La Porta Et Al. (1998) PDF
La Porta Et Al. (1998) PDF
La Porta Et Al. (1998) PDF
Robert W. Vishny
University of Chicago
1113
This content downloaded from 148.088.244.181 on March 21, 2017 06:23:33 AM
All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
1114 journal of political economy
owner to receive dividends. Recent financial research has shown that
this is far from the whole story and that the defining feature of vari-
ous securities is the rights that they bring to their owners (Hart
1995). Thus shares typically give their owners the right to vote for
directors of companies, whereas debt entitles creditors to the power,
for example, to repossess collateral when the company fails to make
promised payments.
The rights attached to securities become critical when managers
of companies act in their own interest. These rights give investors
the power to extract from managers the returns on their investment.
Shareholders receive dividends because they can vote out the direc-
tors who do not pay them, and creditors are paid because they have
the power to repossess collateral. Without these rights, investors
would not be able to get paid, and therefore firms would find it
harder to raise external finance.
But the view that securities are inherently characterized by some
intrinsic rights is incomplete as well. It ignores the fact that these
rights depend on the legal rules of the jurisdictions in which securi-
ties are issued. Does being a shareholder in France give an investor
the same privileges as being a shareholder in the United States, In-
dia, or Mexico? Would a secured creditor in Germany fare as well
when the borrower defaults as one in Sri Lanka or Italy, with the
value of the collateral assumed the same in all cases? Law and the
quality of its enforcement are potentially important determinants
of what rights security holders have and how well these rights are
protected. Since the protection investors receive determines their
readiness to finance firms, corporate finance may critically turn on
these legal rules and their enforcement.
The differences in legal protections of investors might help ex-
plain why firms are financed and owned so differently in different
countries. Why do Italian companies rarely go public (Pagano, Pa-
netta, and Zingales 1998)? Why does Germany have such a small
stock market but also maintain very large and powerful banks (Ed-
wards and Fischer 1994)? Why is the voting premiumthe price of
shares with high voting rights relative to that of shares with low vot-
ing rightssmall in Sweden and the United States, and much larger
in Italy and Israel (Levy 1983; Rydquist 1987; Zingales 1994, 1995)?
Indeed, why were Russian stocks nearly worthless immediately after
privatizationby some estimates 100 times cheaper than Western
stocks backed by comparable assetsand why did Russian compa-
nies have virtually no access to external finance (Boycko, Shleifer,
and Vishny 1993)? Why is ownership of large American and British
companies so widely dispersed (Berle and Means 1932)? The con-
Legal Families
Comparative legal scholars agree that, even though no two nations
laws are exactly alike, some national legal systems are sufficiently
similar in certain critical respects to permit classification of national
legal systems into major families of law. Although there is no una-
nimity among legal scholars on how to define legal families,
1
The religious traditions, such as Jewish law, Canon law, Hindu law, and Muslim
law, appear to be less relevant in matters of investor protection. Thus the Arabian
countries unquestionably belong to Islamic law as far as family and inheritance law
is concerned, just as India belongs to Hindu law, but economic law of these countries
(including commercial law and the law of contract and tort) is heavily impressed
by the legal thinking of the colonial and mandatory powersthe Common Law in
the case of India, French law in the case of most of the Arab States (Zweigert and
Kotz 1987, p. 66). We focus on the principal secular legal traditions in this study.
2
The European Community is currently attempting to harmonize West European
laws, including those pertaining to corporate governance, by issuing directives (An-
denas and Kenyon-Slade 1993; Werlauff 1993). Several countries have changed parts
of their laws to adhere to E.C. directives. However, in most instances, the directives
are not mandatory, and the countries are given some time to change their laws.
Moreover, the E.C. directives accommodate a great deal of diversity among coun-
tries. As of 199394the point in time for which we examine the legal rules of the
countries in our sampleE.C. harmonization has not generally affected the legal
rules that we focus on. The one area in which the E.C. impact has been large, namely
mergers and acquisitions, is not an area that we examine in this paper (see below).
3
Several readers have pointed to the U.S. state antitakeover laws as evidence of
an antiminority shareholder position in the U.S. legal system that our data do not
capture. Even with all these antitakeover laws, the United States and the United
Kingdom still have by far the most takeovers of any country in the world, so their
laws are evidently not nearly as antitakeover as those elsewhere.
1122
before meeting quire that shareholders deposit their shares prior to a general shareholders cial code
meeting, thus preventing them from selling those shares for a number of
days, and zero otherwise
Cumulative voting or Equals one if the company law or commercial code allows shareholders to Company law or commer-
proportional repre- cast all their votes for one candidate standing for election to the board of cial code
sentation directors (cumulative voting) or if the company law or commercial code
allows a mechanism of proportional representation in the board by which
minority interests may name a proportional number of directors to the
board, and zero otherwise
Oppressed minorities Equals one if the company law or commercial code grants minority sharehold- Company law or commer-
mechanism ers either a judicial venue to challenge the decisions of management or of cial code
the assembly or the right to step out of the company by requiring the com-
pany to purchase their shares when they object to certain fundamental
changes, such as mergers, asset dispositions, and changes in the articles of
incorporation. The variable equals zero otherwise. Minority shareholders
All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
Preemptive rights Equals one when the company law or commercial code grants shareholders Company law or commer-
the first opportunity to buy new issues of stock, and this right can be cial code
waived only by a shareholders vote; equals zero otherwise
Percentage of share capi- The minimum percentage of ownership of share capital that entitles a share- Company law or commer-
tal to call an extraordi- holder to call for an extraordinary shareholders meeting; it ranges from 1 cial code
nary shareholders to 33 percent
meeting
Antidirector rights An index aggregating the shareholder rights we labeled as antidirector Company law or commer-
rights. The index is formed by adding 1 when (1) the country allows cial code
shareholders to mail their proxy vote to the firm, (2) shareholders are not
required to deposit their shares prior to the general shareholders meeting,
(3) cumulative voting or proportional representation of minorities in the
board of directors is allowed, (4) an oppressed minorities mechanism is in
place, (5) the minimum percentage of share capital that entitles a share-
holder to call for an extraordinary shareholders meeting is less than or
equal to 10 percent (the sample median), or (6) shareholders have pre-
emptive rights that can be waived only by a shareholders vote. The index
ranges from zero to six
Mandatory dividend Equals the percentage of net income that the company law or commercial Company law or commer-
1123
code requires firms to distribute as dividends among ordinary stockholders. cial code
It takes a value of zero for countries without such a restriction
Restrictions for going Equals one if the reorganization procedure imposes restrictions, such as credi- Bankruptcy and reorganiza-
into reorganization tors consent, to file for reorganization; equals zero if there are no such re- tion laws
strictions
No automatic stay on Equals one if the reorganization procedure does not impose an automatic Bankruptcy and reorganiza-
secured assets stay on the assets of the firm on filing the reorganization petition. Auto- tion laws
matic stay prevents secured creditors from gaining possession of their secu-
rity. It equals zero if such a restriction does exist in the law
Secured creditors first Equals one if secured creditors are ranked first in the distribution of the pro- Bankruptcy and reorganiza-
ceeds that result from the disposition of the assets of a bankrupt firm. tion laws
Equals zero if nonsecured creditors, such as the government and workers,
are given absolute priority
Management does not Equals one when an official appointed by the court, or by the creditors, is re- Bankruptcy and reorganiza-
stay sponsible for the operation of the business during reorganization. Equiva- tion laws
lently, this variable equals one if the debtor does not keep the administra-
tion of its property pending the resolution of the reorganization process.
All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
TABLE 1 (Continued )
Creditor rights An index aggregating different creditor rights. The index is formed by add- Bankruptcy and reorganiza-
ing 1 when (1) the country imposes restrictions, such as creditors consent tion laws
or minimum dividends to file for reorganization; (2) secured creditors are
able to gain possession of their security once the reorganization petition
has been approved (no automatic stay); (3) secured creditors are ranked
first in the distribution of the proceeds that result from the disposition of
the assets of a bankrupt firm; and (4) the debtor does not retain the ad-
ministration of its property pending the resolution of the reorganization.
The index ranges from zero to four
Legal reserve The minimum percentage of total share capital mandated by corporate law to Company law or commer-
avoid the dissolution of an existing firm. It takes a value of zero for coun- cial code
tries without such a restriction
Efficiency of judicial Assessment of the efficiency and integrity of the legal environment as it af- Business International
system fects business, particularly foreign firms produced by the country risk rat- Corp.
1124
ing agency Business International Corp. It may be taken to represent in-
vestors assessments of conditions in the country in question. Average
between 1980 and 1983. Scale from zero to 10; with lower scores, lower ef-
ficiency levels
Rule of law Assessment of the law and order tradition in the country produced by the International Country Risk
country risk rating agency International Country Risk (ICR). Average of the guide
months of April and October of the monthly index between 1982 and
1995. Scale from zero to 10, with lower scores for less tradition for law and
order (we changed the scale from its original range going from zero to six)
Corruption ICRs assessment of the corruption in government. Lower scores indicate that International Country Risk
high government officials are likely to demand special payments and il- guide
legal payments are generally expected throughout lower levels of govern-
ment in the form of bribes connected with import and export licenses,
exchange controls, tax assessment, policy protection, or loans. Average of
the months of April and October of the monthly index between 1982 and
All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
Risk of expropriation ICRs assessment of the risk of outright confiscation or forced nationaliza- International Country Risk
tion. Average of the months of April and October of the monthly index guide
between 1982 and 1995. Scale from zero to 10, with lower scores for higher
risks
Repudiation of contracts ICRs assessment of the risk of a modification in a contract taking the form International Country Risk
by government of a repudiation, postponement, or scaling down due to budget cut- guide
backs, indigenization pressure, a change in government, or a change in gov-
ernment economic and social priorities. Average of the months of April
and October of the monthly index between 1982 and 1995. Scale from
zero to 10, with lower scores for higher risks
Accounting standards Index created by examining and rating companies 1990 annual reports on International accounting
their inclusion or omission of 90 items. These items fall into seven catego- and auditing trends,
ries (general information, income statements, balance sheets, funds flow Center for International
statement, accounting standards, stock data, and special items). A mini- Financial Analysis and
mum of three companies in each country were studied. The companies rep- Research
resent a cross section of various industry groups; industrial companies rep-
resented 70 percent, and financial companies represented the remaining
30 percent
1125
Ownership, 10 largest The average percentage of common shares owned by the three largest share- Moodys International,
private firms holders in the 10 largest nonfinancial, privately owned domestic firms in a CIFAR, EXTEL,
given country. A firm is considered privately owned if the state is not a WorldScope, 20-Fs,
known shareholder in it Price-Waterhouse, and
various country sources
GNP and GNP per Gross national product and gross national product per capita in constant dol- World Bank and Interna-
capita lars of 1994 tional Monetary Fund
Gini coefficient Gini coefficient for income inequality in each country. When the 1990 coeffi- Deininger and Squire
cient is not available, we use the most recent available (1996); World Bank
(1993a, 1993b)
4
We made two significant changes: we redefined the cumulative voting variable
to also cover the right of minority shareholders for proportional representation,
and we added a variable on preemptive rights of minority shareholders to buy new
issues of stock (see below). In this and the following sections, all dummies have
been defined so that 1 means more protective.
5
One of the E.C. directives recommends the adoption of one-shareone-vote rules
throughout the Community. It does not appear that this directive is being incorpo-
rated into national laws too rapidly.
6
For the United States, our reliance on Delaware presents a problem since the
state leaves up to corporations the percentage of shares needed to call an extraordi-
nary shareholder meeting. We use 10 percent for the United States because the
majority of U.S. states (27) use this number.
Percentage
of Share
Capital
Shares Not Cumulative to Call an
Proxy by Blocked Voting/ Preemptive Extraordinary
One Share Mail before Proportional Oppressed Right to Shareholder Antidirector Mandatory
Country One Vote Allowed Meeting Representation Minority New Issues Meeting Rights Dividend
A. Shareholder Rights (1 Investor Protection Is in the Law)
Australia 0 1 1 0 1 0 .05 a 4 .00
Canada 0 1 1 1 1 0 .05 5 .00
Hong Kong 0 1 1 0 1 1 .10 5 .00
India 0 0 1 1 1 1 .10 5 .00
Ireland 0 0 1 0 1 1 .10 4 .00
Israel 0 0 1 0 1 0 .10 3 .00
Kenya 0 0 1 0 1 0 .10 3 .00
Malaysia 1 0 1 0 1 1 .10 4 .00
1130
New Zealand 0 1 1 0 1 0 .05 4 .00
Nigeria 0 0 1 0 1 0 .10 3 .00
Pakistan 1 0 1 1 1 1 .10 5 .00
Singapore 1 0 1 0 1 1 .10 4 .00
South Africa 0 1 1 0 1 1 .05 5 .00
Sri Lanka 0 0 1 0 1 0 .10 3 .00
Thailand 0 0 1 1 0 0 .20 b 2 .00
United Kingdom 0 1 1 0 1 1 .10 5 .00
United States 0 1 1 1 1 0 .10 5 .00
Zimbabwe 0 0 1 0 1 0 .05 3 .00
English-origin average .17 .39 1.00 .28 .94 .44 .09 4.00 .00
Argentina 0 0 0 1 1 1 .05 4 .00
Belgium 0 0 0 0 0 0 .20 0 .00
Brazil 1 0 1 0 1 0 .05 3 .50
Chile 1 0 1 1 1 1 .10 5 .30
Colombia 0 0 1 1 0 1 .25 3 .50
All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
Indonesia 0 0 1 0 0 0 .10 2 .00
Italy 0 0 0 0 0 1 .20 1 .00
Jordan 1 0 1 0 0 0 .25 1 .00
Mexico 0 0 0 0 0 1 .33 1 .00
Netherlands 0 0 0 0 0 1 .10 2 .00
Peru 1 0 1 1 0 1 .20 3 .00
Philippines 0 0 1 1 1 0 open 3 .00
Portugal 0 0 1 0 0 1 .05 3 .00
Spain 0 0 0 1 1 1 .05 4 .00
Turkey 0 0 1 0 0 0 .10 2 .00
Uruguay 1 0 0 0 1 1 .20 2 .20
Venezuela 0 0 1 0 0 0 .20 1 .00
French-origin average .29 .05 .57 .29 .29 .62 .15 2.33 .11
Austria 0 0 0 0 0 1 .05 2 .00
Germany 0 0 0 0 0 0 .05 1 .00
Japan 1 0 1 1 1 0 .03 4 .00
South Korea 1 0 0 0 1 0 .05 2 .00
Switzerland 0 0 0 0 0 1 .10 2 .00
Taiwan 0 0 0 1 1 0 .03 3 .00
German-origin average .33 .00 .17 .33 .50 .33 .05 2.33 .00
Denmark 0 0 1 0 0 0 .10 2 .00
Finland 0 0 1 0 0 1 .10 3 .00
1131
Norway 0 1 1 0 0 1 .10 4 .00
Sweden 0 0 1 0 0 1 .10 b 3 .00
Scandinavian-origin average .00 .25 1.00 .00 .00 .75 .10 3.00 .00
Sample average .22 .18 .71 .27 .53 .53 .11 3.00 .05
All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
1132 journal of political economy
dic countries) incidence of laws protecting oppressed minorities,
and the highest (15 percent) percentage of share capital needed to
call an extraordinary shareholders meeting. The aggregate antidi-
rector rights score is the lowest (2.33) for the French-civil-law coun-
tries. The difference in this score between French civil law and com-
mon law is large and statistically significant. It is interesting to note
that France itself, except for allowing proxy voting by mail and hav-
ing a preemptive right to new share issues, does not have strong legal
protections of shareholders. These results suggest that shareholders
in the two most widely spread legal regimescommon law and
French civil lawoperate in very different legal environments.
The German-civil-law countries are not particularly protective of
shareholders either. They have a relatively high frequency of one-
shareone-vote rules (because of East Asia), require few votes to call
an extraordinary meeting, and offer preemptive rights in a third of
the cases. But they usually block shares before shareholder meetings,
never allow voting by mail, and have oppressed minority mecha-
nisms in only half of the countries. The average antidirector score
for this family is 2.33, exactly the same as that for the French family.
In Scandinavia, no country has oppressed minority protections, a
one-shareone-vote restriction, or a cumulative voting/propor-
tional representation mechanism, and only Norway allows voting by
mail. At the same time, no country blocks shares before a share-
holder meeting, and three out of four give shareholders preemptive
rights. The average Scandinavian antidirector rights score is 3.
The one remedial measure in table 2, namely mandatory divi-
dend, shows that mandatory dividends are used only in French-civil-
law countries. This result is broadly consistent with the rest of our
evidence and suggests that mandatory dividends are indeed a reme-
dial legal protection for shareholders who have relatively few other
legal rights.
The results in panel B of table 2 suggest that the differences in
the various measures of shareholder rights between different legal
families are often significant and almost always significant when
common- and civil-law families are compared. One further question
is whether the difference in scores by legal origin just reflects differ-
ences in per capita income levels. To address this question, table 3
divides all countries into the bottom 25 percent, middle 50 percent,
and top 25 percent by gross national product per capita. The results
show, in particular, that antidirector rights scores are independent
of per capita income, rejecting the notion that legal rules that are
more protective of investors are a reflection of higher per capita
income.
In sum, common-law countries have the relatively strongest, and
Legal Reserve as
Countries Sorted by GNP per One Share Antidirector Mandatory Creditor a Percentage
GNP per Capita Capita One Vote Rights Dividend Rights of Capital
A. Means
Bottom 25% 705 .17 2.92 .08 3.18 .15
Mid 50% 9,465 .32 3.16 .05 2.13 .16
Highest 25% 25,130 .08 2.75 .00 1.83 .15
Total average 11,156 .22 3.00 .05 2.30 .15
B. Tests of Means (t-Statistics)
Bottom 25% vs. mid 50% 4.59* .97 .56 .54 2.08** .20
Bottom 25% vs. top 25% 18.63* .60 .30 1.48 2.49** .05
Mid 50% vs. top 25% 7.44* 1.58 .85 2.02*** .69 .16
Legal Reserve
Secured Restrictions Management Required as a
No Automatic Creditors for Going into Does Not Stay in Creditor Percentage
Country Stay on Assets First Paid Reorganization Reorganization Rights of Capital
A. Creditor Rights (1 Creditor Protection Is the Law)
Australia 0 1 0 0 1 .00
Canada 0 1 0 0 1 .00
Hong Kong 1 1 1 1 4 .00
India 1 1 1 1 4 .00
Ireland 0 1 0 0 1 .00
Israel 1 1 1 1 4 .00
Kenya 1 1 1 1 4 .00
Malaysia 1 1 1 1 4 .00
New Zealand 1 0 1 1 3 .00
Nigeria 1 1 1 1 4 .00
1136
Pakistan 1 1 1 1 4 .00
Singapore 1 1 1 1 4 .00
South Africa 0 1 1 1 3 .00
Sri Lanka 1 0 1 1 3 .00
Thailand 1 1 0 1 3 .10
United Kingdom 1 1 1 1 4 .00
United States 0 1 0 0 1 .00
Zimbabwe 1 1 1 1 4 .00
English-origin average .72 .89 .72 .78 3.11 .01
Argentina 0 1 0 0 1 .20
Belgium 1 1 0 0 2 .10
Brazil 0 0 1 0 1 .20
Chile 0 1 1 0 2 .20
Colombia 0 0 0 0 0 .50
Ecuador 1 1 1 1 4 .50
All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
Italy 0 1 1 0 2 .20
Jordan na na na na na .25
Mexico 0 0 0 0 0 .20
Netherlands 0 1 1 0 2 .00
Peru 0 0 0 0 0 .20
Philippines 0 0 0 0 0 .00
Portugal 0 1 0 0 1 .20
Spain 1 1 0 0 2 .20
Turkey 0 1 1 0 2 .20
Uruguay 0 1 0 1 2 .20
Venezuela na 1 na na na .10
French-origin average .26 .65 .42 .26 1.58 .21
Austria 1 1 1 0 3 .10
Germany 1 1 1 0 3 .10
Japan 0 1 0 1 2 .25
South Korea 1 1 0 1 3 .50
Switzerland 0 1 0 0 1 .50
Taiwan 1 1 0 0 2 1.00
German-origin average .67 1.00 .33 .33 2.33 .41
1137
Denmark 1 1 1 0 3 .25
Finland 0 1 0 0 1 .00
Norway 0 1 1 0 2 .20
Sweden 0 1 1 0 2 .20
Scandinavian-origin average .25 1.00 .75 .00 2.00 .16
Sample average .49 .81 .55 .45 2.30 .15
Common vs. civil law 2.65* 1.04 1.86*** 4.13* 3.61* 4.82*
English vs. French origin 3.06* 1.75** 1.89*** 3.55* 3.61* 5.75*
English vs. German origin .25 1.46 1.74*** 2.10** 1.43 5.21*
English vs. Scandinavian origin 1.83*** 1.46 .11 7.71* 1.71*** 5.90*
French vs. German origin 1.85*** 3.20* .37 .32 1.29 2.14**
French vs. Scandinavian origin .05 3.20* 1.18 2.54** .60 .59
German vs. Scandinavian origin 1.27 .00 1.26 1.58 .63 1.37
All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
1138 journal of political economy
pay secured creditors first, over half restrict the managers right to
seek protection from creditors unilaterally, and 45 percent remove
management in reorganization proceedings.
As in table 2, we see that, for many creditor rights, the legal origin
matters. Common-law countries offer creditors stronger legal protec-
tions against managers. They have the highest (72 percent) inci-
dence of no automatic stay on assets; with two exceptions, they guar-
antee that secured creditors are paid first (the German-civil-law and
Scandinavian families have no exceptions); they frequently (72 per-
cent, behind only Scandinavia) preclude managers from unilaterally
seeking court protection from creditors; and they have far and away
the highest (78 percent) incidence of removing managers in reorga-
nization proceedings. The United States is actually one of the most
anticreditor common-law countries: it permits automatic stay on
assets, allows unimpeded petition for reorganization, and lets man-
agers keep their jobs in reorganization. The average aggregate credi-
tor rights score for common-law countries is 3.11by far the highest
among the four familiesbut this score is only 1 for the United
States.
The French-civil-law countries offer creditors the weakest protec-
tions. Few of them (26 percent, tied with Scandinavia) have no auto-
matic stay on assets; relatively few (65 percent) assure that secured
creditors are paid first; few (42 percentstill more than German-
civil-law countries) place restrictions on managers seeking court pro-
tection from creditors; and relatively few (26 percent) remove man-
agers in reorganization proceedings. The average aggregate creditor
rights score for the French-civil-law countries is 1.58, or roughly half
of that for the common-law family.
On some measures, countries in the German-civil-law family are
strongly pro-creditor. For instance, 67 percent of them have no auto-
matic stay, and secured creditors in all of them are paid first. On the
other hand, relatively few of these countries (33 percent) prevent
managers from getting protection from creditors unilaterally, and
most (67 percent) allow managers to stay in reorganization. One
view of this evidence is that the German-civil-law countries are very
responsive to secured creditors by not allowing automatic stay and
by letting them pull collateral. As a consequence of making liquida-
tion easy, these countries rely less on reorganization of defaulting
firms, and hence being soft on such firms by letting managers stay
may not be a big problem. The overall average creditor rights score
of 2.33 for the German family may therefore understate the extent
to which secured creditors are protected.
Finally, Scandinavia has an overall average score of 2.00, which is
7
We have also examined whether investor rights are a consequence of geography
by dividing the world into Australia, Europe, Africa, Asia, and America. They do not
appear to be.
V. Enforcement
In principle, a strong system of legal enforcement could substitute
for weak rules since active and well-functioning courts can step in
and rescue investors abused by the management. To address these
issues, we examine proxies for the quality of enforcement of these
rights, namely estimates of law and order in different countries
compiled by private credit risk agencies for the use of foreign invest-
ors interested in doing business in the respective countries. We use
five of these measures: efficiency of the judicial system, rule of law,
corruption, risk of expropriationmeaning outright confiscation
or forced nationalizationby the government, and likelihood of
contract repudiation by the government. The first two of these mea-
sures obviously pertain to law enforcement proper; the last three
deal more generally with the governments stance toward business.
Some of these measures have been previously shown to affect na-
tional growth rates (Knack and Keefer 1995).
In addition, we use an estimate of the quality of a countrys ac-
counting standards. Accounting plays a potentially crucial role in
corporate governance. For investors to know anything about the
companies they invest in, basic accounting standards are needed to
render company disclosures interpretable. Even more important,
contracts between managers and investors typically rely on the veri-
fiability in court of some measures of firms income or assets. If a
bond covenant stipulates immediate repayment when income falls
below a certain level, this level of income must be verifiable for the
bond contract to be enforceable in court even in principle. Account-
ing standards might then be necessary for financial contracting, es-
pecially if investor rights are weak (Hay et al. 1996). The measure
of accounting standards we use, like the rule of law measures, is a
privately constructed index based on examination of company re-
ports from different countries. Unfortunately, it is available for only
44 countries, 41 of which are in our sample.8
8
The measure of accounting standards we use was published in 1991. At around
the same time, European countries began to harmonize their accounting standards
under pressure from the European Community. Over time, accounting standards
may converge in Europe. However, for the purposes of our analysis of country differ-
ences and of determinants of ownership, historical differences in the quality of stan-
dards are obviously more important than the future convergence.
9
We have also estimated these equations using Tobits, with very similar results.
One difference is that the Tobit procedure does not produce a standard error on
the Scandinavian dummy because all Scandinavian countries have the same values
for some of the variables.
Enforcement Variables
Accounting:
Risk of Rating on GNP
Efficiency of Rule of Risk of Contract Accounting per Capita
Country Judicial System Law Corruption Expropriation Repudiation Standards (U.S. $)
A. Country Scores
1142
Pakistan 5.00 3.03 2.98 5.62 4.87 na 430
Singapore 10.00 8.57 8.22 9.30 8.86 78 19,850
South Africa 6.00 4.42 8.92 6.88 7.27 70 2,980
Sri Lanka 7.00 1.90 5.00 6.05 5.25 na 600
Thailand 3.25 6.25 5.18 7.42 7.57 64 2,110
United Kingdom 10.00 8.57 9.10 9.71 9.63 78 18,060
United States 10.00 10.00 8.63 9.98 9.00 71 24,740
Zimbabwe 7.50 3.68 5.42 5.61 5.04 na 520
English-origin average 8.15 6.46 7.06 7.91 7.41 69.62 9,353
Argentina 6.00 5.35 6.02 5.91 4.91 45 7,220
Belgium 9.50 10.00 8.82 9.63 9.48 61 21,650
Brazil 5.75 6.32 6.32 7.62 6.30 54 2,930
Chile 7.25 7.02 5.30 7.50 6.80 52 3,170
Colombia 7.25 2.08 5.00 6.95 7.02 50 1,400
All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
Indonesia 2.50 3.98 2.15 7.16 6.09 na 740
Italy 6.75 8.33 6.13 9.35 9.17 62 19,840
Jordan 8.66 4.35 5.48 6.07 4.86 na 1,190
Mexico 6.00 5.35 4.77 7.29 6.55 60 3,610
Netherlands 10.00 10.00 10.00 9.98 9.35 64 20,950
Peru 6.75 2.50 4.70 5.54 4.68 38 1,490
Philippines 4.75 2.73 2.92 5.22 4.80 65 850
Portugal 5.50 8.68 7.38 8.90 8.57 36 9,130
Spain 6.25 7.80 7.38 9.52 8.40 64 13,590
Turkey 4.00 5.18 5.18 7.00 5.95 51 2,970
Uruguay 6.50 5.00 5.00 6.58 7.29 31 3,830
Venezuela 6.50 6.37 4.70 6.89 6.30 40 2,840
French-origin average 6.56 6.05 5.84 7.46 6.84 51.17 7,102
Austria 9.50 10.00 8.57 9.69 9.60 54 23,510
Germany 9.00 9.23 8.93 9.90 9.77 62 23,560
Japan 10.00 8.98 8.52 9.67 9.69 65 31,490
South Korea 6.00 5.35 5.30 8.31 8.59 62 7,660
Switzerland 10.00 10.00 10.00 9.98 9.98 68 35,760
Taiwan 6.75 8.52 6.85 9.12 9.16 65 10,425
German-origin average 8.54 8.68 8.03 9.45 9.47 62.67 22,067
1143
Denmark 10.00 10.00 10.00 9.67 9.31 62 26,730
Finland 10.00 10.00 10.00 9.67 9.15 77 19,300
Norway 10.00 10.00 10.00 9.88 9.71 74 25,970
Sweden 10.00 10.00 10.00 9.40 9.58 83 24,740
Scandinavian-origin average 10.00 10.00 10.00 9.66 9.44 74.00 24,185
Sample average 7.67 6.85 6.90 8.05 7.58 60.93 11,156
All use subject to University of Chicago Press Terms and Conditions (http://www.journals.uchicago.edu/t-and-c).
TABLE 6
Ordinary Least Squares Regressions: Cross Section of 49 Countries
Dependent Variable
Repudiation
Efficiency of Risk of of Contracts
Judiciary Rule of Law Corruption Expropriation by Government Accounting Standards
System (N 49) (N 49) (N 49) (N 49) (N 49) (N 41)
Independent
Variable (1) (2) (1) (2) (1) (2) (1) (2) (1) (2) (1) (2)
Log of GNP per .8421* .9763* 1.4761* 1.5541* 1.3088* 1.4020* .9099* .9679* .9951* 1.0976* 4.3348* 5.7747*
capita (.1450) (.1355) (.1584) (.1379) (.1138) (.0993) (.0932) (.0772) (.0832) (.0734) (1.2453) (1.2908)
Civil-law dummy a 1.3774* .3642 1.1388* .3855*** .4111*** 14.331*
(.4235) (.4290) (.3024) (.2132) (.2228) (2.7407)
French origin 1.6609* .5250 1.3236* .5164** .6459** 17.366*
(.4796) (.4563) (.3190) (.2518) (.2520) (2.9445)
German origin 1.0305*** .2715 1.2422* .0009 .3803*** 11.890*
(.6033) (.6312) (.4749) (.2097) (.1946) (2.9104)
Scandinavian .2392 .7174 .4369 .0054 .1300 1.5272
origin (.3550) (.4681) (.3152) (.2242) (.2095) (4.7556)
Intercept 1.2677 .1702 5.6050* 6.2421* 3.6367* 4.3986* .4732 .0018 .7290 1.5671* 31.807* 19.249
(1.3598) (1.2862) (1.3600) (1.2087) (.9881) (.8711) (.8431) (.7181) (.7250) (.6493) (10.844) (11.442)
Adjusted R 2 .5719 .5185 .7744 .7605 .8442 .8056 .8120 .7998 .8465 .8146 .6125 .5131
VI. Ownership
In this section, we explore the hypothesis that companies in coun-
tries with poor investor protection have more concentrated owner-
ship of their shares. There are at least two reasons why ownership
in such countries would be more concentrated. First, large, or even
dominant, shareholders who monitor the managers might need to
own more capital, ceteris paribus, to exercise their control rights
and thus to avoid being expropriated by the managers. This would
be especially true when there are some legal or economic reasons
for large shareholders to own significant cash flow rights as well as
votes. Second, when they are poorly protected, small investors might
be willing to buy corporate shares only at such low prices that make
it unattractive for corporations to issue new shares to the public.
Such low demand for corporate shares by minority investors would
indirectly stimulate ownership concentration. Of course, it is often
efficient to have some ownership concentration in companies since
large shareholders might monitor managers and thus increase the
value of the firm (Shleifer and Vishny 1986). But with poor investor
protection, ownership concentration becomes a substitute for legal
protection, because only large shareholders can hope to receive a
return on their investment.
To evaluate this hypothesis, we have assembled a database of up
to the 10 largest (by market capitalization) nonfinancial (i.e., no
banks or insurance companies), domestic (i.e., no foreign multina-
tionals), totally private (i.e., no government ownership), publicly
traded (i.e., not 100 percent privately held) companies in each coun-
try in our sample. For some countries, including Egypt, India, Nige-
ria, Philippines, and Zimbabwe, we could not find 10 such compa-
nies and settled for at least five.
For each company, we collected data on its three largest share-
holders and computed the combined (cash flow) ownership stake
Note.A firm is considered privately owned if the state is not a known shareholder in it.
* Significant at the 1 percent level.
** Significant at the 5 percent level.
*** Significant at the 10 percent level.
Shareholder and
Independent Variable Basic Regression Creditor Rights
Log of GNP per capita .0077 .0397
(.0097) (.0242)
Log of GNP .0442* .0428*
(.0119) (.0118)
Gini coefficient .0024*** .0027
(.0014) (.0023)
Rule of law .0143
(.0115)
Accounting .0029***
(.0016)
French origin .1296* .0733
(.0261) (.0802)
German origin .0113 .0025
(.0666) (.0728)
Scandinavian origin .0496 .0430
(.0371) (.0473)
Antidirector rights .0315**
(.0150)
One shareone vote .0497
(.0406)
Mandatory dividend .2197***
(.1113)
Creditor rights .0128
(.0171)
Legal reserve required .2237**
(.0766)
Intercept .7785* .8686*
(.1505) (.2952)
Number of observations 45 39
Adjusted R 2 .5582 .7348
Note.Variables are defined in table 1. Robust standard errors are in parentheses.
* Significant at the 1 percent level.
** Significant at the 5 percent level.
*** Significant at the 10 percent level.
VII. Conclusion
In this paper, we have examined laws governing investor protection,
the quality of enforcement of these laws, and ownership concentra-
tion in 49 countries around the world. The analysis suggests three
broad conclusions.
First, laws differ markedly around the world, though in most
places they tend to give investors a rather limited bundle of rights.
In particular, countries whose legal rules originate in the common-
law tradition tend to protect investors considerably more than the
countries whose laws originate in the civil-law, and especially the
French-civil-law, tradition. The German-civil-law and the Scandina-
vian countries take an intermediate stance toward investor protec-
tions. There is no clear evidence that different countries favor differ-
ent types of investors; the evidence rather points to a relatively
stronger stance favoring all investors in common-law countries. This
evidence confirms our basic hypothesis that being a shareholder, or
a creditor, in different legal jurisdictions entitles an investor to very
different bundles of rights. These rights are determined by laws; they
are not inherent in securities themselves.
Second, law enforcement differs a great deal around the world.
German-civil-law and Scandinavian countries have the best quality
of law enforcement. Law enforcement is strong in common-law
countries as well, whereas it is the weakest in the French-civil-law
countries. These rankings also hold for one critical input into law
enforcement in the area of investor protections: the accounting stan-
References
Aghion, Philippe; Hart, Oliver; and Moore, John. The Economics of Bank-
ruptcy Reform. J. Law, Econ., and Organization 8 (October 1992): 523
46.
10
La Porta et al. (1997) use the original La Porta et al. (1996) data. We have
reconfirmed their results using the refined measures presented in this paper.