The Economic Consequences of Legal Origins

Journal of Economic Literature 2008, 46:2, 285–332 http:www.aeaweb.org/articles.php?doi10.1257/jel.46.2.285

Rafael La Porta, Florencio Lopez-de-Silanes, and Andrei Shleifer*

In the last decade, economists have produced a considerable body of research suggesting that the historical origin of a country’s laws is highly correlated with a broad range of its legal rules and regulations, as well as with economic outcomes. We summarize this evidence and attempt a unified interpretation. We also address several objections to the empirical claim that legal origins matter. Finally, we assess the implications of this research for economic reform.

1. Introduction financial development.1 From there, LLSV bout a decade ago, the three of us, made two contributions. First, they showed Atogether with Robert Vishny, published that legal rules governing investor protection a pair of articles dealing with legal protection can be measured and coded for many coun- of investors and its consequences (La Porta et tries using national commercial (primar- al. or LLSV, 1997, 1998). These articles gen- ily corporate and bankruptcy) laws. LLSV erated a fair amount of follow-up research coded such rules for both the protection of and a good deal of controversy. This paper is outside shareholders, and the protection of our attempt to summarize the main findings outside senior creditors, for forty-nine coun- and, more importantly, to interpret them in tries. The coding showed that some coun- a unified way. tries offer much stronger legal protection of LLSV started from a proposition, standard outside investors’ interests than others. in corporate law (e.g., Robert Clark 1986) and Second, LLSV documented empirically emphasized by Shleifer and Vishny (1997), that legal rules protecting investors vary sys- that legal protection of outside investors limits tematically among legal traditions or origins, the extent of expropriation of such investors by corporate insiders, and thereby promotes 1 This argument followed naturally from the contractual view of the firm (Michael C. Jensen and William H. Meckling 1976, Sanford J. Grossman and Oliver D.

* La Porta: Tuck School Of Business At Dartmouth. Hart 1988, Hart 1995), which sees the protection of the Lopez-de-Silanes: EDHEC Graduate School of Manage- property rights of the financiers as essential to assure the ment. Shleifer: Harvard University. We are grateful to flow of capital to firms. Financial economists have often Daniel Berkowitz, Olivier Blanchard, Simeon Djankov, argued, in contrast, that financial markets are sustained Nicola Gennaioli, Roger Gordon, Oliver Hart, Paul by “market forces” such as competition and reputation Mahoney, Elias Papaioannou, Katharina Pistor, Mark (Hayne E. Leland and David H. Pyle 1977, Eugene F. Roe, René Stulz, students in Ec 2470, four reviewers, and Fama 1980). Comparative research emphasized the role especially Louis Kaplow for extremely helpful comments. of banks (Franklin Allen and Douglas Gale 2000). 285 286 Journal of Economic Literature, Vol. XLVI (June 2008) with the laws of common law countries (orig- Why is its influence so pervasive? How can inating in English law) being more protective the superior performance of common law in of outside investors than the laws of civil law many areas be reconciled with the high costs (originating in Roman law) and particularly of litigation, and well-known judicial arbi- French civil law countries. LLSV further trariness, in common law countries? argued that legal traditions were typically In this paper, we adopt a broad conception introduced into various countries through of legal origin as a style of social control of conquest and colonization and, as such, were economic life (and maybe of other aspects of largely exogenous. LLSV then used legal ori- life as well). In strong form (later to be supple- gins of commercial laws as an instrument for mented by a variety of caveats), we argue that legal rules in a two stage procedure, where common law stands for the strategy of social the second stage explained financial devel- control that seeks to support private market opment. The evidence showed that legal outcomes, whereas civil law seeks to replace investor protection is a strong predictor of such outcomes with state-desired alloca- financial development. tions. In words of one legal scholar, civil law Subsequent research showed that the is “policy implementing,” while common law influence of legal origins on laws and is “dispute resolving” (Mirjan R. Damaška regulations is not restricted to finance. 1986). In words of another, French civil law In several studies conducted jointly with embraces “socially-conditioned private con- Simeon Djankov and others, we found that tracting,” in contrast to common law’s sup- such outcomes as government ownership port for “unconditioned private contracting” of banks (La Porta et al. 2002), the bur- (Katharina Pistor 2006). We develop an den of entry regulations (Djankov et al. interpretation of the evidence, which we call 2002), regulation of labor markets (Juan the Legal Origins Theory, based on these C. Botero et al. 2004), incidence of mili- fundamental differences. tary conscription (Casey B. Mulligan and Legal Origin Theory traces the different Shleifer 2005a, 2005b), and government strategies of common and civil law to dif- ownership of the media (Djankov et al. ferent ideas about law and its purpose that 2003a) vary across legal families. In all England and France developed centuries these spheres, civil law is associated with ago. These broad ideas and strategies were a heavier hand of government ownership incorporated into specific legal rules, but and regulation than common law. Many of also into the organization of the legal system, these indicators of government ownership as well as the human capital and beliefs of and regulation are associated with adverse its participants. When common and civil law impacts on markets, such as greater cor- were transplanted into much of the world ruption, larger unofficial economy, and through conquest and colonization, the higher unemployment. rules, but also human capital and legal ide- In still other studies, we have found that ologies, were transplanted as well. Despite common law is associated with lower formal- much local legal evolution, the fundamen- ism of judicial procedures (Djankov et al. tal strategies and assumptions of each legal 2003b) and greater judicial independence system survived and have continued to exert (La Porta et al. 2004) than civil law. These substantial influence on economic outcomes. indicators are in turn associated with better As the leading comparative legal scholars contract enforcement and greater security of Konrad Zweigert and Hein Kötz (1998) note, property rights. “the style of a legal system may be marked Assuming that this evidence is correct, it by an ideology, that is, a religious or politi- raises an enormous challenge of interpreta- cal conception of how economic or social life tion. What is the meaning of legal origin? should be organized” (p. 72). In this paper, La Porta, Lopez-de-Silanes, and Shleifer: Economic Consequences of Legal Origins 287 we show how these styles of different legal and controlling shareholders. This research, systems have developed, survived over the however, is not covered in our paper. years, and continued to have substantial economic consequences. In our conception, 2. Background on Legal Origins legal origins are central to understanding the varieties of capitalism. In their remarkable three-hundred- The paper is organized as follows. In sec- page survey of human history, The Human tion 2, we describe the principal legal tradi- Web, John Robert McNeill and William H. tions. In section 3, we document the strong McNeill (2003) show how the transmission and pervasive effects of legal origins on of information across space shapes human diverse areas of law and regulation, which societies. Information is transmitted through in turn influence a variety of economic out- trade, conquest, colonization, mission- comes. In section 4, we outline the Legal ary work, migration, and so on. The bits of Origins Theory, and interpret the findings information transmitted through these chan- from that perspective. In sections 5–7, we nels include technology, language, religion, deal with three lines of criticism of our sports, but also law and legal systems. Some research, all organized around the idea that of these bits of information are transplanted legal origin is a proxy for something else. voluntarily, as when people adopt technolo- The three alternatives we consider are cul- gies they need. This makes it difficult to ture, politics, and history. Our strong conclu- study the consequences of adoption because sion is that, while all these factors influence we do not know whether to attribute these laws, regulations, and economic outcomes, consequences to what is adopted or to the it is almost certainly false that legal origin conditions that invited the adoption. In other is merely a proxy for any of them. Section 8 instances, the transplantation of information briefly considers the implications of our work is involuntary, as in the cases of forced reli- for economic reform and describes some gious conversion, conquest, or colonization. of the reforms that had taken place. Many These conditions, unfavorable as they are, developing countries today find themselves make it easier to identify the consequences heavily overregulated in crucial spheres of of specific information being transplanted. economic life, in part because of their legal Legal origins or traditions present a key origin heritage. Legal Origin Theory, and example of such often involuntary trans- the associated measurement of legal and mission of different bundles of information regulatory institutions, provides some guid- across human populations. Legal scholars ance to reforms. Section 9 concludes the believe that some national legal systems are paper. sufficiently similar in some critical respects We note that this paper is not a survey and, to others to permit classification of national therefore, only introduces particular papers legal systems into major families of law (Rene in so far as they enter the discussion of the David and John Brierley 1985, Thomas meaning and the consequences of legal ori- Reynolds and Arturo Flores 1989, Mary gins. The last decade has witnessed an explo- Ann Glendon, Michael Wallace Gordon, and sion of research on corporate governance that Christopher Osakwe 1982, 1994, Zweigert uses the investor protection framework. This and Kötz 1998). “The following factors seem research has successfully replaced the tradi- to us to be those which are crucial for the tional Berle–Means conception of a public style of a legal system or a legal family: (1) corporation with a much more realistic for its historical background and development, most of the world model of family-run firms, (2) its predominant and characteristic mode pyramidal and group structures, and tre- of thought in legal matters, (3) especially mendous conflicts between outside investors distinctive institutions, (4) the kind of legal 288 Journal of Economic Literature, Vol. XLVI (June 2008) sources it acknowledges and the way it han- Europe.2 It was also imitated by some social- dles them, and (5) its ideology” (Zweigert and ist states, such as Mongolia and China. After Kötz 1998, p. 68). the fall of the Berlin Wall, the countries of Most writers identify two main secular the former Soviet Union and Eastern Europe legal traditions: common law and civil law, reverted to their pre–Russian Revolution or and several subtraditions—French, German, pre–World War II legal systems, which were socialist, and Scandinavian—within civil French or German civil law. In our work law. Occasionally, countries adopt some laws based on data from the 1990s, we have often from one legal tradition and other laws from classified transition economies as having the another, and researchers need to keep track socialist legal system. However, today, aca- of such hybrids, but generally a particular demics and officials from these countries tradition dominates in each country. object to such classification, so, in the pres- The key feature of legal traditions is that ent paper, we classify them according to the they have been transplanted, typically though main influence on their new commercial not always through conquest or colonization, laws. Several countries, such as Cuba, still from relatively few mother countries to most maintain the socialist legal system, and await of the rest of the world (Alan Watson 1974). liberation and reclassification. These coun- Such transplantation covers specific laws and tries typically lack other data, so no socialist codes and the more general styles or ideolo- legal origin countries appear in the analysis gies of the legal system, as well as individuals in the present paper. with mother-country training, human capi- Figure 1 shows the distribution of legal ori- tal, and legal outlook. gins of commercial laws throughout the world. Of course, following the transplantation The common-law legal tradition includes the of some basic legal infrastructure, such as law of England and its former colonies. The the legal codes, legal principles and ideolo- common law is formed by appellate judges gies, and elements of the organization of the who establish precedents by solving specific judiciary, the national laws of various coun- legal disputes. Dispute resolution tends to be tries changed, evolved, and adapted to local adversarial rather than inquisitorial. Judicial circumstances. Cultural, political, and eco- independence from both the executive and nomic conditions of every society came to be legislature are central. “English common law reflected in their national laws, so that legal developed because landed aristocrats and and regulatory systems of no two countries merchants wanted a system of law that would are literally identical. This adaptation and provide strong protections for property and individualization, however, was incomplete. contract rights, and limit the crown’s ability Enough of the basic transplanted elements to interfere in markets” (Paul G. Mahoney have remained and persisted (Paul A. David 2001, p. 504). Common law has spread to the 1985) to allow the classification into legal British colonies, including the United States, traditions. As a consequence, legal trans- Canada, Australia, India, South Africa, and plantation represents the kind of involuntary many other countries. Of the maximal sam- information transmission that the McNeills ple of 150 countries used in our studies, there have emphasized, which enables us to study are forty-two common law countries. the consequences of legal origins. Before discussing the legal traditions of market economies, we briefly comment 2 The socialist legal tradition illustrates the signifi- on socialist law. The socialist legal tradi- cance of ideologies for legal styles. “. . . the socialist con- tion originates in the Soviet Union, and was cept of law can be directly traced to the movement of legal positivism. The movement . . . sees law as an expression of spread by the Soviet armies first to the for- the will of the legislators, supreme interpreters of justice” mer Soviet republics and later to Eastern (David and Brierley 1985, p. 69). La Porta, Lopez-de-Silanes, and Shleifer: Economic Consequences of Legal Origins 289

Legal Origins English French German Scandinavian Socialist

Figure 1. The Distribution of Legal Origin

The civil law tradition is the oldest, the it did because the revolutionary generation, most influential, and the most widely dis- and Napoleon after it, wished to use state tributed around the world, especially after so power to alter property rights and attempted many transition economies have returned to to insure that judges did not interfere. Thus, it. It originates in Roman law, uses statutes quite apart from the substance of legal rules, and comprehensive codes as a primary means there is a sharp difference between the ide- of ordering legal material, and relies heavily ologies underlying common and civil law, on legal scholars to ascertain and formulate with the latter notably more comfortable rules (John Henry Merryman 1969). Dispute with the centralized and activist govern- resolution tends to be inquisitorial rather ment” (Mahoney 2001, p. 505). than adversarial. Roman law was rediscov- Napoleon’s armies introduced his codes ered in the Middle Ages in Italy, adopted by into Belgium, the Netherlands, Italy, and the Catholic Church for its purposes, and parts of Germany. In the colonial era, France from there formed the basis of secular laws extended her legal influence to the Near in many European countries. East and Northern and Sub-Saharan Africa, Although the origins of civil law are Indochina, Oceania, and French Caribbean ancient, the French civil law tradition is usu- Islands. Napoleonic influence was also sig- ally identified with the French Revolution nificant in Luxembourg, Portugal, Spain, and Napoleon’s codes, which were written in and some Swiss cantons. When the Spanish the early nineteenth century. In contrast to and Portuguese empires in Latin America common law, “French civil law developed as dissolved in the nineteenth century, it was 290 Journal of Economic Literature, Vol. XLVI (June 2008) mainly the French civil law that the lawmak- ration for Bonaparte, partly language, and ers of the new nations looked to for inspira- partly Napoleonic influence on the Spanish tion. In the nineteenth century, the French and Portuguese codes. In this instance, the civil code was also adopted, with many exogeneity assumption from the viewpoint of modifications, by the Russian Empire, and studying economic outcomes is still appropri- through Russia by the neighboring regions ate. The nineteenth century influence of the it influenced and occupied. These countries French civil law in Russia and Turkey was adopted the socialist law after the Russian largely voluntary, as both countries sought Revolution, but typically reverted to the to modernize. But the French and German French civil law after the fall of the Berlin civil law traditions in the rest of the countries Wall. There are eighty-four French legal ori- in Eastern Europe, the Middle East, and gin countries in the sample. Central Asia are the result of the conquests The German legal tradition also has by the Russian, Austro–Hungarian, Ottoman, its basis in Roman law, but the German and German empires. The return by these Commercial Code was written in 1897, countries to their pre-Soviet legal traditions after Bismarck’s unification of Germany. during the transition from socialism is volun- It shares many procedural characteristics tary but shaped largely by history. with the French system but accommodates Second, because Scandinavian countries greater judicial law making. The German did not have any colonies, and Germany’s legal tradition influenced Austria, the for- colonial influence was short-lived and mer Czechoslovakia, Greece, Hungary, Italy, abruptly erased by World War I, there are Switzerland, Yugoslavia, Japan, Korea, and relatively few countries in these two tradi- a few countries of the former Soviet Union. tions. As a consequence, while we occa- Taiwan’s laws came from China, which relied sionally speak of the comparison between heavily on German laws during moderniza- common and civil law, most of the discussion tion. There are nineteen German legal origin compares common law to the French civil countries in the sample. law. This is largely because each tradition The Scandinavian family is usually viewed includes a large number of countries, but also as part of the civil law tradition, although because they represent the two most distinct its law is less derivative of Roman law than approaches to law and regulation. the French and German families (Zweigert Third, although we often speak of common and Kötz 1998). Most writers describe the law and French civil law in terms of pure Scandinavian laws as distinct from others, types, in reality there has been a great deal and we have kept them as a separate family of mutual influence and in some areas con- (with five members) in our research. vergence. There is a good deal of legislation Before turning to the presentation of in common law countries, and a good deal of results, five points about this classification judicial interpretation in civil law countries. are in order. First, although the majority of But the fact that the actual laws of real coun- legal transplantation is the product of con- tries are not pure types does not mean that quest and colonization, there are important there are no systematic differences. exceptions. Japan adopted the German legal Fourth, some have noted the growing system voluntarily. Latin American former importance of legislation in common law Spanish and Portuguese colonies ended up countries as proof that judicial law mak- with codifications heavily influenced by the ing no longer matters. This is incorrect for French legal tradition after gaining indepen- a number of reasons. Statutes in common dence. Beyond the fact that Napoleon had law countries often follow and reflect judicial invaded the Iberian Peninsula, the reasons rulings, so jurisprudence remains the basis were partly the new military leaders’ admi- of statutory law. Even when legislation in La Porta, Lopez-de-Silanes, and Shleifer: Economic Consequences of Legal Origins 291 common law countries runs ahead of judi- the data used in this paper, and a good deal cial law making, it often must coexist with, more, are available at http://www.econom- and therefore reflects, preexisting common ics.harvard.edu/faculty/shleifer/data.html. law rules. Indeed, statutes in common law We do not discuss the data in detail, but countries are often highly imprecise, with the descriptions are available in the original an expectation that courts will spell out the papers presenting the data. rules as they begin to be applied. Finally, and Second, the basic evidence we present most crucially, because legal origins shape takes the form of cross-country studies. An fundamental approaches to social control important feature of these studies is that all of business, even legislation in common law countries receive the same weight. There is countries expresses the common law way of no special treatment of mother countries, of doing things. For all these reasons, the uni- rich countries, etc. This design may obscure versal growth of legislation in no way implies the differences, discussed below, within the irrelevance of legal origins. legal origins, such as the greater dynamism Fifth, with the reclassification of transition of law in mother countries than in former economies from socialist into the French and colonies. German civil law families, one might worry Third, the sources of data on legal rules that the differences among legal origins and institutions vary significantly across described below are driven by the transi- studies. Some rules, such as many indicators tion economies. They are not. None of our of investor protection and of various govern- substantive results change if we exclude the ment regulations, come from national laws. transition economies. Those tend to be “laws on the books.” Other With these points in mind, we can turn to indicators are mixtures of national laws and the evidence. actual experiences, and tend to combine substantive and procedural rules. These variables are often constructed through collaborative 3. Basic Facts efforts with law firms around the world and yield summary indicators of legal rules and 3.1 The Evidence in Brief their enforcement. For example, the study Figure 2 organizes some of our own and of legal formalism (Djankov et al. 2003b) related research on the economic conse- reflects the lawyers’ characterization of pro- quences of legal origins. It shows the links cedural rules that would typically apply to a from legal origins to particular legal rules, specific legal dispute; the study of the effi- and then to economic outcomes. Figure 2 ciency of debt enforcement (Djankov et al. immediately suggests several concerns for 2006) incorporates estimates of time, cost, empirical work. First, in our framework, legal and resolution of a standardized insolvency origins influence many spheres of law mak- case. The data used in each study have their ing and regulation, which makes it dangerous advantages and problems. An important fact, to use them as instruments. Second, we have however, is the consistency of results across drawn a rather clean picture pointing from both data collection procedures and spheres particular legal rules to outcomes. In reality, of activity that we document below. a variety of legal rules (e.g., those governing Fourth, over the years, various writers both investor protection and legal procedure) have criticized both the conceptual founda- can influence the protection of outside inves- tions of LLSV variables such as shareholder tors and hence financial markets. This, again, rights indices (John C. Coffee 1999) and makes empirical work less clean. the particular values we have assigned to Before turning to the evidence, we make these variables, in part because of concep- four comments about the data. First, all tual ambiguity (Holger Spamann 2006b). 292 Journal of Economic Literature, Vol. XLVI (June 2008)

Institution Outcomes

Time to Evict Nonpaying Tenant Procedural Formalism Time to Collect a Bounced Check

Judicial Independence Property Rights

Corruption Regulation of Entry Unofficial Economy

Government Ownership of the Media

Participation Rates Legal Origin Labor Laws Unemployment

Conscription

Stock Market Development Company Law Firm Valuation Securities Law Ownership Structure Control Premium

Bankruptcy Law Private credit

Government Ownership Interest Rate Spread of Banks

Figure 2. Legal Origin, Institutions, and Outcomes

We have corrected our mistakes and have The available studies can be divided into also moved on to conceptually less ambigu- three categories. First, several studies follow- ous measures (Djankov et al. 2008). These ing LLSV (1997, 1998) examine the effects of improvements have strengthened the origi- legal origins on investor protection and then nal results. The findings we discuss below the effect of investor protection on financial use the most recent data. development. Some of these studies look at To organize the discussion, we do not pro- stock markets. The LLSV measure of antidi- vide a full survey of the available evidence rector rights has been replaced by a measure but rather a sampling with an emphasis on the of shareholder protection through securi- breadth of the findings. The available studies ties laws in the offerings of new issues (La have followed a similar pattern, shown in fig- Porta et al. 2006) and by another measure ure 2. They first consider the effect of legal of shareholder protection from self-deal- origins on particular laws and regulations, ing by corporate insiders through corporate and then the effects of these laws and regu- law (Djankov et al. 2008). As outcomes, lations on the economic outcomes that they these studies use such measures as the ratio might influence most directly. of stock market capitalization to GDP, the La Porta, Lopez-de-Silanes, and Shleifer: Economic Consequences of Legal Origins 293 pace of public offering activity, the voting tion and the size of the unofficial economy. premium (see Alexander Dyck and Luigi Botero et al. (2004) construct indices of labor Zingales 2004), dividend payouts (La Porta market regulation and examine their effect et al. 2000), Tobin’s Q (La Porta et al. 2002), on labor force participation rates and unem- and ownership dispersion (La Porta et al. ployment. Djankov et al. (2003c) examine 1999). Predictions for each of these variables government ownership of the media, which emerge from a standard agency model of remains extensive around the world, par- corporate governance, in which investor pro- ticularly for television. Mulligan and Shleifer tection shapes external finance (e.g., Shleifer (2005a, 2005b) look at one of the ultimate and Daniel Wolfenzon 2002). 3 forms of government intervention in private Other studies in this category look at cred- life, military conscription. itor rights. The LLSV (1997, 1998) measure The third category of papers investigates from bankruptcy laws has been updated by the effects of legal origins on the character- Djankov et al. (2007). Djankov et al. (2006) istics of the judiciary (and other government take a different approach to creditor protec- institutions), and then the effect of those on tion by looking at the actual efficiency of the security of property rights and contract debt enforcement, as measured by creditor enforcement. Djankov et al. (2003b) look recovery rates in a hypothetical case of an at the formalism of judicial procedures in insolvent firm. The latter study addresses a various countries and its effect on the time common criticism that it is law enforcement, it takes to evict a nonpaying tenant or to col- rather than rules on the books, that matters lect a bounced check. This variable can be for investor protection by integrating legal interpreted more broadly as the efficiency rules and characteristics of enforcement in of contract enforcement by courts, and in the efficiency measure. La Porta et al. (2002) fact turns out to be highly correlated with focus on state involvement in financial mar- the efficiency of debt collection obtained in kets by looking at government ownership of an entirely different way by Djankov et al. banks. These studies typically consider the (2006). La Porta et al. (2004) adopt a very size of debt markets as an outcome measure, different strategy and collect information although Djankov et al. (2007) also examine from national constitutions on judicial inde- several subjective assessments of the quality pendence (as measured by judicial tenure) of private debt markets. and the acceptance of appellate court rul- In the second category, several papers ings as a source of law. They then ask directly consider government regulation, or even whether judicial independence contributes ownership, of particular economic activities. to the quality of contract enforcement and Djankov et al. (2002) look at the number the security of property rights. of steps an entrepreneur must complete in Tables 1–3 show a sampling of results from order to begin operating a business legally, each category of studies. In each table, the top a number that in 1999 ranged from two in panel presents the regressions of legal or regu- Australia and Canada to twenty-one in the latory institutions on legal origins, controlling Dominican Republic. They examine the only for per capita income. In the original impact of such entry regulation on corrup- papers, many more controls and robustness checks are included, but here we present the stripped down regressions. The bottom panel 3 The theoretical prediction that investor protection of each table then presents some results of leads to greater ownership dispersion is not unambiguous, regressions of outcomes on legal rules. We and the data on ownership around the world is less clean could of course combine the two panels in and satisfactory than that on other variables. Nonetheless, much of the criticism of LLSV has focused on ownership an instrumental variables specification, but, dispersion. as we indicated previously, we do not recom- 294 Journal of Economic Literature, Vol. XLVI (June 2008) mend such specifications since legal origins Investor protection is associated with more influence a broad range of rules and regula- developed financial markets (panel B). The tions and we cannot guarantee that the rel- estimated coefficients imply that a two-stan- evant ones are not omitted in the first stage. dard deviation increase in the anti-self-deal- Begin with table 1. Higher income per ing index is associated with an increase in capita is associated with better shareholder the stock-market-to-GDP ratio of 42 per- and creditor protection, more efficient debt centage points, an increase in listed firms enforcement, and lower government owner- per capita of 38 percent, and a reduction ship of banks (panel A). Civil law is generally in ownership concentration of 6 percentage associated with lower shareholder and credi- points. A two-standard deviation improve- tor protection, less efficient debt enforcement in prospectus disclosure is associated ment, and higher government ownership with a reduction in the control premium of of banks. The estimated coefficients imply 15 percentage points (the mean premium is that, compared to common law, French legal 11 percent). The effect of legal rules on debt origin is associated with a reduction of 0.33 markets is also large. A two-standard devia- in the anti-self-dealing index (which ranges tion increase in creditor rights is associated between 0.1 and 1), of 0.33 in the index of with an increase of 15 percentage points in prospectus disclosure (which ranges between the private-credit-to-GDP ratio. A two-stan- 0 and 1), of 0.84 in the creditor rights index dard deviation increase in the efficiency of (which ranges from 0 to 4), of 13.6 points in debt collection is associated with an increase the efficiency of debt collection (out of 100), of 27 percentage points in the private-credit- and a rise of 33 percentage points in gov- to-GDP ratio. A two-standard deviation ernment ownership of banks (which ranges increase in government ownership of banks between 0 and 100 percent). The effect is associated with a 16 percentage point rise of legal origins on legal rules and financial in the spread between lending and borrow- institutions is statistically significant and ing rates (the mean spread is 20).5 economically large. Table 2 presents the results on regulation. Higher income per capita is generally Higher income per capita is correlated with associated with more developed financial lower entry regulation and government own- markets, as reflected in a higher stock-mar- ership of the media, but not with labor regu- ket-capitalization-to-GDP ratio, more firms lation or conscription (panel A). Relative to per capita, less ownership concentration, common law countries, French legal origin and a higher private-credit-to-GDP ratio.4 countries have more entry and labor regulation, higher state ownership of the media, and heavier reliance on conscription.6 The 4 Recent research has looked at additional outcome coefficients imply that, compared to common variables as well as measures of credit market regulation. Benjamin C. Esty and William L. Megginson (2003) find that creditor protection shapes foreign bank lending, while Steven Ongena and David C. Smith (2000) show 5 Paola Sapienza (2004) shows that government-owned it influences the number of banking relationships. Jun banks in Italy lend to big enterprises rather than small Qian and Philip E. Strahan (2007) find that better credi- ones. I. Serdar Dinc (2005) finds that government-owned tor protection lowers interest rates that lenders charge. banks sharply increase lending in election years. Asim Ijaz Pablo Casas-Arce and Albert Saiz (2007) find that costly Khwaja and Atif Mian (2005) present evidence that politi- enforcement of rental contracts hampers the development cally connected firms in Pakistan get preferential treat- of the rental housing market in a cross-section of coun- ment from government-owned banks. They borrow 45 tries. Krishnamurthy Subramanian, Frederick Tung, and percent more and have 50 percent higher default rates. Xue Wang (2007) find that stronger creditor rights mitigate 6 In a similar spirit, Avi Ben-Bassat and Momi Dahan the effect of managerial self-dealing on project finance. (2008) show that constitutional commitments to “social James R. Barth, Gerard Caprio, and Ross Levine (2004) rights” (the right to social security, education, health, introduce measures of banking regulation and show that housing, and workers rights) are less prevalent in common they vary systematically by legal origin. law countries than in the French civil law ones. La Porta, Lopez-de-Silanes, and Shleifer: Economic Consequences of Legal Origins 295

TABLE 1 Financial Institutions and Capital Markets Development

Panel A: Financial Institutions and Legal Origin (1) (2) (3) (4) (5)

Anti-Self-Dealing Prospectus Creditor Rights Debt Bank Government Index Disclosure Index (2003) Enforcement Owernship (1970) French Legal –0.3334a –0.3298a –0.8394a –13.6361b 0.3316a Origin (0.0511) (0.0577) (0.2251) (5.6535) (0.0755) German Legal –0.3454a –0.2370b –0.1714 –8.8577 0.3456a Origin (0.0736) (0.0966) (0.2579) (5.8022) (0.1060) Scandinavian –0.3820a –0.2867a –0.9435c 5.2707 0.3109 Legal Origin (0.0642) (0.0478) (0.4865) (5.8212) (0.1545) Ln (GDP per 0.0728a 0.0618b 0.2022b 19.8980a –0.1808a capita) (0.0263) (0.0261) (0.0875) (2.7517) (0.0377) Constant 0.0177 0.2102 0.6043 –124.6692a 1.6206a (0.2433) (0.2422) (0.7560) (26.9421) (0.2876)

Observations 71 49 130 85 74 R-squared 45% 0.45 18% 0.57 37%

Panel B: Financial Institutions and Capital Markets Development (1) (2) (3) (4) (5) (6)

Stock-market-to- Ln(Firms/Pop) Ownership Control Private-credit-to- Interest spread GDP (1999–2003) (1999–2003) Concentration Premium GDP (1999–2003) (1970–95)

Anti-self-dealing 0.8940b 0.8004c –0.1277c Index (0.3674) (0.4750) (0.0724) Prospectus –0.3254a Disclosure (0.0807) Credit Rigths Index 0.0645c (0.0336) Debt Enforcement 0.0053a (0.0015) Government Ownership 22.0813a of Banks—1970 (7.3675) Ln (GDP per capita) 0.3204a 0.9794a –0.0495b –0.0273 0.2546a 1.8522 (0.0601) (0.1346) (0.0200) (0.0238) (0.0604) (3.0169) Constant –2.7604a –6.9496a 0.9844a 0.5524b –2.1494a –4.4219 (0.5558) (1.2352) (0.1761) (0.2202) (0.4912) (23.0311)

Observations 72 72 49 37 85 57 R-squared 40% 47% 20% 36% 52% 10%

Notes: a Significant at the 1 percent level. b Significant at the 5 percent level. c Significant at the 10 percent level. 296 Journal of Economic Literature, Vol. XLVI (June 2008)

TABLE 2 Government Regulation

Panel A: Government Regulation and Legal Origin (1) (2) (3) (4)

Regulation of Entry Regulation of Labor Press Government Conscription (1999) (1997) Ownership (1999) (2000)

French Legal Origin 0.6927a 0.2654a 0.2095a 0.5468a (0.0929) (0.0362) (0.0834) (0.0772) German Legal Origin 0.5224a 0.2337a 0.1100 0.8281a (0.1206) (0.0473) (0.0926) (0.0794) Scandinavian Legal Origin –0.1922 0.3978a 0.1308b 0.7219a (0.1352) (0.0443) (0.0555) (0.2015) Ln (GDP per capita) –0.1963a –0.0083 –0.1753a –0.0382 (0.0367) (0.0164) (0.0307) (0.0331) Constant 3.4367a 0.3703b 1.6565a 0.4702c (0.3037) (0.1520) (0.3024) (0.2802)

Observations 85 84 95 146 R-squared 61% 42% 37% 34%

Panel B: Government Regulation, Corruption, Unofficial Economy, and Labor Market Outcomes (1) (2) (3) (4) (5)

Corruption Index Employment Labor Participation Unemployment Unemployment Rate (1996–2000) Unofficial Economy Male Rate (1991–2000) for Men Aged 20–24 Regulation of Entry –0.6733a 13.2601a (0.0998) (4.4569) Regulation of –5.2009a 6.0738b 14.8363a Labor (1.7319) (2.7868) (4.2699) Ln (GDP per capita) 0.6194a –5.7288a –1.9305a –0.9913c –1.1890 (0.0537) (2.0969) (0.3982) (0.5795) (1.1308) Constant –3.6273a 58.7496b 102.5096a 14.8245b 18.4049 (0.5800) (25.8820) (3.3120) (6.0449) (11.4316)

Observations 85 46 78 65 52 R-squared 80% 42% 32% 11% 15%

Notes: a Significant at the 1 percent level. b Significant at the 5 percent level. c Significant at the 10 percent level. law, French legal origin is associated with an According to the estimated coefficients in increase of 0.69 in the (log) number of steps to panel B, a two-standard deviation increase open a new business (which ranges from 0.69 in the (log) number of steps to open a new to 3.0), a rise of 0.26 in the index of labor regu- business is associated with a 0.71 worsening lation (which ranges from 0.15 to 0.83), a 0.21 of the corruption index and a 14 percent- rise in government ownership of the media age point rise in employment in the unof- (which ranges from 0 to 1), and a 0.55 increase ficial economy. A two-standard deviation in conscription (which ranges from 0 to 1). increase in the regulation of labor implies La Porta, Lopez-de-Silanes, and Shleifer: Economic Consequences of Legal Origins 297

TABLE 3 Judicial Institutions

Panel A: Legal Origin and Judicial Institutions (1) (2) (3) Formalism Check Collection Tenure of Judges Case Law French Legal Origin 1.4945a –0.2375a –0.6733a (0.1841) (0.0620) (0.0951) German Legal Origin 0.9917a –0.4627a –0.2874 (0.2013) (0.1459) (0.2156) Scandinavian Legal Origin 0.7623a –0.0636 0.0460 (0.2966) (0.0470) (0.0727) Ln (GDP per capita) –0.2610a 0.0412 –0.0004 (0.0707) (0.0295) (0.0337) Constant 5.0505a 0.6514a 0.9578a (0.6103) (0.2500) (0.2850)

Observations 109 65 65 R-squared 45% 25% 44%

Panel B: Judicial Instititutions and Outcomes (1) (2) (3) (4)

Time to Collect on Contract Property Rights Property Rights Bounced Check Enforcement (2004) (2004)

Legal Formalism — 0.3095a –0.5099a Bounced Check (0.0519) (0.0966) Tenure of judges 1.2066a (0.2547) Case law 0.5596a (0.2035) Ln (GDP per capita) 0.0402 1.0544a 0.8673a 0.8767a (0.0546) (0.1410) (0.0818) (0.0886) Constant 3.7354 –1.7313 –5.6499a –5.0485a (0.5251) (1.5253) (0.7439) (0.8288)

Observations 109 52 64 64 R-squared 12% 74% 69% 67%

Notes: a Significant at the 1 percent level. b Significant at the 5 percent level. c Significant at the 10 percent level. a 1.99 percentage point reduction in the with less legal formalism but not with longer male labor force participation rate, a 2.32 judicial tenure or the acceptance of case law percentage point increase in the unemploy- (panel A). Here again, legal origin has a pro- ment rate, and a 5.67 percentage point rise nounced effect on institutions. Compared to in the unemployment rate of young males. common law countries, civil law countries Table 3 shows the results on judicial institu- generally have more legal formalism, lower tions. Higher income per capita is associated judicial tenure, and sharply lower constitu- 298 Journal of Economic Literature, Vol. XLVI (June 2008) tional acceptance of case law.7 The estimated turn associated with more secure property coefficients imply that French legal origin rights and better contract enforcement. is associated with an increase of 1.49 in the To us, the most important aspect of these index of legal formalism, a reduction of 0.24 in results is how pervasive is the influence of judicial tenure, and of 0.67 in case law. These legal origins. Many objections have been are large effects since legal formalism ranges raised with respect to individual pieces of from 0.73 to 6.0, and both judicial tenure and this evidence. We address later the most far- case law range from 0 to 1. reaching criticism, that legal origin is a proxy Judicial institutions matter for both the for something else, but deal here with more efficiency of contract enforcement and the parochial concerns. The key point to start security of property rights (panel B). The esti- with, however, is that objections rarely come mated coefficients imply that a two-standard to grips with the breadth of the influence of deviation increase in legal formalism is asso- legal origins on economic outcomes. ciated with a 67 percentage point increase in We focus on objections to the law and the time to collect on a check and a reduction finance evidence. The most immediate objec- of 1.1 in the index of contract enforcement tion is reverse causality: countries improve (the latter ranges from 3.5 to 8.9). Moreover, their laws protecting investors as their finan- a two-standard deviation increase in judicial cial markets develop, perhaps under political tenure is associated with a 0.8 point rise in pressure from those investors. If instrumen- the property rights index. Finally, a two stan- tal variable techniques were appropriate in dard deviation increase in case law is asso- this context, a two stage procedure, in which ciated with an improvement of 0.6 points in in the first stage the rules are instrumented the property rights index, which ranges from by legal origins, would address this objec- 1 to 5. tion. LLSV (1997, 1998) pursue this strategy. But even if instrumental variable techniques 3.2 Initial Criticisms are inappropriate because legal origin influ- So, what do we learn from these tables? ences finance through channels other than The economic consequences of legal ori- rules protecting investors, legal origins are gins are pervasive. Compared to French still exogenous, and to the extent that they civil law, common law is associated with (a) shape legal rules protecting investors, these better investor protection, which in turn is rules cannot be just responding to market associated with improved financial develop- development. Moreover, this criticism in no ment, better access to finance, and higher way rejects the significance of legal origins in ownership dispersion, (b) lighter government shaping outcomes; it speaks only to the dif- ownership and regulation, which are in turn ficulty of identifying the channel. associated with less corruption, better func- Recent evidence has gone beyond cross- tioning labor markets, and smaller unofficial section to look at changes in financial devel- economies, and (c) less formalized and more opment in response to changes in legal rules, independent judicial systems, which are in thereby relieving the reverse causality concerns. Michael Greenstone, Paul Oyer, and Annette Vissing-Jørgensen (2006) examine the effects of the 1964 Securities Act 7 Daniel Berkowitz and Karen B. Clay (2005, 2006, Amendments, which increased the disclo- 2007) exploit the fact that ten U.S. states were initially settled by France, Spain, or Mexico to examine the effects sure requirements for U.S. over-the-coun- of legal origin. They find that states initially settled by civil ter firms. They find that firms subject to the law countries granted less independence to their judicia- new disclosure requirements had a statisti- ries as recently as 1970–90, had lower quality courts in 2001–03, and used different procedures for setting judi- cally significant abnormal excess return of cial budgets as late as 1960–2000. about 10 percent over the year and a half La Porta, Lopez-de-Silanes, and Shleifer: Economic Consequences of Legal Origins 299 that the law was debated and passed relative A second concern about the law and to a comparison group of unaffected NYSE/ finance evidence is omitted variables—the AMEX firms. Brian J. Bushee and Christian very reason IV techniques are not suitable Leuz (2005) obtain similar findings using for identifying the channels of influence. a regulatory change in the U.S. over-the- How do we know that legal origin influences counter markets. Nadia Linciano (2003) financial development through legal rules, examines the impact of the Draghi reforms rather than some other channel (or perhaps in Italy, which improved shareholder protec- even other rules)? The most cogent version of tion. The voting premium steadily declined this critique holds that legal origin influences over the period that the Draghi commit- contract enforcement and the quality of the tee was in operation, culminating in a drop judiciary, and it is through this channel that of 7 percent in the premium at the time it effects financial development. Indeed, we of the passage of the law. Tatiana Nenova know from La Porta et al. (1999, 2004) and (2006) analyzes how the control premium is Djankov et al. (2003b), as illustrated in table affected by changes in shareholder protec- 3 above, that common law is associated with tion in Brazil. She documents that the con- better contract enforcement. trol value more than doubled in the second This objection is significant since, in reality, half of 1997 in response to the introduction enforcement and rules are not entirely sepa- of Law 9457/1997, which weakened minor- rable. A formalistic judiciary might be better ity shareholder protection. Moreover, control able to enforce bright line rules than broad values dropped to pre-1997 levels when in legal standards; an independent judiciary the beginning of 1999 some of the minor- might have a comparative advantage at enforc- ity protection rules scrapped by the previous ing standards. One way to address this concern legal change were reinstated. is to control for contract enforcement as best Turning to the evidence on credit markets, we can. In the regressions above, we control Djankov et al. (2007) show that private credit for per capita income, which is a crude proxy of rises after improvements in creditor rights the quality of the judiciary. More recent stud- and in information sharing in a sample of 129 ies, such as Djankov et al. (2008) and La Porta countries. For a sample of twelve transition et al. (2006), also control for the quality of con- economies, Rainer F. H. Haselmann, Pistor, tract enforcement from Djankov et al. (2003b), and Vikrant Vig (2006) report that lending with the result that both the actual legal rules volume responds positively to improvements and the quality of contract enforcement mat- in creditor rights. Sujata Visaria (2006) esti- ter. For the case of credit markets, Mehnaz mates the impact of introducing specialized Safavian and Siddharth Sharma (2007) show tribunals in India aimed at accelerating that creditor rights benefit debt markets if the banks’ recovery of nonperforming loans. country has a good enough court system, but She finds that the establishment of tribunals not if it does not. Tullio Jappelli, Marco Pagano, reduces delinquency in loan repayment by and Magda Bianco (2005) find that credit is between 3 and 10 percentage points. Aldo less widely available in Italian regions with lon- Musacchio (2008) finds that the development ger trials and larger backlogs of pending trials.8 of bond markets in Brazil is correlated with Djankov et al. (2006) combine the rules and changes in creditors’ rights. Mario Gamboa- their actual enforcement into an integrated Cavazos and Frank Schneider (2007), in measure of debt enforcement efficiency. This an exhaustive study of recent bankruptcy measure (see table 1 above) is highly predictive reform in Mexico, show that changes in legal rules lowered the time it takes firms to go 8 Matthieu Chemin (2007a, 2007b) shows that reforms through bankruptcy proceedings and raised of the judiciary improved firms’ access to finance and per- recovery rates. formance in India and Pakistan, respectively. 300 Journal of Economic Literature, Vol. XLVI (June 2008) of debt market development. The available evi- A second version of the flexibility thesis dence suggests that both good rules and their stresses the ability of common law courts enforcement matter, and that the combination to use broad standards rather than specific of the two is generally most effective. rules in rendering their decisions. This abil- Another relevant distinction is between ity enables judges to “catch” self-dealing legal rules and their interpretation. One view or tunneling, and thereby discourages it. is that the actual legal rules, which might Coffee (1999) has famously called this the have come from legislation, from appel- smell-test of common law. Simon Johnson late decisions, or from legislation codifying et al. (2000) examine several legal cases previous appellate decisions, are shaped by concerning tunneling of assets by corporate legal origins and in turn shape finance. For insiders in civil law countries, and find that example, the extensive approval and disclo- the bright line rules of civil law allow corpo- sure procedures for self-dealing transactions rate insiders to structure legal transactions discourage them in common law countries, that expropriate outside investors. In con- as compared to the French civil law coun- trast, the broader standards of common law, tries (Djankov et al. 2008). such as fiduciary duty, discourage tunneling Other writers emphasize the flexibility more effectively. of judicial decision making under common At this point, there is evidence support- law. One version of this argument suggests ing both the “laws on the books” and the that common law judges are able or willing “judicial flexibility” theories. As we argue to enforce more flexible financial contracts, in section 4, both interpretations are also and that such flexibility promotes financial consistent with the fundamental differences development (Nicola Gennaioli 2007). Josh between common and civil law. Lerner and Antoinette Schoar (2005) and 3.3 Recent Findings on Resource Nittai K. Bergman and Daniel Nicolaievsky Allocation (2007) present some evidence in support of this view. Craig G. Doidge, Andrew Karolyi, Recent years have seen an explosion of and René M. Stulz (2007) and Reena research on the consequences of legal rules Aggarwal et al. (2008) also find that national and regulations, many of which are related legal rules protecting investors improve the to legal origins, for resource allocation. We ability of firms to develop company-specific briefly review this evidence before turning corporate governance mechanisms. In the to the interpretation. context of labor markets, Beth Ahlering and Perhaps the largest body of work con- Simon Deakin (2005) likewise argue that in cerns the effect of financial development on civil law countries, unlike in common law resource allocation. Many of these papers ones, freedom of contract is counterbalanced use LLSV indicators of investor protection, by the exercise of public power for the pro- as well as legal origins as instruments for tection of workers in the French tradition, financial development. The central paper and the communitarian conception of the here is Raghuram Rajan and Zingales (1998), enterprise in the German one. Pistor (2006) who find that, in financially developed coun- presents a legal and historical account of the tries, sectors which for technological rea- greater contractual flexibility in common law, sons depend more on external finance grow the reason being that contractual freedom is faster. In a similar spirit, Jeffrey Wurgler unencumbered by social conditionality.9 (2000) finds that financially developed

9 Naomi R. Lamoreaux and Jean-Laurent Rosenthal (2005) dispute the flexibility hypothesis by pointing to organization in France than in the United States in the the broader range of legally acceptable forms of business nineteenth century. La Porta, Lopez-de-Silanes, and Shleifer: Economic Consequences of Legal Origins 301 countries exhibit a higher responsiveness of enforcement specialize in the production of investment to growth opportunities.10 goods for which relationship-specific invest- Several studies use the data from Botero ments are more important. Pol Antras, Mihir et al. (2004) to examine the effects of labor A. Desai, and C. Fritz Foley (2007) find that regulation on resource allocation. Ricardo J. weak investor protections limit the scale of Caballero et al. (2004) find that, in countries multinational firm activity, increase the reli- with strong rule of law, higher job security is ance on FDI flows, and alter the decision to associated with slower adjustment to shocks deploy technology through FDI rather than and lower productivity growth. Gaëlle Pierre licensing. and Stefano Scarpetta (2007) show that Finally, a growing body of research shows employment regulations lead to substitution that costly regulation can reduce the benefits from permanent to temporary employment. of international trade. Jose Ernesto López- John Haltiwanger, Scarpetta, and Helena Córdova (2007) finds that exporting firms Schweiger (2006) and Alejandro Micco and grow 4 percentage points faster after trade Carmen Pagés (2006) find that legal employ- liberalization in countries with less burden- ment protections reduce labor turnover. some labor regulations. Using cross-coun- Alejandro Cuñat and Marc J. Melitz (2006) try data, Caroline Freund and Bineswaree find that countries with light labor mar- Bolaky (2007) show that trade openness has ket regulations specialize in volatile indus- a positive impact on per capita income only tries. Francine Lafontaine and Jagadeesh in countries with low regulation of entry. Sivadasan (2007) study one firm operating in Norman Loayza, Linda Kaltani, and Roberto forty-three countries, and find that employ- Chang (2005) present a similar finding for ment protections lead to labor misalloca- labor market regulation. Elhanan Helpman, tion, delayed entry, and operation of fewer Melitz, and Yona Rubinstein (2008) find that outlets. the probability that two countries trade is Entry regulations, another sphere influ- smaller when the cost of entry regulation is enced by legal origins, also affect resource high in both countries. allocation. Fisman and Virginia Sarria- All this evidence suggests that, through Allende (2004) find that entry regulations their effect on finance, labor markets, and distort industry structure and promote con- competition, legal origins indeed influence centration. Leora Klapper, Laeven, and Rajan resource allocation. This raises the question (2006) and Djankov et al. (2007b) show that of whether one can take the next step and such regulations stifle entry. Ciccone and connect legal origins to aggregate economic Papaioannou (2006c) report that countries growth. This, however, has proved to be dif- with lower entry regulations see more entry ficult, as we explain next. in industries that experience expanding Mahoney (2001) shows that, in the recent global demand and technology shifts. period, common law countries have grown Two other papers use our variables to faster than French legal origin countries. examine the effects of contract enforcement Mahoney is indeed correct: during 1960– on the structure of production. Nathan Nunn 2000, compared to the common law coun- (2007) finds that countries with good contract tries, GDP per capita in the French legal origin countries has grown about 0.6 percentage point slower per year. On the other hand, 10 Other papers in this area include Stijn Claessens and Luc Laeven (2003), Matías Braun (2003), Raymond German legal origin countries grew faster Fisman and Inessa Love (2004), Thorsten Beck, Asli than the common law countries. Depending Demirguc-Kunt, and Levine (2005), Enrico C. Perotti on the specification, there are similar differ- and Paolo F. Volpin (2004), Antonio Ciccone and Elias Papaioannou (2006a, 2006b), Alain De Serres et al. ences in the growth rates of GDP per worker, (2006), and Geert Bekaert et al. (2007). capital stock per worker, and productivity. 302 Journal of Economic Literature, Vol. XLVI (June 2008)

These results, however, are not particu- The evidence also shows that legal origins larly robust. The growth effects of legal ori- influence patterns of growth within indus- gins become weaker once we control for a tries, but it is less clear that legal origins measure of human capital, namely average predict aggregate growth. The last finding years of schooling in 1960—a standard con- resonates with the obvious observation made trol in such regressions. Indeed, throughout by LLSV (1998) that countries like France the 1960–2000 period, years of schooling and Belgium achieved high living standards are sharply higher in common law countries despite their legal origin. One possible expla- than in French legal origin ones, even hold- nation of the aggregate growth evidence is ing per capita income constant. Interestingly, that civil law countries have found compen- Jacek Rostowski and Bogdan Stacescu (2006) sating mechanisms to overcome the bag- argue that legal origins should enter the gage of their legal tradition in the long run. growth equation precisely through educa- Another possibility is that the last forty years tion because England pursued more enlight- have been unrepresentative and that, in the ened educational policies in its colonies than long run, there are periods that advantage did France. French colonial education was civil law regimes (such as state-led growth). largely guided by the idea of assimilation, We do not know which of these, or some with French textbooks, French teachers, other explanation, is correct.11 and instruction in French. The British, in All this evidence leaves us with a major contrast, adapted colonial education to local question: why do legal origins matter, and conditions and taught in vernacular. This why do they matter in such a pervasive way is a very original theory, and we hope it is for both rules and economics outcomes? developed. What are the historical and structural differ- The most obvious potential channel of ences among common and civil law countries influence of legal origins on growth is finan- that have such pervasive consequences for cial development, since legal origins have both the specific legal and regulatory rules such strong effects on finance. Using legal and major economic outcomes? We attempt origins as instruments, Beck, Levine, and to answer this question in the next section. Loayza (2000) and Levine, Loayza, and Beck (2000) find that private debt market 4. Explaining the Facts development is a statistically significant and quantitatively important predictor of growth. The correlations documented in the pre- Again, however, one needs to be careful, vious section require an explanation. LLSV both because (as we argued above) the exclu- (1997, 1998) do not advance such an expla- sion restriction is unlikely to be satisfied and nation, although in a broader study of gov- because the results are often sensitive to the ernment institutions, LLSV (1999) follow inclusion of other variables, such as alterna- Friedrich A. Hayek (1960) and suggest that tive measures of human capital. common law countries are more protective In sum, there is by now a great deal of of private property than French legal origin evidence that legal origins influence legal ones. In the ensuing years, many academ- rules and regulations, which in turn have ics, ourselves included, used the historical substantial impact on important economic narrative to provide a theoretical founda- outcomes—from financial development, to tion for the empirical evidence (see Glaeser unemployment, to investment and entry, to the size of unofficial economy, to international trade. Much of this evidence suggests 11 We note, however, that the evidence on the relationship between institutions and aggregate growth more gen- that common law is associated with better erally, which seemed substantial a few years ago, has been economic outcomes than French civil law. crumbling (see Edward L. Glaeser et al. 2004). La Porta, Lopez-de-Silanes, and Shleifer: Economic Consequences of Legal Origins 303 and Shleifer 2002, Djankov et al. 2003a, the ability of appellate common law courts and Mulligan and Shleifer 2005b). In this to make legal rules, thereby becoming an section, we begin with the alternative his- independent source of legal change separate torical explanations and then try to revise, from Parliament. Judicial independence and synthesize, and advance previous theoretical law-making powers in turn made judging a accounts into the Legal Origins Theory.12 highly attractive and prestigious occupation. In contrast, the French judiciary was 4.1 Revolutionary Explanations largely monarchist in the eighteenth cen- The standard explanation of the differ- tury (many judges bought offices from the ences between common law and French king) and ended up on the wrong side of civil law in particular, and to a lesser extent the French Revolution. The revolutionaries German law, focuses on seventeenth–nine- reacted by seeking to deprive judges of inde- teenth century developments (Merryman pendence and law-making powers, to turn 1969, Zweigert and Kötz 1998, Daniel them into automata in Napoleon’s felicitous Klerman and Mahoney 2007). According to phrase. Following Charles M. de Secondat this theory, the English lawyers were on the Montesquieu’s (1748) doctrine of separation same winning side as the property owners in of powers, the revolutionaries proclaimed the Glorious Revolution, and in opposition to legislation as the sole valid source of law the Crown and to its courts of royal preroga- and explicitly denied the acceptability of tive. As a consequence, the English judges judge-made law. “For the first time, it was gained considerable independence from the admitted that the sovereign is capable of Crown, including lifetime appointments in defining law and of reforming it as a whole. the 1701 Act of Settlement. A key corollary True, this power is accorded to him in order of such independence was the respect for to expound the principles of natural law. private property in English law, especially But as Cambaceres, principal legal adviser against possible encroachments by the sover- to Napoleon, once admitted, it was easy to eign. Indeed, common law courts acquired change this purpose, and legislators, outside the power to review administrative acts: of any consideration for “natural laws” were the same principles applied to the depriva- to use this power to transform the basis of tion of property by public and private actors society” (David and Brierley 1985, p. 60). (Mahoney 2001, p. 513). Another corollary is Hayek (1960) traces the differences respect for the freedom of contract, includ- between common and civil law to distinct ing the ability of judges to interpret contracts conceptions of freedom. He distinguishes without a reference to public interest (Pistor two views of freedom directly traceable to 2006). Still another was the reassertion of the predominance of an essentially empiri- cist view of the world in England and a rationalist approach in France: “One finds 12 Legal Origins Theory is intimately related to the the essence of freedom in spontaneity and discussion of the varieties of capitalism, which (typically the absence of coercion, the other believes in the context of the OECD economies) distinguishes it to be realized only in the pursuit and between liberal and coordinated market economies, the latter having firms that “depend more heavily on non- attainment of an absolute social purpose; market relationships to coordinate their endeavors with one stands for organic, slow, self-conscious other actors to construct their core competencies” (Peter growth, the other for doctrinaire deliberate- A. Hall and David Soskice 2001, p. 8). As Pistor (2006) points out, all the liberal market economies in the OECD ness; one for trial and error procedure, the are common law countries, and all the coordinated ones other for the enforced solely valid pattern” are civil law ones. The literature on the variety of capital- (p. 56). To Hayek, the differences in legal isms has long looked for an objective measure of different types; perhaps it should have looked no further than legal systems reflect these profound differences in origins. philosophies of freedom. 304 Journal of Economic Literature, Vol. XLVI (June 2008)

To implement his strategy, Napoleon pro- Over decades, new French courts were cre- mulgated several codes of law and procedure ated, and they as well as older courts increas- intended to control judicial decisions in all ingly became involved in the interpretation circumstances. Judges became bureaucrats of codes, which amounted to the creation employed by the State; their positions were of new legal rules. Even so, the law-making seen as largely administrative, low-prestige role of French courts was never explicitly occupations. The ordinary courts had no acknowledged, and never achieved the scope authority to review government action, ren- of their English counterparts. dering them useless as guarantors of prop- Perhaps more importantly for cross-coun- erty against the state. try analysis, the developing countries into The diminution of the judiciary was also which the French legal system was trans- accompanied by the growth of the admin- planted apparently adhered faithfully to istrative, as Napoleon created a huge and the Napoleonic vision. In those countries, invasive bureaucracy to implement the state’s judges stuck to the letter of the code, resolv- regulatory policies (Isser Woloch 1994). ing disputes based on formalities even when Under Napoleon, “the command orders the law needed refinement. Luca Enriques were now unity of direction, hierarchically (2002) shows that, even today, Italian mag- defined participation in public affairs, and istrates let corporate insiders expropriate above all the leading role assigned to the investors with impunity, as long as formally executive bureaucracy, whose duty was to correct corporate decision-making proce- force the pace and orient society through the dures are followed. In the transplant and to application from above of increasingly com- some extent even in the origin countries, leg- prehensive administrative regulations and islation remained, at least approximately, the practices” (Stuart Woolf 1992, p. 95). sole source of law, judicial law making stayed Merryman (1969) explains the logic of cod- close to nonexistent, and judges retained their ification: “If the legislature alone could make bureaucratic status. Merryman (1996) memo- laws and the judiciary could only apply them rably writes that “when the French exported (or, at a later time, interpret and apply them), their system, they did not include the infor- such legislation had to be complete, coherent, mation that it really does not work that way, and clear. If a judge were required to decide and failed to include the blueprint of how it a case for which there was no legislative pro- actually does work” (p. 116). This analysis vision, he would in effect make law and thus of the “French deviation” may explain the violate the principle of rigid separation of considerable dynamism of the French law as powers. Hence it was necessary that the legis- compared to its transplant countries, where lature draft a code without gaps. Similarly, if legal development stagnated. The French there were conflicting provisions in the code, emphasis on centralized bureaucratic control the judge would make law by choosing one may have been the most enduring influence rather than another as more applicable to the of transplantation. situation. Hence there could be no conflicting Although less has been written about provisions. Finally, if a judge were allowed to German law, it is fair to say that it is a bit decide what meaning to give to an ambiguous of a hybrid (John P. Dawson 1960, 1968, provision or an obscure statement, he would Merryman 1969, Zweigert and Kötz 1998). again be making law. Hence the code had to Like the French courts, German courts be clear” (p. 30). had little independence. However, they had Yet, according to Merryman (1996), greater power to review administrative acts, Napoleon’s experiment failed in France, as and jurisprudence was explicitly recognized the notion that legislation can foresee all as a source of law, accommodating greater future circumstances proved unworkable. legal change. La Porta, Lopez-de-Silanes, and Shleifer: Economic Consequences of Legal Origins 305

The historical analysis has three key impli- (2007) argue in the spirit of Benjamin N. cations for the economic consequences of Cardozo (1921) and Julius Stone (1985) that legal origins. First, the built-in judicial inde- the central strategy of judicial law mak- pendence of common law, particularly in the ing is distinguishing cases from precedents, cases of administrative acts affecting indi- which has an unintended benefit that the viduals, suggests that common law is likely to law responds to a changing environment. be more respectful of private property and The quality of law improves on average even contract than civil law. when judges pursue their policy preferences; Second, common law’s emphasis on judi- law making does not need to be benevolent. cial resolution of private disputes, as opposed The theoretical research on the adaptability to legislation, as a solution to social problems, of common law has received some empirical suggests that we are likely to see greater support in the work of Beck, Demirguc- emphasis on private contracts and orderings, Kunt, and Levine (2003), who show that the and less emphasis on government regulation, acceptability of case law variable from La in common law countries. To the extent that Porta et al. (2004) captures many of the ben- there is regulation, it aims to facilitate pri- efits of common law for financial and other vate contracting rather than to direct par- outcomes. On the other hand, a recent study ticular outcomes. Pistor (2006) describes of the evolution of legal doctrine governing French legal origin as embracing socially construction disputes in the United States conditioned private contracting, in contrast over the period of 1970–2005 finds little evi- to common law’s support for unconditioned dence either that legal rules converge over private contracting. Damaska (1986) calls time, or that they move toward efficient solu- civil law “policy-implementing,” and com- tions (Anthony Niblett, Posner, and Shleifer mon law “dispute resolving.” 2008). Third, the greater respect for jurispru- 4.2 Medieval Explanations dence as a source of law in the common law countries, especially as compared to The idea that the differences between the French civil law countries, suggests that common and civil law manifest themselves common law will be more adaptable to the for the first time during the Enlightenment changing circumstances, a point emphasized seems a bit strange to anyone who has heard by Hayek (1960) and more recently Levine of Magna Carta. Some of the differences (2005). These adaptability benefits of com- were surely sharpened, or even created, by mon law have also been noted by scholars the English and the French Revolutions. in law and economics (Richard Posner 1973, For example, judges looked to past judicial Paul H. Rubin 1977, George L. Priest 1977, decisions for centuries in both England Giacomo A. M. Ponzetto and Patricio A. and France prior to the revolutions (Gino Fernandez forthcoming), who have made Gorla and Luigi Moccia 1981). However, the the stronger claim that, through sequential explicit reliance on precedent as a source of decisions by appellate courts, common law law (and the term precedent itself) is only a evolves not only for the better but actually seventeenth and eighteenth century develop- toward efficient legal rules. The extreme ment in England (Harold J. Berman 2003). hypothesis of common law’s efficiency is Likewise, the denial of the legal status of difficult to sustain either theoretically or precedent in France is a Napoleonic rather empirically, but recent research does suggest than an earlier development. that the ability of judges to react to changing But in other respects, important differ- circumstances—the adaptability of common ences predate the revolutions. The English law—tends to improve the law’s quality over judges fought the royal prerogative, used time. For example, Gennaioli and Shleifer juries to try criminal cases, and pressed the 306 Journal of Economic Literature, Vol. XLVI (June 2008) argument that the King (James) was not Edward Coke’s early seventeenth century above the law early in the seventeenth cen- insistence that the king is not above the law tury. They looked down on the inquisitorial is neither a continental nor a postrevolution- system that flourished on the Catholic con- ary phenomenon. The Glorious Revolution tinent. In light of such history, it is hard to eliminated the courts of royal prerogative, sustain the argument that the differences and eventually enshrined the principles between common and civil law only emerged of judicial independence in several acts of through revolutions. Parliament. Several distinguished legal historians, Glaeser and Shleifer (2002) present a including Dawson (1960) and Berman theoretical model intended to capture this (1983), trace the divergence between French comparative twelfth and thirteenth century and English law to a much earlier period, narrative, but with an economic twist. They namely the twelfth and thirteenth centuries. argue that England was a relatively peaceful According to this view, the French Crown, country during this period, in which decen- which barely had full control over the Ile tralized dispute resolution on the testimony de France let alone other parts of France, of independent knights (juries) was efficient. adopted the bureaucratic inquisitorial sys- France was a less peaceful country, in which tem of the Roman Church as a way to unify high nobles had the power to subvert decen- and perhaps control the country. The system tralized justice, and hence a much more cen- persisted in this form through the centuries, tralized system, organized, maintained, and although judicial independence at times protected by the sovereign, was required to increased as judges bought their offices from administer the law. Roman law provided the the Crown. Napoleonic bureaucratization backbone of such a system. This view sees the and centralization of the judiciary is seen as developments of seventeenth and eighteenth a culmination of a centuries-old tug of war centuries as reinforcing the structures that between the center and the regions. evolved over the previous centuries. England, in contrast, developed jury tri- Regardless of whether the revolutionary or als as far back as the twelfth century and the medieval story is correct, they have similar enshrined the idea that the Crown cannot empirical predictions. In the medieval narra- take the life or property of the nobles with- tive, as in the revolutionary one, common law out due process in the Magna Carta in 1215. exhibits greater judicial independence than The Magna Carta stated: “No freeman shall civil law, as well as greater sympathy of the be taken or imprisoned or disseised or out- judiciary toward private property and con- lawed or exiled or in any way ruined, nor tract, especially against infringements by the will we go or send against him, except by the executive. In both narratives, judicial law mak- lawful judgment of his peers or by the law of ing and adaptation play a greater role in com- the land.” The Magna Carta established the mon than in civil law, although this particular foundations of the English legal order. As in difference might have been greatly expanded France, such independence was continuously in the Age of Revolutions. The historical challenged by the Crown, and the courts of accounts may differ in detail, but they lead to royal prerogative, subordinate to the Crown, the same place as to the fundamental features grew in importance in the sixteenth century, of law. These features, then, carry through the during the reign of Queen Elizabeth. Yet, as process of transplantation, and appear in the we indicated earlier, even during Elizabeth’s differences among legal families. reign, and much more so during those of 4.3 Legal Origins Theory James I and Charles I, Parliament and courts repeatedly reaffirmed the rights of indi- Legal Origins Theory has three basic viduals against royal demands. Chief Judge ingredients. First, regardless or whether the La Porta, Lopez-de-Silanes, and Shleifer: Economic Consequences of Legal Origins 307 medieval or the revolutionary narrative is the was so different from what the French or the correct one, by the eighteenth or nineteenth Spaniards brought, and that had such per- centuries England and Continental Europe, sistent consequences? They key point is that particularly France, have developed very dif- transplantation involves not just specific legal ferent styles of social control of business, and rules (many of which actually change later) institutions supporting these styles. Second, but also legal institutions (or which judicial these styles of social control, as well as legal independence might be the most important), institutions supporting them, were trans- human capital of the participants in the legal planted by the origin countries to most of system, and crucially the strategy of the law the world, rather than written from scratch. for dealing with new problems. Successive Third, although a great deal of legal and generations of judges, lawyers, and politi- regulatory change has occurred, these styles cians all learn the same broad ideas of how have proved persistent in addressing social the law and the state should work. The legal problems. system supplies the fundamental tools for Djankov et al. (2003a) propose a particu- addressing social concerns and it is that sys- lar way of thinking about the alternative legal tem, as defined by Zweigert and Kötz, with styles. All legal systems seek to simultane- its codes, distinctive institutions, modes of ously address twin problems: the problem of thought and even ideologies, that is very slow disorder or market failure and the problem of to change. dictatorship or state abuse. There is an inher- The fact that legal system is slow to change ent trade-off in addressing these twin prob- does not mean that specific legal rules and lems: as the state becomes more assertive in regulations never change. As we discuss dealing with disorder, it may also become below, governments in both common and more abusive. We can think of the French civil law countries entered many new spheres civil law family as a system of social control of social control in the twentieth century, but of economic life that is relatively more con- typically in ways consistent with their legal cerned with disorder, and relatively less with traditions. In some more stable areas of law, dictatorship, in finding solutions to social and such as legal procedure, there is sometimes economic problems. In contrast, the common a great deal of rigidity even in the specific law family is relatively more concerned with rules. Aron Balas et al. (2008) compute the dictatorship and less with disorder. These index of the formalism of legal procedure, are the basic attitudes or styles of the legal considered in table 3, for twenty common and regulatory systems, which influence the law and twenty civil law countries over the “tools” they use to deal with social concerns. period 1950–2000. Consistent with Djankov Of course, common law does not mean anar- et al. (2003b), they find that formalism is chy, as the government has always maintained higher in common than in civil law countries a heavy hand of social control; nor does civil in 2000, but also in 1950. Perhaps more sur- law mean dictatorship. Indeed, both systems prisingly, formalism is extraordinarily stable. seek a balance between private disorder and Among common law countries, the average public abuse of power. But they seek it in of the ratio of 2000 to 1950 procedural for- different ways: common law by shoring up malism is 0.90; among civil law countries, markets, civil law by restricting them or even the average of this ratio is 1.10. The data replacing them with state commands. reflects significant persistence of the differ- Legal Origins Theory raises the obvious ences among legal origins, with no evidence question of how the influence of legal ori- of convergence. gins has persisted over the decades or even The reader might wonder at this point centuries. Why so much hysteresis? What is whether Legal Origin Theory simply identi- it that the British brought on the boats that fies legal families with different “ideologies” 308 Journal of Economic Literature, Vol. XLVI (June 2008) or “cultures.” To the extent that ideologies or According to Randall K. Morck and Lloyd cultures refer to the beliefs about how the Steier (2005), “the responses of the Dutch, law should deal with social problems, Legal Italian, Japanese, and Swedish governments Origin Theory clearly accepts the view that to the financial crisis of the 1920s and 1930s ideologies and cultures are crucial for the were to substitute various mechanisms of persistent influence of legal families. But state-controlled capital allocation for their the central point is that the reason for per- stock markets” (p. 39). “A similar succes- sistence is that the beliefs and ideologies sion of financial manias, panics, and crises become incorporated in legal rules, institu- in Britain, Canada, and the United States tions, and education and, as such, are trans- ultimately strengthened shareholder rights” mitted from one generation to the next. It is (p. 39). The United States responded to the this incorporation of beliefs and ideologies Great Depression by introducing securi- into the legal and political infrastructure that ties regulation and deposit insurance. These enables legal origins to have such persistent strategies intended to rehabilitate and sup- consequences for rules, regulations, and eco- port markets, not to replace them. Even nomic outcomes. Roosevelt’s most radical aspirations fell short The account of legal origins has implica- of nationalization. This contrast between the tions for how the government responds to replacement of markets by state solutions in new needs both across activities and over civil law countries and the rehabilitation of time. Essentially, the toolkit of civil law markets in common law countries appears features more prominently such policies quite pervasive. as nationalization and direct state control; One form of government reaction to new the toolkit of common law features more circumstances is the expansion of public litigation and market-supporting regulation. involvement into new spheres. Economic his- Mulligan and Shleifer (2005b) argue that, torians have sometimes argued that, because by specializing in such “policy-implement- legal origins have differed for centuries, one ing” solutions, the civil law system tends to should observe equally sharp differences in expand the scope of government control to rules and regulations in the nineteenth cen- new activities when a need arises. Perhaps tury as well. This, of course, does not follow. the best known historical example of this is To the extent that public intervention in mar- the vast expansion of military conscription kets changes over time and responds to social in France under Napoleon, made possible needs or political imperatives, laws and regu- by the already existing presence of govern- lations will change as well, but in ways that ment bureaucracy that could administer the are consistent with legal traditions. Both draft in every French village (Woloch 1994). labor laws and securities laws are creatures of Because the state’s presence on the ground the twentieth century; they were introduced is less pervasive under the common law, it as a response to perceived social needs. Yet, tends not to rely as extensively on adminis- as the evidence in section 3 shows, these laws trative solutions and more on “market-sup- took different forms in countries from differ- porting” or “dispute-resolving” ones. ent legal traditions, consistent with broad Likewise, one can argue that, when the strategies of how the state intervenes. market system gets into trouble or into a Ahlering and Deakin (2005) elaborate crisis, the civil law approach is to repress it this point in the context of labor laws. They or to replace it with state mandates, while argue that the current differences between the common law approach is to shore it up. the labor laws of Britain and Continental One place to see this might be the regula- Europe can be traced to the differences in tory response to the Great Depression and the ways common and civil law systems saw financial crises of the twentieth century. the role of the enterprise as far back as the La Porta, Lopez-de-Silanes, and Shleifer: Economic Consequences of Legal Origins 309

Industrial Revolution. Common law saw the in a country might bring into one domain a enterprise as an unencumbered property or set of tools that has been used in another, the employer, with the workers relegated to based on either philosophical outlook or a contractual claims on the surplus from pro- desire for consistency, with adverse results. duction. In contrast, civil law saw property For example, the strategy of extensive inter- and responsibility as two sides of the same locutory appeals that is standard in a civil law coin. Thus, the support provided by the system can slow down a bankruptcy proceed- legal system to the freedom of contract and ing, where time is of the essence, and lead property rights was counterbalanced in the to a large loss of value (Djankov et al. 2006). French tradition by the exercise of public Third, additional inefficiencies may arise power for the protection of workers, and in from transplantation. A regulatory approach the Germanic tradition by the communitar- that works well in France may become little ian conception of the enterprise. Ahlering but a source of corruption and delay in a poor and Deakin suggest that these differences in West African country. As we show in section “legal cultures” persist even today. 8, an understanding of regulatory inefficien- Crucially, the Legal Origins Theory does cies afforded by the Legal Origins Theory not say that common law always works bet- can form the basis of reform. ter for the economy. As Glaeser and Shleifer To reiterate, no country exhibits a system (2002, 2003) show, regulation and state con- of social control that is an ideal type; all trol may well be efficient responses to dis- countries mix the two approaches. Common order, where common-law solutions fail to law countries are quite capable of civil law sustain markets.13 Indeed, all countries effi- solutions, and vice versa. Nonetheless, the ciently resort to the quintessentially civil law empirical prediction of the Legal Origin solution of planning in time of war and add Theory is that the differences between legal good dollops of state intervention and control origins are deep enough that we observe in response to major threats to order, such them expressed in the different strategies as terrorism. Glaeser and Shleifer (2003) of social control of economic life even after interpret the early twentieth century rise of centuries of legal and regulatory evolution. the regulatory state in the United States as Perhaps because the legal system is such a an efficient response to the subversion of the difficult-to-change element of social order, justice system by large corporations. supported by legal institutions, human capi- Legal Origin Theory also does not imply tal, and expectations, legal origins survive that the outcomes we observe are always or both time and transplantation. This, we sub- even typically efficient within a given legal mit, is what gives them explanatory power. family. There are several reasons for ineffi- 4.4 Interpretation of the Evidence ciency, quite aside from interest group politics. First, at the most basic level, the tools used In interpreting the evidence in light of the by a legal system may lead to outcomes that Legal Origins Theory, it is easiest to pro- are worse than the initial problem. Excessive ceed in reverse: from judicial independence regulation of entry in civil law countries is a to government regulation to finance. The good example. Second, courts or legislators evidence on judicial independence directly confirms the predictions. As we saw in table 3, compared to French civil law, common 13 Critically from the perspective of the developing law countries have less formalized contract countries, Glaeser and Shleifer (2003) show that, when all enforcement, longer constitutional tenure mechanisms of state action can be subverted by private of Supreme Court judges (a direct indicator interests, the best policy might be to do nothing and leave the markets alone, even in the presence of substantial of independence), and greater recognition market failure. of case law as a source of law, which Beck, 310 Journal of Economic Literature, Vol. XLVI (June 2008)

Demirguc-Kunt, and Levine (2003) use as are higher in common law countries, predict an indicator of adaptability. Also consistent financial development. Is this evidence con- with the Legal Origins Theory, these char- sistent with Legal Origins Theory? acteristics of legal systems predict both the The answer, we believe, is yes. Common efficiency of contract enforcement—mea- law countries succeed in finance because sured objectively and subjectively—and the their regulatory strategies seek to sustain security of property rights. markets rather than replace them. Returning The evidence on government regulation to the examples of securities regulation and is consistent with the Legal Origins Theory of the often-statutory regulation of self-deal- as well. The historical evidence suggests that ing transactions, the statutory requirements civil law countries are more likely to address of disclosure originate in the common law social problems through government own- of fiduciary relationships. Market forces on ership and mandates, whereas common law their own are not strong enough, and contract countries are more likely to do so through claims not cheap enough to pursue, to pro- private contract and litigation. When com- tect investors from being cheated. A regula- mon law countries regulate, we expect their tory framework that offers and enforces such regulation to support private contracting protection, and makes it easier for investors rather than dictate outcomes. We see those to seek legal remedies to rectify the wrongs differences across a broad range of activi- even when doing so relies on public action, ties—from entry and labor regulation to allows more extensive financial contracting. recruitment of armies. We also see that civil The form of statutory protection of investors law countries exhibit heavier government in common law countries, as compared to ownership of both the media and banks. civil law countries, is consistent with Legal The theory is also consistent with the evi- Origins Theory. Finance falls into line with dence on finance. The better protection of other evidence. both shareholders and creditors in common law countries than especially in the French 5. Legal Origins and Culture civil law ones is consistent with the principal historical narrative of the greater security of In this section and the next two, we address private property and better contract enforce- the central criticism of research on legal ori- ment under common law. Moreover, as noted gins: that they are merely proxy for other by Beck, Demirguc-Kunt, and Levine (2003), factors influencing legal rules and outcomes. financial markets may be an area where the The three factors we consider are culture, adaptability of judge-made rules, as exem- history, and politics. We stress from the out- plified by the American Delaware courts, is set that it is not our position, nor our objec- especially beneficial. tive in these sections, to show that culture, Mark J. Roe (2006) points out that many of history, or politics are unimportant for legal the legal rules protecting investors in common and regulatory rules. All of them are clearly law countries are statutory rather than judge- important, and there is a great deal of evi- made, so in many crucial respects regulation dence confirming their roles (see, e.g., Luigi rather than judge-made law is responsible for Guiso, Sapienza, and Zingales 2004, 2006 investor protection. Securities laws in gen- on the role of culture). Our point is rather eral, and disclosure rules in particular, which to establish, as clearly as possible, a much La Porta et al. (2006) show to provide some more modest yet central to the Legal Origins of the most effective investor protections, are Theory proposition that legal origins are not entirely statutory. Howell Jackson and Roe proxies for something else. (2007) further argue that the budgets and We begin with culture, which has been staffing levels of securities regulators, which considered as a potential explanation of the La Porta, Lopez-de-Silanes, and Shleifer: Economic Consequences of Legal Origins 311 evidence on legal origins. Stulz and Rohan (2007). Although the papers differ in detail, Williamson (2003) suggest that, in light of they have a common theme, so we take the the hostility of some of the religious tradi- liberty of providing an integrated account. tions to lending on interest, religion may be a Also, while some of the papers cover devel- more fundamental determinant of legal rules oping countries, virtually all of them deal governing creditor protection than legal tra- with Western Europe, or the Wealthy West, dition. Amir N. Licht, Chanan Goldschmidt, a point we return to below. and Shalom H. Schwartz (2005) present a According to the political theories, some- more sweeping case, using psychological time in the middle of the twentieth century, measures of cultural attitudes to predict legal Continental European countries formed alli- rules. So are legal origins merely proxies for ances between families that controlled firms cultural variables? and (typically organized) labor. In many cases, Table 4 shows the facts. First, religion is not these alliances were a response to crises from nearly as important a determinant of credi- hyperinflation, depression, or defeat in war. tor rights as legal origin (see Djankov et al. These political alliances sought to win elec- 2007). Second, most indices of cultural atti- tions in order to secure the economic rents of tudes do not influence creditor rights holding the insiders, and to keep them from the “out- legal origin constant. There is some evidence siders,” such as unorganized labor, minor- that a nation’s masculinity (defined as “the ity shareholders, corporate challengers, or degree to which the society reinforces, or potential entrants. When these alliances won does not reinforce, the traditional masculine elections, they wrote legal rules to benefit work role model of achievement, control, and themselves. The families secured poor pro- power”) is not conducive to creditor protec- tection of outside shareholders, so they could tion, while belief in the independence of hold on to the private benefits of control. children is, but neither variable makes much Labor got social security and worker protec- of a dent in the effect of legal origin on credi- tion laws, which maintained employment tor rights. and wages of the insiders. Both the families Cultural variables, then, do not make much and labor secured the laws protecting them of a dent in the explanatory power of legal against product market competition, such as origins. We note, however, that the notions regulation of entry. The legal rules observed of culture we consider focus on religion and in the data, then, are outcomes of this demo- broad social attitudes. One can alternatively cratic process and not of any “permanent” include in culture beliefs about the law, regu- conditions, such as legal origins. lation, and the role of the state. This theory The political story is part of a broader nar- of culture is of course broadly consistent with rative of Continental European history in our views. the twentieth century, in which the response to crisis is variously characterized by the rise of proportional representation (Alberto 6. Legal Origins and Politics Alesina and Glaeser 2004, Torsten Persson A broader challenge to the explanatory and Guido Tabellini 2003), socialist poli- power of legal origins has been posed by polit- tics (Alesina and Glaeser 2004), and social ical theories of corporate finance. There are democracy (Roe 2000). The United States now many papers in this literature, including was spared these political developments and, Martin Hellwig (2000), Rajan and Zingales therefore, did not get the laws adopted on the (2003), Pagano and Volpin (2005, 2006), Continent. Some implications of these theo- Perotti and Ernst-Ludwig Von Thadden ries are broadly consistent with the evidence: (2006), and Roe (2000, 2006), and even a countries that have strong shareholder pro- recent survey by Stephen Haber and Perotti tection indeed have weak protection of labor 312 Journal of Economic Literature, Vol. XLVI (June 2008)

TABLE 4 Creditor rights, Culture, and Legal Origin

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11)

Dependent Variable: Creditor’s Rights in 2002

% catholic –0.2561 [0.2283] Power Distance 0.0013 Index [0.0096] Individualism –0.0073 [0.0079] Uncertainty –0.0100 Avoidance Index [0.0062] Masculinity –0.0198c [0.0099] % agree child –0.4113 obedience is [0.7531] important % agree child 1.3655b independence is [0.6010] important % agree parents –0.5432 must do their [0.9007] best for children % agree that –1.3109 parents must be [0.8417] respected regardless % agree family life 0.0726 is very important [1.2854] % agree strangers 0.6841 can generally be [0.8051] trusted French Legal –0.7585a –0.8578b –0.9374a –0.4519 –1.0133a –0.8542b –0.7470b –0.8351b –0.7563b –0.7979b –0.8246b Origin [0.2383] [0.3431] [0.3417] [0.3917] [0.3669] [0.3361] [0.3447] [0.3519] [0.3468] [0.3534] [0.3302] German Legal –0.1320 –0.5119 –0.5528 –0.2347 –0.2764 –0.2798 –0.3270 –0.2318 –0.1893 –0.1542 –0.2004 Origin [0.2603] [0.4472] [0.4197] [0.4485] [0.4253] [0.3913] [0.3470] [0.3816] [0.3519] [0.3960] [0.3623] Scandinavian Legal –1.0091b –0.8831 –0.9013 –0.9597c –1.7406b –0.7378 –0.9349c –0.6631 –1.0181b –0.6091 –0.8950 Origin [0.4804] [0.5768] [0.5625] [0.5382] [0.6865] [0.5724] [0.5316] [0.5773] [0.4938] [0.5908] [0.6500] Log(GDP per 0.2415a 0.2573 0.3920c 0.2770 0.2480 –0.0823 –0.0991 –0.0835 –0.1771 –0.0685 –0.0780 capita in 2002) [0.0893] [0.2349] [0.1956] [0.1856] [0.1887] [0.1249] [0.1214] [0.1225] [0.1409] [0.1246] [0.1190] Constant 0.2311a 0.0177 –0.7691 0.3359 1.2775 3.3971a 2.6663b 3.6212a 5.0362a 2.9833c 2.9900a [0.0882] [2.4440] [1.6375] [1.5286] [1.7662] [1.2091] [1.1083] [1.2349] [1.6400] [1.5731] [1.0425]

Observations 131 52 52 52 52 73 73 71 73 72 73 R-squared 21% 15% 16% 17% 20% 14% 19% 14% 17% 13% 15%

Notes: a Significant at the 1 percent level. b Significant at the 5 percent level. c Significant at the 10 percent level. and low regulation of entry. The suggestion on the exact paper—social democracy, left- of this research is that legal origin enters the ist politics, or proportional representation. various regressions summarized in section 3 If politics were appropriately controlled for spuriously, with French (and German) legal in the regressions, legal origin would not origins serving as proxies for—depending matter. La Porta, Lopez-de-Silanes, and Shleifer: Economic Consequences of Legal Origins 313

The political story is plausible, since we see collected by Botero et al. (2004) for eighty- social democracies in Continental Europe five countries, is the share of years between but not in the United States. For this reason, 1928–95 when the chief executive and the we consider it in some detail. We do so in largest party in the legislature were leftist or three steps. First, we briefly look at the logic centrist. The third variable is union density, of the story. Second, we show what happens defined as the percentage of the total work when some of the political variables pro- force affiliated to labor unions in 1997. The posed in this literature are actually added regressions in table 5 cover the whole sample to the regressions. Third, we test an implica- and are not confined to Western Europe or tion of the available political models, namely the OECD. that the formation of laws is a consequence For all three variables, the results in of democratic politics.14 This prediction table 5 are straightforward. Political variables implies, most immediately, that the relation- explain the variation in legal rules only occa- ship between legal origins and laws should sionally. In contrast, legal origins continue to not hold outside democracies. explain the variation even with political vari- With respect to the logic of the story, it ables in the regression, and the difference is hard to understand why organized labor between common law and French civil law accepts rules that facilitate the diversion of remains highly statistically significant. This corporate wealth, or tunneling—something is true for all three political variables aim- we see on a fairly large scale in, say, Italy or ing to get at the political explanation of legal Belgium. We can see the argument for the rules. While each political variable is surely Swedish system, in which the leading fam- measured with error, and our specifications ilies stay in control but are kept on a tight surely do not capture the full subtlety of leash through norms and regulations and the political theories, political variables are certainly not allowed to expropriate inves- rarely significant. In contrast, legal origins tors. Sweden indeed has a valuable stock are consistently significant, even with politi- market and low private benefits of control. cal variables in the regression. It is harder to accept the notion that orga- We next ask whether legal origins only nized labor endorses tunneling of corporate have an effect in democracies, which wealth, since presumably such wealth could would be the case if they were proxies for be taxed or shared with the workers. the political sentiment of the majority. In But what do the data say? Table 5 pres- this scenario, legal origins would not pre- ents regressions of the legal and institutional dict legal rules in autocracies. In contrast, rules on three variables considered by the under Legal Origins Theory, they should political theories. The first one is propor- predict legal rules in autocracies as well. In tional representation, the form of democracy table 6, we focus on autocracies (countries seen as an adaptation to political demands of with a positive autocracy score from Adam labor in the early twentieth century (Alesina Przeworski et al. 2000). For nearly all our and Glaeser 2004, Persson and Tabellini variables, the differences between common 2003). We obviously run these regressions law and French legal origin remain signifi- for democracies only. The second variable, cant among autocracies. This result holds for other measures of nondemocratic government as well. We see this evidence as a 14 Haber and Perotti (2007) write: “Recent explana- direct rejection of the hypothesis that legal tions suggest that a democratic majority in countries origins are proxies for the political senti- hit by a major redistribution of wealth may shift in favor ment of the democratic majority. Political of low minority investor protection and less corporate restructuring and competition to protect established labor theories can perhaps be adjusted to incor- rents” (p. 4). porate autocracies but the data suggest that 314 Journal of Economic Literature, Vol. XLVI (June 2008)

TABLE 5 Legal Origin and Politics

Panel A: Legal Origin and Proportional Representation (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) Disclosure Govt Anti-Self- in Creditor Debt Ownership Labor Media Judicial Judicial Tenure Case Dealing Prospectus Rights Enforcement Banks Ln(Steps) laws ownership Draft Formalism Independence Judges Law French 20.2976a 20.2439a 21.0218b 212.4951c 0.3155b 0.8880a 0.2373a 0.0555 0.4590a 1.3404a 20.4281a 20.1939b 20.8993a Legal 30.0711430.0632430.4142437.0568430.1257430.1548430.0582430.0818430.1587430.3077430.0606430.0856430.08004 Origin German 20.3827a 20.2140b 20.1538 24.8148 0.2281c 0.8272a 0.1190 0.0583 0.6707a 0.9227a 20.2729a 20.3367a 20.1527 Legal 30.0698430.0835430.5264438.0829430.1234430.1594430.0847430.0584430.2336430.2388430.0665430.1130430.18014 Origin Scandinavian 20.3096a 20.1431b 20.8423 10.4188 0.2448 20.0602 0.3010a 0.1534c 0.6385b 0.7470c 20.0431 20.0798 0.0314 Legal 30.0782430.0644430.6446436.4080430.1935430.1854430.0629430.0817430.2718430.4158430.0459430.0726430.09114 Origin Proportional 20.0259 20.0678b 0.0286 26.0155a 0.0362 0.0153 0.0394c 20.0640c 0.0334 0.0706 0.0303 0.0352 0.0206 Represen- 30.0301430.0258430.1666432.1895430.0420430.0591430.0224430.0332430.0607430.1242430.0268430.0338430.04984 tation Ln(GDP per 0.0432 0.0477 0.0321 21.9633a 20.1590b 20.2857a 0.0166 20.1381a 20.0560 20.4443a 20.0301 0.0113 20.1145 capita) 30.0437430.0432430.2280434.2290430.0661430.0691430.0254430.0418430.0685430.1248430.0388430.0377430.07664 Constant 0.3283 0.4163 2.1746 2133.1648a 1.3978b 4.1702a 0.0967 1.4464a 0.6356 6.7233a 1.2535a 0.8605b 2.0535a 30.4268430.4453432.09614342.1285430.5318430.6242430.2423430.4384430.6184431.1629430.3476430.3489430.70734 Observations 39 34 45 41 45 38 38 38 53 49 35 35 35 R-squared 0.51 0.62 0.18 0.69 0.4 0.78 0.63 0.48 0.3 0.5 0.63 0.32 0.71

Panel B: Legal Origin and Power of the Left (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) Govt Anti-Self- Disclosure in Creditor Debt Ownership Media Judicial Tenure Dealing Prospectus Rights Enforcement Banks Ln(Steps) Labor Laws Ownership Draft Formalism Judges Case Law French Legal –0.3356a –0.3318a –0.9337a –18.2437a 0.3432a 0.7018a 0.2606a 0.0699 0.6459a 1.6322a –0.2234a –0.6417a Origin [0.0478] [0.0581] [0.2932] [5.7738] [0.0737] [0.0907] [0.0357] [0.0737] [0.1041] [0.2033] [0.0705] [0.1148] German Legal –0.3255a –0.2395b –0.2227 –12.9735b 0.3417a 0.5590a 0.2058a 0.0357 0.7322a 1.0687a –0.4615a –0.2718 Origin [0.0753] [0.0967] [0.3249] [5.4856] [0.0870] [0.1226] [0.0496] [0.0877] [0.1165] [0.2227] [0.1461] [0.2264] Scandinavian –0.2935a –0.2763a –0.7540 0.7023 0.0062 –0.1003 0.3365a 0.0189 0.7661a 0.5668b 0.0694 0.1708 Legal Origin [0.0605] [0.0659] [0.5658] [5.5078] [0.1179] [0.1552] [0.0529] [0.0680] [0.1358] [0.2564] [0.0501] [0.1127] Left Power –0.1518b –0.0248 –0.3157 –2.7732 0.3668a –0.1782 0.1212b 0.0787 0.2905a 0.1608 0.3011b –0.2380 [0.0727] [0.0966] [0.3662] [7.9727] [0.1127] [0.1302] [0.0598] [0.1091] [0.1454] [0.2493] [0.1259] [0.2236] Ln(GDP per 0.0665b 0.0596b 0.0752 20.7717a –0.1333a –0.2244a 0.0104 –0.1058a –0.0058 –0.2133b 0.0051 –0.0397 capita) [0.0274] [0.0284] [0.1302] [3.0620] [0.0336] [0.0417] [0.0173] [0.0330] [0.0425] [0.0813] [0.0275] [0.0537] Constant 0.1488 0.2419 1.9708 –127.0101a 1.1137a 3.7683a 0.1487 1.6322a 0.0687 1.6322a 1.1123a 1.6322a [0.2701] [0.2802] [1.2386] [31.9668] [0.2521] [0.3815] [0.1674] [0.3470] [0.3970] [0.7284] [0.2402] [0.5728] Observations 68 49 85 65 60 86 85 71 83 79 54 54 R-squared 0.5 0.45 0.17 0.64 0.49 0.62 0.45 0.23 0.49 0.51 0.35 0.4

Panel C: Legal Origin and Union Density (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) Govt Anti-Self- Disclosure in Creditor Debt Ownership Media Judicial Tenure Dealing Prospectus Rights Enforcement Banks Ln(Steps) Labor Laws Ownership Draft Formalism Judges Case Law

French Legal –0.3652a –0.3417a –1.1527a –14.1876b 0.3340a 0.6856a 0.2306a 0.0527 0.6006a 1.5479a –0.1952a –0.5835a Origin [0.0491] [0.0558] [0.3483] [5.8526] [0.0851] [0.1018] [0.0395] [0.0751] [0.1226] [0.2280] [0.0644] [0.1220] German Legal –0.3465a –0.2336a –0.4663 –9.8379c 0.3237a 0.5845a 0.1891a 0.0795 0.7490a 1.1105a –0.4707a –0.2458 Origin [0.0720] [0.0836] [0.3824] [5.3885] [0.1069] [0.1160] [0.0473] [0.0975] [0.1311] [0.2358] [0.1511] [0.2198] Scandinavian –0.2261a –0.1017 –1.1557 6.7209 0.0200 –0.2346 0.3363a 0.0438 0.8458a 0.9283b 0.0865 0.4935b Legal Origin [0.0813] [0.1063] [0.7030] [7.2616] [0.1813] [0.1845] [0.0633] [0.0945] [0.1745] [0.3883] [0.1882] [0.2430] Union Density –0.2786b –0.3567c 0.3122 –13.8014 0.2637 0.1718 0.0751 0.0371 0.0579 –0.4384 –0.3235 –0.7425c [0.1100] [0.1789] [0.7056] [9.9588] [0.2666] [0.2108] [0.0888] [0.1274] [0.1993] [0.5227] [0.3683] [0.3918] Ln(GDP per 0.0926a 0.0810a 0.0877 25.2792a –0.1830a –0.2432a –0.0107 –0.1139a –0.0623 –0.3230a 0.0912c 0.0175 capita) [0.0272] [0.0264] [0.1528] [3.3672] [0.0420] [0.0356] [0.0208] [0.0405] [0.0547] [0.0886] [0.0527] [0.0459] Constant –0.0729 0.1247 1.7383 –170.3391a 1.5766a 3.8136a 0.3932b 1.0863b 0.7448 5.6368a 0.2690 0.9510b [0.2576] [0.2366] [1.3894] [33.9071] [0.3010] [0.2857] [0.1956] [0.4153] [0.4974] [0.7204] [0.4138] [0.4098] Observations 64 49 70 58 58 71 70 61 68 69 51 51 R–squared 0.56 0.5 0.19 0.69 0.41 0.66 0.4 0.19 0.41 0.51 0.33 0.41

Notes: a Significant at the 1 percent level. b Significant at the 5 percent level. c Significant at the 10 percent level.

La Porta, Lopez-de-Silanes, and Shleifer: Economic Consequences of Legal Origins 315

TABLE 6 Legal Origin in Countries with Autocratic Governments

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) Govt Anti-Self- Disclosure in Creditor Debt Ownership Labor Media Judicial Tenure Dealing Prospectus Rights Enforcement Banks Ln(Steps) Laws Ownership Draft Formalism Judges Case Law

French Legal –0.3421a –0.3642a –1.1816a –14.3174b 0.2822b 0.4438a 0.2040a 0.3632a 0.5135a 1.5754a –0.2245a –0.5494a Origin [0.0792] [0.0858] [0.2685] [6.6720] [0.1172] [0.0925] [0.0464] [0.1157] [0.1041] [0.2511] [0.0781] [0.1481] German Legal –0.2508 –0.1145c –0.7960b –3.4763 0.3852b 0.0936 0.1333b 0.2438 0.8059a 0.6624c –0.7610a –0.4503 Origin [0.1487] [0.0639] [0.3729] [7.9660] [0.1514] [0.1618] [0.0559] [0.1711] [0.1045] [0.3676] [0.1834] [0.3774] Ln(GDP per 0.1074b 0.0907b 0.2571b 21.8679a –0.1259c –0.1023b 0.0011 –0.2153a 0.0185 –0.1181 0.0116 0.0288 capita) [0.0445] [0.0401] [0.0989] [4.3514] [0.0657] [0.0392] [0.0257] [0.0435] [0.0522] [0.1121] [0.0371] [0.0695] Constant –0.2647 –0.0156 0.3189 –141.9287a 1.2749a 2.8843a 0.3142 1.8839a 0.0311 3.9626a 0.9107a 0.7015 [0.3658] [0.3398] [0.8444] [39.5086] [0.4261] [0.3254] [0.2157] [0.3860] [0.3999] [0.8497] [0.2851] [0.5060]

Observations 37 26 78 39 47 47 46 52 84 51 38 38 R-squared 0.36 0.46 0.22 0.51 0.18 0.32 0.28 0.36 0.27 0.46 0.3 0.26

Notes: a Significant at the 1 percent level. b Significant at the 5 percent level. c Significant at the 10 percent level.

legal origins are quite distinct from political twentieth century, they cannot, say the crit- sentiment. ics, account for the differences in financial None of this is to say that politics is unimport- development. ant for either legal rules or economic outcomes. It is useful to break down the historical Indeed, political change may provide the impe- argument into three component parts and to tus for countries to revise their laws and regula- address them sequentially. This also allows tions. But the thrust of Legal Origins Theory us to consider several influential papers. is that, even in response to political demands, First, Rajan and Zingales (2003) present countries will design reforms consistently with evidence showing that in 1913, French civil their legal traditions. Legal origins are not law countries had more developed financial proxies for leftist politics. markets than common law countries. In their sample, as of 1913, the five common law countries had the average stock market to GDP 7. Legal Origins and History ratio of 53 percent, compared to 66 percent Perhaps the most difficult challenge for the ten French civil law countries. to the hypothesis that legal origins cause Second, several writers maintain that outcomes has been posed by historical shareholder protection in Britain at the arguments. Because virtually all of these beginning of the twentieth century was min- arguments focus on finance, we likewise imal. The evidence that Britain was finan- focus on finance in this section, but bear- cially developed at the time, including having ing in mind that an alternative theory must some ownership dispersion, must therefore address all the evidence. At the broadest be accounted for not by law but by alterna- level, historical arguments suggest that the tive mechanisms, such as trust and finan- positive correlation between common law cial intermediaries (Brian R. Cheffins 2001, and finance is a twentieth century phenom- Julian Franks, Colin Mayer, and Stefano enon. According to the critics, if one looks Rossi 2005). at historical data, particularly from the early Third, the historical critique holds that the twentieth century, the correlation does not correlation between common law and finan- exist. Because legal traditions predate the cial development emerges over the twentieth 316 Journal of Economic Literature, Vol. XLVI (June 2008)

1.40 English Legal Origin French Legal Origin 1.20 German Legal Origin Scandinavian Legal Origin

1.00

0.80

0.60

0.40

0.20

0.00 1913 1923 1933 1943 1953 1963 1973 1983 1993

Figure 3. Stock Market Capitalization over GDP (Based on Rajan and Zingales 2003) century, a finding it sees as inconsistent with law and eighteen civil law countries (ten of LLSV. In contrast to the superiority of finan- them French civil law) starting in 1913. To do cial development in the French legal origin so, they find a separate data source for each countries, as compared to the common law country that reports aggregate stock market countries, circa 1913, Rajan and Zingales find capitalization. Their findings of a higher ratio that the respective average stock market for of stock market value to GDP in civil than in common law and French civil law countries common law countries (the variable used by were 130 percent and 74 percent by 1999. They LLSV 1997, 1998), reproduced in table 7 and call this the Great Reversal (see figure 3). illustrated in figure 3, is the starting point of Critics propose two explanations of how most historical critiques of LLSV, as well as common law countries came to excel in of political accounts of finance in the twen- finance. The first is the political argument, tieth century. namely that common law countries hap- pened to have more favorable democratic We have looked at some of the Rajan and politics, which we have already discussed. In Zingales’s data using their own sources. addition, according to Roe (2006), civil law Here we focus on stock market capitaliza- countries suffered greater destruction dur- tion as a measure of financial development. ing World War II, which radicalized their Conceptually, the measure of a country’s politics and in this way led to pro-labor and stock market capitalization relevant for test- anti-capital laws and regulations. ing the influence of legal origins is the capi- It is easiest to take up the three pieces of talization of equities listed on that country’s the historical critique in turn. stock exchange(s) whose shareholders are subject to protection of that country’s laws. Impressively, Rajan and Zingales undertook 7.1 Stock Markets at the Start of the to find such numbers, but doing so for the Twentieth Century early twentieth century is especially diffi- Rajan and Zingales (2003) present data on cult for two reasons. First, many—perhaps stock market development for six common even most—securities that traded on stock La Porta, Lopez-de-Silanes, and Shleifer: Economic Consequences of Legal Origins 317

TABLE 7 Stock Market Capitalization over GDP

Country Legal Origin 1913 1929 1938 1950 1960 1970 1980 1990 1999 Australia English 0.39 0.50 0.91 0.75 0.94 0.76 0.38 0.37 1.13 Canada English 0.74 1.00 0.57 1.59 1.75 0.46 1.22 1.22 India English 0.02 0.07 0.07 0.07 0.07 0.06 0.05 0.16 0.46 South Africa English 0.68 0.91 1.97 1.23 1.33 1.20 United Kingdom English 1.09 1.38 1.14 0.77 1.06 1.63 0.38 0.81 2.25 United States English 0.39 0.75 0.56 0.33 0.61 0.66 0.46 0.54 1.52

Avg Common Law 0.53 0.68 0.74 0.53 0.86 1.14 0.49 0.74 1.30 Argentina French 0.17 0.05 0.03 0.11 0.15 Belgium French 0.99 1.31 0.32 0.23 0.09 0.31 0.82 Brazil French 0.25 0.05 0.08 0.45 Chile French 0.17 0.12 0.00 0.34 0.50 1.05 Cuba French 2.19 Egypt, Arab Rep. French 1.09 0.16 0.01 0.06 0.29 France French 0.78 0.19 0.08 0.28 0.16 0.09 0.24 1.17 Italy French 0.17 0.23 0.26 0.07 0.42 0.14 0.07 0.13 0.68 Netherlands French 0.56 0.74 0.25 0.67 0.42 0.19 0.50 2.03 Russian Federation French 0.18 0.11 Spain French 0.17 0.41 0.69

Avg French Law 0.66 0.77 0.40 0.13 0.29 0.16 0.12 0.28 0.74 Austria German 0.76 0.09 0.03 0.17 0.17 Germany German 0.44 0.35 0.18 0.15 0.35 0.16 0.09 0.20 0.67 Japan German 0.49 1.20 1.81 0.05 0.36 0.23 0.33 1.64 0.95 Switzerland German 0.58 0.50 0.44 1.93 3.23 Avg German Law 0.57 0.78 1.00 0.10 0.36 0.25 0.22 0.99 1.26 Denmark Scandinavian 0.36 0.17 0.25 0.10 0.14 0.17 0.09 0.67 0.67 Norway Scandinavian 0.16 0.22 0.18 0.21 0.26 0.23 0.54 0.23 0.70 Sweden Scandinavian 0.47 0.41 0.30 0.18 0.24 0.14 0.11 0.39 1.77

Avg Scandinavian Law 0.33 0.27 0.24 0.16 0.21 0.18 0.25 0.43 1.05

Source: Rajan and Zingales (2003). exchanges were bonds rather than stocks, also an American colony, with a reported and most of those were government bonds. stock market capitalization to GDP ratio of Second, many of the companies listed on 219 percent. We have looked at this obser- the exchanges of developing countries were vation and discovered that, if one excludes incorporated (and therefore subject to share- bonds and only looks at stocks, the actual holder protection rules), and even had their ratio falls to 33 percent. Moreover, by far the primary listings, in Europe or the United largest company with its stock listed in Cuba States (see Mira Wilkins and Harm Schroter is Havana Electric, a company incorporated 1998). For a developing country, both of in New Jersey, subject to New Jersey laws, these factors may lead to an overestimate of and with a primary listing in New York. We market value of equities subject to national suspect that concerns of Havana Electric shareholder protection laws. shareholders would have been addressed by Take a few examples. In 1913, the most either New Jersey courts or the U.S. marines. financially developed country in the Rajan Many other companies listed in Cuba appear and Zingales sample is Cuba. Cuba at that to be like Havana Electric; indeed—and per- time is a French legal origin country, but haps not surprisingly—there does not seem 318 Journal of Economic Literature, Vol. XLVI (June 2008) to be much of an indigenously Cuban stock Zingales on comparative financial develop- market capitalization at all. Given the small ment cannot be drawn from their sample. size of their sample, the elimination of bonds Perhaps a better way to get at this issue is to from the Cuban data point by itself reduces compare the two mother countries: England the Rajan and Zingales 1913 average French and France. Rajan and Zingales recognize civil law stock market to capitalization ratio that England was more financially developed from 66 percent to 47 percent, below their than France at the start of the century, but the common law estimated average. comparison can be expanded because Bozio The second most financially developed (2002) reports new numbers for France and country in the 1913 Rajan and Zingales adequate data are available for Britain from sample is also a French civil law country, Ranald Michie (1999). Michie’s numbers of namely Egypt, with a stock market to GDP the value of the stock market include corpo- ratio of 109 percent. It appears from Robert rate bonds, so we correct them using data L. Tignor (1984) that this ratio, like that for from Goldsmith (1985). Cuba, includes debt. Moreover, virtually In figure 4, we present Bozio’s numbers all of the largest companies listed in Egypt for France and adjusted numbers for domes- were incorporated in England or in France, tic stocks in Britain. The results show that and many were listed there as well. (Egypt Britain always had a higher stock market in 1913 was under British protection.) We capitalization to GDP ratio than France, estimate that a correct observation for Egypt often by a wide margin. This is true in 1913, (specifically, a stock market to GDP ratio of but also before and after. at most 40 percent) would further reduce the We can also look at Goldsmith’s (1985) Rajan–Zingales French civil law average in data on the ratio of stock market to GDP, 1913 by 6 percentage points. reproduced in table 8.15 The first point that Some corrections appear to be in order for emerges from the table is that, consistent the rich countries as well. For France, Rajan with Charles P. Kindleberger’s (1984) assess- and Zingales estimate a ratio of 78 percent. A ment of Paris as a financial backwater, Britain more recent estimate by Antoine Bozio (2002) is ahead of France as far back as the middle puts this number at 54 percent. Richard Sylla of the nineteenth century and perhaps even (2006) criticizes Rajan and Zingales for pre- earlier. So, interestingly, is the United States. senting too low a number of 39 percent for Goldsmith’s sample allows also for a more the United States and proposes the alterna- general comparison of common and civil tive 95 percent from Raymond W. Goldsmith law countries in 1913. If we pull in the U.S. (1985). Both of these corrections favor the observation from 1912, Goldsmith only has common law countries. The various correc- four common law countries and seven civil tions together, especially the one for Cuba, law ones. Even so, with India pulling the put the common law average stock market to common law average sharply down and no GDP ratio comfortably ahead of the French poor civil law countries in the sample, the civil law one in 1913. common law average in 1913 is 88 percent, To be sure, we have selected Cuba and the French legal origin average based only Egypt nonrandomly as two obviously bizarre on France and Belgium is 77 percent, and observations. A more systematic treatment of the overall civil law average is 69 percent. the data would reveal overestimates in common law, and not just civil law, countries. Some such errors are inevitable, and we have 15 Goldsmith’s (1985) data for corporate stock includes ourselves made many even with more recent unlisted firms. In practice, information on corporate shares “. . . is generally limited to securities listed on exchanges, data. What is beyond doubt, however, is that so that comprehensive figures must be derived, if at all, by the strong conclusions reached by Rajan and a blowup, often on a precarious basis” (p. 337). La Porta, Lopez-de-Silanes, and Shleifer: Economic Consequences of Legal Origins 319

% 160

140

120 France Great Britain

100

0 1800 1810 1820 1830 1840 1850 1860 1870 1880 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990

Figure 4. Stock Market Capitalization over GDP, France and Great Britain (Bozio 2002, Michie 1999)

Goldsmith’s data have many problems of only weakly protected. With the law playing their own and we have not examined them a minor role, the researchers credit financial closely. But they confirm independently the development in England to other mecha- point that the relative financial underdevel- nisms, such as the bonding role of interme- opment of common law countries at the start diaries and trust. of the twentieth century is a myth. The position that British shareholders We conclude that common law countries were utterly unprotected has proved contro- appear to be more financially developed than versial. Several authors, for example, argue civil law ones at the start of the twentieth that Britain led the world in securities regu- century and, in particular, Britain is ahead lation in general, and corporate disclosure of France. Over the course of the twentieth in particular (Coffee 2001, Laurence Gower century, the differences widen, a divergence 1954, Sylla and George Smith 1995). Britain that needs to be explained. But the puzzle is passed the Directors Liability Act in 1890 and divergence, not reversal. Companies Act in 1900, with the effects of both mandating significant disclosure in the 7.2 Britain at the Start of the prospectus and of holding directors account- Twentieth Century able for inaccuracies. Subsequent legislation A small but lively historical literature in the early twentieth century, according to argues that Britain had a well developed Coffee (2001), mandated on-going financial stock market at the beginning of the twen- disclosure and addressed some abuses in the tieth century, with beginnings of ownership new issues market. Britain also had perhaps dispersion, but that this had nothing to do the best commercial courts in the world, with with the law (Cheffins 2001, Franks, Mayer, most professional and least corrupt judges, and Rossi 2005). Looking both at the LLSV with centuries of precedents and experience indices of shareholder protection and at legal in dealing with fraud. rulings, this research sees the rights of minor- This small literature is at a standstill, with ity shareholders in the United Kingdom as some writers arguing the British shareholder 320 Journal of Economic Literature, Vol. XLVI (June 2008) 7 71 23 54 95 139 0.5 5.67 5.59 15 1 2 12 58 12 14 51 182 156 3 154 61 47 33 43 11 28 14 105 26

TABLE 8 Stock market capitalization over GDP French LegalFrench Origin German Legal Origin Scandinavian Legal Origin English Legal Origin (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) Goldsmith (1985). Year Belgium France Italy Mexico Japan Germany Switzerland Denmark Norway Sweden Australia Canada Britain Great India Israel Africa South USA 1805 Source: 1815 1830 18501860 1861 69187518801881 64 12188518951899 1900 581912 3819131914 88 11 1927 19291930 65 7 1937 691938 1119391940 231947 33 26 1948 6 1950 3 6 1951 321953 1955 4 17 1956 411957 2 321960 25 371962 801963 251965 1231966 631969 75 24 88 19 1970 29 47 111 821972 44 14 1973 401975 137 118 201976 74 1977 17 1261978 17 57 63 37 149 39 24 13 28 30 66 107 25 46 46 31 39 10 121 137 29 5 116 53 37 39 27 33 92 24 30 9 130 9 102 26 48 7 110 27 44 85 46 59 193 5 36 83 65 21 14 21 6.30 41 108 124 85 76 113 83 37 57 La Porta, Lopez-de-Silanes, and Shleifer: Economic Consequences of Legal Origins 321 protection glass was half empty and others shareholders—is available for thirty-four of countering that it was half full. What makes the countries with GDP growth data. When this debate utterly frustrating is that it is not we use the larger sample, the correlation comparative, so except with a few remarks reported by Roe disappears, as illustrated in on Britain versus the United States (Coffee figure 5 (right graph). This may not be sur- 2001), we know very little of how the British prising: many developing countries stayed shareholders were protected compared to out of World War II, yet remained financially the French and German ones. To the extent underdeveloped. Continue with alterna- that the literature has a bottom line, it is that tive measures of financial development. The shareholder rights have improved enormously pairwise correlations between GDP growth in Britain over the course of the twentieth 1913–45 and stock market capitalization, century, parallel to the growth of its markets. block premium, listed firms per capita, initial Explaining this parallel growth is a challenge public offerings to GDP, and private credit to the Legal Origins Theory. are either insignificant or have the wrong sign for both the twenty-four and thirty-four 7.3 World War II Destruction observation samples. Finally, consider Roe’s Roe (2006) claims that poor economic other prediction that World War II devasta- performance, particularly associated with tion leads to pro-labor laws. This only is true the destruction of capital stocks in World in univariate regressions using the sample of War II, radicalized continental European twenty-four observations as illustrated in fig- politics, leading to legal rules that were hos- ure 6 (left panel), but not when controlling tile to financial markets and favorable to for common law (right panel). This is also labor. To test this theory, Roe regresses mod- not true in univariate regressions using the ern ownership concentration—as proxied sample of thirty-four observations, as illus- by the percentage of widely held medium- trated in figure 7. The data are inconsistent sized firms—on Angus Maddison’s (2003) with the theory that World War II destruc- estimates of GDP growth between 1913 tion explains LLSV evidence. and 1945 in a sample of twenty-four mostly 7.4 Explaining Divergence developed countries. Figure 5 illustrates the strong positive relationship between owner- Although we do not see any evidence for ship dispersion and 1913–45 GDP growth the reversal of rankings between common (see left graph). Countries with worse eco- and civil law countries in financial devel- nomic growth have higher ownership con- opment over the course of the twentieth centration.16 However, these results fall apart century, the historical research yields two if we use a broader sample of countries, if we important findings that require an explana- use alternative measures of financial devel- tion. First, as shown by Rajan and Zingales opment, or if we look at other predictions of (2003) and in figure 3, common law coun- Roe’s theory. tries appear to have moved sharply ahead of Begin with ownership. An alternative civil law ones in financial development over measure of ownership dispersion—the the course of the twentieth century. Second, percentage not owned by the three largest investor protection improved sharply in the common law countries over the same time period (Coffee 1999, Cheffins 2001, Franks, 16 Legal origin continues to have a large and statisti- Mayer, and Rossi 2005). We suggest that cally significant effect on ownership concentration after Legal Origins Theory naturally accounts for controlling for 1913–45 GDP growth. In contrast, growth these findings. is not significant when controlling for French legal origin (although it is significant when controlling for common The twentieth century represented a law). period of explosive growth of the world econ- 322 Journal of Economic Literature, Vol. XLVI (June 2008)

0.4 0.6

0.2 0.4

0.2 0

0 0.2

0.2 Percent not owned by three largest shareholders Percent widely-held firms among medium-sized 0.4

1 0.5 0 0.5 1 1.5 2 0 2 4 6 GDP growth 1913–45 GDP growth 1913–45 coef 0.18698655, (robust) se 0.08753752, t 2.14 coef 0.00529849, (robust) se 0.01612937, t 0.33

Figure 5. Ownership Concentration and GDP Growth, 1913–45

omy, including of countries that were the companies to promote economic growth and wealthiest at the beginning of that century. resolve crises. These were the standard civil That growth relied to a significant extent law solutions to addressing social problems, on outside capital. That growth was also far going back at least to Napoleon. Common from smooth: it was punctuated by World law countries, particularly the United States Wars, the Great Depression, and significant and the United Kingdom, in contrast, relied economic and financial crises. The coun- more heavily on market-supporting regula- tries that grew successfully found their own tions, such as securities laws, deposit insur- ways to deliver capital to firms and to survive ance, and court-led improvements in the the crises. For some countries, such success corporate law. These differences were not involved massive state involvement in finance absolute, with nationalizations in common and development. For other countries, such law countries and many market-support- success to a much greater extent relied on ing reforms in civil law ones, but they were shoring up markets. pronounced nonetheless. We saw this, for Here is where legal origins come in. As example, in the La Porta et al. (2002) data on Morck and Steier (2005) make clear, civil government ownership of banks. law countries in the middle of the century In these very different ways, both some relied heavily on state supply of finance, of the civil law countries and some of the bank nationalization, and state investment common law ones successfully solved their La Porta, Lopez-de-Silanes, and Shleifer: Economic Consequences of Legal Origins 323

0.4 0.2

0.2

0 Index of labor laws Index of labor laws 0.2

0.2

0.4 0.4 1 0.5 0 0.5 1 1.5 1.5 1 0.5 0 0.5 1 GDP growth 1913–45 GDP growth 1913–45 coef 0.09302556, (robust) se 0.05266563, t 1.77 coef 0.0185974, (robust) se 0.04088877, t 0.45

Figure 6. Labor Laws and GDP Growth, 1913–45 (sample of 24 observations)

problems. In the second half of the century, 8. A Blueprint for Policy Reform however, the world became a good deal more peaceful and orderly. In such a world, the Legal Origins Theory points to three market-supporting solutions of the common important ways in which prevailing legal and law system, whether in the form of judicial regulatory rules might be inefficient. First, to decisions, regulations, or market adaptations the extent that a country has a particular legal worked better than the policy-implementing or regulatory style shaped by its legal tradi- solutions of the civil law system. As a conse- tion, it might apply the tools characteristic of quence of their twentieth century legal and that style to areas of regulation where they regulatory evolution, common law countries are inappropriate. A good example of this that ended up with sharply better investor pro- we already mentioned is the reliance on fre- tection. Their financial markets ran away quent interlocutory appeals in civil law bank- from the civil law ones, as we see in the data. ruptcy procedures. Such appeals are central Looking back over the course of the twen- to the civil procedure of civil law countries, tieth century, we see the basic differences yet result in massive destruction of value in in the legal traditions and regulatory strate- bankruptcy (Djankov et al. 2006, Gamboa- gies playing out in how both the laws and the Cavazos and Schneider 2007). Second, a markets evolve. country that introduces legal and regulatory 324 Journal of Economic Literature, Vol. XLVI (June 2008)

0.4

0.2

0 Index of labor laws

0.2

0.4

2 0 2 4 6 GDP growth, 1913–45 coef 0.00750986, (robust) se 0.02814714, t 0.27

Figure 7. Labor Laws and GDP Growth, 1913–45 (sample of 34 observations) rules in a situation of extreme disorder may torship and disorder in ways compatible with fail to dismantle them when the situation each country’s level of economic develop- returns to normal. Heavy government own- ment and legal tradition. In many instances, ership of banks, which might have a purpose the direction of such reforms is simply less at the time of extreme financial underdevel- government intervention. Neither underde- opment, becomes a burden under normal velopment nor the legal tradition justifies circumstances (La Porta et al. 2002). Third, heavy regulation of entry, so the reduction transplantation of legal and regulatory rules in those barriers is uncontroversial from the might itself become an important source of efficiency perspective. Likewise, aspects of inefficiency, as rules suitable for developed the formalism of bankruptcy procedures, economies become a source of massive delay which probably are the heritage of civil law, and corruption in the developing countries appear detrimental to efficiency at all lev- that copy them (Pistor et al. 2003a, 2003b, els of economic development and could be Spamann 2006a). reduced without impinging the foundations The inefficiency of the prevailing legal of legal order. In other instances, the best and regulatory rules points to a blueprint for solutions might differ across legal systems. reforms. Such reforms would focus on the For example, while common law countries design of what Djankov et al. (2003a) called depend on investor protection to support “appropriate institutions,” those that seek to their debt markets, many civil law coun- achieve the optimal trade-off between dicta- tries have successfully relied on information La Porta, Lopez-de-Silanes, and Shleifer: Economic Consequences of Legal Origins 325 sharing institutions, such as credit bureaus, mon law and civil law countries since 1970, for the same purpose (Djankov et al. 2007). with the U.K. deregulating and France doing Finally, in situations of extreme disorder, the opposite. James J. Heckman and Pagés such as participation or recovery from war, (2004) see no tendency for liberalization in even more aggressive government interven- Latin America during the 1990s. tions might be appropriate. With respect to investor protection, Pagano The crucial requirement of reform is the and Volpin (2005) report gains in shareholder availability of objective data on legal and rights in OECD countries during the 1990s. regulatory rules, preferably in a compara- Enriques and Volpin (2007) describe a ten- tive form so that the consequences of par- dency toward improving shareholder rights ticular rules can be evaluated. Perhaps the in the European Union. At the same time, most useful contribution of our research has they note that “far too little has been done been to establish the possibility of collecting to resolve the problem of related-party trans- such data in a broad range of areas. More actions, which is the most common form of recently, the data collection project has made self-dealing in Europe.” We are aware of no substantial strides through a World Bank systematic evidence for emerging markets, Doing Business initiative, which assembles although there are examples of improvement, and updates much of the information on such as the Mexican bankruptcy reform laws and regulations discussed in this paper, (Gamboa-Cavazos and Schneider 2007). as well as some additional indicators. Even The use of our indicators of laws and reg- the publication of this report has proved ulations, with their clear correlations with controversial, with the French government legal origins, for policy analysis has stimu- accusing its authors of an Anglo–Saxon bias. lated two objections. Some accuse us of Nonetheless, the report has proved popular, claiming that legal origin is destiny, so any and has encouraged regulatory reforms in reform of investor protection or of other dozens of countries. regulations short of wholesale replacement The pace of legal and regulatory reform of the legal system is futile. This is not what stimulated by the evidence is quicken- Legal Origin Theory says. The theory indeed ing. Perhaps the greatest progress has been holds that some aspects of the legal tradition made in the reductions of entry regula- are so hard-wired that changing them would tions. According to the 2006 Doing Business be extremely costly and that reforms must report, fifty-five countries undertook reforms be sensitive to legal traditions. Nonetheless, in 2005 and 2006 that lowered administra- many legal and regulatory rules, such as tive costs of starting a business and obtain- entry regulations, disclosure requirements, ing a license. Evgeny Yakovlev and Ekaterina or some procedural rules in litigation, can be Zhuravskaya (2008) for the case of Russia reformed without disturbing the fundamen- and David S. Kaplan, Eduardo Piedra, and tals of the legal tradition. Enrique Seira (2007) for the case of Mexico Some critics also argue that the legal rules find that reductions in entry regulations we measure are not the right ones. Even if increase new business start-ups. these rules capture the broad stance of the The picture is more mixed for labor mar- law toward investor or worker protection, the kets. OECD (2006) reports that labor mar- most relevant legal rules, doctrines, or even kets were liberalized in OECD countries in patterns of judicial behavior responsible for the last fifteen years, although most reforms the observed outcomes might be different pertained to temporary rather than perma- from what we measure. Focusing the reforms nent employment. Deakin, Priya P. Lele, formalistically on our subindices will then be and Mathias M. Siems (2007) actually find futile. For example, if judges are reluctant some divergence in labor laws between com- to take on corporate self-dealing cases and 326 Journal of Economic Literature, Vol. XLVI (June 2008) find technical or procedural excuses to throw to us to still be standing, perhaps even taller them out, changing the rules of approval of than a decade ago. And that is the idea that self-dealing transactions will be futile. As legal origins—broadly interpreted as highly Berkowitz, Pistor, and Jean-Francois Richard persistent systems of social control of eco- (2003) and Paolo Mauro, Nathan Sussman, nomic life—have significant consequences and Yishay Yafeh (2006) find, reforms are for the legal and regulatory framework of the more likely to succeed when people they society, as well as for economic outcomes. affect choose to accept them. The range of empirically documented legal, We definitely agree with this point and economic, and social spheres where legal ori- believe that legal or regulatory reform in any gins have consequences has expanded over country must be sensitive to the actual legal or the past decade. regulatory bottlenecks. Understanding what At the end of our overview, we believe that actually happens on the ground is essential. four propositions are correct, at least given So if judges throw out self-dealing cases, one the current state of our knowledge. First, legal might want to find out why they do so and rules and regulations differ systematically focus on how to get them to change. If labor across countries, and these differences can courts rule for employees regardless of what be measured and quantified. Second, these the law says, labor market reformers should differences in legal rules and regulations are take note. Having said this, in many circum- accounted for to a significant extent by legal stances the actual laws on the books that we origins. Third, the basic historical divergence measure are indeed the reason for inefficient in the styles of legal traditions—the policy- outcomes. The heavy regulations of entry are implementing focus of civil law versus the one such example, procedural formalism is market-supporting focus of common law— another. And even when the legal rules we explains well why legal rules differ. Fourth, measure are not the entire problem, and the measured differences in legal rules mat- thoughtless formalistic reforms are likely to ter for economic and social outcomes. fail, the rules can point the reformer closer The fact that the outlines of a coherent the- to where the problem actually lies. In either ory have emerged over the last decade does case, the measured rules provide highly rel- not mean that all, or most, of the empirical evant data. issues have been settled or, for that matter, Although the evidence on reforms is just that the theory will survive further scrutiny. beginning to come in and much of it is unfor- From our perspective, the crucial open ques- tunately confined to the developed world, tions deal with the evolution of legal systems: many countries seem to be moving toward How do they deal with crises? How do they market-friendlier government interventions. enter new spheres of regulation? How do If the world remains peaceful and orderly, the they approach reforms? We have offered attraction of such reforms will only grow. many illustrations from the historical record, but a comprehensive account of legal and regulatory evolution under common and civil 9. Conclusion law does not exist. Since their publication a decade ago, the Such an account might clarify an issue two LLSV articles have taken some bumps. that has generated tremendous heat, and We now use different measures of shareholder not much light, throughout this research, protection and are skeptical about the use of namely the circumstances under which each instrumental variables. Our interpretation legal tradition “works better.” Legal Origins of the meaning of legal origins has evolved Theory does not point to the overall supe- considerably over time. But the bumps not- riority of common law; to the contrary, it withstanding, the basic contribution appears points to the superiority of civil law and reg- La Porta, Lopez-de-Silanes, and Shleifer: Economic Consequences of Legal Origins 327 ulatory solutions when the problem of disor- problems, such as the thirty-five hour work- der is sufficiently (but not too) severe. On the week in France, is huge. other hand, our attempt to find evidence for All this, of course, leaves open the question the commonly made defense of civil law that of which legal rules and regulations the coun- it provides greater fairness or better access tries are likely to move toward, even if they do to justice have failed; the data suggest the not converge. So, in conclusion, we again rely opposite (Djankov et al. 2003b). on theory to make a prediction. The world A deeper understanding of the dynam- economy in the last quarter century has been ics of legal traditions may also inform the surprisingly calm, and has moved sharply crucial question of whether the differences toward capitalism and markets. In that envi- between common and civil law will persist ronment, our framework suggests that the into the future. Since we have shown legal common law approach to social control of origins to be closely related to the types of economic life performs better than the civil capitalism, this question can be rephrased law approach. When markets do or can work as follows: what kind of capitalism is likely well, it is better to support than to replace to prevail in the long run? Will it be the them. As long as the world economy remains more market-focused Anglo–Saxon capital- free of war, major financial crises, or order ism, or the more state-centered capitalism of extraordinary disturbances, the competitive Continental Europe and perhaps Asia? pressures for market-supporting regulation There are many arguments for conver- will remain strong and we are likely to see gence. Globalization leads to a much faster continued liberalization. Of course, underly- exchange of ideas, including ideas about laws ing this prediction is a hopeful assumption and regulations, and therefore encourages that nothing like World War II or the Great the transfer of legal knowledge. Globalization Depression will repeat itself. If it does, coun- also encourages competition among countries are likely to embrace civil law solutions, tries for foreign direct investment, for capi- just as they did back then. tal, and for business in general, which must as well put some pressure toward the adop- References tion of good legal rules and regulations. Aggarwal, Reena, Isil Erel, René M. Stulz, and Rohan The convergence is working both by civil Williamson. 2008. “Differences in Governance Prac- law countries increasingly accepting com- tices between U.S. and Foreign Firms: Measure- ment, Causes, and Consequences.” National Bureau mon law solutions, and vice versa. In one of Economic Research Working Paper 13288. area where heavy regulation appears patently Ahlering, Beth, and Simon Deakin. 2005. “Labour absurd—the entry of new firms—coun- Regulation, Corporate Governance, and Legal Ori- gin: A Case of Institutional Complementarity?” Uni- tries are rapidly tearing down the barriers. versity of Cambridge Centre for Business Research In Europe at least, there are some reduc- Working Paper 312. tions in labor regulations, as well as gains in Alesina, Alberto, and Edward L. Glaeser. 2004. Fight- ing Poverty in the US and Europe: A World of Dif- shareholder rights. At the same time, com- ference. Oxford and New York: Oxford University mon law countries are increasingly resort- Press. ing to legislation to address social problems, Allen, Franklin, and Douglas Gale. 2000. Comparing Financial Systems. Cambridge and London: MIT the Sarbanes–Oxley Act being the most Press. recent example of such financial regulation Antras, Pol, Mihir A. Desai, and C. Fritz Foley. 2007. in the United States. Mediating against con- “Multinational Firms, FDI Flows and Imperfect Capital Markets.” National Bureau of Economic vergence is the fact that civil law countries Research Working Paper 12855. continue to resort to “policy-implementing” Balas, Aron, Rafael La Porta, Florencio Lopez-de- solutions to newly arising problems. The bias Silanes, and Andrei Shleifer. 2008. “The Divergence of Legal Procedures.” National Bureau of Economic toward using state mandates to solve social Research Working Paper 13809. Barth, James R., Gerard Caprio, and Ross Levine. 328 Journal of Economic Literature, Vol. XLVI (June 2008)

2004. “Bank Regulation and Supervision: What Press. Works Best?” Journal of Financial Intermediation, Casas-Arce, Pablo, and Albert Saiz. 2007. “Owning 13(2): 205–48. versus Leasing: Do Courts Matter?” Unpublished. Beck, Thorsten, Asli Demirguc-Kunt, and Ross Levine. Cheffins, Brian R. 2001. “Does Law Matter? The 2003. “Law and Finance: Why Does Legal Origin Separation of Ownership and Control in the United Matter?” Journal of Comparative Economics, 31(4): Kingdom.” Journal of Legal Studies, 30(2): 459–84. 653–75. Chemin, Matthieu. 2007a. “Decoding the Code of Beck, Thorsten, Asli Demirguc-Kunt, and Ross Levine. Civil Procedure: Do Judiciaries Matter for Growth?” 2005. “Law and Firms’ Access to Finance.” Ameri- Centre interuniversitaire sur le Risque, les Politiques can Law and Economics Review, 7(1): 211–52. Economiques et l’Emploi Working Paper 07-26. Beck, Thorsten, Ross Levine, and Norman Loayza. Chemin, Matthieu. 2007b. “The Impact of the Judi- 2000. “Finance and the Sources of Growth.” Journal ciary on Entrepreneurship: Evaluation of Pakistan’s of Financial Economics, 58(1–2): 261–300. Access to Justice Programme.” Centre interuniver- Bekaert, Geert, Campbell R. Harvey, Christian Lun- sitaire sur le Risque, les Politiques Economiques et dblad, and Stephan Siegel. 2007. “Global Growth l’Emploi Working Paper 07-27. Opportunities and Market Integration.” Journal of Ciccone, Antonio, and Elias Papaioannou. 2006a. Finance, 62(3): 1081–1137. “Adjustment to Target Capital, Finance and Growth.” Ben-Bassat, Avi, and Momi Dahan. 2008. “Social Centre for Economic Policy Research Discussion Rights in the Constitution and in Practice.” Journal Paper 5969. of Comparative Economics, 36(1): 103–119. Ciccone, Antonio, and Elias Papaioannou. 2006b. Bergman, Nittai K., and Daniel Nicolaievsky. 2007. “Financial Development and Inter-sectoral Invest- “Investor Protection and the Coasian View.” Journal ment: New Estimates and Evidence.” Unpublished. of Financial Economics, 84(3): 738–71. Ciccone, Antonio, and Elias Papaioannou. 2006c. “Red Berkowitz, Daniel, and Karen B. Clay. 2005. “Ameri- Tape and Delayed Entry.” Centre for Economic Pol- can Civil Law Origins: Implications for State Con- icy Research Discussion Paper 5996. stitutions.” American Law and Economics Review, Claessens, Stijn, and Luc Laeven. 2003. “Financial 7(1): 62–84. Development, Property Rights, and Growth.” Jour- Berkowitz, Daniel, and Karen B. Clay. 2006. “The nal of Finance, 58(6): 2401–36. Effect of Judicial Independence on Courts: Evi- Clark, Robert. 1986. Corporate Law. New York: Aspen dence from the American States.” Journal of Legal Publishers. Studies, 35(2): 399–440. Coffee, John C., Jr. 1999. “The Future as History: The Berkowitz, Daniel, and Karen B. Clay. 2007. “Legal Prospects for Global Convergence in Corporate Origins and the Evolution of Institutions: Evidence Governance and Its Implications.” Northwestern from American State Courts.” Unpublished. University Law Review, 93(3): 641–708. Berkowitz, Daniel, Katharina Pistor, and Jean-Francois Coffee, John C., Jr. 2001. “The Rise of Dispersed Richard. 2003. “Economic Development, Legality, Ownership: The Roles of Law and the State in the and the Transplant Effect.” European Economic Separation of Ownership and Control.” Yale Law Review, 47(1): 165–95. Journal, 111(1): 1–82. Berman, Harold J. 1983. Law and Revolution: The Cuñat, Alejandro, and Marc J. Melitz. 2006. “Vola- Formation of the Western Legal Tradition. Cam- tility, Labor Market Flexibility, and the Pattern of bridge and London: Harvard University Press. Comparative Advantage.” Unpublished. Berman, Harold J. 2003. Law and Revolution II: The Damaška, Mirjan R. 1986. The Faces of Justice and Impact of the Protestant Reformations on the West- State Authority: A Comparative Approach to the ern Legal Tradition. Cambridge and London: Har- Legal Process. New Haven and London: Yale Uni- vard University Press, Belknap Press. versity Press. Botero, Juan C., Simeon Djankov, Rafael La Porta, David, Paul A. 1985. “Clio and the Economics of Florencio Lopez-de-Silanes, and Andrei Shleifer. Qwerty.” American Economic Review, 75(2): 2004. “The Regulation of Labor.” Quarterly Journal 332–37. of Economics, 119(4): 1339–82. David, Rene, and John Brierley. 1985. Major Legal Sys- Bozio, Antoine. 2002. “La capitalisation boursière en tems in the World Today. London: Stevens and Sons. France au XXe siècle.” Ecole Normale Supérieure. Dawson, John P. 1960. A History of Lay Judges. Cam- Mémoire de DEA. bridge and London: Harvard University Press. Braun, Matías. 2003. “Financial Contractibility and Dawson, John P. 1968. The Oracles of the Law. Ann Assets’ Hardness.” Unpublished. Arbor: University of Michigan Press. Bushee, Brian J., and Christian Leuz. 2005. “Eco- Deakin, Simon, Priya P. Lele, and Mathias M. Siems. nomic Consequences of SEC Disclosure Regulation: 2007. “The Evolution of Labour Law: Calibrating Evidence from the OTC Bulletin Board.” Journal of and Comparing Regulatory Regimes.” University of Accounting and Economics, 39(2): 233–64. Cambridge Centre for Business Research Working Caballero, Ricardo J., Kevin N. Cowan, Eduardo M. Paper 352. R. A. Engel, and Alejandro Micco. 2004. “Effective De Serres, Alain, Shuji Kobayakawa, Torsten Sløk, and Labor Regulation and Microeconomic Flexibility.” Laura Vartia. 2006. “Regulation of Financial Sys- Cowles Foundation Discussion Paper 1480. tems and Economic Growth.” Organisation for Eco- Cardozo, Benjamin N. 1921. The Nature of the Judicial nomic Co-operation and Development Economics Process. New Haven and London: Yale University Department Working Paper 506. La Porta, Lopez-de-Silanes, and Shleifer: Economic Consequences of Legal Origins 329

Dinc, I. Serdar. 2005. “Politicians and Banks: Political Corporate Governance Institute Finance Working Influences on Government-Owned Banks in Emerg- Paper 09/2003. ing Markets.” Journal of Financial Economics, 77(2): Freund, Caroline, and Bineswaree Bolaki. 2007. “When 453–79. Does Trade Promote Growth?” Unpublished. Djankov, Simeon, Tim Ganser, Caralee McLiesh, Rita Gamboa-Cavazos, Mario, and Frank Schneider. 2007. Ramalho, and Andrei Shleifer. 2008. “The Effect of “Bankruptcy as a Legal Process.” Unpublished. Corporate Taxes on Investment and Entrepreneur- Gennaioli, Nicola. 2007. “Optimal Contracts with ship.” National Bureau of Economic Research Work- Enforcement Risk.” Unpublished. ing Paper 13756. Gennaioli, Nicola, and Andrei Shleifer. 2007. “The Djankov, Simeon, Edward L. Glaeser, Rafael La Porta, Evolution of Common Law.” Journal of Political Florencio Lopez-de-Silanes, and Andrei Shleifer. Economy, 115(1): 43–68. 2003a. “The New Comparative Economics.” Journal Glaeser, Edward L., Rafael La Porta, Florencio Lopez- of Comparative Economics, 31(4): 595-619. de-Silanes, and Andrei Shleifer. 2004. “Do Institu- Djankov, Simeon, Oliver D. Hart, Caralee McLiesh, tions Cause Growth?” Journal of Economic Growth, and Andrei Shleifer. 2006. “Debt Enforcement 9(3): 271–303. around the World.” National Bureau of Economic Glaeser, Edward L., and Andrei Shleifer. 2002. “Legal Research Working Paper 12807. Origins.” Quarterly Journal of Economics, 117(4): Djankov, Simeon, Rafael La Porta, Florencio Lopez- 1193–1229. de-Silanes, and Andrei Shleifer. 2002. “The Regu- Glaeser, Edward L., and Andrei Shleifer. 2003. “The lation of Entry.” Quarterly Journal of Economics, Rise of the Regulatory State.” Journal of Economic 117(1): 1–37. Literature, 41(2): 401–25. Djankov, Simeon, Rafael La Porta, Florencio Lopez- Glendon, Mary Ann, Michael Wallace Gordon, and de-Silanes, and Andrei Shleifer. 2003b. “Courts.” Christopher Osakwe. 1982. Comparative Legal Tra- Quarterly Journal of Economics, 118(2): 453–517. ditions in a Nutshell. St. Paul: West Publishing. Djankov, Simeon, Rafael La Porta, Florencio Lopez- Glendon, Mary Ann, Michael Wallace Gordon, and de-Silanes, and Andrei Shleifer. Forthcoming. “The Christopher Osakwe. 1994. Comparative Legal Tra- Law and Economics of Self-Dealing.” Journal of ditions: Text, Materials, and Cases on the Civil and Financial Economics. Common Law Traditions, with Special Refrence to Djankov, Simeon, Caralee McLiesh, Tatiana Nenova, French, German, English, and European Law. St. and Andrei Shleifer. 2003c. “Who Owns the Media?” Paul: West Publishing. Journal of Law and Economics, 46(2): 341–81. Goldsmith, Raymond W. 1985. Comparative National Djankov, Simeon, Caralee McLiesh, and Andrei Balance Sheets: A Study of Twenty Countries, Shleifer. 2007. “Private Credit in 129 Countries.” 1688–1978. Chicago and London: University of Chi- Journal of Financial Economics, 84(2): 299–329. cago Press. Doidge, Craig, G. Andrew Karolyi, and René M. Stulz. Gorla, Gino, and Luigi Moccia. 1981. “A ‘Revisiting’ 2007. “Why Do Countries Matter So Much for Cor- of the Comparison between ‘Continental Law’ and porate Governance?” Journal of Financial Econom- ‘English Law’ (16th–19th Century).” Journal of ics, 86(1): 1–39. Legal History, 2(2): 143–56. Dyck, Alexander, and Luigi Zingales. 2004. “Private Gower, Laurence. 1954. The Principles of Modern Benefits of Control: An International Comparison.” Company Law. London: Stevens and Sons. Journal of Finance, 59(2): 537–600. Greenstone, Michael, Paul Oyer, and Annette Viss- Enriques. Luca. 2002. “Do Corporate Law Judges ing-Jørgensen. 2006. “Mandated Disclosure, Stock Matter? Some Evidence from Milan.” European Returns, and the 1964 Securities Acts Amendments.” Business Organization Law Review, 3(4): 756–821. Quarterly Journal of Economics, 121(2): 399–460. Enriques, Luca, and Paolo F. Volpin. 2007. “Corporate Grossman, Sanford J., and Oliver D. Hart. 1988. “One Governance Reforms in Continental Europe.” Jour- Share–One Vote and the Market for Corporate nal of Economic Perspectives, 21(1): 117–40. Control.” Journal of Financial Economics, 20(1–2): Esty, Benjamin C., and William L. Megginson. 2003. 175–202. “Creditor Rights, Enforcement, and Debt Owner- Guiso, Luigi, Paola Sapienza, and Luigi Zingales. 2004. ship Structure: Evidence from the Global Syndicated “The Role of Social Capital in Financial Develop- Loan Market.” Journal of Financial and Quantita- ment.” American Economic Review, 94(3): 526–56. tive Analysis, 38(1): 37–59. Guiso, Luigi, Paola Sapienza, and Luigi Zingales. Fama, Eugene F. 1980. “Agency Problems and the 2006. “Does Culture Affect Economic Outcomes?” Theory of the Firm.” Journal of Political Economy, Journal of Economic Perspectives, 20(2): 23–48. 88(2): 288–307. Haber, Stephen, and Enrico C. Perotti. 2007. “The Fisman, Raymond, and Inessa Love. 2004. “Financial Political Economy of Finance.” Unpublished. Development and Intersectoral Allocation: A New Hall, Peter A., and David Soskice, eds. 2001. Variet- Approach.” Journal of Finance, 59(6): 2785–2807. ies of Capitalism: The Institutional Foundations Fisman, Raymond, and Virginia Sarria-Allende. 2004. of Comparative Advantage. Oxford and New York: “Regulation of Entry and the Distortion of Indus- Oxford University Press. trial Organization.” National Bureau of Economic Haltiwanger, John, Stefano Scarpetta, and Helena Sch- Research Working Paper 10929. weiger. 2006. “Assessing Job Flows across Countries: Franks, Julian, Colin Mayer, and Stefano Rossi. 2005. The Role of Industry, Firm Size, and Regulations.” “Ownership: Evolution and Regulation.” European World Bank Policy Research Working Paper 4070. 330 Journal of Economic Literature, Vol. XLVI (June 2008)

Hart, Oliver D. 1995. Firms, Contracts, and Financial La Porta, Rafael, Florencio Lopez-de-Silanes, Cristian Structure. Oxford and New York: Oxford University Pop-Eleches, and Andrei Shleifer. 2004. “Judicial Press, Clarendon Press. Checks and Balances.” Journal of Political Economy, Haselmann, Rainer F. H., Katharina Pistor, and Vikrant 112(2): 445–70. Vig. 2006. “How Law Affects Lending.” Columbia La Porta, Rafael, Florencio Lopez-de-Silanes, and Law and Economics Working Paper 285. Andrei Shleifer. 1999. “Corporate Ownership around Hayek, Friedrich A. 1960. The Constitution of Liberty. the World.” Journal of Finance, 54(2): 471–517. Chicago: University of Chicago Press. La Porta, Rafael, Florencio Lopez-de-Silanes, and Heckman, James J., and Carmen Pagés. 2004. “Law Andrei Shleifer. 2002. “Government Ownership of and Employment: Introduction.” In Law and Banks.” Journal of Finance, 57(1): 265–301. Employment: Lessons from Latin America and La Porta, Rafael, Florencio Lopez-de-Silanes, and the Caribbean, ed. James J. Heckman and Carmen Andrei Shleifer. 2006. “What Works in Securities Pagés, 1–107. Chicago and London: University of Laws?” Journal of Finance, 61(1): 1–32. Chicago Press. La Porta, Rafael, Florencio Lopez-de-Silanes, Andrei Hellwig, Martin. 2000. “On the Economics and Politics Shleifer, and Robert W. Vishny. 1997. “Legal Deter- of Corporate Finance and Corporate Control.” In Cor- minants of External Finance.” Journal of Finance, porate Governance, ed. Xavier Vives, 95–134. Cam- 52(3): 1131–50. bridge and New York: Cambridge University Press. La Porta, Rafael, Florencio Lopez-de-Silanes, Andrei Helpman, Elhanan, Marc J. Melitz, and Yona Rubin- Shleifer, and Robert W. Vishny. 1998. “Law and stein. 2008. “Estimating Trade Flows: Trading Part- Finance.” Journal of Political Economy, 106(6): ners and Trading Volumes.” Quarterly Journal of 1113–55. Economics, 123(2): 441–487. La Porta, Rafael, Florencio Lopez-de-Silanes, Andrei Jackson, Howell, and Mark J. Roe. 2007. “Public Shleifer, and Robert W. Vishny. 1999. “The Quality Enforcement of Securities Laws: Preliminary Evi- of Government.” Journal of Law, Economics, and dence.” Unpublished. Organization, 15(1): 222–79. Jappelli, Tullio, Marco Pagano, and Magda Bianco. La Porta, Rafael, Florencio Lopez-de-Silanes, Andrei 2005. “Courts and Banks: Effects of Judicial Shleifer, and Robert W. Vishny. 2000. “Agency Prob- Enforcement on Credit Markets.” Journal of Money, lems and Dividend Policies around the World.” Jour- Credit, and Banking, 37(2): 223–45. nal of Finance, 55(1): 1–33. Jensen, Michael C., and William H. Meckling. 1976. La Porta, Rafael, Florencio Lopez-de-Silanes, Andrei “Theory of the Firm: Managerial Behavior, Agency Shleifer, and Robert W. Vishny. 2002. “Investor Costs and Ownership Structure.” Journal of Finan- Protection and Corporate Valuation.” Journal of cial Economics, 3(4): 305–60. Finance, 57(3): 1147–70. Johnson, Simon, Rafael La Porta, Florencio Lopez- Leland, Hayne E., and David H. Pyle. 1977. “Infor- de-Silanes, and Andrei Shleifer. 2000. “Tunneling.” mational Asymmetries, Financial Structure, and American Economic Review, 90(2): 22–27. Financial Intermediation.” Journal of Finance, Kaplan, David S., Eduardo Piedra, and Enrique 32(2): 371–87. Seira. 2007. “Entry Regulation and Business Lerner, Josh, and Antoinette Schoar. 2005. “Does Start-Ups: Evidence from Mexico.” http://ssrn. Legal Enforcement Affect Financial Transactions? com/abstract=978863. The Contractual Channel in Private Equity.” Quar- Khwaja, Asim Ijaz, and Atif Mian. 2005. “Do Lenders terly Journal of Economics, 120(1): 223–46. Favor Politically Connected Firms? Rent Provision Levine, Ross. 2005. “Law, Endowments and Property in an Emerging Financial Market.” Quarterly Jour- Rights.” Journal of Economic Perspectives, 19(3): nal of Economics, 120(4): 1371–1411. 61–88. Kindleberger, Charles P. 1984. A Financial History Levine, Ross, Norman Loayza, and Thorsten Beck. of Western Europe. Crows Nest, New South Wales: 2000. “Financial Intermediation and Growth: Cau- Allen and Unwin. sality and Causes.” Journal of Monetary Economics, Klapper, Leora, Luc Laeven, and Raghuram Rajan. 46(1): 31–77. 2006. “Entry Regulation as a Barrier to Entrepre- Licht, Amir N., Chanan Goldschmidt, and Shalom H. neurship.” Journal of Financial Economics, 82(3): Schwartz. 2005. “Culture, Law, and Corporate Gov- 591–629. ernance.” International Review of Law and Eco- Klerman, Daniel, and Paul G. Mahoney. 2007. “Legal nomics, 25(2): 229–55. Origin?” Journal of Comparative Economics, 35(2): Linciano, Nadia. 2003. “Non-voting Shares and the 278–93. Value of Control: The Impact of Corporate Regula- Lafontaine, Francine, and Jagadeesh Sivadasan. 2007. tion in Italy.” http://ssrn.com/abstract=410191. “The Microeconomic Implications of Input Market Loayza, Norman, Linda Kaltani, and Roberto Chang. Regulations: Cross-Country Evidence from Within 2005. “Openness Can Be Good for Growth: The the Firm.” Ross School of Business Paper 1069. Role of Policy Complementarities.” World Bank Pol- Lamoreaux, Naomi R., and Jean-Laurent Rosenthal. icy Research Working Paper 3763. 2005. “Legal Regime and Contractual Flexibility: López-Córdova, Jose Ernesto. 2007. “Labor Regula- A Comparison of Business’s Organizational Choices tion and the Allocative Efficiency of International in France and the United States during the Era of Trade.” Unpublished. Industrialization.” American Law and Economics Maddison, Angus. 2003. The World Economy: Histori- Review, 7(1): 28–61. cal Statistics. Paris and Washington, D.C.: Organisa- La Porta, Lopez-de-Silanes, and Shleifer: Economic Consequences of Legal Origins 331

tion for Economic Co-operation and Development. Development. Mahoney, Paul G. 2001. “The Common Law and Eco- Pagano, Marco, and Paolo F. Volpin. 2005. “The Politi- nomic Growth: Hayek Might Be Right.” Journal of cal Economy of Corporate Governance.” American Legal Studies, 30(2): 503–25. Economic Review, 95(4): 1005–30. Mauro, Paolo, Nathan Sussman, and Yishay Yafeh. Pagano, Marco, and Paolo F. Volpin. 2006. “Share- 2006. Emerging Markets and Financial Globaliza- holder Protection, Stock Market Development, and tion: Sovereign Bond Spreads in 1870–1913 and Politics.” Journal of the European Economic Asso- Today. Oxford and New York: Oxford University ciation, 4(2–3): 315–41. Press. Perotti, Enrico C., and Paolo F. Volpin. 2004. “Lobby- McNeill, John Robert, and William H. McNeill. 2003. ing on Entry.” Centre for Economic Policy Research The Human Web: A Bird’s-Eye View of World His- Discussion Paper 4519. tory. New York and London: Norton. Perotti, Enrico C., and Ernst-Ludwig Von Thadden. Merryman, John Henry. 1969. The Civil Law Tradi- 2006. “The Political Economy of Corporate Control tion: An Introduction to the Legal Systems of West- and Labor Rents.” Journal of Political Economy, ern Europe and Latin America. Stanford: Stanford 114(1): 145–74. University Press. Persson, Torsten, and Guido Tabellini. 2003. The Eco- Merryman, John Henry. 1996. “The French Devia- nomic Effects of Constitutions. Cambridge and Lon- tion.” American Journal of Comparative Law, 44(1): don: MIT Press. 109–19. Pierre, Gaëlle, and Stefano Scarpetta. 2007. “Do Micco, Alejandro, and Carmen Pagés. 2006. “The Eco- Firms Make Greater Use of Training and Temporary nomic Effects of Employment Protection: Evidence Employment When Labor Adjustment Costs Are from International Industry-Level Data.” Institute High?” Unpublished. for the Study of Labor Discussion Paper 2433. Pistor, Katharina. 2006. “Legal Ground Rules in Michie, Ranald. 1999. The London Stock Exchange: Coordinated and Liberal Market Economies.” In A History. Oxford and New York: Oxford University Corporate Governance in Context: Corporations, Press. States, and Markets in Europe, Japan, and the US, Montesquieu, Charles M. de Secondat. 1984 [1748]. ed. Klaus J. Hopt, Eddy Wymeersch, Hideki Kanda, The Spirit of Laws, Special Edition. The Legal and Harald Baum, 249–280. Oxford and New York: Classics Library, Birmingham, AL: Lucas Printing Oxford University Press. Company. Pistor, Katharina, Yoram Keinan, Jan Kleinheister- Morck, Randall K., and Lloyd Steier. 2005. “The Global kamp, and Mark D. West. 2003a. “Evolution of Cor- History of Corporate Governance: An Introduction.” porate Law and the Transplant Effect: Lessons from In A History of Corporate Governance around the Six Countries.” World Bank Research Observer, World: Family Business Groups to Professional 18(1): 89–112. Managers, ed. Randall K. Morck, 1–64. Chicago and Pistor, Katharina, Yoram Keinan, Jan Kleinheister- London: University of Chicago Press. kamp, and Mark D. West. 2003b. “The Evolution Mulligan, Casey B., and Andrei Shleifer. 2005a. “Con- of Corporate Law: A Cross-Country Comparison.” scription as Regulation.” American Law and Eco- University of Pennsylvania Journal of International nomics Review, 7(1): 85–111. Economic Law, 23(4): 791–871. Mulligan, Casey B., and Andrei Shleifer. 2005b. “The Ponzetto, Giacomo A. M., and Patricio A. Fernandez. Extent of the Market and the Supply of Regulation.” Forthcoming. “Case Law vs. Statute Law: An Evolu- Quarterly Journal of Economics, 120(4): 1445–73. tionary Comparison.” Journal of Legal Studies. Musacchio, Aldo. 2008. “Can Civil Law Countries Posner, Richard. 1973. Economic Analysis of Law. Get Good Institutions? Lessons from the History of Boston: Little-Brown. Creditor Rights and Bond Markets in Brazil.” Jour- Priest, George L. 1977. “The Common Law Process nal of Economic History, 68(1): 80–108. and the Selection of Efficient Rules.” Journal of Nenova, Tatiana. 2006. “Control Values and Changes Legal Studies, 6(1): 65–82. in Corporate Law in Brazil.” Latin American Busi- Przeworski, Adam, Michael E. Alvarez, José Antonio ness Review, 6(3): 1–37. Cheibub, and Fernando Limongi. 2000. Democracy Niblett, Anthony, Richard Posner, and Andrei Shleifer. and Development: Political Institutions and Well- 2008. “The Evolution of a Legal Rule.” National Being in the World, 1950–1990. Cambridge and Bureau of Economic Research Working Paper New York: Cambridge University Press. 13856. Qian, Jun, and Philip E. Strahan. 2007. “How Laws and Nunn, Nathan. 2007. “Relationship-Specificity, Incom- Institutions Shape Financial Contracts: The Case of plete Contracts, and the Pattern of Trade.” Quar- Bank Loans.” Journal of Finance, 62(6): 2803–34. terly Journal of Economics, 122(2): 569–600. Rajan, Raghuram G., and Luigi Zingales. 1998. “Finan- Ongena, Steven, and David C. Smith. 2000. “What cial Dependence and Growth.” American Economic Determines the Number of Bank Relationships? Review, 88(3): 559–86. Cross-Country Evidence.” Journal of Financial Rajan, Raghuram G., and Luigi Zingales. 2003. “The Intermediation, 9(1): 26–56. Great Reversals: The Politics of Financial Develop- Organisation for Economic Co-operation and Devel- ment in the Twentieth Century.” Journal of Finan- opment. 2006. OECD Employment Outlook: Boost- cial Economics, 69(1): 5–50. ing Jobs and Incomes: 2006. Paris and Washington, Ramseyer, J. Mark, and Eric B. Rasmusen. 1997. “Judi- D.C.: Organisation for Economic Co-operation and cial Independence in a Civil Law Regime: The Evi- 332 Journal of Economic Literature, Vol. XLVI (June 2008)

dence from Japan.” Journal of Law, Economics, and Subramanian, Krishnamurthy, Frederick Tung, and Organization, 13(2): 259–86. Xue Wang. 2007. “Law, Agency Costs and Project Reynolds, Thomas, and Arturo Flores. 1989. Foreign Finance: An Empirical Analysis.” Unpublished. Law: Current Sources of Basic Legislation in Juris- Sylla, Richard. 2006. “Schumpeter Redux: A Review dictions of the World. Littleton, Co.: Rothman and of Raghuram G. Rajan and Luigi Zingales’s Saving Company. Capitalism from the Capitalists.” Journal of Eco- Roe, Mark J. 2000. “Political Preconditions to Separat- nomic Literature, 44(2): 391–404. ing Ownership from Corporate Control.” Stanford Sylla, Richard, and George Smith. 1995. “Information Law Review, 53(3): 539–606. and Capital Market Regulation in Anglo-American Roe, Mark, J. 2006. “Legal Origins, Politics, and Mod- Finance.” In Anglo-American Financial Systems: ern Stock Markets.” Harvard Law Review, 120(2): Institutions and Markets in the Twentieth Century, 460–527. ed. Michael Bordo and Richard Sylla, 179–205. Rostowski, Jacek, and Bogdan Stacescu. 2006. “The Burr-Ridge, IL: Irwin Professional. Wig and the Pith Helmut—The Impact of ‘Legal Tignor, Robert L. 1984. State, Private Enterprise and School’ and Colonial Institutions on Economic Economic Change in Egypt, 1918–1952. Princeton: Performance.” Center for Social and Economic Princeton University Press. Research Network Studies and Analyses 300. Visaria, Sujata. 2006. “Legal Reform and Loan Repay- Rubin, Paul H. 1977. “Why Is the Common Law Effi- ment: The Microeconomic Impact of Debt Recovery cient?” Journal of Legal Studies, 6(1): 51–63. Tribunals in India.” Boston University Institute for Safavian, Mehnaz, and Siddharth Sharma. 2007. Economic Development Discussion Paper 157. “When Do Creditor Rights Work?” Journal of Com- Watson, Alan. 1974. Legal Transplants: An Approach parative Economics, 35(3): 484–508. to Comparative Law. Athens: University of Georgia Sapienza, Paola. 2004. “The Effects of Government Press. Ownership on Bank Lending.” Journal of Financial Wilkins, Mira, and Harm Schroter, eds. 1998. The Economics, 72(2): 357–84. Free-Standing Company in the World Economy, Shleifer, Andrei, and Robert W. Vishny. 1997. “A Sur- 1830–1996. Oxford and New York: Oxford Univer- vey of Corporate Governance.” Journal of Finance, sity Press. 52(2): 737–83. Woloch, Isser. 1994. The New Regime: Transforma- Shleifer, Andrei, and Daniel Wolfenzon. 2002. “Inves- tions of the French Civic Order, 1789–1820s. New tor Protection and Equity Markets.” Journal of York and London: Norton. Financial Economics, 66(1): 3–27. Woolf, Stuart. 1992. “The Construction of a European Spamann, Holger. 2006a. “Contemporary Legal Trans- World-View in the Revolutionary-Napoleonic Years.” plants: Legal Families and the Diffusion of Corpo- Past and Present, 137(1): 72–101. rate Law.” Unpublished. Wurgler, Jeffrey. 2000. “Financial Markets and the Spamann, Holger. 2006b. “On the Insignificance and/ Allocation of Capital.” Journal of Financial Econom- or Endogeneity of La Porta et al.’s ‘Anti-director ics, 58(1–2): 187–214. Rights Index’ under Consistent Coding.” Harvard Yakovlev, Evgeny, and Ekaterina Zhuravskaya. 2008. Law School Olin Center Discussion Paper 7. “Reforms in Business Regulation: Evidence from Stone, Julius. 1985. Precedent and Law: Dynamics of Russia.” Unpublished. Common Law Growth. Sydney: Butterworths. Zweigert, Konrad, and Hein Kötz. 1998. An Introduc- Stulz, René M., and Rohan Williamson. 2003. “Cul- tion to Comparative Law. Third edition. Oxford ture, Openness, and Finance.” Journal of Financial and New York: Oxford University Press, Clarendon Economics, 70(3): 313–49. Press.