DAVIS TYPE.DOC 3/17/2005 8:41 AM WHAT CAN THE RULE OF LAW VARIABLE TELL US ABOUT RULE OF LAW REFORMS? Kevin E. Davis* I. Introduction......................................................................... 141 II. Potential Methodological Issues .................................... 145 III. Illustrations......................................................................... 148 A. Rule of Law........................................................................ 148 B. Property Rights.................................................................. 151 C. Contract Enforcement ....................................................... 156 1. Contract Intensive Money........................................... 157 2. ICRG........................................................................... 157 3. World Business Environment Survey ......................... 158 4. The Lex Mundi Project ............................................... 159 D. Legal Heritage................................................................... 160 IV. Conclusion ............................................................................ 160 I. Introduction In 2001 per capita income in Haiti was $480, the infant mortality rate was seventy-nine per 1000 live births and the illiteracy rate (age fifteen and over) hovered around fifty percent. By comparison, in the United States, less than two hours flying time away, the per capita income was $34,280, the infant mortality rate was seven per 1000 live 1 births, and the illiteracy rate was negligible. Understanding the reasons why these sorts of disparities in important measures of development arise and persist is one of the greatest challenges in all of the social sciences. For a few brief years in the 1960s and the early 1970s, American legal academics joined with their colleagues in other disciplines to investigate these issues in an exercise now known as the “law and development” movement. Eventually, however, the principal participants in the exercise became disillusioned and lost their faith in the ability of law, much less legal scholarship, to contribute to development. Thus, the law * Professor of Law, New York University School of Law. E-mail: ked2@nyu.edu. This paper draws upon research conducted with Michael Trebilcock as part of a larger study of the relationship between law and development. I am grateful for helpful comments on an earlier draft from GuyLaine Charles, Barry Friedman, Henry Hansmann, Helen Hershkoff, Lewis Kornhauser, Daryl Levinson, Richard Messick, Barry Metzger, Frank Upham, Katrina Wyman and participants in an NYU Institute for Law and Society Brown Bag Luncheon and the Globalization, Law and Development Conference sponsored by the University of Michigan Law School and the William Davidson Institute. 1. World Bank, World Development Indicators 2003 (2003). 141 DAVIS TYPE.DOC 142 3/17/2005 8:41 AM Michigan Journal of International Law [Vol. 26:141 and development movement in the United States died a noisy and welldocumented death.2 Between approximately the mid-1970s and the mid-1990s the American legal academy turned its back on development studies. At the same time, mainstream development studies also largely turned its back on law. During this period development issues were viewed as the proper domain of economists and political scientists. The economists largely focused on topics such as interest rates, tariffs, subsidies and wage and price controls.3 Meanwhile, political scientists focused on meta-issues such as whether developing countries were better off adopting democracy or authoritarianism.4 Now in the post-Cold War era, the situation is completely different. Both inside and outside the academy, capitalism has triumphed and little or no effort is being made to revive communism. Virtually all attention is now focused upon methods of establishing or enhancing the performance of capitalist economies. Laws and legal institutions feature prominently in this new intellectual terrain. One reason for this is that economists have become more attuned to the potential economic functions of legal institutions. It is now widely accepted that markets are unlikely to function in the absence of bodies of contract law and systems of property rights (defined broadly to include corporate and commercial law) that encourage exchange and investment, the two main wealth-creating activities in a capitalist system.5 Similarly, political scientists are now well aware of the fact that the political arrangements that they tend to regard as being characteristic of developed societies—such as democracy, respect for human rights and the welfare state—are almost invariably defined and protected by legal norms and institutions, especially constitutions and the courts that inter- 2. Law and Crisis in the Third World (Sammy Adelman & Abdul Paliwala eds. 1993); John Henry Merryman, Comparative Law and Social Change: On the Origins, Style, Decline & Revival of the Law and Development Movement, 25 Am. J. Comp. L. 457 (1977); Brian Z. Tamanaha, The Lessons of Law-and-Development Studies, 89 Am. J. Int’l L. 470 (1995) (reviewing Law and Development (Anthony Carty ed., vol. 2, 1992)); David M. Trubek & Marc Galanter, Scholars in Self-Estrangement: Some Reflections on the Crisis in Law and Development Studies in the United States, 1974 Wis. L. Rev. 1062. 3. For overviews of the relevant literature in development economics see generally, James M. Cypher & James L. Dietz, The Process of Economic Development (1997); Michael P. Todaro, Economic Development (1999). 4. For examples of work in this vein see Seymour Lipset, Some Social Requisites of Democracy: Economic Development and Political Legitimacy, 53 Am. Pol. Sci. Rev. 69 (1959); Adam Przeworski & Fernando Limongi, Political Regimes and Economic Growth, 7 J. Econ. Persp. 51 (1993). 5. Douglass North, Institutions, Institutional Change and Economic Performance (1990); World Bank, World Development Report 1997: The State in a Changing World (1997). DAVIS TYPE.DOC 3/17/2005 8:41 AM Rule of Law Reforms Fall 2004] 143 6 pret them. These insights have led to a consensus that the overall “quality” of legal institutions is an important determinant of development.7 This view is typically expressed in shorthand form by the assertion that 8 the ”rule of law” is an essential pre-condition to development. Motivated by this consensus, multilateral institutions and developed countries have invested tremendous amounts of resources in efforts to reform legal institutions in developing countries.9 Optimistic claims about the role of legal institutions in achieving development have a long lineage. In fact, most contemporary versions of this theory can be traced back to the turn-of-the-century writings of Max Weber.10 But theoretical analyses that are skeptical about whether legal institutions play an independent role in achieving development, or any other form of social change, are equally easy to find. For example, Engels understood a society’s level of economic development as a factor explaining, rather than being explained by, the characteristics of its laws and legal institutions.11 Similarly but more recently, in declaring the end of the first law and development movement, Trubek and Galanter questioned whether legal institutions (and in particular the courts) played a central role at all in developing countries.12 Given that both optimistic and skeptical claims about law and development seem equally coherent, what explains the recent abundance of optimism? I would suggest that part of the explanation lies in the fact that in recent years the law optimists have achieved a form of methodological superiority over the skeptics by more effectively deploying empirical analyses in support of their claims. Historically, the empirical 6. See, e.g., Mancur Olson, Power and Prosperity: Outgrowing Communist and Capitalist Dictatorships 39 (2000) (discussing the role of an independent judiciary in sustaining democracy). 7. World Bank, supra note 5; see also, Hernando De Soto, The Other Path: The Invisible Revolution in the Third World 185–87 (1989). 8. See Thomas Carothers, The Rule of Law Revival, 77 Foreign Aff. 95 (Mar./Apr. 1998); Ibrahim F.I. Shihata, Complementary Reform: Essays On Legal, Judicial and Other Institutions Reforms Supported by the World Bank (1997). 9. Carothers, supra note 8. 10. Tom Ginsburg, Does Law Matter for Economic Development? Evidence from East Asia, 34 Law & Soc’y Rev. 829 (2000); David Trubek, Max Weber on Law and the Rise of Capitalism, 1972 Wis. L. Rev. 720. See generally Max Weber, Economy and Society: An Outline of Interpretive Sociology (1978). 11. Friedrich Engels, Speech at the Graveside of Karl Marx, in The Marx-Engels Reader 603 (Robert C. Tucker ed., 1972). 12. Trubek & Galanter, supra note 2. For a more recent example of a skeptical approach to law and development see Frank Upham, Mythmaking in the Rule of Law Orthodoxy (Carnegie Endowment for Int’l Peace, Rule of Law Series, Working Paper No. 30, 2002). For a more extensive discussion of the recent intellectual history of law and development see Kevin Davis & Michael Trebilcock, The Recent Intellectual History of Law and Development (2004) (unpublished). DAVIS TYPE.DOC 144 3/17/2005 8:41 AM Michigan Journal of International Law [Vol. 26:141 basis for optimistic claims about the role of law was, at least by the standards of most modern social scientists, shaky. Weber, for example, based his conclusions upon qualitative analyses of a wide range of societies, both historical and contemporary, that were of dubious accuracy. My impression, however, is that the empirical bases for skeptical claims were typically just as anecdotal and subjective. This all changed in the early 1990s, when an intrepid group of economists began to conduct statistical analyses of cross-country data to establish the existence of causal connections between quantitative measures of the characteristics of legal institutions and measures of social and economic development. The most prominent examples of these studies are ones that claim to have found a causal relationship between respect for the rule of law and levels of economic growth.13 Other studies, however, have also claimed to find causal relationships between legal variables—by which I mean simply any variable that purports to measure characteristics of a legal system—such as protection of property rights and the enforceability of contracts and various measures of the level of development.14 Still other studies have established relationships between countries’ legal heritage (meaning the historical origins of their legal system) and their levels of economic development.15 This literature is often interpreted as providing a strong empirical 16 grounding for optimistic perspectives on law and development. For example, in a well-known series of papers, Clague, Keefer, Knack, and Olson devised an innovative variable called Contract Intensive Money (CIM) that they claim measures the enforceability of contracts. After conducting a variety of regression analyses and finding a positive relationship between CIM, investment, and growth, the authors go on to state: Countries with low CIM (or low scores on other, more subjective measures of contract rights and property security) would be advised to examine closely government policies related to enforcement of contracts between private economic actors and 13. See Robert J. Barro, Determinants of Economic Growth: A Cross-Country Empirical Study (1997); Daniel Kaufmann et al., Governance Matters, (The World Bank, Policy Research Working Paper No. 2196, 1999). 14. Christopher Clague et al., Institutions and Economic Performance: Property Rights and Contract Enforcement, in Institutions and Economic Development: Growth and Governance in Less-Developed and Post-Socialist Countries 67–90 (Christopher Clague ed., 1997). 15. Paul G. Mahoney, The Common Law and Economic Growth: Hayek Might be Right, 30 J. Legal Stud. 503 (2001). 16. For a brief non-technical survey of this literature, see Frank B. Cross, Law and Economic Growth, 80 Tex. L. Rev. 1737 (2002). DAVIS TYPE.DOC Fall 2004] 3/17/2005 8:41 AM Rule of Law Reforms 145 the due process guarantees that governments afford firms and individuals when they create and implement policies.17 This understanding of the significance of cross-country statistical 18 analyses appears to be particularly prevalent within the World Bank. The World Bank has conducted, or at least sponsored, several crosscountry statistical analyses of the relationship between legal variables and measures of development. One prominent set of studies uses a large number of subjective measures of institutional quality—meaning data obtained from either polls of country experts or surveys of residents— grouped into six clusters, “voice and accountability,” “political stability,” “government effectiveness,” “regulatory quality,” “rule of law” (which includes variables measuring respect for property rights), and “control of corruption.”19 Other important studies use data derived from surveys of lawyers (the Lex Mundi study) or business enterprises (the World Business Environment Survey) based in the relevant countries.20 All of these studies generally find positive correlations between the legal variables that they employ and measures of development, and their results are commonly cited in key World Bank publications advocating legal re21 forms in developing countries. Over the past decade, in a strong display of legal optimism, the World Bank has made significant investments in funding these types of reforms.22 II. Potential Methodological Issues Statistical analyses that attempt to use legal variables to explain measures of development are open to at least three basic sorts of objec23 tions. First, one might question the choice of dependent variable. In 17. Christopher Clague et al., Contract-Intensive Money, 4 J. Econ. Growth 185, 205 (1999). 18. For an articulate statement of the role of empirical analysis as a guide to legal reform see Daniel Kaufmann, Rethinking Governance: Empirical Lessons Challenge Orthodoxy (2003), available at http://www.worldbank.org/wbi/governance/pdf/rethink_gov_stanford.pdf. 19. Kaufmann et al., supra note 13. 20. Daron Acemoglu & Simon Johnson, Unbundling Institutions (October 2004), available at http://econ-www.mit.edu/faculty/download_pdf.php?id=660. 21. World Bank, World Development Report 2002: Building Institutions for Markets 19 (2002); World Bank, World Development Report 2000/2001: Attacking Poverty 50, 103 (2000); World Bank, Doing Business in 2004: Understanding Regulation (2004). 22. See generally, World Bank, Initiatives in Legal and Judicial Reform: 2004 Edition (2004). 23. For a similar methodological discussion with greater emphasis on concerns about the reliability of cross-country legal variables, see Amanda Perry-Kessaris, Finding and Facing Facts about Legal Systems and Foreign Direct Investment in South Asia, 23 Legal Stud. 649 (2003). For methodological critiques of studies relying on other cross-country data on DAVIS TYPE.DOC 146 3/17/2005 8:41 AM Michigan Journal of International Law [Vol. 26:141 other words, one might question whether the variables chosen to measure the level of development accurately capture all of the features of a society that are relevant to promoting development. For instance, one might argue that a society’s per capita income, perhaps the most commonly used measure of development, is a poor measure of whether a society embodies the kind of tolerance, respect, and broadly distributed opportunities for human flourishing that characterize a truly developed society. My focus here is on problems that stem from the legal variables used in these studies and so, although important, this line of argument, which inevitably leads to the controversial question—what is development?—will not be pursued here. A second line of attack would be to question the choice of legal variables and ask whether they actually provide useful information about societies’ legal systems. In this context useful information must be defined to mean information that identifies specific aspects of a legal system that are amenable to reforms designed to promote development. No other type of information seems capable of supporting the optimistic claim that it is possible to promote a society’s development by modifying specific features of its legal system. Broadly speaking there are at least three ways in which a variable that purports to measure characteristics of a legal system might in fact fail to capture useful information of the sort that it purports to capture. First, the variable in question may not refer exclusively to aspects of the legal system. A legal system is typically defined as some sort of combination of norms and personnel, with the relevant norms usually being limited to those that are applied or enforced by agents of the State. At the margins there may be doubt about which norms and which personnel fall within the scope of this definition and thus qualify as part of the legal system, but it is clear that this way of defining a legal system serves to exclude at least some features of a society and its inhabitants. So for example, a variable that essentially measures the wealth of a society does not seem to qualify as a measure of any characteristic of its legal system. Similarly, variables that refer exclusively to characteristics of non-legal norms or non-legal personnel should not qualify as legal variables. Finally, there are some variables—the recurring examples in this context are crime rates—that describe the products of interaction between both legal and non-legal factors. These types of variables are of limited use as guides to legal reform unless it is clear how they respond to purely legal institutional performance see Kurt Weyland, “Good Governance” and Development: A Skeptical View (2004) (unpublished manuscript), and Michael Johnston, Measuring the New Corruption Rankings: Implications for Analysis and Reform, in Political Corruption 865 (A. J. Heidenheimer & M. Johnston eds., 2002). DAVIS TYPE.DOC Fall 2004] 3/17/2005 8:41 AM Rule of Law Reforms 147 reforms. Of course, variables that do not relate exclusively to the legal system might provide information about the potential effects of nonlegal reforms. Such information, however, is of little use to those who press the claim that legal reforms are of crucial significance or who specialize in conducting legal reforms. A second reason why a legal variable might prove to be unhelpful is if it lacks precision. Sometimes this occurs because the variable captures too broad a range of characteristics of a legal system to provide practical guidance for reformers. For example, a variable measuring the “quality of the legal system” is simply too highly aggregated to be of great assistance to reformers with finite resources. On other occasions the lack of precision stems from the fact that the label assigned to the variable does not correspond, at least in ordinary usage, to what it in fact measures. This frequently occurs because the variable is labeled to suggest that it captures information about the functioning of a legal system in a broad range of circumstances, such as enforceability of contracts simpliciter, when in fact it captures a much narrower set of information, such as enforceability of government contracts, which is not necessarily representative of the broader set of information. A third way in which a putative legal variable might prove to be unhelpful to legal reformers is if it refers to aspects of a society that are not amenable to reform. The clearest example of such a variable would be one that captures purely historical information—which obviously cannot be changed—without any analogue in any contemporary society. Whether a legal system can trace its roots back to England or France does not directly reveal anything about the present characteristics of its legal system and so showing an association between French legal heritage and poor development outcomes provides little guidance for reformers. Less clear-cut examples would be variables that are theoretically amenable to reform, but in practice can only be manipulated with difficulty. For example, in some societies fundamental constitutional norms are extremely difficult to amend. A final general approach to challenging optimistic statistical analyses of the role of law in development is to question the inferences about the relationship between law and development that researchers have drawn from their data. For instance, studies of this kind are typically vulnerable to the claim that the observed correlations simply reflect the existence of an as yet unobserved factor that causes both the characteristics of legal institutions and levels of development to vary without implying any causal connection between the latter sets of variables.24 24. See generally, Janine Aron, Growth and Institutions: A Review of the Evidence, 15 World Bank Research Observer 99 (2000); T.N. Srinivasan & Jagdish Bhagwati, DAVIS TYPE.DOC 148 3/17/2005 8:41 AM Michigan Journal of International Law [Vol. 26:141 Another concern is that the researchers have failed to capture important structural features of the relationship between the variables. For example, they might characterize a relationship as linear when it is actually non-linear. It may be the case, for instance, that judicial efficiency matters up to a point, but beyond that point its importance diminishes. It is possible to criticize the inferences that researchers draw about the relationship between legal systems and development even when they use appropriate data to measure dimensions of both legal systems and development. It is often even easier to mount such a critique, however, when they use inappropriate data. For example, one way of misstating the relationship between law and development would be to attribute high levels of development principally to legal reforms when in fact development is explained, at least to some extent, by interactions between the legal system and other aspects of a society such as morality and economic inequality. This type of error is particularly likely to occur if one purports to measure the quality of a legal system by using data that actually captures interactions between the legal system and, for example, inequality.25 III. Illustrations The methodological concerns raised in the preceding Part are not merely theoretical. In fact, a number of the most commonly used legal variables are vulnerable to complaints that they do not capture useful information or are incapable of supporting inferences about the relationship between law and development. A. Rule of Law As mentioned above, a significant number of studies have found what appear to be causal connections between variables that measure respect for the rule of law and measures of social and economic development.26 The most commonly used source of data on respect for the rule of law is a private publication known as the International Country Risk Guide (ICRG), which provides quantitative assessments by unidentified experts of the strength of the law and order tradition in various counOutward-Orientation and Development: Are Revisionists Right? in Trade Development and Political Economy: Essays in Honour of Anne O. Krueger 3, 20–22 (Deepak Lal & Richard H. Snape eds., 2001). 25. For a similar complaint that widely used measures of the quality of political institutions are actually “outcome measures,” see Glaeser et al., Do Institutions Cause Growth?, 9 J. Econ. Growth 271 (2004). 26. See sources cited supra in note 13. DAVIS TYPE.DOC Fall 2004] 3/17/2005 8:41 AM Rule of Law Reforms 149 27 tries. According to the ICRG, the law and order variable is constructed as follows: Law and Order are assessed separately, with each subcomponent comprising zero to three points. The Law subcomponent is an assessment of the strength and impartiality of the legal system, while the Order sub-component is an assessment of popular observance of the law. Thus, a country can enjoy a high rating—3—in terms of its judicial system, but a low rating—1—if it suffers from a very high crime rate or if the law is routinely ignored without effective sanction (for example, widespread illegal strikes).28 For our purposes the first point to notice is that this variable clearly includes information about both the behavior of legal personnel and the behavior of members of the general public. Consequently, it is difficult to describe this measure of the rule of law as a purely legal variable. This in turn means that studies showing a correlation between the rule of law and development tell us little or nothing about the relationship between the design of a legal system and development outcomes. Given the way that this variable is defined, the widely observed statistical relationship between the rule of law and development may boil down to a relationship between crime rates and development. Some might argue that crime rates and more generally rates of noncompliance with the law are strongly determined by, and thus represent reasonable proxies for, the presence of specific features in a society’s legal system. For example, one might presume that low crime rates in a given society reflect the existence of a harsh criminal code or a large and dedicated police force. Those rates might, however, also reflect the fact that members of the local population are unusually willing to report crimes, appear as witnesses in legal proceedings and shun offenders, or that they have particularly deep ethical commitments to law abiding behavior. More generally, prevailing ethical norms and the extent of lawabidingness are likely to depend upon an array of factors such as the strength of families and other informal social networks, the extent of poverty and economic inequality, and the degree to which state officials 27. The ICRG is one of two sources of cross-country institutional data whose use in published studies is described as “High” by the World Bank. The other source is Freedom House. However, the Freedom House data being referred to here are the data on political freedoms and civil liberties, which is clearly not designed to provide exclusively legal information. See World Bank, Indicators of Governance and Institutional Quality, available at http://www1.worldbank.org/publicsector/indicators.htm#research%20papers. 28. International Country Risk Guide, About ICRG, at http:// www.icrgonline.com/page.aspx?page+icrgmethods#_The_Composite_Risk. DAVIS TYPE.DOC 150 3/17/2005 8:41 AM Michigan Journal of International Law [Vol. 26:141 are perceived as exercising legitimate rather than illegitimate authority. Legal rules and the manner in which they are enforced may influence some or all of these features of a society, but any causal relationships are likely to be subtle, indirect, and context-specific. Theoretically speaking, it is even possible that strengthening laws or improving law enforcement will undermine the ability of moral values and family structures to serve as alternative methods of inducing compliance with the law. Consequently, it is difficult to see how information about a society’s crime rates qualifies as a reasonable proxy for information about the characteristics of its legal system. There is also a second and more subtle reason for believing that the ICRG’s rule of law variable captures a combination of legal and nonlegal information. This concern stems from the fact that the variable is based on subjective assessments made by individuals who are not necessarily resident in the society to which their assessments pertain. Those assessments will naturally be a function of the information available to the assessors about the phenomena they are assessing. The nature of that information may in turn, however, be a function of non-legal variables. Most notably, unless the assessors rely exclusively on firsthand observations or official statistics, the nature of the information they obtain about the behavior of both legal personnel and members of the general public seems likely to depend heavily on the behavior of the local news media. For example, perceptions of the integrity of state officials will depend in part on the willingness of investigative journalists to uncover political corruption. Similarly, perceptions of the frequency of various types of crimes will typically depend on the media’s willingness to report those crimes. In principle, therefore, the apparent connection between low levels of respect for the rule of law and underdevelopment could simply reflect a relationship between aggressive media reporting and underdevelopment.29 Alternatively, if aggressive media activity is more common in developed countries then existing studies might under-state the relationship between respect for the rule of law and development. Even if the ICRG’s rule of law variable were properly characterized as a legal variable it would still be of limited utility. This is because as a measure of “strength and impartiality” it does not refer to specific features of the legal system that can be modified to promote development. In addition, on account of biases and limits on the information available to the ICRG’s experts, the ICRG variable probably does not provide a 29. This possibility is not completely far-fetched. As Carothers points out, as part of a more general trend towards democratization, many developing societies have recently become more open and have begun to demonstrate greater tolerance for freedom of the press. See Carothers, supra note 8. DAVIS TYPE.DOC Fall 2004] 3/17/2005 8:41 AM Rule of Law Reforms 151 holistic assessment of any country’s legal system. For example, the selection of illegal strikes as opposed violations of health and safety legislation to illustrate the concept of routine non-compliance with law is telling.30 The significance of this point—and this general mode of criticism—should not be overstated though. A study that suggests that further work is warranted on methods of enhancing the strength and impartiality of a legal system deserves to be regarded as a contribution to knowledge about the relationship between law and development; at a minimum any such study contradicts claims that weak and biased legal systems are either irrelevant or even conducive to development. It is worth noting briefly that although the ICRG appears to be the most popular source of rule of law data, there are other sources. The variables constructed by some of these other sources give rise to the same kinds of concerns about inclusion of non-legal information and aggregation as the ICRG data. For instance, researchers at the World Bank have constructed a rule of law index that combines variables drawn from a number of different sources, including the ICRG.31 That index clearly captures subjective non-legal information embedded in crime data because in addition to the ICRG’s law and order variable, it includes variables such as “losses and costs of crime,” “kidnapping of foreigners,” “corruption in banking,” “crime,” “theft and crime,” “crime and theft as obstacles to business,” “extent of tax evasion,” and “costs of organized crime for business.” B. Property Rights There are a number of different sources of cross-national data on the protection of property rights. In fact, the term property rights is used so loosely by non-lawyers (as well as by some lawyers) that it often seems to refer to virtually every possible type of legal norm and it is not uncommon to see variables such as the ICRG’s rule of law variable described as measures of the security of property rights.32 Among the variables explicitly defined as measures of the strength of property rights, however, the best known is probably the one contained in the Heritage Foundation/Wall Street Journal Index of Economic Freedom.33 30. I owe this observation to Frank Upham. 31. See Kaufmann et al. supra note 13. 32. See, e.g., Stijn Claessens & Luc Laeven, Financial Development, Property Rights, and Growth, 58 J. Fin. 2401 (2003). For a similar complaint, see Donald C. Clarke, Economic Development and the Rights Hypothesis: The China Problem, 51 Am J. Comp. L. 90, 99 (2003). 33. The Heritage Foundation describes itself as a think tank “whose mission is to formulate and promote conservative public policies based on the principles of free enterprise, limited government, individual freedom, traditional American values, and a strong national DAVIS TYPE.DOC 152 3/17/2005 8:41 AM Michigan Journal of International Law [Vol. 26:141 The property rights component of the Index of Economic Freedom is based upon information derived from publications by the Economist Intelligence Unit and the State Department (which tend to use locally based analysts). The calculation of countries’ grades for property rights is described in the following terms: This factor scores the degree to which a country’s laws protect private property rights and the degree to which its government enforces those laws. It also accounts for the possibility that private property will be expropriated. In addition, it analyzes the independence of the judiciary, the existence of corruption within the judiciary, and the ability of individuals and businesses to enforce contracts. The less legal protection of property, the higher a country’s score; similarly, the greater the chances of government expropriation of property, the higher a country’s score.34 The actual grading scale used is as follows: SCORE PROTECTION OF PRIVATE PROPERTY 1 VERY HIGH 2 HIGH 3 MODERATE CRITERIA Private property guaranteed by the government; court system efficiently enforces contracts; justice system punishes those who unlawfully confiscate private property; corruption nearly nonexistent and expropriation unlikely. Private property guaranteed by the government; court system suffers delays and is lax in enforcing contracts; corruption possible but rare; expropriation unlikely. Court system inefficient and subject to delays; corruption may be present; judiciary may be influenced by other branches of government; expropriation possible but rare. defense.” The Wall Street Journal is a well-known daily newspaper. The World Bank describes the use in published studies of the Heritage Foundation’s property rights variable as “Low” but no other variables described as measures of property rights are described as being used more frequently. See World Bank, Indicators of Governance and Institutional Quality, supra note 27. 34. William W. Beach & Marc A. Miles, Explaining the Factors of the Index of Economic Freedom, in 2004 Index Of Economic Freedom (Marc A. Miles et al. eds., 2004). DAVIS TYPE.DOC 3/17/2005 8:41 AM SCORE 153 Rule of Law Reforms Fall 2004] PROTECTION OF PRIVATE PROPERTY 4 LOW 5 VERY LOW CRITERIA Property ownership weakly protected; court system inefficient; corruption present; judiciary influenced by other branches of government; expropriation possible. Private property outlawed or not protected; almost all property belongs to the state; country in such chaos (for example, because of ongoing war) that property protection is nonexistent; judiciary so corrupt that property is not effectively protected; expropriation frequent. In addition to reporting numerical grades for property rights (and other variables), the Heritage Foundation provides brief discussions of the factors influencing the assigned grades. For example, here is the discussion of the property rights grade assigned to Jamaica in 2004: Property Rights Score: 3.0 The likelihood of expropriation is remote, and private property is protected. The U.S. Department of State reports that “Jamaica’s legal system is based on English common law principles and the rules in relation to the enforceability of contracts are therefore based on the English common law. The Jamaican judicial system therefore recognizes and upholds the sanctity of contracts.” However, the judiciary lacks adequate resources, and this creates delays. In some cases, according to the U.S. Department of State, “trials . . . are delayed for years, and other cases are dismissed because files cannot be located.” An inadequate police force further weakens the security of property rights; the same source reports that “crime poses a greater threat to foreign investment than do politically motivated activities.” Similarly, here is the discussion of the grade assigned to India in 2004: DAVIS TYPE.DOC 154 3/17/2005 8:41 AM Michigan Journal of International Law [Vol. 26:141 Property Rights Score: 3.0 Protection of property rights is applied unevenly in India. The Economist Intelligence Unit reports that “large backlogs create delays—sometimes years long—in reaching decisions. Consequently, foreign corporations often include clauses for international arbitration in their contracts.” According to the U.S. Department of State, “Critics say that liquidating a bankrupt company may take as long as 20 years.” Protection of property for local investors, particularly the smallest ones, is weak. For present purposes there are a number of striking features of the Heritage Foundation’s property rights variable. The first is the fact that, like the ICRG’s rule of law variable, it seems to be picking up both legal and non-legal information. For instance, the discussion of the Jamaican grade suggests that in some countries at least the Heritage Foundation uses crime rates as a factor in calculating the property rights grade. As outlined above there are a number of potential non-legal determinants of crime rates. A similar concern arises to the extent that the property rights variable captures information about the likelihood of expropriation.35 At first glance this variable seems to be inherently legal in nature because it is simply designed to capture information about the behavior of governmental actors. The complicating factor, however, is that the behavior of those governmental actors is often influenced by non-legal factors.36 For instance, officials’ willingness to expropriate property is likely to be determined by levels of popular support for their actions which may in turn be shaped by non-legal factors such as whether ownership of property is 37 concentrated in the hands of foreigners or particular ethnic groups. There is yet another sense in which the utility of the Heritage Foundation’s property rights variable is arguably compromised by the inclusion of information pertaining to non-legal factors. The discussions accompanying the grades for India and Jamaica reveal that the property rights variable takes into account whether the capacities of legal organizations such as the police or the civil courts are adequate in light of the demand for their services. It seems incorrect to presume that the demand for the services of legal organizations is wholly dependent on legal fac35. It is unclear from the Heritage Foundation’s documentation whether they are concerned with measuring the risk of expropriation or the risk of expropriation without compensation. 36. Cf. Glaeser et al., supra note 25. 37. See Amy L. Chua, The Privatization-Nationalization Cycle: The Link Between Markets and Ethnicity in Developing Countries, 95 Colum. L. Rev. 223 (1995). DAVIS TYPE.DOC Fall 2004] 3/17/2005 8:41 AM Rule of Law Reforms 155 tors. For example, the demand for policing depends in part on the volume of crime. Similarly, the demand for civil trials depends in part on the volume of disputes in a society, which in turn depends, among other things, on culturally determined attitudes towards resolving disputes formally as opposed to informally. The upshot of the preceding three points is that, along a number of dimensions, the Heritage Foundation variable seems to capture information about the interaction between legal and non-legal aspects of society, rather than purely legal variables. Generally speaking, it is not clear that this kind of information is useful because it makes it difficult to analyze the merits of legal reforms. To illustrate, consider the fact that by taking into account whether the courts are given enough resources to avoid delays, the Heritage Foundation variable appears, at least in part, to provide information on the extent to which legal institutions respond satisfactorily to local demand for protection of property rights. On the one hand, this kind of data can be used to justify claims that investments in hiring and training additional judges will be conducive to development. On the other hand, this information will be misleading if it turns out to be most cost-effective to invest in altering the non-legal rather than the legal side of the equation. For example, the easiest way of easing under-capacity in the civil courts might be to encourage non-legal forms of dispute resolution. Another striking feature of the Heritage Foundation’s property rights variable is the range of purely legal information that it appears to capture. In terms of legal categories, the country-specific discussions reproduced above suggest that the property rights variable encompasses virtually every aspect of the legal system, ranging from criminal law, to bankruptcy law, to contract law, to civil procedure. The aggregation of information about the protection of property rights from incursions by both the government and private individuals may be particularly problematic. At a purely doctrinal level, in some legal systems the question of whether there is an inviolable right to protection from expropriation without compensation is regarded as a matter of constitutional law, while the right to compensation for and protection from private actors’ infringements of property rights is governed by norms that do not originate from the constitution. From a functional perspective this difference in sources of law will be significant if, as is frequently the case, it is more difficult to reform constitutional norms than it is to reform other types of norms. Consequently, even if the property rights variable only captures legal information, because it aggregates information about a number of aspects of the legal system which cannot necessarily be manipulated DAVIS TYPE.DOC 156 3/17/2005 8:41 AM Michigan Journal of International Law [Vol. 26:141 through the same mechanisms or with equal ease, the variable may be of limited use as a guide to legal reform. Other variables purporting to measure the extent of protection of property rights suffer from some or all of the defects of the Heritage Foundation data. For instance, in connection with the preparation of the 1997 World Development Report, the World Bank conducted a survey of approximately 3000 enterprises in sixty-nine countries. The purpose of the survey was to understand private firms’ perceptions of the constraints that government action places on them. The two questions described by the survey’s authors as having to do with “security of property rights” were as follows: “Theft and crime are serious problems that can substantially increase the costs of doing business.” To what degree do you agree with this statement? “I am not confident that the state authorities protect my person and my property from criminal actions.” To what degree do you agree with this statement?38 Both these questions seem to invite responses that describe existing levels of crime and the state’s capacity to deal with them. Again, it is not clear to what extent crime levels are influenced by either the content of legal norms or the manner in which they are enforced. Consequently, it is not clear to what extent data derived from these survey questions can be used to test claims about the independent role that legal reforms might play in development. Another popular source of cross-country data on property rights is the ICRG’s measure of the risk of government expropriation.39 This variable is much more narrowly focused than the Heritage Fund’s property rights variable. As discussed above, however, it is not clear that the risk of governmental expropriation is properly regarded as a purely legal variable. C. Contract Enforcement Aside from protection of property rights, enforcement of contracts is generally considered to be one of the most crucial economic functions that a legal system can perform.40 Unfortunately, much of the cross- 38. World Bank, World Development Report 1997, Private Sector Survey, available at http:// www.worldbank.org/wbi/governance/wdr97data.html. 39. World Bank, Indicators of Governance and Institutional Quality, supra note 27. 40. See Christopher Clague et al., supra note 17. DAVIS TYPE.DOC 3/17/2005 8:41 AM Rule of Law Reforms Fall 2004] 157 national data that purports to measure the enforceability of contracts is as unsatisfactory as the data on protection of property rights. 1. Contract Intensive Money One prominent cross-country measure of the enforceability of contracts is Clague’s “contract-intensive money.”41 CIM is defined as the ratio of non-currency money to the total money supply. The logic behind this definition is that the more effective legal methods of enforcing contracts and securing property rights are, the more willing actors in a society should be to hold money in a form other than currency. This is because the value of non-currency money depends upon both depositors’ ability to enforce claims against banks and banks’ ability to enforce claims against borrowers. Furthermore, because of its traceability, noncurrency money is not likely to be an attractive medium of exchange in societies where the government not only declines to enforce but actively prohibits a large number of transactions. Thus, its creators suggest that CIM is a good measure of the quality of a society’s legal institutions and, as we have seen, recommend that societies with relatively low amounts of CIM strive to improve the quality of their legal institutions. One of the first things, however, to notice about CIM is that it does not purport to be a highly specific measure of the enforceability of contracts. In fact, its creators describe CIM as a measure of both the enforceability of contracts and the security of property rights. This aggregation across legal doctrines naturally limits CIM’s potential utility for legal reformers. More fundamentally, CIM seems to be a measure of the combined strength of both legal and non-legal factors that influence people’s willingness to hold non-currency money. There is simply no way to determine whether societies with high CIM are societies in which actors have faith in legal methods of recovering the value embodied in noncurrency forms of money, or are societies in which they have faith that some combination of moral norms and social sanctions will discourage other actors from reneging on their obligations. 2. ICRG Another widely used measure of the enforceability of contracts is the ICRG’s measure of the risk that government will repudiate a contract. This is defined as an assessment of “the risk of a modification in a contract taking the form of a repudiation, postponement, or scaling down” due to “budget cutbacks, indigenization pressure, a change in 41. Id. DAVIS TYPE.DOC 158 3/17/2005 8:41 AM Michigan Journal of International Law [Vol. 26:141 42 government or a change in government economic and social priorities.” This variable is narrowly focused but, like the ICRG’s measure of the risk of expropriation, it is not clear that it captures useful legal information because many of the factors that determine whether the government decides to repudiate a contract may be non-legal in nature. 3. World Business Environment Survey A broader and perhaps more useful cross-country measure of the enforceability of contracts can be found in the World Business Environment Survey, which is administered to representatives of various types of private enterprises by the World Bank in eighty countries and one territory. One of the questions on that survey is as follows: 8. “I am confident that the legal system will uphold my contract and property rights in business disputes.” To what degree do you agree with this statement? Now 3 years ago (1) fully agree (2) agree in most cases (3) tend to agree (4) tend to disagree (5) disagree in most cases 43 (6) fully disagree This question requests data on a broad range of business disputes as opposed to just contractual disputes. Also, because it does not ask respondents to elaborate upon the grounds for their assessments, further research will be required to determine how various characteristics of the legal system influence those assessments. For example, one might be interested in whether confidence in the legal system is most greatly influenced by concerns about judicial integrity, or the time it takes a case to get to trial, or the social distance between judges and litigants. Nevertheless, the data provided by this survey question seems capable of providing more guidance to prospective legal reformers than do the ICRG or CIM variables. 42. Rafael La Porta et al., Law and Finance, 106 J. Pol. Econ. 1113 (1998). 43. World Business Environment Survey, Measuring Conditions for Business Operation and Growth, Private Enterprise Questionnaire, available at http:// info.worldbank.org/governance/wbes/wbes_questions.asp. DAVIS TYPE.DOC Fall 2004] 3/17/2005 8:41 AM Rule of Law Reforms 159 4. The Lex Mundi Project An even more useful source of data on the enforceability of contracts is a paper entitled Courts derived from a study known as the Lex Mundi project.44 The authors of that study collected data on the procedural steps required to enforce two simple types of contracts in 109 countries. Among other things they asked law firms in those countries to describe all of the steps required to evict a tenant and collect on a bad check in their country’s largest city as well as the estimated time required to complete each step. They then demonstrate that the survey responses can be coded in such a way as to enable statistical analyses of the relationship between the norms and practices described by the respondents and various indicators of the overall quality of the legal system. Because the surveys poses highly specific questions and the norms and practices described by the respondents may be readily modified, these data ought to be eminently useful to would-be legal reformers.45 It is worth noting, however, that despite claims to the contrary by the designers of the Lex Mundi study and others who have subsequently used the study for the purposes of testing the relationship between law and development, it is misleading to characterize data derived from this study as an overall measure of the enforceability of contracts in any given jurisdiction. At most, the data can be described as measures of the enforceability of particular types of contracts in each jurisdiction as there is no reason to presume that any given legal system treats all contractual claims similarly. For example, in many jurisdictions residential tenants are granted significant levels of protection from eviction for ideological reasons that have no application in cases involving commercial contracts. As for cases involving bounced checks, in some societies checks are rarely used as methods of payment and so it may not be reasonable to assume that proceedings based upon nonpayment of a check will be routine and thus straightforward. This general concern seems to be inherent in the methodology used in this study because in order to obtain data that is comparable across countries from a survey of professionals it is important to question them about a specific scenario and there will inevitably be limits on the extent to which it is possible to solicit information about variants of that scenario, even in contexts where those variants are more relevant than the baseline scenario. 44. Simeon Djankov et al., Courts, 118 Q.J. Econ. 453 (2003). 45. For other examples of such surveys of legal professionals, see also World Bank, Doing Business in 2004: Understanding Regulation, supra note 21. DAVIS TYPE.DOC 160 3/17/2005 8:41 AM Michigan Journal of International Law [Vol. 26:141 D. Legal Heritage It is now well established that there is a correlation between a country’s legal heritage and its levels of economic growth.46 Even if one accepts the notion that these studies succeed in demonstrating a connection between legal heritage and development,47 the studies are of little direct use to prospective legal reformers, simply because a society’s legal heritage cannot be changed. From a reformer’s perspective, the principal lesson to be taken from these studies is that an effort ought to be made to identify the distinctive contemporary characteristics of legal systems associated with each legal heritage so that the connections between those characteristics and indicia of development can be investigated directly. This is the approach taken by Lee, who posits that the essential structural differences between common law and civil law systems can be captured by variables measuring the length of judges’ tenure, whether precedent is regarded as a source of law, and whether the ordinary courts have the ultimate authority to control administrative action.48 Unlike legal heritage these variables represent specific features of a society’s legal system that ought to be amenable to reform. Consequently, from the perspective of prospective legal reformers, Lee’s variables seem to capture more useful information than the legal heritage variables. IV. Conclusion Cross-country empirical analyses that purport to establish the existence of a causal relationship between legal reforms and development are premised on the assumption that the legal variables that they use reasonably capture characteristics of legal systems that are amenable to reform. As we have seen, that assumption is not necessarily valid. Many 46. See generally Mahoney, supra note 15. 47. In a well-known series of articles, a group of scholars whose core consists of Simeon Djankov, Rafael La Porta, Florencio Lopez-De Silanes, Andrei Shleifer and Robert Vishny has uncovered relationships between countries’ legal heritages—common law, French civil law, German civil law, etc.—and the extent to which their legal systems protect shareholders’ and creditors’ rights, regulate entry of new firms, regulate labor markets, and adopt formalistic judicial procedures. See Simeon Djankov et al., The Regulation of Entry, 117 Q. J. Econ. 1 (2002); Djankov et al., supra note 44; Rafael La Porta et al., Legal Determinants of External Finance, 52 J. Fin. 1131 (1997); La Porta et al., supra note 42; Rafael La Porta et al., Agency Problems and Dividend Policies Around the World, 55 J. Fin. 1 (2000); Juan Botero et al., The Regulation of Labor, 119 Q. J. Econ. 1339 (2004). My understanding of these studies is that they are not designed to investigate the connection between legal heritage and development. If this interpretation is correct then these studies are beyond the scope of the present discussion. 48. Injae Lee, Essays on Legal System and Economic Performance (Jan. 2003) (unpublished Ph.D. dissertation, New York University, Department of Economics) (on file with author). DAVIS TYPE.DOC Fall 2004] 3/17/2005 8:41 AM Rule of Law Reforms 161 of the variables that are most commonly used to measure respect for the rule of law, the protection of property rights and the enforceability of contracts, fail this test because they do not necessarily describe purely legal information. Meanwhile, the legal heritage variables do not seem to capture information about aspects of contemporary legal systems that are amenable to reform. Under the circumstances, the cross-country studies that use this data provide little basis for favoring an optimistic rather than a skeptical attitude towards the role of law in development. This suggests that additional empirical research employing more useful legal variables will be required in order to resolve this crucially important debate. Fortunately, the state of the art in collection of this sort of data continues to improve—witness, for instance, the advances reflected in the data from the relatively recent World Business Environment Survey and the Lex Mundi project (as compared to the ICRG or CIM data)— and so eventually empirical research in this field may come to fulfill its promise.