the inclusion or omission of 90 items in seven categories. We also use binary variables
that represent the legal origin of the different countries (English, French, German and Scandinavian legal
regimes). The equity data for developed countries are from Morgan Stanley Capital International, and the
equity data for emerging markets are from International Financial Corporation. t-statistics are reported in the
parentheses.

(Turnover/Volatility), with volatility defined as the standard deviation of daily returns

computed each month. In Model 6 we compute Market Depth using the ratio of
market turnover to the absolute value of monthly returns. We control for liberalization
in all models. The panel regressions use data for the 32 countries for which we have
daily return data. We correct for country-specific heteroskedasticity and country-
specific autocorrelation.
582 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593

Table 8
Effect of earnings opacity on stock synchronicity
Dependent variable Stock Price Synchronicity
Independent variables Model 1 Model 2 Model 3 Model 4 Model 5
CMG 0.0224
(2.05)
Insider Trading 0.0354 0.0252
(2.04) (1.14)
Earning Opacity 0.0013 0.0044
(0.19) (0.52)
Short-Selling 0.0065 0.0034
(1.62) (0.62)
Acctstand 0.0708 0.0167  0.0373 0.0262 0.0322
(0.86) (0.23) (0.54) (0.40) (0.40)
Antidir 0.1113 0.0324 0.0200 0.0424 0.0581
(1.59) (0.51) (0.32) (0.70) (0.80)
Judsys 0.0057 0.0004  0.0224 0.0295 0.0111
(0.09) (0.01) (0.36) (0.50) (0.17)
French Law 0.0062 0.0039 0.0013 0.0020 0.0047
(2.20) (1.51) (0.54) (0.88) (1.55)
German Law 0.0070 0.0062  0.0210 0.0223 0.0027
(0.31) (0.30) (1.08) (1.25) (0.11)
Scandinavian Law 0.0290 0.0293  0.0299 0.0240 0.0271
(2.55) (2.58) (2.68) (2.18) (2.19)
Country fixed effects
not reported
No. of observations 28 28 32 32 28
2
The dependent variable, Stock Price Synchronicity, is measured as the average R of firm-level regressions of bi-
weekly stock returns on local and U.S. market indexes in each country. The first independent variable is bCMGQ.
bCMGQ is capital market governance and it consists of three different elements: 1—the level of earning opacity,
2—enforcement of insider trading laws, and 3—short-selling restrictions. The next three independent variables
are the three components of CMG. We control for three variables that were featured in La Porta et al. (1998) as
measures of the level of corporate governance and protection of minority shareholder rights. Judsys is a measure
of the efficacy of the judicial system, ranging from 0 (least efficient) to 10 (most efficient). Antidir is an aggregate
index developed by La Porta et al. (1998) to capture shareholder rights within a country. Acctstand is a crude
measure of the quality of financial reporting in a country, based on the inclusion or omission of 90 items in seven
categories. We also use binary variables that represent the legal origin of the different countries (English, French,
German and Scandinavian legal regimes). The equity data for developed countries are from Morgan Stanley
Capital International, and the equity data for emerging markets are from International Financial Corporation. t-
statistics are reported in the parentheses.

Table 6 shows that the coefficient of the CMG index is positive and statistically
significant at the 1% level, indicating that improvements in CMG increase market depth.
An increase in CMG of one standard deviation is associated with a 9.3% increase in
market depth.11 The results from Models 2 to 7 are also in line with our hypothesis.12

11
This is calculated as 0.08  0.86 / 0.74 where 0.86 is average time-series standard deviation of CMG index,
and 0.74 is the overall average of market depth.
12
We also use the conditional volatility of monthly returns computed from the multivariate ARCH model used
in the International Asset Pricing Model described above. The results remain the same.
 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593 583

Table 9
Effect of earnings opacity on IPO returns
Dependent variable IPO returns
Independent variables Model 1 Model 2 Model 3 Model 4 Model 5
CMG 0.0586
(3.01)
Insider Trading 0.0771 0.0395
(2.37) (1.03)
Earning Opacity 0.0101 0.0044
(0.84) (0.30)
Short-Selling 0.0189 0.0187
( 2.65) (1.91)
Acctstand 0.0914 0.2421 0.2414 0.1946 0.2270
(0.66) (1.84) (2.08) ( 1.85) (1.72)
Antidir 0.1097 0.0788 0.0129 0.0465 0.0244
(0.90) (0.67) (0.12) (0.46) (0.20)
Judsys 0.1435 0.1749 0.2037 0.1998 0.2165
(1.31) (1.53) (1.79) ( 1.98) (1.92)
French Law 0.0130 0.0071 0.0091 0.0090 0.0068
(2.60) (1.46) (1.86) (2.13) (1.28)
German Law 0.0257 0.0667 0.0639 0.0740 0.0750
(0.66) (1.78) (1.85) ( 2.42) (1.89)
Scandinavian Law 0.0342 0.0352 0.0485 0.0333 0.0209
(1.78) (1.71) (2.49) ( 1.80) (0.99)
Country fixed effects
not reported
No. of observations 30 30 33 33 30
The dependent variable, IPO returns, is the initial returns accruing to investors in IPOs of common stock. This
variable was downloaded from Jay Ritter’s web site (http://bear.cba.ufl.edu/ritter/). The first independent variable
is bCMGQ. bCMGQ is capital market governance and it consists of three different elements: 1—the level of earning
opacity, 2—enforcement of insider trading laws, and 3—short-selling restrictions. The next three independent
variables are the three components of CMG. We control for three variables that were featured in La Porta et al.
(1998) as measures of the level of corporate governance and protection of minority shareholder rights. Judsys is a
measure of the efficacy of the judicial system, ranging from 0 (least efficient) to 10 (most efficient). Antidir is an
aggregate index developed by La Porta et al. (1998) to capture shareholder rights within a country. Acctstand is a
crude measure of the quality of financial reporting in a country, based on the inclusion or omission of 90 items in
seven categories. We also use binary variables that represent the legal origin of the different countries (English,
French, German and Scandinavian legal regimes). The equity data for developed countries are from Morgan
Stanley Capital International, and the equity data for emerging markets are from International Financial
Corporation. t-statistics are reported in the parentheses.

Model 7 indicates that improved CMG is associated with increased volatility. However,
the results from Models 1 through 6 show that overall, improved CMG increases
market depth—i.e., is associated with increased volume per unit of volatility. In short,
our evidence shows that the increase in trading volume due to better governance more
than compensates for the increase in volatility, leading to a net improvement in market
depth.
In addition to volume and depth, we also use the U.S. holdings as an indirect measure
of liquidity. We use as our dependent variable log ratio of the percentage of U.S.
stockholdings in each country to the percentage of the market capitalization of the country
584 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593

in the global portfolio as the dependent variable. Five cross-sectional regressions are
conducted.13
Table 7 presents the results from these regressions. In Model 1, the coefficient of the
capital market governance variable is positive and statistically significant at the 1% level.
This is consistent with our hypothesis that our capital market governance measure
increases the percentage of holdings by U.S. citizens relative to the size of the market. An
increase in CMG of one standard deviation is associated with a 12.7% increase in the
holdings ratio.14 These findings are supported in Models 2 and 5. Our finding is related to
Giannetti and Koskinen (2004) which propose in a theoretical model that investors are
biased towards investing in countries with good governance. Dahlquist et al. (2002) find
that most firms in countries with poor investor protection are closely held. They argue that
this finding might explain part of the home bias of U.S. investors.

5.2.3. Pricing efficiency (stock price synchronicity and IPO underpricing)

We have two measures of pricing efficiency, stock price synchronicity and IPO
underpricing. Using stock price synchronicity and IPO underpicing as the dependent
variable, we run five cross-sectional regressions in Tables 8 and 9. All models control for
accounting standard, anti-director rights, efficiency of the judicial system, and legal origin.
Table 8 presents the results on stock price synchronicity. In Model 1, the coefficient of
the capital market governance variable is negative and statistically significant at the 1%
level. An increase in CMG of one standard deviation is associated with a 1.9% decrease in
the stock price synchronicity (the R 2 measure described above). The results from Models 2
to 5 are all in the same directions. These findings are consistent with, and generalizes, Bris
et al.’s (2004) finding that a relaxation of short sales constraints is associated with
improved market pricing efficiency.
Table 9 presents the results from the IPO regressions. In Model 1, the coefficient of the
capital market governance variable is negative and statistically significant at the 1% level.
An increase in CMG of one standard deviation is associated with a 5.2% decrease in the
IPO underpricing.15 This is consistent with our hypothesis that our capital market go-
vernance measure improves pricing efficiency, and reduces the cost to firm insiders of
obtaining capital via going public. The results from Models 2 to 5 are qualitatively similar.

6. Additional robustness checks

Although we have already conducted a number of robustness checks, the interpretation

of our empirical results can still be complicated by potential correlated omitted variables
and the endogeneity of capital market governance. We have attempted to deal with these
concerns through the use of country-level fixed-effect models. In addition, we now run
two additional robustness tests.

13
We also used the ratio itself without taking log. The result remains the same.
14
Computed as exp(0.1386  0.86)  1 where 0.86 is average time-series standard deviation of CMG index.
15
Calculated as exp(0.0586  0.86)  1 where 0.86 is average time-series standard deviation of CMG index.
 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593 585

First, we reconstruct the CMG index by incorporating an extra factor. We use an

aggregate country corruption index to proxy for other governance factors in that country’s
legal environment. This corruption index published by International Country Risk Guide
(ICRG) is more general in scope, and is intended to capture governance effects that are not
included in our previous measures (see Lee and Ng, 2004). We reran all the regressions in
the paper and find that all the results are robust to the inclusion of this measure.
Second, we estimated a Vector Auto Regression (VAR) model proposed by Sims
(1980). In this model, we treat capital market governance, the cost of equity, and market
depth as endogenously determined dependent variables. These three endogenous variables
are modeled as linear functions of lagged endogenous variables and all exogenous
variables in the system. The system of equations in the VAR is estimated jointly. This
means that the effect of the independent variables on each endogenous variable takes into
account the endogenous nature of the other endogenous variables.
Formally, the system of equations to estimate the effect on the cost of equity is:

cost of  equityit ¼ a10 þ bi11 CMGi;t1 þ bi12 Forexi;t þ bi13 Liquidityi;t1

þ bi14 Liberalizationi;t þ u1i;t

and

CMGi;t ¼ a20 þ bi21 cost of  equityit1 þ bi22 Forexi;t þ bi23 Liquidityi;t1

þ bi24 Liberalizationi;t þ u2i;t

The system of equations on market depth (Liquidity) is:

Liquidityit ¼ a10 þ bi11 CMGi;t1 þ bi12 Liberalizationi;t þ u1i;t

and
CMGi;t1 ¼ a20 þ bi21 þ Liquidityi;t1 ¼ þ bi22 Liberalizationi;t þ u2i;t :

The system of equations is estimated jointly using Seemingly Unrelated Regressions

(SUR). SUR computes estimates using the technique of joint GLS (Generalized Least
Squares). The two error terms u 1i,t and u 2i,t are allowed to be correlated (see Enders, 1996
for further details). The estimation allows for country fixed-effects, for country-specific
heteroskedasticity, and for country-specific autocorrelation.
We find that endogeneity does exist. Overall capital market governance is negatively
affected by the cost of equity using the dividend yield method. Also, governance is
affected by liquidity. However, despite the endogeneity that we explicitly account for, our
previous inferences on the effect of capital market governance variables on the cost of
equity or trading volume are not affected.16

16
We use the implied cost of capital for this test. Because of the two-stage nature of the estimation of the cost of
capital using the international asset pricing model method, it was not possible for us to run a VAR for this method.
586 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593

7. Conclusion

This paper explores the link between capital market governance (CMG) and market
performance in a broad cross-section of countries. We use variables related to three
dimensions of capital market governance-earning opacity, enforcement of insider trading
laws, and short-selling. We combine these three dimensions to obtain an overall capital
market governance time-series measure per country. We then examine the association
between changes in CMG and changes in: (1) the cost of capital (both realized and implied);
(2) market liquidity (i.e., trading volume and market depth), and U.S. foreign stockowner-
ship); and (3) pricing efficiency (i.e., stock price synchronicity and IPO underpricing).
We document in our tests that, after controlling for other influences, an increase in
overall capital market governance in a country is linked to a decrease in the cost of equity,
an increase in market liquidity, and an increase in pricing efficiency. Specifically,
improved security laws are associated with decreased cost of capital, higher trading
volume, greater market depth, increased U.S. ownership, lower price synchronicity, and
reduced IPO underpricing. These results hold for the overall CMG index, and are
directionally consistent for each of the three individual CMG components.
We believe our analyses have important implications for investors, securities regulators
and financial academics. Collectively, our evidence points to the importance of security
laws to the proper functioning of stock markets. Specifically, our results indicate that
increased enforcement of insider trading laws, improved accounting standards (through
more stringent auditing and disclosure standards) and a relaxation of short-selling
constraints are all associated with market performance. Specifically, improvements in
these CMG variables lead to decreases in the cost of capital, increases in market liquidity,
and increases in market pricing efficiency. In general, the magnitude of these relations
suggests they are economically and statistically important.
These findings are consistent with the view that investors associate bad capital market
governance with increased risk. Specifically, our findings suggest heightened investor
concerns over capital market governance can prompt investors to reduce their trading
activity and demand greater premiums for holding equity securities. Interestingly, we also
find that bad security laws result in lower market pricing efficiency. To the extent that
regulators prefer lower costs of capital, greater market liquidity, and improved pricing
efficiency, the prescriptive implications of our results are unambiguous.

Acknowledgements

We thank Johnny Fong and Justin Zhang for excellent research assistance. We also
thank Charles Chang and Brian Henderson, as well as participants at the Cornell
University research workshop, the 2004 Georgia Tech International Finance Conference,
and the 2004 Taiwan Accounting Association Annual Conference, for helpful suggestions
and comments. Thomson Financial Services Inc. provided the earnings per share forecast
data, available through the Institutional Brokers Estimate System (I/B/E/S). These data
have been provided as part of a broad academic program to encourage earnings
expectation research. All errors are our own.
 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593 587

Appendix A. Description of earnings opacity measures

Earnings opacity is inherently difficult to measure, particularly across countries,

because it is not possible to pinpoint management’s motives, and it is difficult to
compare accounting standards and the enforcement of these accounting standards. In
addition, it is not possible to capture all factors that might influence earnings opacity,
or to model how they interact to produce more or less opaque earnings. Therefore,
rather than studying the inputs that determine earnings opacity, Bhattacharya et al.
(BDW, 2003) analyze the outcome: the distributional properties of reported
accounting numbers across countries and across time that suggest earnings opacity.
We adapt the BDW approach in this study, and explain our estimation process in
detail below.
We use measures that are intended to capture three attributes of earnings numbers that
could lead to earnings opacity: earnings aggressiveness, loss avoidance, and earnings
smoothing. We focus on these three dimensions because prior literature has suggested
that these three dimensions may weaken the link between accounting performance and
the true economic performance of a firm. We limit our analysis of earnings opacity to
industrial firms, so that differences in the underlying earnings process across different
industry groups, and differences in the proportion of firms in various industry groups
across countries and across time, do not affect the dimensions of reported earnings we
examine.
Finally, given the above mentioned difficulties in measuring earnings opacity, all our
tests are inherently joint tests of two hypotheses: one, our three measures, or a
composite of all three, are associated with uninformative or opaque earnings and, two,
earnings opacity creates an informational risk that affects the cost of equity and liquidity.
We construct a panel data set for each of these three measures of the three dimensions of
earnings opacity–earnings aggressiveness, loss avoidance, and earnings smoothing–and
then combine them to obtain an overall earnings opacity time-series measure per
country.

A.1. Earnings aggressiveness measure

Our first measure of earnings opacity is earnings aggressiveness. Ball et al. (2000)
argue that the opposite of aggressiveness, accounting conservatism, which is the more
timely incorporation of economic losses versus economic gains into accounting earnings,
arises to reduce information asymmetry. Specifically, they argue that three factors are
expected to lead to accounting conservatism. First, accountants are aware that managers
would like to report economic gains and suppress information about economic losses.
Hence, accountants find negative information more credible, and are more likely to
incorporate it into accounting income. Second, lenders are important users of financial
statements, and lenders are more impacted by economic losses than by economic gains.
Third, the timely incorporation of economic losses provides an important corporate
governance role, providing quick feedback about bad investment decisions and strategies
that managers may not wish to disclose. The first and third of these factors suggest that
accounting conservatism is related to informativeness, since conservative accounting is
588 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593

expected to provide information that management may have incentives to withhold

otherwise.17
We follow Givoly and Hayn (2000) and Ahmed et al. (2002) and use accruals to measure
earnings aggressiveness. As earnings aggressiveness is the tendency to delay the recognition
of losses and speed the recognition of gains, it implies that, if cash flow realizations are held
equal, we would expect accruals to increase as earnings aggressiveness increases.
Aggressive accounting would be characterized by fewer such negative accruals which
capture economic losses, and more positive accruals which capture economic gains,
increasing the overall level of accruals. Although unrealized gains and unrealized losses will
eventually be recognized in accounting earnings in any clean surplus accounting system, a
more conservative accounting system is expected to result in more negative accruals at any
given point in time. This result arises because a greater proportion of economic losses
relative to economic gains will be reflected in accounting earnings at any point in time.
This observation provides the impetus for us to measure the earnings aggressiveness of
a country at a point in time as the median for country i, year t, of accruals divided by
lagged total assets. We use the median observation of scaled accruals to minimize the
influence of extreme observations. The higher is the median observation of scaled accruals
of country i in year t, the higher is the earnings aggressiveness.
Consistent with much of the past literature (e.g., Healy, 1985; Jones, 1991; Dechow et al.,
1995; Leuz et al., 2002), we compute scaled accruals from balance sheet and income statement
information, and then compute scaled cash flows as scaled operating income minus scaled
accruals. We do not use information from the cash flow statement because of differences in the
presentation of cash flow information across countries and time. In fact, many of our sample
countries do not require the preparation or presentation of a statement of cash flows. We define
scaled accruals as
ACCkt ¼ ðCAkt  CLkt  CASHkt þ STDkt  DEPkt þ TPkt Þ=TAkt1 ðA1Þ
where

ACCkt = Scaled accruals for firm k, year t

CAkt = Change in total current assets for firm k, year t
CLkt = Change in total current liabilities for firm k, year t
CASHkt = Change in cash for firm k, year t
STDkt = Change in current portion of long-term debt included in total current liabilities
for firm k, year t
DEPkt = Depreciation and amortization expense for firm k, year t
TPkt = Change in income taxes payable for firm k, year t
TAkt1 = Total assets for firm k, year t  1.
17
It is possible that earnings aggressiveness does not necessarily lead to earnings opacity. It could be argued that
conservative accounting prevents good news from being transmitted quickly, thus adding noise. However, given
that one might reasonably expect managerial incentives to overstate rather than understate earnings on average,
our belief is that aggressive earnings are more opaque earnings, because such accounting reports are more likely
to reflect biased and optimistic reporting on the part of management, adding noise to reported earnings and,
hence, increasing earnings opacity. To understand these managerial motives, see, for example, Rangan (1998),
Teoh et al. (1998), Shivakumar (2000), Healy (1985), and Barth et al. (1999). Ultimately, whether earnings
aggressiveness leads to earnings opacity or not is an empirical issue.
 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593 589

A.2. Loss avoidance measure

Our second measure of earnings opacity is loss avoidance behavior. Burgstahler and
Dichev (1997) present persuasive evidence that U.S. firms engage in earnings
management to avoid reporting negative earnings. Degeorge et al. (1999) provide
evidence that suggests that the following hierarchy exists among three earnings thresholds:
1) avoiding negative earnings, 2) reporting increases in quarterly earnings, and 3) meeting
analysts’ earnings forecasts. As Burgstahler and Dichev (1997) and Degeorge et al. (1999)
discuss, these results indicate that incentives to report positive earnings (i.e., beat a
benchmark of 0 earnings) exist for some sample firms. Such loss avoidance behavior
obscures the relationship between earnings and economic performance, thus increasing
earnings opacity.
We define firms with small positive earnings (small negative earnings) as firms with net
income scaled by lagged total assets between 0% and 1% (between 0% and  1%). We
find the ratio of the number of firms with small positive earnings minus the number of
firms with small negative earnings divided by their sum. The higher is this ratio in country
i, year t, the higher is the loss avoidance.

A.3. Earnings smoothing measure

Our third measure of earnings opacity is earnings smoothing. Some accounting

standards (for example, cases of high book/tax conformity) or some managerial motives
may lead to smooth earnings over time. If accounting earnings are artificially smooth, they
fail to depict the true swings in underlying firm performance, thus decreasing the
informativeness of reported earnings and, hence, increasing earnings opacity. This is
consistent with the view of earnings smoothing taken in Leuz et al. (2002). An alternative
view is that earnings smoothing can be used by management as a means to convey
information, potentially decreasing earnings opacity. While we believe that earnings
smoothing at the country level is indicative of accounting that obscures information about
economic volatility, whether or not earnings smoothing leads to earnings opacity and
adverse capital market consequences is again an empirical issue.
Following Leuz et al. (2002), we find the cross-sectional correlation between the
change in accruals and the change in cash flows, both scaled by lagged total assets,
in country i, year t. Cash flows are obtained by subtracting accruals (which were
obtained in (A1)) from operating earnings. Because some degree of earnings smoothing is
a natural outcome of any accrual accounting process, this measure is expected to be
negative on average. However, the more negative this correlation, the more likely it is that
earnings smoothing is obscuring the variability in underlying economic performance, and
the greater is the earnings opacity.

A.4. Combining the measures

We construct a panel data set for each of these three measures of the three dimensions of
earnings opacity, and combine them to create an aggregate Earnings Opacity score.
Specifically, we rank each country-month into deciles for each measure, and take the average
590 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593

of these three rankings for each country-month to create the index. The descriptive statistic
reported in Table 1 is the average score across all available months for each country.

Appendix B. International asset pricing model

In computing the cost of capital using average realized returns, we adopt a simplified
version of the international asset pricing model in Bekaert and Harvey (1995). This
appendix describes the model and our estimation procedure.
The basic model we use is expressed as follows:
     
ri;t  rf ;t ¼ a0 þ /i;t kcov hi;w;t þ 1 þ /i;t kvar hi;t þ ei;t ðB1Þ
where

r i,t is the dollar monthly return of the stock market index of country i at time t,
r f,t is the monthly return of the 1 month U.S. T-Bill at time t,
a 0 is a constant that would be estimated,
/ i,t is a measure of the level of integration of country i at time t, 0 b / i,t b 1,
k cov is the price of the covariance risk that would be estimated,
h i,w,t is the conditional covariance of the monthly return of the stock market index of
country i with the monthly return of the world index at time t,
k var is the price of own country variance risk that would be estimated (which we are
restricting to be the same across all countries),
h i,t is the conditional variance of the monthly return of the stock market index of
country i at time t, and
e i,t is the residual error term.

The independent variables in model (B1)–conditional covariance h i,w,t and conditional

variance h i,t –are separately estimated pair-wise for each country i and world pair from the
multivariate ARCH model specified below:

ri;t ¼ c1 þ ei;t ;

rw;t ¼ c2 þ ew;t ;
 
1 2 1 1
hi;t ¼ b1 þ a1 ei;t1 þ e2i;t2 þ e2i;t3 ;
2 3 6
 
1 2 1 1
hw;t ¼ b2 þ a2 ew;t1 þ e2w;t2 þ e2w;t3 ;
2 3 6
 
1 1 1
hi;w;t ¼ b3 þ a3 ei;t1 ew;t1 þ ei;t2 ew;t1 þ ei;t3 ew;t1 ;
2 3 6
   
0 h hi;w;t
ei;t1 ; ew;t1 fN ; i;t ðB2Þ
0 hi;w;t hw;t
 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593 591

where

r w,t is the dollar monthly return of the stock market index of the world at time t,
e i,tj is the innovation in monthly return of the stock market index of country i at time
t  j, j {0,1,2,3},
e w,tj is the innovation in monthly return of the stock market index of the world at time
t  j, j {0,1,2,3}, and
h w,t is the conditional variance of the monthly return of the stock market index of the
world at time t.

Model (B2) was first introduced by Bollerslev et al. (1988). As in Engle, Lilien, and
Robins (1987), the weights of the lagged residual vectors are taken to be 1 / 2, 1 / 3, and 1 /
6, respectively. The constants a 2, b 2, and c 2 are constrained to be identical for all country–
world pairs. Maximum likelihood is used to estimate model (B2).
The other independent variable in Model 1–/ i,t –measures the level of integration of
country i at time t. We define it as follows:
!!
exportsi;t þ importsi;t
exp a1
gdpi;t
/i;t ¼ !! ðB3Þ
exportsi;t þ importsi;t
1 þ exp a1
gdpi;t
The definition of / i,t in (B3) implies that it is a function of the ratio of the sum of exports
and imports to gross domestic product. It is designed to take on values between 0 and 1.
When its value is 0, the country is not integrated with world equity markets, and its equity is
exposed only to local risk (own variance). When its value is 1, the country is fully integrated
with world equity markets, and its equity is exposed only to global risk (covariance with
world factor). Bekaert and Harvey (1997) find that increases in this ratio are empirically
associated with increased importance of the world factor relative to local risk factors.
We use a two-step procedure (first remove the effect of risk, and then test the effect on
residuals) instead of using a one-step procedure (include all independent variables in
model (B1) directly). We do so because of technical convergence problems in the one-step
non-linear estimation procedure. If the capital market governance variables have no
incremental effect on the cost of equity, then those variables will be orthogonal to the
residuals from the model in (B1). We control for other influences on this residual. The
advantage of using a well-specified asset pricing factor model like (B1) to measure cost of
equity is that we explicitly account for risk. This comes at a price. Recall that all the
independent variables in model (B1) are estimates from other models. This introduces
estimation error, which reduces power and may introduce bias.

References

Ahearne, A., Griever, W., Warnock, F., 2001. Information costs and home bias: an analysis of U.S. holdings of
foreign equities. FRB International Finance Discussion Paper No. 691.
592 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593

Ahmed, A., Billings, B., Morton, R., Stanford-Harris, M., 2002. The role of accounting conservatism in
mitigating bondholder–shareholder conflicts over dividend policy and in reducing debt costs. The Accounting
Review 77, 867 – 890.
Amihud, Y., 2002. Illiquidity and stock returns: cross-section and time-series effects. Journal of Financial Markets
5, 31 – 56.
Amihud, Y., Mendelson, H., 1986. Asset pricing and the bid–ask spread. Journal of Financial Economics 15,
223 – 249.
Bainbridge, S., 2000. Insider trading. In: Encyclopedia of Law and Economics III. Edward Elgar Publishing,
Cheltenham, U.K., pp. 772 – 812.
Ball, R., Kothari, S.P., Robin, A., 2000. The effect of international institutional factors on properties of accounting
earnings. Journal of Accounting and Economics 29, 1 – 51.
Barth, M.E., Elliott, J.A., Finn, M.W., 1999. Market rewards associated with patterns of increasing earnings.
Journal of Accounting Research 37, 387 – 413 (Autumn).
Bekaert, G., Harvey, C., 1995. Time varying world market integration. Journal of Finance 50, 403 – 444.
Bekaert, G., Harvey, C., 1997. Emerging equity market volatility. Journal of Financial Economics 43, 29 – 77.
Bekaert, G., Harvey, C., 2000. Foreign speculators and emerging equity markets. Journal of Finance 55,
565 – 613.
Bhattacharya, U., Daouk, H., 2002. The world price of insider trading. Journal of Finance 57, 75 – 108.
Bhattacharya, U., Groznik, P., 2003. Melting pot or salad bowl: some evidence from U.S. investments abroad.
EFA 2003 Annual Conference Paper No. 650.
Bhattacharya, U., Daouk, H., Jorgenson, B., Kehr, C., 2000. When an event is not an event: the curious case of an
emerging market. Journal of Financial Economics 55, 69 – 101.
Bhattacharya, U., Daouk, H., Welker, M., 2003. The world price of earnings opacity. Accounting Review 78,
641 – 678.
Black, B., Jang, H., Kim, W., 2006-this issue. Predicting firms’ corporate governance choices: evidence from
Korea. Journal of Corporate Finance 12, 660–691 (http://papers.ssrn.com/abstract=428662).
Bollerslev, T., Engle, R., Wooldrige, J., 1988. A capital asset pricing model with time-varying covariances.
Journal of Political Economy 96, 116 – 131.
Brennan, M., Subrahmanyam, A., 1996. Market microstructure and asset pricing: on the compensation for
illiquidity in stock returns. Journal of Financial Economics 41, 441 – 464.
Bris, A., Goetzmann, W., Zhu, N., 2004. Efficiency and the bear: short sales and markets around the world.
Working paper, Yale University.
Burgstahler, D., Dichev, I., 1997. Earnings management to avoid earnings decreases and losses. Journal of
Accounting and Economics 24, 99 – 126.
Bushman, R., Smith, A., 2001. Financial accounting information and corporate governance. Journal of
Accounting and Economics 32, 237 – 333.
Charoenrook, A., Daouk, H., 2003. Market-wide short-selling restrictions. Working paper, Vanderbilt University
and Cornell University.
Dahlquist, M., Pinkowitz, L., Stulz, R., Williamson, R., 2002. Corporate governance, investor protection, and the
home bias. Journal of Financial and Quantitative Analysis 38, 87 – 110.
Dechow, P.M., Sloan, R., Sweeney, A., 1995. Detecting earnings management. Accounting Review 70, 193 – 226.
Degeorge, F., Patel, J., Zeckhauser, R., 1999. Earnings management to exceed thresholds. Journal of Business 72,
1 – 33.
Dumas, B., Solnik, B., 1995. The world price of foreign exchange risk. Journal of Finance 50, 445 – 479.
Enders, W., 1996. RATS Handbook for Econometric Time Series. John Wiley and Sons, New York, NY.
Engle, R., Lilien, D., Robins, R., 1987. Estimating time-varying risk premia in the term structure: The ARCH-M
model. Econometrica 55, 391 – 408.
Ferson, W., Harvey, C., 1993. The risk and predictability of international equity returns. Review of Financial
Studies 6, 527 – 566.
Giannetti, M., Koskinen, Y., 2004. Investor protection and the demand for equity, SSE/EFI Working Paper Series
in Economics and Finance No. 526.
Givoly, D., Hayn, C., 2000. The changing time-series properties of earnings, cash flows and accruals: has
financial reporting become more conservative? Journal of Accounting and Economics 29, 287 – 320.
 H. Daouk et al. / Journal of Corporate Finance 12 (2006) 560–593 593

Glosten, L., Milgrom, P., 1985. Bid, ask and transaction prices in a specialist model with heterogeneously
informed traders. Journal of Financial Economics 14, 71 – 100.
Hail, L., Leuz, C., 2003. International differences in the cost of equity capital: do legal institutions and securities
regulation matter? Working paper, University of Zurich and University of Pennsylvania. November.
Hasbrouck, J., 2003. Trading costs and returns for US equities: the evidence from daily data. NYU Stern School
Department of Finance Working Paper.
Healy, P., 1985. The effect of bonus schemes on accounting decisions. Journal of Accounting and Economics 7,
85 – 107.
Henry, P., 2000. Stock market liberalization, economic reform, and emerging market equity prices. Journal of
Finance 55, 529 – 564.
Jacoby, G., Fowler, D., Gottesman, A., 2000. The capital asset pricing model and the liquidity effect: a theoretical
approach. Journal of Financial Markets 3, 69 – 81.
Jones, J.J., 1991. Earnings management during import relief investigations. Journal of Accounting Research 29,
193 – 228 (Autumn).
Klapper, L., Love, I., 2004. Corporate governance, investor protection and performance in emerging markets.
Journal of Corporate Finance 10, 703 – 728.
Kyle, A., 1985. Continuous auctions and insider trading. Econometrica 53, 1315 – 1335.
La Porta, R., Lopez-de-silanes, F., Shleifer, A., Vishny, R.W., 1998. Law and finance. Journal of Political
Economy 106, 1113 – 1155.
La Porta, R., Lopez-de-silanes, F., Shleifer, A., 2002. What works in securities laws? Working paper, Harvard
University and Yale University. October.
Lee, C.M.C., Ng, D., 2004. Corruption and international valuation: does virtue pay? Working paper, Cornell
University.
Lee, C.M.C., Ng, D., Swaminathan, B., 2003. The cross-section of international cost of capital. Working paper,
Cornell University.
Leuz, C., Nanda, D., Wysocki, P., 2002. Earnings management and institutional factors: an international
comparison. Journal of Financial Economics 69, 505 – 527.
Lombardo, D., Pagano, M., 2000. Law and equity markets: a simple model. Working paper, Stanford Law School
working paper no. 194. March.
Loughran, T., Ritter, J., Rydqvist, K., 1994. Initial public offerings: international insights. Pacific Basin Finance
Journal 2, 165 – 199.
Manne, H., 1966. Insider Trading and the Stock Market. Free Press.
Miller, E., 1977. Risk, uncertainty and divergence of opinion. Journal of Finance 32, 1151 – 1168.
Morck, R., Yeung, B., Yu, W., 2000. The information content of stock markets: why do emerging markets have
synchronous stock price movements? Journal of Financial Economics 58, 215 – 260.
Rangan, S., 1998. Earnings management and the performance of seasoned equity offerings. Journal of Financial
Economics 50, 101 – 122.
Ritter, J.R., Welch, I., 2002. A review of IPO activity, pricing, and allocations. Journal of Finance 57,
1795 – 1828.
Shivakumar, L., 2000. Do firms mislead investors by overstating earnings before seasoned equity offerings?
Journal of Accounting and Economics 29, 339 – 371.
Sims, C., 1980. Macroeconomics and reality. Econometrica 48, 1 – 49.
Stulz, R., 1999a. Globalization of equity markets and the cost of capital. Working paper, National Bureau of
Economic Research.
Stulz, R., 1999b. Globalization, corporate finance, and the cost of capital. Journal of Applied Corporate Finance,
8 – 25 (Fall).
Teoh, S.H., Welch, I., Wong, T.J., 1998. Earnings management and the underperformance of seasoned equity
offerings. Journal of Financial Economics 50, 63 – 99.