The Bulging Pocket and the Rule of Law: Corruption, Inequality, and Trust*

The Bulging Pocket and the Rule of Law: Corruption, Inequality, and Trust Eric M. Uslaner Department of Government and Politics University of Maryland–College Park College Park, MD 20742 [email protected] Corruption is generally considered to be a curse. It flouts rules of fairness and gives some people advantages that others don’t have. Corruption transfers resources from the mass public to the elites–and generally from the poor to the rich. It acts as an extra tax on citizens, leaving less money for public expenditures. Economic inequality provides a fertile breeding ground for corruption–and, in turn, it leads to further inequalities. Most of the models of corruption focus on the institutional determinants of government dishonesty. However, such accounts are problematic. Corruption is remarkably sticky over time. There is a very powerful correlation between cross-national measures corruption in 1980 and in 2004. In contrast, measures of democracy such as the Freedom House scores are not so strongly correlated over time–and changes in corruption are unrelated to changes in institutional design. On the other hand, inequality and trust–like corruption–are also sticky over time. The connection between inequality and the quality of government is not necessarily so simple: As the former Communist nations of Central and Eastern Europe show, you can have plenty of corruption without economic inequality. The aggregate relationships between inequality and corruption are not strong. The path from inequality to corruption may be indirect–through generalized trust–but the connection is key to understanding why some societies are more corrupt than others. I estimate a set of simultaneous equations cross-nationally and show that there is strong support for the linkage from high inequality to low generalized trust to high levels of corruption–and then back to higher levels of inequality, leading to an “inequality trap.” I also develop a new measure of government effectivness from the World Economic Forum Executive Opinion Survey and show that corrupt governments are less effective. High levels of economic inequality lead to low levels of generalized trust, which in turn lead to greater corruption. Poverty and ethnic tensions also lead to less effective government. The simultaneous equation estimations for corruption, effective government, trust, inequality, a country’s risk level, and strangling regulation suggests that the roots of bad government and corruption are largely economic and cultural. The one institutional factor that clearly matters is the fairness of the legal system, which leads to greater corruption and to more strangling regulation. Regulatory policy does have a powerful effect on corruption and on the level of risk that determine how much money countries can borrow. In turn, higher level of corruption leads to more inequality, completing the vicious cycle of the inequality trap. They tell you that the best in life is mental-Just to starve yourself and do a lot of reading Up in some garret where the rats are breeding. Should you survive it’s purely accidental.... Now once I used to think it would be worthy To be a brave and sacrificing person. I soon found out it wasn’t reimbursin’ Decided to continue being earthy..... Where’s the percentage? asks Mack the Knife. The bulging pocket makes the easy life. From “The Ballad of the Easy Life,” Berthold Brecht and Kurt Weill, The Threepenny Opera Uslaner, “The Bulging Pocket and the Rule of Law” (4) “I think,” said Martin Lomasny, “that there’s got to be in every ward somebody that any bloke can come to–no matter what’s done–and get help. Help, you understand, none of your law and your justice, but help.” So the Boston ward leader for the Democratic party told political reformer (or “muckraker”) Lincoln Steffens (1931, 618) in the 1920s. Lomasny’s constituents were poor immigrants who often found themselves on the wrong side of the law. They came to the leader–or if they were not out on bail, their families came to him–and sought freedom. They had little faith in the legal system, which was clearly biased in favor of people with money to hire high-priced lawyers. They had little faith in (or interest in) due process of law. They had to feed their families. Sure, Lomasny was corrupt himself. But he was their lifeline against a system that was stacked against them. Lomasny accumulated power by accumulating debts from the constituents he helped. He had a commodity–the ability to intervene on behalf of a citizen–that ordinary people did not possess. Lomasny, like other patrons, was not merely one among equals of his clients. Patronclient relations rest upon a foundation of inequality. The corruption of the leaders was simply their fee for providing services to the poor–and often undeserving: After all, if you came to Martin Lomasny to ask him to get a family member out of jail, you were usually not in a position to claim moral superiority. The problems of corruption arise, paraphrasing Wittgenstein, when morality goes on holiday. The lines of virtue and vice were blurred in Lomasny’s Boston. Both 1 Uslaner, “The Bulging Pocket and the Rule of Law” (5) the churches and Lomasny’s political machine “provided help and counsel and a hiding place in emergencies for friendless men, women, and children who were in dire need, who were in guilty need, with the mob of justice after them” (Steffens, 1931, 618). Corruption is the story of how the rich exploit the poor–and how the poor have neither the political or moral resources to rebel. Corruption is thus part of an inequality trap that saps people (especially at the bottom) of the belief that it is safe to trust others. Corruption flouts rules of fairness and gives some people advantages that others don’t have. Corruption transfers resources from the mass public to the elites–and generally from the poor to the rich (Tanzi, 1998). It acts as an extra tax on citizens, leaving less money for public expenditures (Mauro, 1997, 7). Corrupt governments have less money to spend on their own projects, pushing down the salaries of public employees. In turn, these lower-level staffers will be more likely to extort funds from the public purse. Government employees in corrupt societies will thus spend more time lining their own pockets than serving the public. Corruption thus leads to lower levels of economic growth and to ineffective government (Mauro, 1997, 5). Indeed, corruption is often considered the most widely accepted measure of the quality of government (since there is no objective cross-national measure of what good government is). Does corruption, like a rotting fish, start at the head and spread downward? Are bad leaders and weak institutions the cause of corruption? Or does corruption stem from a society’s culture and the distribution of resources? I argue for the latter: The roots of corruption lie in the unequal distribution of resources in a society. When we look back at leaders of “political machines” such as Martin Lomasny and the patronage systems that sustained them, we see a system of corruption that thrived on economic inequality. Economic inequality provides a fertile Uslaner, “The Bulging Pocket and the Rule of Law” (6) breeding ground for corruption–and, in turn, it leads to further inequalities. The connection between inequality and the quality of government is not necessarily so simple: As the former Communist nations of Central and Eastern Europe show, you can have plenty of corruption without economic inequality. The path from inequality to corruption may be indirect–through generalized trust–but the connection is key to understanding why some societies are more corrupt than others. The argument from inequality to low trust to corruption–and back again both to low trust and greater inequality–stands in contrast to the more common approach to explaining corruption as stemming from deficient institutions. I shall argue that the roots of corruption are largely not institutional, but rather stem from economic inequality and a mistrusting culture, which itself stems from an unequal distribution of wealth. There is one key institutional structure that I posit to affect corruption, both directly and indirectly. It is the fairness of the legal system. The fairness of the legal system is distinct from the efficiency of the legal system (Rothstein and Stolle, 2002). I offer a cross-national model of inequality, trust, corruption, regulation, a country’s risk rating (for potential investors), and the quality of government –and argue that the roots of corruption lie largely in culture and policy choices. I find no direct effect for inequality, but rather an indirect one through generalized trust. The fairness of the legal system has both direct and indirect effects on corruption. The indirect effects come through the impact of the fairness of the legal system on strangling regulation–which, in turn, leads to greater corruption. My argument stands in contrast to more traditional institutional accounts of corruption, which often suggest that the cure for malfeasance is to put the bad guys in jail. If we do so (and Uslaner, “The Bulging Pocket and the Rule of Law” (7) we ought to do so), they will be replaced by other corrupt leaders. Nor do we need a reformed system of government that either centralizes power to herd in independent “entrepreneurs” who extort businesses or average citizens (Treisman, 1999) or decentralizes power to prevent an allpowerful “grabbing hand” (DiFrancesco and Gitelman, 1984, 618; Fisman and Gatti, 2000; Kunicova and Rose-Ackerman, 2004). The inequality trap is hard to break. I posit a model where inequality, mistrust, and corruption are mutually reinforcing: inequality –> low trust –> corruption –> more inequality. My argument on the sources of corruption is largely pessimistic: Corruption is not easy to eradicate if it is largely based upon the distribution of resources (economic inequality) and a society’s culture (trust in people who may be different from yourself). Changing institutions may not be easy, but its difficulty pales by comparison with reshaping a society’s culture or its distribution of wealth (and power). While my account is largely cultural and economic, institutions do matter. Institutions shape the policies that can reduce inequality. Yet, institutions don’t matter because they are somehow “unmoved movers.” The corrupt state, the weak state, the overcompensating state, and even the good state are the creatures of their social structures. Aside from the fairness of the legal system, institutional quality seems to be more the effect rather than the cause of trust, inequality, and corruption. I shall argue below that institutions are more than end products–they might not have big effects on corruption or trust, but policies that might reduce inequality are, after all, the choices of legislatures and executives. Corruption has consequences: Dishonesty leads to a less stable economic and political Uslaner, “The Bulging Pocket and the Rule of Law” (8) environment–as reflected in the ratings of overall risk by the Interforecasting firms. These ratings guide decisions by investors and lending agencies. Governments considered very risky overcompensate by trying to control too much–enacting regulations that are so strict that they provide greater opportunities for state capture and further corruption. Even more critically, corruption leads to less effective government, but effective government does not lead to either less corruption, more trust, or less inequality. I present a new measure of government effectiveness derived from cross-national surveys of business executives (the 2004 Executive Opinion Survey of the World Economic Forum). Rothstein and Toerell (2005) see as fair institutions as the foundation of good government. There is much to be said for this argument: Kurer (2003) argues that fairness lies at the heart of honest government. Yet, this equation of honesty and government quality conflates corruption with government performance. I shall argue below that corruption does not generally depend upon structural factors–with the exception of the fairness of the legal system. There is a wide literature linking corruption to institutional structures, especially the electoral and constitutional systems. Even more important should be how well these institutions function–and this shall be a focus of my analysis. I shall argue that well-functioning governments are no more immune from corruption than are poorly-functioning states. Equally critical is the lack of a direct connection between institutional design--Presidential versus parliamentary system, proportional representation versus plurality electoral systems, the responsiveness of the bureaucracy, or even how democratic a country is–determine corruption levels or even government effectiveness. Good government, instead, occurs where trust is high, corruption is low, few people live in poverty, and there are minimal ethnic tensions. Uslaner, “The Bulging Pocket and the Rule of Law” (9) Corruption matters, then, because it leads to less effective government, less trust, and greater inequality (Rose-Ackerman, 2004, 6. 14). The path backward from corruption to greater inequality is direct, even though the path forward is not (through trust). Rose-Ackerman (2004, 6) argues that the route from corruption to inequality may be complex: Corrupt officials spend too much on big projects (infrastructure), where they can channel contracts to their cronies. As corrupt officials empty the treasury, there is less money available for investments in programs that benefit those at the bottom of the economic ladder such as education. Corruption leads to less effective government. While there is no path from ineffective government to larger disparities in wealth, effective government might lead to poilcy choices that promote policies that can break the poverty trap–if we could only get good government. Corruption rests upon a foundation of economic inequality and low trust. The same sort of factors that lead to greater corruption also lead to less effective government, to lower trust, and to more inequality. Yet, the idealized path from good government to greater equality is, as Samuel Johnson said of second marriages, a triumph of hope over experience. Good government, I shall 2 show, has lots of positive effects, but policies promoting greater equality are not among them. Corruption, Inequality, Trust, and Institutional Design: Deep Roots and Crabgrass Why is corruption pervasive in some societies and not in others? Can we reform our political systems to reduce corruption? What lessons, in particular, can we learn from the experiences of transition countries about the causes and consequences of Communism? Rothstein (2001) argues that corruption stems from venal leaders and weak institutions. A strong legal system would create a sense of social insurance for ordinary citizens: Neither their fellow citizens nor the government could exploit them if there were an independent and honest Uslaner, “The Bulging Pocket and the Rule of Law” (10) judiciary that ensured compliance with the law. Trusting others would be less risky. Stronger institutions such as courts, independent media, decentralization, property rights, and political stability have loomed large as both explanations for why some societies are less corrupt–and as prescriptions for reducing corruption (Adsera, Boix, and Payne, 2000; Treisman, 2000, among others). Most studies of corruption as well as virtually every anti-corruption agency focus on institutional sources of malfeasance. Such accounts are attractive for two reasons. First, institutional explanations are very much in vogue in the social sciences and they seem very reasonable, even attractive, in explaining coruption. If corruption is a sign of bad government, then “good” institutions seem to be a reasonable remedy. Authoritarian governments are not accountable to their citizens, so democratization offers great hope for more responsive governance and less pilfering from the public purse (You and Khagram, 2005), as does greater political stability (Leite and Weidmann, 1999, 20; Treisman, 2000). A free press with large circulation can expose malfeasance by people in power (Adsera, Boix, and Payne, 2000). Local control increases the number of access points for corruption, so federalism (as opposed to unitary) government leads to greater corruption–as does the share of government revenues spent at the local level (Treisman, 1999; Fisman and Gatti, 2000). The structure of the governmental and electoral systems also determine the level of corruption in a country, Kunicova and Rose-Ackerman (2004) argue: Presidential systems centralize power in a single executive, who is better situated to extort rents than the large number of actors in a parliamentary system. The effect is particularly pronounced in Presidential systems with closed electoral lists that give great power over nominations and elections to political Uslaner, “The Bulging Pocket and the Rule of Law” (11) parties, which are more likely sources of corruption than individual legislators. Second, anti-corruption organizations such as Transparency International place great emphasis on structural reform because it seems to be more feasible. Political institutions may not change so readily, but they are not nearly as sticky as are values such as trust and economic conditions such as inequality. A top-down, or institutional, approach offers greater hope for anticorruption efforts. In some arguments, it is simply a matter of putting the evil doers in jail and a stronger court system or fewer temptations for the “grabbing class” such as higher salaries for public employees (LaPorta et al., 1999; Mauro, 1997, 5; Tanzi, 1998, 573; Treisman, 2000; but cf. Rose-Ackerman, 1978, 90-91). However, institutional explanations for corruption face four insurmountable difficulties. First, they are not always internally consistent. Does centralization (a single chief executive) or decentralization (federalism, local spending rates) promote malfeasance? Second, institutions are generally the products of their cultures. Almost 60 percent of the Presidential systesms with closed party lists in the analysis of Kunicova and Rose-Ackerman are in Latin America or in transition countries–so this institutional factor may reflect the low levels of generalized trust in these countries. Third, institutions may do a good job of curbing corruption in the short run. Courts can put dishonest politicians or business people in jail, as the Russians have done with Mikhail Khodorskovsy. And the media can publicize malfeasance and put an end to it (Adsera, Boix, and Payne, 2000; Brunetti and Weder, 2003). Convictions and press campaigns have two weaknesses. They are selective, focusing on the most visible--but not always, as the case of Russian oil entrepreneur Mikhail Khodorkovsky shows, the most guilty. Press campaigns Uslaner, “The Bulging Pocket and the Rule of Law” (12) against corruption are like crabgrass control: You can pull out the offending weeds, but new growth will come back before you know it. Quick fixes gets rid of a handful of corrupt officials or businesspeople. But they don’t “solve” the problem of endemic corruption. McMillan and Zoido (2004) tell a fascinating story of how Vladimir Montesinos Torres, chief of Peru’s secret police and the right hand man of President Alberto Fujimori, bribed judges, politicians, and especially the owners of television stations in the 1990s. Montesinos not only kept meticulous records of these bribes–and the written receipts that he insisted each recipient give him, but he also videotaped the negotiations. Ultimately, an opposition politician was able to get a copy of one of these videos and played it at a press conference in 2000. One small cable channel, which Montesinos had not bribed, began playing the video again and again. Other networks soon followed suit, even though they had been part of the conspiracy. Montesinos and Fujimori were soon indicted, and Montesinos’s corrupt circle of more than 1600 people collapsed (McMillan and Zoido, 2004). These revelations toppled Fujimori’s government as the President fled to Japan, the home of his ancestors. Did the revelations lead to cleaner government? McMillan and Zoido (2004, 89-91) caution that media exposure of misdeeds may not itself be sufficient to end corruption: Ukraine and Russia in the 1990s serve as potent counterexamples. Peruvians had already begun to sour on the Fujimori administration’s poor performance on the economy. Yet, they still argue (McMillan and Zoiodo, 2004, 91): “Safeguards for the media—ensuring they are protected from political influence and are credible to the public—may be the crucial policies for shoring up democracy.” Uslaner, “The Bulging Pocket and the Rule of Law” (13) Yet, the media revelations seem to have had little effect in either the short or the immediate term. In 1998, while Montesinos’s operations were in full swing, Peru’s score on Transparency International’s (TI) Corruption Percpetions Index was 4.5 (with higher scores indicating greater “transparency” or honesty)–about the level of Italy or Uruguay. Peru ranked 41 out of 88 countries in the TI ratings. By 2002, its score fell to 4.0 (more corrupt), ranking st 45 out of 102 and tied with Brazil, Bulgaria, Poland, and Jamaica. By 2005, the index had th fallen again to 3.5, tied for 65 out of 160 countries rated (and tied with Mexico, Panama, th Turkey, and Ghana). The increase in corruption from 1998 to 2005 for Peru was the seventh greatest among the 85 countries sampled in both years–ranking behind only Zimbabwe, Costa Rica, Belarus, Malawi, Côte d'Ivoire, and Poland (tied with Namibia. The ratio of the 2005 to the 1998 score was the ninth smallest of 85 countries. The exposure of corrupt media and the ensuing media campaign that felled the government failed to have lasting effects. Five years later, corruption seems to have gotten worse rather than better. The videos of the chief of secret police negotiating bribe prices seem to be no more lasting than episodes of CSI. Fourth, and most critically, institutional explanations may not be sufficient to explain the persistent stickiness of corruption. Institutional structures are far more malleable than is corruption. Two of the most widely used indicators of democratization, Freedom House’s political rights and civil liberties indices, are not so stable over time. The r for political rights 2 from 1973 to 2003 is .165 and for civil liberties it is .263 (both N = 77). Even excluding countries that were Communist in 1973, the respective r values increase only to .264 and .375 2 (N = 67). If we restrict the comparison period to begin in 1988, the year before the revolutions in Uslaner, “The Bulging Pocket and the Rule of Law” (14) Central and Eastern Europe, the picture does not look much brighter. The political rights index of Freedom House showed considerable change from 1988 to 2003: On the 7 point scale (with lower scores indicating greater political rights), the mean score improved from 3.138 in 1988 to 2.554 in 2003. For former and present Communist nations, the score moved from 5.789 to 2.917. More critically, there was virtually no correlation between the 1988 and 2003 scores. For 62 nations, there is a significant, though not powerful, coefficient of earlier scores on the later ones (t = 2.22, R = .076). For the 17 Eastern bloc countries, the coefficient is not at all 2 significant (t = -.05. R = .0001). If political structures change but trust, inequality, and 2 especially corruption persist, it is hardly surprising that institutional reform will be insufficient to generate transparency. Even more troubling, the changes in political rights and civil liberties from 1973 to 2003 are unrelated to changes in corruption from 1980-85 to 2004 ( r = .007 and .038 respectively, N 2 = 38). Moving the democratization measures forward to 1988 does not improve the fit with changes in corruption ( r = .004 and .0005 for political rights and civil liberties, N = 39). 2 Institutions are not nearly as sticky as corruption–and structural change does not track the level of transparency (cf. Uslaner, 2004a). Strong institutions do not emerge from constitutional or institutional engineering. Strong institutions reflect the underlying culture of a society. In former Communist countries in particular, corrupt officials and business people are rarely held to account. While crime spiralled in Russia after the fall of Communism and the number of trials for malfeasance increased dramatically, conviction rates plummeted (Varese, 1997). The dilemma is simple: You can’t create trust (or reduce corruption) by creating new courts. A strong legal system depends upon a Uslaner, “The Bulging Pocket and the Rule of Law” (15) respect for the law, not on the number of courts or the number of trials. Instead, I argue that corruption is largely, though not exclusively. a bottom-up phenomenon. It reflects low levels of generalized trust and high levels of economic inequality. This argument is powerful for two reasons. First, it suggests strong limits on what institutional engineering (such as anti-corruption campaigns to put unscrupulous politicians in jail or to increase the penalties for malfeasance) can accomplish. Second, it creates a clear link among perceptions of corruption and increasing inequality, pessimism for the future, declining trust in other people (and in government), and opposition to market reforms in the transition countries that I find in public opinion surveys. Most critically, corruption is sticky. There is little evidence that countries can escape the curse of corruption easily–or at all. The r between the 2004 Transparency International 2 estimates of corruption and those of the ICRG (International Country Risk Guide) in 1980-85 across 52 countries is .742. Any theoretical perspective on corruption must take into account its persistence over time. Instead of focusing on institutional reform, I suggest that the roots of corruption lie in economic and legal inequality and low levels of generalized trust (which are not readily changed) and poor policy choices (which may be more likely to change). The close connections between inequality and trust and trust and corruption produce the “inequality trap,” in which countries that have high levels of inequality will persist in low shares of generalized trusters and high corruption–and are unlikely to move “upward” regardless of any institutional reforms they undertake. Economic inequality provides a fertile breeding ground for corruption–and, in turn, it Uslaner, “The Bulging Pocket and the Rule of Law” (16) leads to further inequalities. Just as corruption is “sticky,” inequality and trust do not change much over time, either. The r for the most commonly used measures of economic inequality 2 (Deininger and Squire, 1996) between 1980 and 1990 is not quite as strong as the connection with trust over time, but it is still substantial at .676 for a sample of 42 countries. A new inequality data base developed by James Galbraith extends measures of inequality further back in time and across more countries. The r between economic inequality in 1963 and economic 2 inequality in 1996 is .706 (for 37 countries). The r between generalized trust, as measured in 2 the 1981, 1990-1995 World Values Surveys across between 1980 and the 1990s is .81 for the 22 nations included in both waves—the r between generalized trust in 1990 and 1995 is also robust 2 (.851, N = 28). Corruption persists, then, because its roots–economic inequality and low trust–change little over time. Moreover, high levels of inequality, low levels of trust, and high levels of corruption form a syndrome that leads back to more inequality, less trust, and more corruption. Yet, all is not lost: As I shall argue below in my cross-national analysis, there are more malleable ways to control corruption: Policy choices that countries make also shape corruption. Countries that have very high levels of regulation of business have more corruption. In turn, the level of regulation is shaped by the fairness of the legal system, the openness of the economy, and the stability (risk level) of a government. Inequality and Corruption The link between inequality and corruption seems compelling. Corruption is exploitive. Defining corruption is tricky, but it seems reasonable to argue that the core of “grand” corruption is giving some people advantages that are not available to others–and that these advantages are Uslaner, “The Bulging Pocket and the Rule of Law” (17) seen as unfair. A biased system is an inequitable one. Inequality promotes corruption in many ways. Glaeser, Scheinkman, and Schleifer (2002, 2-3) argue: ...inequality is detrimental to the security of property rights, and therefore to growth, because it enables the rich to subvert the political, regulatory, and legal institutions of society for their own benefit. If one person is sufficiently richer than another, and courts are corruptible, then the legal system will favor the rich, not the just. Likewise, if political and regulatory institutions can be moved by wealth or influence, they will favor the established, not the efficient. This in turn leads the initially well situated to pursue socially harmful acts, recognizing that the legal, political, and regulatory systems will not hold them accountable. Inequality can encourage institutional subversion in two distinct ways. First, the havenots can redistribute from the haves through violence, the political process, or other means. Such Robin Hood redistribution jeopardizes property rights, and deters investment by the rich. Similarly, You and Kaghram (2005) argue: “The rich, as interest groups, firms, or individuals may use bribery or connections to influence law-implementing processes (bureaucratic corruption) and to buy favorable interpretations of the law (judicial corruption).” Inequality breeds corruption by: (1) leading ordinary citizens to see the system as stacked against them (Uslaner, 2002, 181-183); (2) creating a sense of dependency of ordinary citizens and a sense of pessimism for the future, which in turn undermines the moral dictates of treating your neighbors honestly; and (3) distorting the key institutions of fairness in society, the courts, Uslaner, “The Bulging Pocket and the Rule of Law” (18) which ordinary citizens see as their protectors against evil-doers, especially those with more influence than they have (see also Glaeser, Scheinkman, and Schleifer, 2002; and You and Khagram, 2005). Economic inequality creates political leaders such as Martin Lomasny, who will “take care” of his constituents and New York City boss George Washington Plunkitt who made patronage a virtue rather than a vice, since it provided jobs for ordinary citizens. These leaders help their constituents, but more critically they help themselves. Perhaps the most famous line from Plunkitt’s autobiography (as told to a political reporter) explained how the 19 century th political boss who had only a small salary from his post got rich: “I seen my opportunities and I took ‘em” (Riordan, 1948, 4). Inequality breeds corruption–and to a dependency of the poor on the political leaders. Inequality leads to clientelism–leaders establish themselves as monopoly providers of benefits for average citizens. These leaders are not accountable to their constituents as democratic theory would have us believe. There may well be the trappings of democracy, with regularly scheduled elections, so that the link between democratic and honest government may not be as strong as we might initially expect. The political boss is well entrenched in his position–he may (as Plunkitt did) or may not 4 (as Lomansny did not) actually hold elective office. His party reigns supreme in the area. Potential opponents don’t have the resources to mount a real challenge–and, even if they tried, the boss can count on the support of the legions whose jobs he controls through his patronage machine. The political leader may claim to be the benevolent caretaker of his constituents, but he is really the ventriloquist stringing along his marionettes. The jobs that patronage creates do not alleviate inequality–they reinforce it. Uslaner, “The Bulging Pocket and the Rule of Law” (19) Where inequality is high, people of different classes (or other groups) will not believe that they have a shared fate with out-groups. People will also believe that you cannot achieve success by honest means such as hard work. You must have connections. This leads to an insular worldview–and to a willingness to do whatever is necessary to make your way in the world. Unequal wealth leads people to feel less constrained about cheating others (Mauro, 1998, 12) and about evading taxes (Oswiak, 2003, 73; Uslaner, 2003). Where corruption is widespread, people realize that they are not the masters of their own fate–and they lose faith that their future will be bright. Where corruption is widespread, people realize that they are not the masters of their own fate–and they lose faith that their future will be bright. People become resigned to their fate. In the World Values Survey waves 1-3 (1981, 1990, 1995-97), respondents who believed that corruption was widespread in their country were significantly less likely to believe that they could get ahead by hard work rather than by luck or having connections. The zero-order correlation is modest (as we might expect with a sample of almost 60,000, tau-b = .061)–but 34 percent of people in societies where corruption was seen as widespread thought the only way you could get ahead was by luck, compared to 29 percent in honest societies. In turn, 59 percent of respondents who said that they could get ahead by hard work said that the future looked bright, compared to 45 percent who said that you need luck or connections (tau-b = .116). And people living in honest societies are considerably more likely to have a high level of life satisfaction ( r = -.179, for a difference of a full point on a 10 point scale). People who believe that the future looked bright were significantly less likely to condone buying stolen goods or taking bribes. If you live in a more honest society, you are less likely to 5 Uslaner, “The Bulging Pocket and the Rule of Law” (20) condone cheating on taxes: 56 percent in the most corrupt societies said that cheating the government of revenue was always wrong, compared to 61 percent in more honest societies (taub = .06). The connection is even stronger (for a sample twice as large) between cheating on taxes and beliefs about how people get ahead. If you believe that you need luck, your probability of saying that cheating on taxes is never justified is .52, compared to .62 if you believe that hard work pays off. Honest government makes people feel that they are in control of their own fate–and do not need to take advantage of others. Corruption leads to the belief that there is no way to get ahead fairly–so that the ideals underlying what Rawls calls a “well-ordered society” do not develop strong roots. In a “well-ordered” society “everyone accepts and knows that the others accept the same principles of justice, and the basic social institutions satisfy and are known to satisfy these principles” (Rawls, 1971:454). It is prosperous and its government earns the approval of its citizens. A well-ordered society is run through the rule of law. The key to less corruption is an effective system of property rights and the rule of law (Lambsdorff, 1999; Leite and Weidemann, 1999, 20, 23; Treisman, 2000). Tyler (1990, chs. 4, 5) argues that people respect–and obey-the law because they believe that the justice system is fair and that they have been treated fairly. If people feel that they have been treated unfairly by the police or in the courts, they are less likely to have faith in the legal system. Inequality before the law is part of the larger theme of inequality more generally. The justice system is especially important for two reasons. First, a corrupt court system can shield dishonest elites from retribution. Second, the courts, more than any other branch of the polity, is presumed to be neutral and fair. We appeal “unjust” decisions to the judiciary–and Uslaner, “The Bulging Pocket and the Rule of Law” (21) our vernacular includes the phrase “court of last resort,” suggesting that somewhere there must be justice. Rothstein (2001, 491-492) argues: In a civilized society, institutions of law and order have one particularly important task: to detect and punish people who are “traitors,” that is, those who break contracts, steal, murder, and do other such non-cooperative things... Rothstein and Stolle (2002) argue that there are two dimensions to the legal system: fairness and efficiency. Fairness, I argue, is the key to the connection between law and corruption because it reflects the advantages that some people have over others. The efficiency of the courts should not matter so much for corruption–since rounding up the corrupt leaders and putting them in jail only makes room for a new group of miscreants, doing little to address the underlying causes of corruption. The fairness of the legal system is critical because no other political institution is predicated upon equality to such an extent. Elections are formally about equal access and power: Each of us has one vote. But many people do not vote, at least in some countries, and the distribution of participation is not equal across the population. Even more critically, elections are not determined by atomized individuals casting ballots in isolation. Elections require mobilization and in many parts of the world lots of money–and certainly neither of these are distributed equitably (Verba, Schlozman, and Brady, 1995, 190-192). When people have little faith in the fairness of the legal system, there are few incentives to obey the law. When Khodorkovsky confessed his sins of relying on “beeznissmeny” (stealing, lying, and sometimes killing) and promised to become scrupulously honest in early 2003, most Russians regarded this pledge as “startling.” When he was arrested and charged with tax evasion Uslaner, “The Bulging Pocket and the Rule of Law” (22) and extortion under orders from President Vladimir Putin ten months later, the average Russian was unphased: About the same share of people approved of his arrest as disapproved of it (Tavernise, 2003). The arrest of Khodorkovsky stands out as exceptional: Corrupt officials and business people are rarely held to account. While crime spiraled in Russia after the fall of Communism, conviction rates plummeted (Varese, 1997). In a corrupt polity, honesty is the exception to the norm–and professions of midnight conversions bring a mixture of derision and incredulousness. No one who has profited so immensely from a corrupt system could possibly change course so drastically–and pity the poor citizen who took him at his word. Where corruption is rampant, people have little hope for justice–and thus they don’t even seek it. Martin Lomasny’s constituents did not come to him for justice, only help. Lomasny could not run a political machine dispensing justice. Some Preliminary Evidence Fairness of the legal system is not as strongly connected to economic inequality as we might suppose. In Figure 1, I present a plot of a cross-national indicator of fairness of the legal system in 2004 and economic inequality. The legal fairness indicator was developed by the Economist Intelligence Unit; it only covers 60 countries, so I derived estimated values for other countries by imputation. Economic inequality is measured by the Gini index from Deininger 6 and Squire (1996). There are too many countries (especially with values close to each other) to label each one on the graph. Instead, the points are denoted by W for developed Western nations, E for present and former Communist nations, and * for other countries. Overall, the fit between these two indicators of equality (equal treatment before the law and equal distribution of wealth) is not strong. For 88 nations, r = .131. The correlation is depressed by the former and present 2 Uslaner, “The Bulging Pocket and the Rule of Law” (23) Communist nations that largely have unfair legal systems but more equitable distributions of income. For many years, this equality was imposed from above by a command economy–but 7 even as inequality has grown sharply, it has not approached the level of capitalist economies. Overall, we see relatively high economic equality matched with both low and high levels of judicial fairness. When I remove the East bloc countries, the r rises to .279–still rather modest. 2 Fairness of the legal system is not the same as economic inequality. ________________ Figure 1 about here What, then, can we make of the connection between inequality and corruption? I shall offer a multiple-equation estimation focusing on corruption below, but for now, I focus on bivariate plots of inequality and corruption. Measuring corruption is tricky, but there is general agreement that the best measure is that of Transparency International. The TI scores for a wide sample of countries (here I use the measure for 2004) come from a “poll of polls” business executives and the public, as well as rankings by risk analysts and experts on the politics and economics of each country (Treisman, 2000). The ratings range from zero (most corrupt) to 10 (least corrupt). As noted, I use the Deininger-Squire Gini indices to measure economic 8 inequality. The plot of inequality and corruption (see Figure 2) is striking: Across 85 countries, there is a weak (at best) relationship. The r is a paltry .082, suggesting no relationship at all between 2 inequality and corruption. In Figure 3, I offer a lowess plot of inequality and corruption. Lowess is an iterative technique, which fits a spline-like curve, a “locally weighted” regression, smoothed to produce a plot that clarifies the relationship between two variables. The lowess curve 10 Uslaner, “The Bulging Pocket and the Rule of Law” (24) suggests a slight downward slope–more inequality leads to more corruption (or less transparency). But the pattern is not at all clear and we can see from this figure that two groups of countries stand out: the former and present Communist nations, which have lots of corruption and relatively equitable distributions of income (lower left part of the graph), and the Western industrialized nations, which have relatively low inequality and even less corruption than we might expect based upon their distribution of wealth. Figures 2 and 3 suggest no clear relationship between economic inequality and corruption. But they also point to the former and present Communist regimes as outliers. What happens when we remove the former and present Communist regimes? The lowess plot in Figure 4 does suggest a moderate relationship between inequality and corruption. There is a gently tapering off downward slope in the connection between trust and corruption. The lowess plot becomes flat at moderate levels of inequality (a Gini of .4) and then rises a bit at the more extreme values. Overall, there is a moderate fit between the two indicators ( r = .246, N = 2 62) when the former and present Communist countries are excluded. With a bivariate r of this 2 magnitude, it should not take much effort to see it vanish in a multivariate analysis. You and Khagram (2005) report similar findings, but conclude that inequality matters more in democratic countries. They are correct, but this still doesn’t resolve the problem. Regressions predicting corruption by economic inequality yield insignificant coefficients for countries classified as “not free” or “partially free” by Freedom House–and a significant (at p < .01) negative coefficient for the 55 “free” nations: The nation with the highest level of inequality ranks 2.8 points lower on the 10 point corruption index compared to the most equal democracy–but the fit of the regression is rather meager ( r = .10). Democracy does not seem 2 Uslaner, “The Bulging Pocket and the Rule of Law” (25) to be the mechanism that makes inequality increase corruption. ________________________ Figures 2, 3, and 4 about here The connection with fairness of the legal system is far stronger–and this is hardly surprising. While I took care to find an indicator of the fairness of the legal system that is not based upon an underlying measure, it is hardly surprising that corruption flourishes where the courts give special treatment to some over others–and where court procedures are not transparent. I plot the original EIU data and corruption in Figure 5. The least fair legal systems have a mean corruption score of 2.82, while the most fair systems have a mean of 8.78 (high scores indicate greater transparency, less corruption). The fit between legal fairness and corruption is very strong: r = .722 for the 55 cases of the original EIU data and .733 for the 86 cases including the 2 imputed scores. _________________ Figure 5 about here There is some evidence that inequality of treatment by the courts is strongly associated with corruption, but the support for a link with economic inequality is modest at best. There are good theoretical reasons to believe that corruption stems from economic inequality as well as the fairness of the legal system. But the evidence does not seem compelling. Have we reached a dead end? Trust, Inequality, and Corruption Not at all. There is a link between inequality and corruption, but it is not direct. Inequality leads to corruption because it leads to resentment of out-groups and enhanced in-group Uslaner, “The Bulging Pocket and the Rule of Law” (26) identity. Generalized trust, the value that is predicated upon the belief that many others are part of your moral community, is the foundation of the “well-ordered society.” When we believe that “most people can be trusted,” we are more likely to give of ourselves and to look out for the welfare of others. When we believe that “you can’t be too careful in dealing with people,” we are likely to be on our guard and to feel little compunction in taking advantage of others who may not have our best interests in mind. Generalized trust is predicated on the notion of a common bond between classes and races and on egalitarian values (Putnam, 1993, 88, 174; Seligman, 1997, 36-37, 41). Faith in 13 others leads to empathy for those who do not fare well, and ultimately to a redistribution of resources from the well-off the poor. If we believe that we have a shared fate with others, and especially people who are different from ourselves, then gross inequalities in wealth and status will seem to violate norms of fairness. Generalized trust rests upon the psychological foundations of optimism and control and the economic foundation of an equitable distribution of resources. Optimism and control lead people to believe that the world is a good place, it is going to get better, and that you can make it better. Economic equality promotes both optimism and the belief that we all have a shared fate, across races, ethnic groups, and classes. Corruption, of course, depends upon trust–or honor among thieves. As it takes two to tango, it takes at least two to bribe. Corrupt officials need to be sure that their “partners” will deliver on their promises (Lambsdorff, 2002a, 2002b). Lambsdorf (2002a) argues: “...if corrupt deals cannot be enforced, this can act as a deterrent to corruption itself.” Corruption thrives upon trust, but it cannot be based upon the notion of widespread goodwill and common interests in a society underlying generalized trust. If you believe that the world is a good place and we should Uslaner, “The Bulging Pocket and the Rule of Law” (27) do our best to help those with less, we shouldn’t be willing to exploit others through corrupt deals. Entrance into a corruption network is not easy. Members of a conspiracy of graft cannot simply assume that others are trustworthy (as generalized trusters do). Treating strangers as if they were trustworthy (also as trusters do) can be hazardous at best. And believing that people without any ties to the conspiracy are trustworthy (as generalized trusters do) threatens the integrity of the cabal. Instead, corruption thrives on particularized trust, where people only have faith in their own kind (or their own small circle of malefactors). Particularized trusters strongly distrust outsiders. They fear that people of different backgrounds will exploit them–and in a dog-eat-dog world, you have little choice to strike first before someone exploits you. Gambetta (1993) argues that the Mafia took root in Southern Italy because there were strong in-group ties and weak generalized trust there. Where is generalized trust high and where is it low? Across a wide set of nations, across the American states, and over time in the United States–the only country with a long enough time series on the standard survey question on trust –the strongest predictor of trust is the level of 14 economic inequality. As economic inequality increases, trust declines (Uslaner, 2002, chs. 6, 8; Uslaner and Brown, 2005). Optimism for the future makes less sense when there is more economic inequality. People at the bottom of the income distribution will be less sanguine that they too share in society’s bounty. The distribution of resources plays a key role in establishing the belief that people share a common destiny–and have similar fundamental values. When resources are distributed more equally, people are more likely to perceive a common stake with others. If there is a strong skew in wealth, people at each end may feel that they have little in Uslaner, “The Bulging Pocket and the Rule of Law” (28) common with others. In highly unequal societies, people will stick with their own kind. Perceptions of injustice will reinforce negative stereotypes of other groups, making trust and accommodation more difficult (Boix and Posner, 1998, 693). Putnam (1993, 88, 174) argues that trust will not develop in a highly stratified society. And Seligman (1997, 36-37, 41) goes further. Trust can not take root in a hierarchal culture. Such societies have rigid social orders marked by strong class divisions that persist across generations. Feudal systems and societies based on castes dictate what people can and can not do based upon the circumstances of their birth. Social relations are based on expectations of what people must do, not on their talents or personalities. Trust is not the lubricant of cooperation in such traditional societies. The assumption that others share your beliefs is counterintuitive, since strict class divisions make it unlikely that others actually have the same values as people in other classes. A history of poverty with little likelihood of any improvement led to social distrust in the Italian village of Montegrano that Edward Banfield (1958, 110) described in the 1950s: “...any advantage that may be given to another is necessarily at the expense of one’s own family. Therefore, one cannot afford the luxury of charity, which is giving others more than their due, or even justice, which is giving them their due.” Montegrano is a mean world, where daily life is “brutal and senseless” (Banfield, 1958, 109), much like Hobbes’s “nasty, brutish, and short” existence. All who stand outside the immediate family are “potential enemies,” battling for the meager bounty that nature has provided. People seek to protect themselves from the “threat of calamity” (Banfield, 1958, 110). Inequality leads to low levels of trust in strangers. What trust remains is entirely within Uslaner, “The Bulging Pocket and the Rule of Law” (29) your group, so there are few moral sanctions for cheating people of a different background. Inequality thus breeds corruption indirectly–by turning people inward and reducing the sanctions, both external and internal, of taking advantage of others. So I posit an indirect link from inequality to corruption. Recapitulating, the bare bones model is: inequality –> low generalized trust & high in-group trust –> corruption This means that we cannot estimate the effects of inequality on trust directly. A simultaneous equation model is necessary to untangle the effects of inequality, trust, and corruption upon each other. First, let me demonstrate that trust and corruption are linked. I show the connection in Figure 6 below (see also Uslaner, 2004a). The graph is a bit difficult to read, since it is difficult to fit the country abbreviations into the graph since many countries have similar values on both variables. The trust question comes from the World Values Survey (see n. 11)–and to increase the number of cases, I imputed values on this measure as well. Here we see a more robust fit 15 than in the connection between inequality and corruption: r = .420 for 83 cases. 2 16 _________________ Figure 6 about here I now move to a more comprehensive model of the determinants of corruption, trust, and inequality. I add three other factors to the mix: the level of regulation in a society, the overall risk rating by the International Country Risk Guide (ICRG), and a measure of government effectiveness. A highly regulated economy can lead to greater corruption in two ways. First, when the government takes a dominant role in the economy, it creates many access points for Uslaner, “The Bulging Pocket and the Rule of Law” (30) entrepreneurs to “seize the state.” Business leaders “capture” the state by paying off public officials to provide them with private benefits. Businesspeople and bureaucrats work together to profit at the expense of the rest of those in society (by cutting growth rates). Second, high regulation leads business leaders to evade taxes and this widespread (Friedman, Johnson, Kaufmann, and Zoido-Lobaton, 2000; Hellman, Jones, and Kaufmann, 2000). Since regulation is also endogenous to the institutions and policies of the state, I need an equation for the extent of state regulation. The regulation measure comes from the World Bank governance indicators for 2004. Strangling regulation is different from steering economic or environmental policy. The World Bank measure is a composite index including restrictions on imports, exports, ownership of business by non-residents, discriminatory tariffs and protections, burdensome regulations in conducting and starting businesses, wage and price controls, foreign investment restrictions, unfair competition and trade, the efficiency of the tax system, price stability, distortionary taxes, and restrictions on competition (Kaufmann, Kraay, Mastruzzi, 2005, 106-107). Zimbabwe, Belarus, and Iran rank as having the most intrusive regulation, while Luxembourg, Singapore, Iceland, and Finland have the least. Such strangling regulation is the product of an unfair legal system as well as a high level of risk. Investors will be wary of doing business in countries with high levels of corruption. When countries receive a high risk rating, they will rely more upon the unofficial economy, on the one hand (Rose-Ackerman, 2004, 6) and on strangling regulations that make it difficult for firms out of favor to do business in a country. Risk is endogenous since I expect it to stem from corruption–and, through encouraging strangling regulation, to lead to further corruption. But risk Uslaner, “The Bulging Pocket and the Rule of Law” (31) depends upon more than corruption. Inequality should produce social strains that make investing risky–and ethnic tensions and other internal conflicts should also make investors wary. Effective government is important for two simple reasons. First, governments make policy. An ineffective government is likely to adopt policies that may be less responsive to pressing problems than to pressure from powerful groups (Rothstein and Teorell, 2005, 6). Second, governments execute the policies that they have already made. Effective governments deliver the services they promise–and do so fairly (Rothstein and Teorell, 2005, 5). This is not simply reiterating that effective governments are free from corruption. It is, rather, about the ability of government to function well. It includes making laws, enforcing them, making sure that service delivery is fair, efficient, and open to influence by all citizens. Effective government is difficult to measure and often one is tempted to rely upon the maxim of United States Supreme Court Justice Potter Stewart on how to define obscenity: “I shall not today attempt further to define the kinds of material I understand to be embraced . . . [b]ut I know it when I see it...”. There have been several attempts to measure effective 17 government, mostly in the United States. The most prominent is David R. Mayhew’s (1991) measure of the number of important laws passed by the United States Congress over time. Another set of measures of effective government in the American states are performance rankings on financial management, capital management, human resources, “managing for results,” and information technology in a study by the Government Performance Project (GPP) of Governing magazine and the Maxwell School of Citizenship at Syracuse University. The GPP used program information, a survey, and “interviews [with] budget officers, auditors, public managers, auditors, academics, and legislative aides in every state” (Knack, 2002, 775). An Uslaner, “The Bulging Pocket and the Rule of Law” (32) additional measure for the states is the number of Ford Foundation/Kennedy School of Government (Harvard University) awards for innovation a state has won (King, Zeckhauser, and Kim, 2001). These measures have gotten considerable currency in work on American politics– especially Mayhew’s measure of major laws enacted and variations on it. However, the measure of important laws is too specific to the American system of dispersed powers, which allow small numbers of legislators (sometimes even a single individual) to become obstructionists, thus making it difficult to enact legislation. Most other democracies are parliamentary systems, where enacting legislation is far easier. The GPP project is a rather thin measure of the quality of government. Surely, information technology and financial management are important for a government agency–but they are hardly the defining characteristics of what makes one government “effective” and another “ineffective.” Cross-nationally, the World Bank Governance project has a much broader indicator of government effectiveness (Kaufmann, Kraay, and Mastruzzi, 2005, 104-105). The World Bank measure includes, among other items: ratings of administrative and technical skills of the civil service, government instability, the quality of the bureaucracy, policy consistency, management of public debt, the effectivness of the executive, “consensus building,” debt management, trust in government, the consistency of policy-making, global e-government, and debt management. This measure is a remarkable achievement, but it may be too inclusive. Instability and trust in government, for example, are better considered as consequences of effective government rather than as components. I propose a new measure of effective government focusing on government capacity, Uslaner, “The Bulging Pocket and the Rule of Law” (33) efficiency, and inclusiveness. The measure are the factor scores from from a factor analysis of six questions asked in a cross-national survey of business executives. The 2004 Executive Opinion Survey of the World Economic Forum asked businesspeople to rate their country on judicial independence, the efficiency of the legal and legislative systems, the wastefulness of governmen spending, the favoritism of government decision-making, and the transparency of government decision-making. I present the full question wordings and scorings in the Appendix. The government effectiveness measure is a factor score of these six measures. All six measures loaded very highly on a single factor. Five of the six indicators had loadings of .90 or higher; the wastefulness of government spending had a marginally lower loading (.876). All six of the measures had communalities of .8 or greater (see Table 1). ________________ Table 1 about here This measure is attractive since it focuses on the capacity and fairness of government policy-making: An independent judiciary is critical to the rule of law, but independence is not sufficient. The efficiency of the court system matters as well. Long delays in legal affairs can lead to inequities in justice and make firms reluctant to enter into contracts. They may also punish those out of favor with the state, as the interminable Khodorkovsky trial in Russia demonstrates. A strong parliament will be more likely to represent a larger number of interests than we might expect from an all-powerful executive. Wastefulness of government spending and favoritism of government decision-making seem close to corruption–but they are not identical to it and are closer to the other measures in this index than to the TI measure of corruption. Government spending can be wasteful even in Uslaner, “The Bulging Pocket and the Rule of Law” (34) the absence of corruption–and favoritism in decision-making is at best a weak indicator of corruption. In some contexts, this may be quite acceptable–and hardly illegal. Finally, the transparency of government decision-making is important for corruption–but as defined here, it is also central for effective government. People cannot influence their government if they do not know what it is doing and how it is doing it. Perhaps surprising is the modest correlation of these six indicators with a question on red tape in the bureacucracy–how much time does your firm spend in negotiating with government officials. The red tape question correlates only at -.364 (N = 84) and does not load on the same factor. A measure of bureacuratic quality from the ICRG has a hgher correlation ( r = .734 , N 81), but even this measure is not so strongly related to the other quality of governmentas the other six measures are to each other. Effective government is not the same thing as service delivery–and this should be reassuring. The quality of government ought to reflect more than filling out forms and fighting with petty bureacurats–and by this measure, it does. My measure of government effectiveness is related to both fairness and corruption, but it is not based upon such measures. The indicators for the legal system are conceptually distinct from the fairness of the legal system as measured by the Executive Intelligence Unit. The new 18 measure is strongly related to the World Bank effectiveness measure for 2004 ( r = .870), but it seems more straightfoward in interpretation. I present the country scores for the government effectiveness index in Figure 7. I move now to a discussion of the model. _________________ Figure 7 about here The Corruption Model Uslaner, “The Bulging Pocket and the Rule of Law” (35) I estimate a six equation model of corruption across 63 countries–the number of countries on which I have data on all variables in the model. The six endogenous variables are corruption 19 (the TI Index), generalized trust (with imputed values), the level of regulation in a country (from the World Bank Governance data set for 2002), and economic inequality, the overall stability and credit worthiness of a country, and how effective a country’s government is. The key questions 20 I pose are: • Is there a direct relationship between trust and corruption? • Is there a direct relationship between economic inequality and trust–and does it flow from inequality to trust (Uslaner, 2002, ch. 8), trust to inequality (Knack and Keefer, 1997), or both ways? A direct relationship between inequality and trust and a similar connection between trust and corruption would provide support for my argument that inequality has an indirect impact on corruption. • Does corruption in turn lead to more inequality? Corruption slows economic growth (Leite and Weidemann, 1999; Mauro, 1995, 701; Tanzi, 1998, 585). It reduces the amount of money available for various government programs, including the government share of the gross domestic product and expenditures on the public sector, for education, and transfers from the rich to the poor (Mauro, 1998, 269; Tanzi, 1998, 582-586). So corruption should lead to more inequality–even if there is not a direct link from inequality to corruption. • Is the fairness of the legal system an important determinant of corruption? The fairness of the legal system should also shape the level of regulation in a society. An independent and fair judiciary should also lead to less regulation. Political leaders would not attempt Uslaner, “The Bulging Pocket and the Rule of Law” (36) to control business if they believe that the courts would step in and challenge attempts to capture the state. • Does corruption lead to less stable and less effective government? Corruption should lead investors to shy away from a country, for fear of expropriation or being compelled to function in a weak legal enviroment (Rose-Ackerman, 2004, 6). Corruption should also lead to less effective government. When leaders steal from the public purse, they shouldl be less responsive to the broader public. • Does ineffective government lead to greater corruption? Does effective government lead to better policy choices–especially policies that reduce inequality and create greater support for the regime? • Do higher risk ratings lead to less effective government or to poor policy choices? Countries with higher risk ratings should be likely to adopt strangling regulations that distort market competition. These regulations in turn should lead to greater corruption in a vicious cycle. • Do trust, corruption, a country’s risk rating, and effective government rest upon institutional foundations–or upon cleavages within society? I expect that structural factors should not be the key determinants of trust, corruption, a country’s risk rating, or effective government–except for the fairness of the legal system. Rather, corruption should depend upon trust and policy choices; risk ratings and government effectiveness should depend upon corruption and the health of a country’s economy–and on its domestic conflicts. And trust, in turn, depends on economic equality and its historical legacies of culture and conflict. Uslaner, “The Bulging Pocket and the Rule of Law” (37) • Putnam (1993, 111, 180) argues that trust and good government go hand-in-hand: Good government promotes trust (cf. Levi, 1998; Rothstein, 2001) and trust promotes good government. Is there a reciprocal relationship or does it only go one way–and, if so, which way? I expect that trust will encourage good government, but that good government should not lead to greater trust (Uslaner, 2002, chs. 2, 7). The endogenous relationships I test are from (see Figure 8 for a diagrammatic presentation of this model): • trust –> corruption • inequality –> trust –> inequality • government regulation –> corruption • inequality –> overall risk • corruption –> effective government –> policy choices • trust –> effective government –> trust And some key connections involving exogenous factors: • legal fairness –> corruption –> overall risk –> strict regulation –> corruption • legal fairness –> strict regulation –> corruption • internal conflicts –> trust, overall risk, effective government _________________ Figure 8 about here For both corruption and effective government, I estimated both basic and extended equations. The extended equations add structural factors that might loom large in the effectiveness of government institutions. Higher levels of democracy, as reflected in Freedom Uslaner, “The Bulging Pocket and the Rule of Law” (38) House’s index of civil and political rights, should lead to less corruption and more effective government (Adsera, Boix, and Payne, 2000; Treisman, 2000). Higher levels of bureaucratic quality (ICRG) should also be associated with better government and less corruption. So should a proportional representation electoral system and a parliamentary, as opposed to Presidential, legislative system. Linz (1990) has argued that Presidential systems lead to chief executives to centralize power–and to become more corrupt (cf. Kunicova and Rose-Ackerman, 2004, Lambsdorff, 2005). Proportional representation systems disperse accountability, compared to plurality systems, where there is a direct connection between a representative and her constituency. Voters who must choose from a list of candidates, rather than voting for one individual, will have fewer opportunities to punish corruption (Persson, Talbellini, and Trebbi, 2000, 4, 11-12; Rose-Ackerman, 2004, 12). The extended equations test for the impact of structural factors. I shall not report them for a simple reason: With one exception, the structural factors failed to reach statistical significance. In neither equation do the form of electoral or legislative system matter. Nor are there are effects for the quality of the bureaucracy. Democracy does not matter, perhaps surprisingly, for effective government. The only case where democracy matters is for corruption. Hence, I include the Freedom House measure in the basic model for corruption–and “move” the other variables to the list of instruments used for the system. For the corruption equation, I include: the Freedom House democracy measure, trust 21 (imputed), the level of regulation, the fairness of the legal system (imputed), ethnic fractionalization, and following Paldam (2000), the wealth of a country, measured as GDP per capita. Ethnic diversity may lead to a strong sense of ethnic identity, which in turn, Lassen Uslaner, “The Bulging Pocket and the Rule of Law” (39) (2003, 8) argues, may result in “the political process allocat[ing] excludable public goods and transfers based on ethnic characteristics (favoritism).” Lassen (2003) and Alesina, Devleeschauwer, Easterly, Kurlat, and Wacziarg (2003) find strong support for the argument, though Treisman (1999) and Leite and Weidmann (1999) failed to find significant effects for ethnic diversity upon corruption. I thus include the Alesina et al. (2003) measure of ethnic fractionalization in the equation. In the equations for overall risk and government effectiveness, I use a measure of ethnic tensions in a society, but here I follow the existing literature and employ a measure of ethnic fractionalization. The model for corruption may appear thin and different from many in the literature. However, I did many sensitivity tests for variables discussed in the literature and have found no reason to add any of them to my model. Specifically, I did not find significant relationships for the level of public sector wages (LaPorta et al., 1998; Mauro, 1997, 5; Tanzi, 1998, 573; Treisman, 2000; but cf. Rose-Ackerman, 1978, 90-91), per capita income or gross domestic product (Lambsdorff, 1999, 7; Mauro, 1995, 701; 1998, 13; Paldam, 2000, 9); the size of the unofficial economy (Lambsdorff,1999); the level of newspaper readership (Adsera, Boix, and Payne, 2000); federal versus unitary governments or the share of government revenues spent at the local level (Treisman, 1998; Fisman and Gatti, 2000); the level of political stability (Leite and Weidmann, 1999, 20; Treisman, 2000); the level of democracy in a country (You and Khagram, 2005); or the use of closed or open list proportional representation systems (Kunicova and Rose-Ackerman, 2004). All of these other variables faded into insignificance within the model I present below. At the bottom of Table 2, I also list the exogenous variables used as instruments. Uslaner, “The Bulging Pocket and the Rule of Law” (40) For the trust model, I largely follow the cross-national model in Uslaner (2002, ch. 8) in arguing that the level of economic inequality and the share of the population that is Protestant should be key factors shaping trust. I have already laid out the argument on inequality. The “Protestant ethic” is an individualistic creed: To succeed in a competitive world, we need to rely upon other people. In collectivist societies, people can rely upon their peer groups and get by with particularized trust. In individualistic societies, generalized trust becomes essential. Trust is higher in individualistic societies (Triandis, 1995, 126) and Protestant societies are more individualistic (Uslaner, 2002, 232, n. 21) and less hierarchical (Inglehart, 1999, 92-93) than other countries. Civil war can tear a country apart, so I expect that countries that have had civil wars are less likely to be less trusting. While there is little evidence that democracy leads to greater 22 trust, there is ample support for the claim that Communism depresses trust. The repressive Communist system made it treacherous for ordinary citizens to trust each other–at best people had faith in their close friends and family members, and even then there were often risks (Gibson, 2001). I thus include a dummy variable for former and present Communist regimes. Finally, I test whether there is a link from government effectiveness to trust. While this is a prominent theme in much of the literature on trust (see the citations above), I do not expect to find such a connection. Generalized trust develops early in life and is largely resistant to experiences, including those with the government (Uslaner, 2002, chs. 2, 4, 5). The equation for business regulations includes the fairness of the legal system, the openness of the economy to external trade, the growth rate of the gross domestic product, and 23 the overall risk rating by the ICRG. An economy open to foreign trade, on the other hand, Uslaner, “The Bulging Pocket and the Rule of Law” (41) would give foreign investors more of a say in how companies are run. To be able to export to foreign countries, a firm must be free of tight control from above. Imports also lessen central control over the economy. Similarly, when the economy is growing at a fast pace, political leaders might be more wary about scaring investors away with strangling regulations. On the other hand, the more a country is perceived to be a bad risk, the more the leadership may try to regulate the economy. However, a country that already is at risk is likely to overcompensate its regulations–and to make things more complicated than they already are. Leaders may adopt such strangling regulations either to appear to be putting some order into a chaotic economy–or simply as reacting defensively to prevent foreign creditors from gaining too much power in their countries. The equation for inequality includes trust and corruption–and also the dummy for former and present Communist countries and the shares of the population that are Muslim and Protestant. Eastern bloc countries should have lower levels of inequality, since incomes were as should both countries with large Muslim and large Protestant populations. Protestantism stresses individual achievement. Achievement-oriented values stress equality of opportunity rather than equality of results. Yet, the greater wealth of Protestant nations and the higher levels of trust in individualistic nations leads to the expectation that Protestant countries will have lower levels of inequality. Islam has placed greater emphasis on collective goals, especially on one’s economic responsibility to the larger community (as reflected in the prohibition on charging interest on loans). So it should not be surprising to find a powerful coefficient on percent Muslim for economic equality (Esposito and Voll, 1996, 25). The model for overall risk includes economic inequality, corruption, and measures of Uslaner, “The Bulging Pocket and the Rule of Law” (42) internal conflict and ethnic tensions, both from the ICRG. Here my argument is straightforward: The creditworthiness of a country reflects its social conflicts as well as the misdeeds of its public officials. Corruption should have the greatest impact on risk ratings, but social conflicts should not be far behind. A country should be a bad risk if its social fabric is torn. Economic inequality clearly contributes to such strains. I also use two other measures, both from the ICRG, that should contribute to risk: the level of internal conflicts and ethnic tensions. Finally, the government effectiveness includes trust (imputed), corruption, ethnic tensions, the poverty level (from the ICRG), and a legacy of Communism. The extended equation also includes structural factors that have played a prominent role in the literature: the level of democracy (from Freedom House), the quality of the bureaucracy (from the ICRG), whether a country has a Presidential or parliamentary system, and whether a country has a proportional representation or plurality electoral system. Trust, Putnam (1993, 103) argues that “civic regions” in Italy, which are marked by high levels of trust, have better government: Political leaders in civic regions are...readier to compromise than their counterparts in less civic regions....there is no evidence at all that politics in civic regions is any less subject to conflict and controversy, but leaders there are readier to resolve their conflicts. Civic regions are characterized not by an absence of partisanship, but by an openness of partisanship.... Uslaner (1993, ch. 6; 2002, 212-215) links the increasing contentiousness and use of obstructionist tactics in the United States Congress to declining levels of trust. So I expect that trust should be a key factor in shaping effective government, even if there is no evidence for a Uslaner, “The Bulging Pocket and the Rule of Law” (43) linkage going from government effectiveness to trust. The linkage between corruption and ineffective government needs no elaboration. Similarly, the transition from Communism has not been smooth for most countries in Central and Eastern Europe. Not only does corruption remain high and trust remain low, but governments are not generally considered effective. The difficulties in governing that plague transition countries were expressed by an academic from the region, who commented at a conference I attended: “There are two types of governments in Europe, those that always get reelected (Western Europe) and those that never get reelected (Central and Eastern Europe).” The measures of ethnic tensions and poverty from the ICRG are indicators of the social strains that make effective government difficult. High poverty rates places additional demands on government and adds to the social strains in a society–so poverty should lead to less effective government. Evaluating the Model All six of the equations perform very well in accounting for corruption, trust, regulation of the economy, and inequality. Even though R is not strictly appropriate for two-stage least 2 squares estimation, the high values for R and the low values of the standard error of the estimate 2 (see Table 1) give us confidence in the models. The R values range from .533 (for inequality) 2 to .895 (for corruption) and the standard errors of the estimates are, relative to the means, generally quite small. The models, then, fit the data well. The most important result is that there is an indirect linkage between inequality and corruption and it goes through trust. As we move from the low level of inequality in Belgium to the very high level in South Africa, trust declines by 23 percent. This is equivalent to moving Uslaner, “The Bulging Pocket and the Rule of Law” (44) from the low trust level of Serbia, the Czech Republic, South Korea, Spain, or Bulgaria to the high levels of the Netherlands and Canada. As we move from the least trusting country (Brazil) to the most trusting (Norway), corruption decreases on the 10 point TI scale by 3.47 units. This is equivalent to moving from the low transparency of Kenya or Pakistan to that of Taiwan. The impact of inequality on trust is sufficient to produce a considerable effect on corruption. A shift from a country ranking highest on legal fairness to one at the bottom corresponds to a shift of 2.42 on the TI index–from Kenya or Pakistan to the level of Greece or South Korea. A shift from the most strict regulatory regime to the least (from Nigeria to Luxembourg) would bring a country to the level of Portugal (6 on the 10 point scale), while moving from a “not free” to a “free” country has a more modest shift of 1.55 points: More democracy in Pakistan would put it on a par with Ghana, Thailand, and Mexico–still quite far from even the middle of the TI rankings. Ethnic diversity seems to lead to less corruption rather than more: The coefficient is positive rather than negative (the simple correlation is -.386). Trust has the greatest impact on corruption, followed by the regulatory regime, the fairness of the legal system, a nation’s GDP per capita, and finally whether a country is free or not. There are significant impacts for one institutional factor–but they pale by comparison to those for other variables. And the effect of democracy on corruption is only contemporaneous. As I have noted, the correlation between changes in democracy and changes in corruption is effectively zero. Institutional factors do not loom large in determining the level of corruption in a country except insofar as they lead to more or less equitable treatment of citizens before the law or except insofar as they promote economic equality and an economic system free of political interference. This does not imply that governments cannot regulate the economy–the Uslaner, “The Bulging Pocket and the Rule of Law” (45) Western European countries that have the least “regulation” are all welfare states. It does suggest that politicians must take care in how they regulate. The equation for corruption performs extremely well without any standard institutional variables such as centralization, parliamentary system, type of electoral list, the type of executive–each of which fell to insignificance when added. The model for trust confirms the results in Uslaner (2002, ch. 8): Eastern bloc countries are far less trusting (by an average of nine percent). Protestant societies are more trusting: The difference in trust between the country with the largest Protestant population (Norway) and the smallest (among them Turkey) is 17 percent. Countries that have experienced civil wars are nine percent less trusting. But the largest effect comes from economic inequality. As we move from low values on the Gini index (Belgium) to the highest (South Africa), trust declines by 18 percent. The coefficient for government effectiveness on trust is insignificant. In the model for regulation, the levels of risk and the fairness of the legal system are the most important predictors. The state with the greatest risk level will be very likely to have a strangling regulatory regime–going from the lowest risk level to the highest leads to a change in regulatory quality equivalent to the distance from Norway to Kenya or the Ukraine. The gap in legal fairness is somewhat smaller (1.255 on the standardized scale rather than 1.722), but still equal to the gap between Norway and Romania or the Philippines. An open economy also makes strangling regulation less likely–but here the effect is half that of overall risk. Closing Norwegian markets would “only” take that state down to the regulatory level of Bulgaria or Mexico–still quite a bit better than the Ukraine or Kenya. The model for inequality suggests that inequality shapes trust rather than the other way Uslaner, “The Bulging Pocket and the Rule of Law” (46) around (cf. Uslaner, 2002, 232-236). However, the link from trust to inequality is more complex (see below). Even if there is no direct linkage, there is ample support for an indirect connection. Present and former Communist countries are considerably more egalitarian, by an average of .17. The Muslim share of the population is significantly associated with reduced inequality: The inequality gap between countries with no Muslims and the largest Muslim share (Indonesia) is .194; Indonesia is an outlier in this sample (which does not include many Muslim countries). Less extreme is Turkey, and here the difference in effects is reduced to .07. But the share of Protestants in a society is incorrectly signed. Finally, corruption leads to more inequality: The difference between the most and the least corrupt country leads to a change in Gini coefficients of .23–which is the difference in inequality between Belgium and the Philippines or Costa Rica. The equation for inequality is the least successful of the four, according to three different criteria: the lower (though still far from modest) R , the low F ratio, and the high t-value of the constant. 2 Nevertheless, the equation still performs well and the other estimations all support my overall framework. Corruption has the strongest effect on the overall risk rating: The most “honest” country ranks 67.3 points lower on the ICRG scale than does the most corrupt. Finland would be as risky for investors as Argentina if it were as corrupt as Bangladesh. But internal conflicts also have a powerful effect on risk. The highest levels of conflict lead to a 45 point shift in risk ratings–the equivalent of a shift from the Nordic countries to Russia, Slovakia, or Estonia. The effects of inequality (14 points) and ethnic tensions (18 points) are smaller, but still significant. As with corruption, the structural variables (Presidential versus parliamentary system, proportional representation versus plurality elections, quality of the bureaucracy, and status as a Uslaner, “The Bulging Pocket and the Rule of Law” (47) democracy) are all insignficant. The biggest impact, once more, is for corruption. If Bangladesh, the most corrupt country, were to become as honest as Finland, its government would be as effective as any Nordic country’s. Trust and poverty are also very important. Raising the trust level of Brazil, the least trusting country, to Norway’s top rating would raise its quality of government to that of Malaysia or Tunisia (and not too far from the United States or Austria). Reducing Bangladesh’s poverty level to that of the Nordic countries would not give it as effective a government as we find in Northern Europe, but would move it up to the level of Portgual and not too far from Spain. Low trust and high poverty may not be quite as important as corruption, but they have a major effect on the quality of government. Ethnic tensions are also a significant factor shaping the quality of government: Higher tensions mean less effective government, but the impact is modest at .49 on the standardized scale. This effect is slightly larger than I find for former Communist countries (-.39). The effectiveness of government, then, mostly reflects corruption and societal forces. Corrupt governments are by far the least effective. Yet there are powerful effects, both direct and indirect (through corruption) for generalized trust. Trusting countries have better governments, even though there is no evidence of a direct link back from effective government to generalized trust. Moreover, trust works not only directly but also as the most significant predictor of corruption. In 1976, American Presidential candidate Jimmy Carter promised the public “a government as good as its people.” Seemingly, it is difficult to give the people a government much better than they are. Escaping the Inequality Trap? Overall, there is considerable support for my thesis: Inequality shapes trust. Even if there Uslaner, “The Bulging Pocket and the Rule of Law” (48) is no simple relationship, there is a powerful indirect linkage through trust. The fairness of the legal system also shapes corruption, indirectly through the regulatory regime. Social bonds and the distribution of wealth–and justice–play key roles in determining whether a country will be corrupt or transparent. Institutional factors do not matter as much as social bonds, policy choices, and equity. Not only does inequality lead to greater corruption (albeit indirectly), but corruption leads back to more inequality. This inequality trap works both directly–from corruption to inequality–but also indirectly, through an unfair legal system, strangling regulation, a risky credit environment, and ineffective governments. This model shows that it is not simply societal factors that constitute the inequality trap. Bad policy choices–-economic policies that lead to high risk ratings and strangling regulation–are part of this same syndrome. It would be nice if a simple shift in policy choices could get us out of the trap. In theory, it could. But the same forces that lead from high levels of inequality to low trust to high corruption also lead to strangling regulation, high risk ratings, unfair legal systems, and ineffective government. There is a causal spiral from inequality to corruption (and back again) and from both inequality to lower levels of trust and from low levels of out-group trust–and high levels of ingroup (or particularized) trust to corruption (Gambetta, 1993). Perhaps most critical in this vicious circle is the link between inequality and trust. Countries with high levels of trust enact policies that help reduce inequality–they spend more on the poor and more on programs such as education that help equalize opportunities (Uslaner, 2002, 245-246). Yet, as Rothstein and I have argued (Rothstein and Uslaner, 2005), the policies that are most effective in reducing inequalities and in promoting trust are universalistic social welfare programs. Means-tested Uslaner, “The Bulging Pocket and the Rule of Law” (49) programs stigmatize the poor and exclude those who are better off. Recipients of means-tested benefits in both Sweden and the United States, our analyses of surveys has shown, are less trusting. The logic of means-tested programs is clear: Help those most in need and do not reward people not so wanting. However, if there are already strong social tensions within a society, these programs may be little more than masked calls for confiscation of the earnings of those who have adapted to Western market economics (Uslaner and Badescu, 2005). To be sure, many who get rich in poor countries, especially those in transition from Communism, are corrupt–but if we equate all wealth with corruption, markets and democracy cannot take root in these transition states. Each class trusts only its own members, which is a recipe for continued corruption and persistent inequality. Effective government in this model seems to be a dead end–it goes nowhere. It is shaped by trust, but does not affect faith in others or anything else in this model. Are institions doomed to irrelevance? Hardly. Effective governance matters. Good government promotes economic growth, reduces tax evasion, leads to the development of market economies while encouraging both strong environmental protection and more cooperative labor-management relations, and encourages ethical behavior and greater charitable contributions in society. Good government 24 is an important component in the links among inequality, trust, and corruption. However, we should be cautious in interpreting too much power to strong institutions: Effective government seems to be the result of low corruption and high trust–rather than the determinant of either. Moreover, effective government does produce many positive benefits for society, but there is at best a weak relationship between those policies and reduction of inequality. There do Uslaner, “The Bulging Pocket and the Rule of Law” (50) not seem to be direct links to more spending on social welfare or education, nor on levels of inequality. Effective (efficient) government gets people to support the system, to pay their taxes, to do good deeds (give to charity and for businesses to act ethically), to cooperate with people in other sectors of the economy, and to see their incomes rise. It does not pay particular attention to whose incomes rise. Perhaps some of these other outcomes–ethical behavior, better labormanagement relations–will lead, indirectly, to policies designed to reduce social strains. There seem to be few quick fixes out of the inequality trap. Redesigning institutions doesn’t seem up to the job. You can get better government if you end corruption, but you can’t get rid of corruption by changing institutional structures. Enhancing democracy works–but its impacts are less than those of policy choices and cultural factors such as trust. An Indian journalist commented on the sharp cleavages that led to a cycle of unstable coalitions, none of which could form a government: “We have the hardware of democracy, but not the software, and that can’t be borrowed or mimicked” (Constable, 1999, A19). Reprise If corruption stems, either directly or indirectly (or both), from economic inequality, then our efforts to stamp it out must focus on the distribution of resources in society. Corruption cannot be eradicated by simply putting errant leaders in jail. Corruption is a sign of a much larger breakdown of the rule of law. Azfar (2005) and I have independently used the International Crime Victimization Surveys aggregated to the country level to track levels of crime and corruption. Azfar shows that personal theft and homicide rates are strongly linked to corruption. My data (as yet unreported) show that corruption levels are also strongly related to fraud, car theft, the extent of damage to cars, to longer sentences for those who are convicted, Uslaner, “The Bulging Pocket and the Rule of Law” (51) and, perhaps most telling, the rate of pickpocketing. Across 48 countries, the R between 2 corruption perceptions and perceptions of pickpocketing levels is .660. Pickpocketing levels are also predicted by trust (imputed) and the fairness of the legal system (imputed, R .441 and 2 = .572, respectively for N = 48). Economic inequality is also a significant predictor, but only when present and former Communist countries are excluded (R .243 for the Deininger-Squire 2 = measure and .360 for the Galbraith measure in 1994, N = 31 and 26). We see the same syndrome at the street level as at the top. The direction of causality is unclear at least at the present: Do venal leaders make people commit crimes or does petty theft indicate a mistrusting culture that not only tolerates, but encourages corruption? We also need to be careful in linking “crime” to corruption. Not all crime is so clearly linked to malfeasance at the top. Corruption seems to breed economic crimes (theft, car damage, pickpocketing) and not crimes of violence. The frequency of assaults is not at all linked to corruption levels (R .0005) or to the frequency of sexual assaults (R .016, both N = 48). So it 2 = 2 = is not crime per se that is associated with corruption, but rather economic-based crime. And this lends further support to the claim that economic inequality is the foundation for corruption. I have also shown that corruption in turn increases inequality, so we seem to be in a vicious cycle that is difficult to break by anti-corruption campaigns alone. In one sense the question of whether corruption is top-down or bottom up–stemming from venal leaders or a culture of mistrust–may not matter theoretically as it does practically. The persistence of corruption over time, as institutions have changed, suggests that we pay more attention to what is happening at the bottom, to the distribution of resources. Morality goes on holiday, then, when the crush of inequality makes trust in strangers Uslaner, “The Bulging Pocket and the Rule of Law” (52) seem like a bet that is too risky. People turn inward, so that cheating outsiders loses its moral force. Poor people are hardly less moral than rich folks. But the poor may be less able to afford the morality of clean politics–and as long as there are Martin Lomasnys who promise relief from a judicial system that seems unfair, the cost of trusting strangers will always exceed the price of tolerating corruption–and even rewarding the politicians who protect them with their loyalty. The small payoffs that poor folk make to the Martin Lomasnys and other patrons or to doctors, police officers, and even university professors (very common in the formerly Communist nations of Central and Eastern Europe and many poor countries) do not enrich the people who profit from it. Plunkitt called petty corruption--and favoring one’s friends for jobs and contracts--“honest graft.” It is at least morally neutral; both those who make the extra payments and those who receive them often see (see (Kornai 2000: 3, 7, 9; Kramer, 1977, 220; Miller, Grodeland, and Y. Koshechkina, 2001; Uslaner and Badescu, 2005). Many people in societies with lots of corruption see no problem with having to make small “gift payments”. It may depend upon an inequitable distribution of wealth–there should be no need to make “gift” payments in a properly functioning market economy. Yet, it does not exacerbate the gap between the rich and the poor--and may actually narrow it by providing some small benefits to the middle class bureaucrats, teachers, and doctors who benefit from it. People living in corrupt and unequal societies find themselves constrained to accept some corruption as the price of getting by. Low-level corruption is acceptable because everyone does it, it seems impossible to stop, and even necessary in daily life. And citizens in 26 highly corrupt countries–such as Romania–do not let petty corruption–the need for connections to get anything done, extra “gift” payments to local officials, police officers, bank clerks, or the Uslaner, “The Bulging Pocket and the Rule of Law” (53) police–make them lose faith in the state, the market, democracy, or each other. “Bribes in Third World countries are a form of payment for services efficiently rendered,” claims travel writer Jeffrey Tayler (2005, 35, emphasis in original). “Grand” corruption refers to malfeasance of considerable magnitude by people who exploit their positions to get rich (or become richer)–political or business leaders. So grand corruption is all about extending the advantages of those already well endowed. It is high-level corruption–misdeeds by politicians or business people–or low-level corruption by institutions that are supposed to be fair such as the courts–that leads to a loss of trust in government and especially in each other. While the perpetrators of petty corruption don’t get rich by these small gift payments, corrupt politicians and business people do get rich (Uslaner, 2004b). The perception that corruption (or luck) is the only way to get rich is widespread in the transition countries where corruption is high and inequality has been growing apace. While most Westerners believe that the path to wealth stems from hard work, 80 percent of Bulgarians, Hungarians, and Russians say that high incomes reflect dishonesty (Kluegel and Mason, 2000, 167; cf. Orkeny, 2000, 109). Effective government, then, doesn’t generate trust in other people because it doesn’t create more equality. The air may be cleaner, markets may open up, and there will be many signs of economic growth and wealth all around. To overcome the pessimism that inequality generates, government must do more than make the trains run on time. Everyone should ride in a comfortable cabin, if not in the same class. Uslaner, “The Bulging Pocket and the Rule of Law” (54) TABLE 1 Factor Analysis of Government Effectiveness Measures: World Economic Forum Executive Opinion Survey 2004 Variable Loading Communality Judicial independence .919 .908 Efficiency of legal system .976 .971 Efficiency of legislative system .913 .852 Wastefulness of government spending .876 .801 Favoritism of government decision making .942 .901 Transparency of government decision making .934 .882 Uslaner, “The Bulging Pocket and the Rule of Law” (55) TABLE 2 Simultaneous Equation Estimation of Corruption Variable Coefficient Standard Error t ratio Corruption equation Trust (imputed) 5.580**** 1.567 3.57 Regulation of business 1.018*** .351 2.90 Fairness of legal system .605*** .203 2.97 GDP per capita (ICRG) .404*** .158 2.58 Ethnic fractionalization (Alesina) 1.035 .546 1.89 Freedom House 2003 democratization -.776*** .257 -3.01 Constant -.013 .698 -.02 Trust equation Economic inequality (Gini index) -.461*** .195 -2.36 Civil war -.086**** .025 -3.41 Protestant share of population 1980 .174*** .063 2.79 Former Communist nation -.091** .045 -2.01 Government effectiveness .028 .024 1.18 Constant .479**** .077 6.19 Regulation equation Fairness of legal system .251*** .080 3.15 Openness of economy to trade .232*** .091 2.54 Real growth in gross domestic product -.086* .067 -1.28 Overall country risk (ICRG) -.014**** .003 -4.76 Constant .328 .837 .39 Uslaner, “The Bulging Pocket and the Rule of Law” (56) TABLE 2 (continued) Simultaneous Equation Estimation of Corruption Inequality equation Trust (imputed) -.152 .187 -.81 Corruption -.028*** .009 -3.23 Former Communist nation -.166**** .025 -6.59 Protestant share of population 1980 .123 .052 2.38 Muslim percent of population -.001**** .000 -3.77 Constant .583**** .035 16.77 Overall risk equation Economic inequality (GINI) 35.602* 25.180 1.41 Corruption -8.209**** .952 -8.62 Internal conflicts (ICRG) -7.564**** 1.416 -5.34 Ethnic tensions (ICRG) -3.650** 1.607 -2.27 Constant 170.527**** 15.812 10.78 Government effectiveness equation Trust (Imputed) 1.738** .934 1.86 Corruption .436**** .065 6.69 Ethnic tensions (ICRG) -.098** .046 -2.12 Poverty level 2005 (ICRG) -.313**** .091 -3.42 Former Communist nation -.390*** .144 -2.70 Constant -1.475**** .228 -6.47 * p < .10 ** p < .05 *** p < .01 **** p < .0001 (all tests one tailed except for constants) Uslaner, “The Bulging Pocket and the Rule of Law” (57) TABLE 2 (continued) Equation R S.E.E. Mean F Statistic 2 Corruption .893 .859 5.252 79.13 Trust .664 .081 .296 21.35 Regulation of business .821 .388 .612 68.20 Inequality (Gini) .534 .072 .360 12.34 Overall risk (ICRG) .854 13.749 51.714 84.09 Government Effectiveness .831 .431 .146 53.09 Instrumental variables: Religious fractionalization (from Alesina, Devleeschauwer, Easterly, Kurlat, and Wacziarg, 2003); English legal tradition (from the Levine-Loyaza-Beck data set at http://www.worldbank.org/research/growth/llbdata.htm ), GNP per capita (State Failure Data), constraints on the executive branch of government (Glaeser, La Porta, Lopez-de-Silanes, and Shleifer, 2004); military in politics (at www.freetheworld.com ); terrorism risk (ICRG); bureaucratic quality (ICRG), parliamentary system and proportional representation (from the Data Base of Political Institutions at http://www.worldbank.org/wbi/governance/pubs/wps2283.html ) Uslaner, “The Bulging Pocket and the Rule of Law” (58) Figure 1 W = Western bloc E= former and present Communist countries * In neither bloc Uslaner, “The Bulging Pocket and the Rule of Law” (59) Figure 2 Uslaner, “The Bulging Pocket and the Rule of Law” (60) Figure 3 Uslaner, “The Bulging Pocket and the Rule of Law” (61) Figure 4 Uslaner, “The Bulging Pocket and the Rule of Law” (62) Figure 5 Uslaner, “The Bulging Pocket and the Rule of Law” (63) Figure 6 Corruption and Generalized Trust (Imputed) Uslaner, “The Bulging Pocket and the Rule of Law” (64)