Corruption is an archetypal topic for students of Public Choice. It brings together the private search for economic gain with the government’s efforts to supply public goods, correct market failures, and aid the needy. Public Choice’s insistence on viewing politicians and government bureaucrats as motivated by the same economic interests as private individuals and firms provides a background for understanding why corruption occurs and why it is difficult to combat.
Corruption in my formulation is the misuse of public office for private gain. This definition leaves open the issue of just what constitutes misuse, but it recognizes that sometimes public office can legitimately provide private benefits to politicians and bureaucrats. Thus, targeted "pork barrel" projects and special interest legislation are not corrupt. They result from the day-to-day operation of a representative political system. If a legislator works to pass a statute that is favored by his or her legal campaign donors, this is not corrupt even if it violates democratic ideals. Those who seek to discredit government across the board often put the "corruption" label on all kinds of government actions. Although many of these phenomena are indeed proper subjects of study and the loci of reform efforts, it will not help the analysis of democracy to put them all into the corruption pot.
There are several reasons for maintaining a distinction between bribery, fraud, and self-dealing, on the one hand, and quid pro quo politics, on the other. First, a political system that encourages legislators to "bring home the bacon" for their constituents may also be one that encourages voters to monitor their representatives to be sure they are not benefiting personally from their position. Voting systems that limit constituency-based politics may encourage corruption (Kunicova and Rose-Ackerman, 2002). Second, strict rules on legal campaign donations may simply drive contributions underground into a corrupt netherworld. Thus, it is valuable to maintain a distinction between legal donations from wealthy interests and illegal, secret gifts. Third, some reform proposals designed to deal with bureaucratic corruption involve the use of legal incentive payments. Mixing financial incentives with the provision of public services is not invariably corrupt. Often it is an efficient method of service delivery.
This entry concentrates on corruption that involves a public official, either a politician or a bureaucrat. However, corrupt incentives can also arise in purely private interactions. Corruption is, in essence, an agency/principal problem. An agent violates the trust of his or her principal through self-enrichment or through illegally enriching a political party. A public official may take a bribe in return for a favorable decision or may simply steal from the state’s coffers. Clearly, corporate managers can face similar incentives, and with the growing privatization of former state enterprises, the locus of some forms of corruption will shift into the private sector. Private-to-private corruption has been little studied but ought to be the object of future work (for one example see Andvig, 1995).
I proceed as follows. Section 1 outlines the underlying causes of corruption and its consequences from a political-economic point of view. Section 2 discusses reform options in the light of the discussion in section 1 and the broader literature behind the summary presented here. This note provides only a brief overview of both topics. Readers who want to pursue these issues further should consult my two books — Rose-Ackerman (1978, l999), a review article by Pranab Bardhan (1997), the framework presented in Shleifer and Vishny (1993), Robert Klitgaard illustrative case studies (1988), and della Porta and Vannucci’s reflections on the Italian case (1999). Most of these references also include extensive references to the literature. To access current work, the World Bank Institute maintains a website [http://www.worldbank.org/wbi/governance] as does Transparency International (TI) an international nongovernmental organization committed to fighting international bribery [http://www.transparency.org].
1. The Incentives and Consequences of Bribery
I focus on bribery. Ordinary fraud is relatively uninteresting as an analytic matter, and few would argue that stealing from the state is to be encouraged. However, with bribery the story is different. Some economists observe money changing hands and assume that something efficient must be occurring. Some Public Choice scholars who favor a minimal state and who view most state actions as illegitimate exercises of power interpret bribes as a desirable way to avoid the exercise of government power. I want to argue that both of these tolerant views are, as a general matter, mistaken, but to do so requires one to understand the incentives for paying and accepting bribes. My basic message is that even if an individual bribe seems to further efficiency or get around an irrational rule, the systemic effects of widespread tolerance are invariably harmful both for the efficient operation of the economy and for the legitimacy of the state.
The government allocates scarce benefits and imposes costs. Individuals and firms may be willing to pay government agents to gain the former and to avoid the latter. Opportunities for corruption arise whenever the officials’ actions involve the exercise of discretion and are impossible to monitor perfectly (Klitgaard, 1988). The level of benefits under official control can vary from the allocation of a driver’s license to the award of a major public works contract. The costs avoided can be a traffic ticket, a multi-million dollar tax bill, or a prison sentence. Bribes can also improve quality, notably by speeding up service delivery or jumping someone ahead in a queue.
The potential bribe revenues available to any individual politician or bureaucrat depend upon his or her monopoly power. If potential bribe payers have non-corrupt alternatives, bribes, if they are paid at all, will be low. If the chance of being caught and punished is high, corruption may be deterred. Thus, one can think of corruption in cost\benefit terms where payoffs will be deterred if at least one side of the potential deal faces costs that exceed the benefits. If expected penalties increase more than in proportion to the size of the bribe, only small bribes may be paid and accepted. Conversely, if penalties do not rise in proportion to benefits, small bribes are deterred, and large bribes are unaffected (Rose-Ackerman, 1978).
The mere existence of corrupt opportunities, however, says nothing about their welfare implications. In discussing this issue, it is important to recognize that the level of bribe payments is likely to be a poor measure of their social cost. Sometimes very small bribes have large consequences. Bribe may be low, not because the value of the quid pro quo is low, but because the bribe payer has bargaining power relative to the official. For example, if a majority-rule legislature with weak parties is bribed to approve a law favored by a particular firm, no individual politician has much bargaining power; he or she can easily be replaced by another person formerly outside the corrupt coalition. Thus, my focus is not on situations where bribes are high but on those cases where the social costs are severe.
One might suppose that if a government has scarce benefits to distribute, say a number of restaurant licences, then corruption will distribute them to those with the highest willingness-to-pay, and the winners will be the most efficient restaurateurs. There are several responses to this claim. First, corrupt markets are inefficient compared with the aboveboard sale of licences. Bribe-prices are secret, and entry may be blocked. Thus, the government should simply legally sell the scarce rights if its goal is to allocate the service to those who value it the most in dollar terms. Second, the basic purposes of some public programs would be violated by sales to the highest bidders. For example, selling places in public universities and in subsidized housing would undermine the basic goals of those programs. Third, toleration of corruption gives officials an incentive to engage in the creation of more scarce benefits in order to create more corrupt opportunities. For example, corrupt contracting officials have an incentive to support wasteful public projects designed to make payoffs easy to hide.
Similar points can be made about bribes paid to avoid the imposition of costs. Clearly, if a regulation is onerous and inefficient, then paying for an exemption seems efficient. However, permitting such individualized law compliance can be very harmful. First, profit-maximizing firms and individuals will not distinguish between socially efficient and socially inefficient rules. They will want to be exempted from all of them. The rules will only be enforced against those with a low willingness to pay. This includes not just those for whom the rule is not costly but also poor households and marginal businesses. In the case of tax collection, those exempted from taxes generate higher bills or lower services for others. Selective exemption on the basis of willingness to pay is inefficient and unfair. Second, officials will seek to create even more restrictive rules so that they can be paid to decline to enforce them. Empirical work suggests that in countries where corruption is high, red tape is high, and managers spend considerable time dealing with public officials (Kaufmann, 1997). Thus, even if each individual corrupt decision is rational for the bribing firm, the overall costs of doing business in society are high. Investment and entrepreneurship are discouraged.
The costs of corruption are not limited to its impact on the efficacy of public programs taken one by one. In addition, endemic corruption has implications for the legitimacy of the state in the eyes of its citizens. In highly corrupt states, where both day-to-day interactions with officials and high-level deals are riddled with payoffs, people often express great cynicism about political life. This can lead to vicious spirals. The theoretical work on corruption has produced a number of multiple-equilibria models where both high corruption and low corruption solutions exist (Bardhan, 1997; Rose-Ackerman, 1999: 107-108, 124-125). Some countries, particularly a number of the former socialist countries, illustrate these pathologies. To give a flavor of these models consider two variants. First, suppose that there is a fixed supply of law enforcement personnel. If very few transactions are corrupt, the enforcers can catch most of the illegal deals, thus encouraging more people to be honest in the next round and so forth. If most are corrupt, the law enforcement authorities are spread very thin and only catch a few wrongdoers. This encourages more to enter the corrupt arena next period and so on in a vicious spiral. Similar results occur if we assume that the moral stigma of corruption is a function of the number of others who engage in it. If most are corrupt, the stigma is low, and next period more shift to the corrupt side, and so forth. Second, another kind of spiral can affect the character of those who become politicians or bureaucrats. If most officials are corrupt, this will discourage honest people from working for the government and encourage the dishonest to apply, making the government even more corrupt. If government work makes one rich, those who want to get wealthy choose the public sector and do not become entrepreneurs. Their corruption creates a costly environment for business that further discourages private business activities. This self-selection mechanism can produce an equilibrium in which the dishonest and the greedy have disproportionately chosen public sector employment.
Empirical work has begun to shed light on some of the costs of corruption outlined above. Research on corruption is difficult because the perpetrators seek to keep their transactions secret. Nevertheless, scholars have begun to analyze and measure the impact of corruption on economic and political phenomena and to explain how political and economic conditions contribute to corruption. This work, based on cross-country data, is quite consistent in finding that corruption is harmful to growth and development and that corruption is the result of weak economic and political institutions.
The cross-country research uses data that measure perceptions of corruption, such as the composite Transparency International index, developed by Johann Graf Lambsdorff, or the World Bank Institute’s recalculation using similar data. The perceptions are mostly those of international business people and country experts. Studies using these data have found that high levels of corruption are associated with lower levels of investment and growth, and that foreign direct investment is discouraged (e.g., Mauro, 1995; Wei, 2000). Highly corrupt countries tend to under-invest in human capital by spending less on education and to over-invest in public infrastructure relative to private investment (Mauro, 1997; Tanzi and Davoodi, 1997). Corrupt governments lack political legitimacy and hence tend to be smaller than more honest governments, everything else equal (Johnson et al., 2000). Corruption reduces the effectiveness of industrial policies and encourages business to operate in the unofficial sector in violation of tax and regulatory laws (Ades and Di Telia, 1997; Kaufmann, 1997). Turning the causal story around, recent research suggests that autocracies tend to be more corrupt than democracies, but that democracy is not a simple cure. Within the universe of democracies, corruption is facilitated by features of government structure such as presidentialism, closed-list proportional representation, and federalism (Kunicova, 2001; Kunicova and Rose-Ackerman, 2002; Treisman, 2000).
These are important findings, but they are limited by the aggregated nature of the data. Each country is treated as a single data point that is more or less "corrupt." This work shows that corruption is harmful but says little about the precise mechanisms. To counter this weakness, two new types of research are underway: detailed questionnaires that target households, businesses, and public officials; and what might be called "econometric case studies." The questionnaires permit researchers to explore people’s actual experiences. The case studies help one understand how corrupt sectors operate and how malfeasance might be controlled.
Here are some examples of the research I have in mind. Several studies questioned small- and medium-sized businesses about the costs of corruption and red tape. Other researchers have used questionnaires and focus groups to examine household attitudes and behavior. Researchers have studied countries as diverse as those in sub-Saharan Africa and in Central and Eastern Europe. Some of the most comprehensive are a study of four countries in Central and Eastern Europe by William Miller et al. (2001, forthcoming) and work that focuses on the business environment in the same region by Simon Johnson et al. (2000). This research complements the World Bank Institute’s work on "state capture" and administrative corruption in post-socialist countries (Hellman et al., 2000).
Sectoral studies are represented by work on how corruption limits the performance of the judiciary in Latin America (e.g., Buscaglia and Dakolias, 1996). Other examples are Wei Li’s (forthcoming) estimates of the waste and corruption generated when China had a two-price policy for basic raw materials, and research by Rafael di Tella and Ernesto Schagrodsky (forthcoming) on the benchmarking of product prices in the hospital sector in Argentina that shows how monitoring and civil service pay reform can go hand in hand. As an example of research that can make a difference, consider Ritva Reinikka and Jakob Svensson’s (forthcoming) documentation of the severe leakage of federal funds meant for local schools in Uganda. Their study led to a simple, information-based reform that had positive results.
These contributions are very diverse in topic and methodology, but they all share an interest in using detailed data to understand both how corrupt systems operate and which policies have promise. Only if one looks at the fine structure of political and economic systems, can one go beyond a showing that corruption is harmful to an understanding of the way it operates in different contexts. Given that knowledge, reform programs can attack corruption where it has the worst effects.
2. Reform
Reform strategies attack the problem of corruption from several directions: program redesign, law enforcement, improved government performance and accountability. Before presenting this mixture of reform options, however, I begin with a solution that is favored by some Public Choice scholars. Many Public Choice analysts accept the claim that corruption is harmful. However, they argue that the solution should be, not the reform of public programs, but a reduction in the size of government. They argue that the best way to avoid corruption is to shrink government and rely on the market. Of course, this will sometimes be true, but it is not a general solution and would be risky if employed across the board. The most obvious problem with this argument is that it misses the benefits of some, even poorly operating, public programs. Programs to limit external costs, correct for information failures, produce public goods, or aid the needy have no effective private market counterparts. Free rider problems plague efforts at private provision. Furthermore, if a program is reduced in size but not eliminated, corruption may increase instead of decrease. To see this, consider a program to provide public housing to the needy. A cut in the program by half creates scarcity and hence the competition for places. Bribes may increase.
Another form of government "load shedding" has similar difficulties. Privatization is justified as a way of introducing market discipline into the operation of formerly state-owned firms. Competitive pressures and the need to raise capital in the private market will squeeze out waste and encourage a focus on consumer satisfaction. Unfortunately, privatization does not always imply the creation of competitive markets. Sometimes the process of turning over assets has itself been corrupted by collusion between powerful private and public interests. This sometimes implies that public firms are sold too cheaply to insiders and that the terms of the deal give the new owners access to monopoly rents. Corruption in the privatization process in some countries is analogous to corruption in large scale public procurements — powerful politicians and business interests gain at the cost of ordinary citizens. Citizens lose both because the benefits to the state coffers are lower than they should be and because the benefits of expanding the role of competitive markets are lost. Thus, over-enthusiastic efforts to limit the role of government should be avoided, and the cutbacks that are carried out should be carefully designed to avoid the problems outlined here.
If a country faces a vicious spiral of corruption, such as I outlined above, this would seem the best case for the "load shedding" solution. The government is in a dysfunctional low-level trap where piecemeal reform will be ineffective. The state needs a major overhaul in law enforcement and in the recruitment of personnel. However, a simple attempt to shrink the state is unlikely to be effective because it can create a chaotic situation in which a lawless free-for-all replaces the corruption that went before. A new kind of corruption and self-dealing may arise that is based on the attempt to establish some kind of certainty in a situation of fluidity and chaos.
If corruption cannot be countered by single-minded efforts to limit the size of government, then one must also consider ways to reform government from within and to limit the willingness of citizens and firms to pay bribes. Any actual program needs to be adapted to the conditions in a particular country, but the broad outlines can be identified. Anticorruption policies can increase the benefits of being honest, increase the probability of detection and the level of punishment, reduce the corrupt opportunities under the control of public officials, and increase the accountability of government to its citizens. The incentives for corruption are influenced by:
• the level of benefits and costs under the discretionary control of officials,
• the formal laws designed to combat defining corruption, bribery, and conflicts of interest, and to regulate finance spending,
• the credibility of law enforcement against both, those who pay and those who accept bribes,
• the conditions of civil service employment, and the performance incentives officials face,
• the extent of auditing and monitoring within government,
• the ability of citizens to learn about government activities, file complaints, and obtain redress, and
• the level of press freedom and the freedom of individuals to form nongovernmental organizations.
I focus on four broad categories: reductions in the discretion and monopoly power of government officials, enforcement of anticorruption laws, civil service reform, and increased accountability to citizens.
2.1. Reducing the Incentives for Payoffs
The most basic reforms are those that reduce the level of benefits under the control of public officials. As I noted above, the most obvious option is simply to eliminate laws and programs that are permeated with corruption. If the state has no authority to restrict exports or license businesses, no one will pay bribes in those areas. If a subsidy program is eliminated, the bribes that accompanied it will disappear as well. If price controls are lifted, market prices will express scarcity values, not bribes.
In general, any reform that increases the competitiveness of the economy will help reduce corrupt incentives. Thus policies that lower the controls on foreign trade, remove entry barriers for private industry, and privatize state firms in a way that assures competition will all contribute to the fight against corruption.
But any move toward deregulation and privatization must be carried out with care. Deregulating in one area may increase corruption elsewhere. Furthermore, many regulatory and spending programs have strong justifications and ought to be reformed, not eliminated. Corruption in the collection of taxes obviously cannot be solved by failing to collect revenue. One solution is to clarify and streamline the necessary laws in ways that reduce official discretion. Rules could be made more transparent with publicly provided justifications. Governments might favor simple nondiscretionary tax, spending, and regulatory laws as a way of limiting corrupt opportunities. Clear rules of proper behavior could be established so violations can be noticed even if the bribery itself is not. Where possible, procurement decisions could favor standard off-the-shelf items to provide a benchmark and to lower the cost of submitting a bid. Obviously, the value of such reforms depends upon the costs of limiting the flexibility of public officials (Anechiarico and Jacobs, 1996). Sometimes a certain risk of corruption will need to be tolerated because of the benefits of a case-by-case approach to program administration. Transparency and publicity can help overcome corrupt incentives even in such cases, but only if the systems of accountability discussed below exist. If they do not, simple clear rules can simply permit a top ruler more effectively to extract payoffs. This is just one example of the importance of viewing reform in the context of the entire political-economic environment.
Economists have long recommended reforming regulatory laws in such areas as environmental protection by introducing market-based schemes that limit the discretion of regulators. Analysts also recommend user fees for scarce government services. These reforms have the additional advantage of removing corrupt incentives by replacing bribes with legal payments. The sale of water and grazing rights, traceable pollution rights, and the sale of import and export licenses can improve the efficiency of government operations while limiting corruption.
Finally, administrative reforms may lower corrupt incentives. Corruption is often embedded in the hierarchical structure of the bureaucracy. Low level officials collect bribes and pass a share on to higher level officials perhaps in the form of an up-front payment for the job itself. Conversely, higher ups may organize and rationalize the corrupt system to avoid wasteful competition between low-level officials. The top officials may then share the gains of their organizational ability with subordinates, perhaps using them to run errands, transfer funds, and do other risky jobs that expose them to arrest. To break such patterns may require a fundamental reorganization effort.
One possibility is the introduction of competitive pressures within government to lower the bargaining power of individual officials. If bribes are paid for such benefits as licenses and permits, which are not constrained by budgetary limits, overlapping, competitive bureaucratic jurisdictions can reduce corruption. Because clients can apply to any one of a number of officials and can go to a second one if the first turns him down, no one official has much monopoly power. Thus no one can extract a very large payoff. For qualified clients, bribes will be no larger than the cost of reapplication. Unqualified clients will still pay bribes, but even they will not pay much so long as they too can try another official (Rose-Ackerman, 1978). If all officials are corrupt, the outcome is stable. However, if some establish an honest reputation, applicants will prefer those officials, thus reducing the gains to the corrupt. This reduction in benefits may induce some marginal officials to shift to being honest, further reducing the benefits to the remaining corrupt officials and so on. A small number of honest officials can overturn a corrupt system if congestion is not a serious problem. Honesty may drive out dishonesty even if only a few officials are honest on principle (Rose-Ackerman, 1978). If, instead, those who pay bribes are unqualified, the honesty of some officials increases the gains to those who are corrupt, inducing more to become corrupt.
When officials, such as police officers, can impose costs, another type of overlapping jurisdiction model should be considered. Police officers seeking to control illegal businesses can be given overlapping enforcement areas. That way gamblers and drug dealers will not pay much to an individual policeman since a second one may come along later and also demand a payoff. The first one is simply unable to supply protection. Bribes may fall so low that it is not worthwhile for police officers to risk taking them. This system may work better if the law enforcement officers belong to different police forces — state or federal, for example. Then collusion between officers to defeat the system will be less likely (Rose-Ackerman, 1978).
Alternatively, consider the losers in corrupt transactions. The state could introduce ways for the potential losers to appeal unsatisfactory decisions. Sometimes bribe payers view themselves as losers who would be better off in an honest world. They feels themselves to be the victims of extortion. Such bribe payers are potential allies in an anti-corruption effort who will cooperate in efforts to elim-mate payoffs. Conversely, in other cases bribery makes both payer and receiver better off with respect to a no-bribery world. Thus control incentives must rest with outsiders not in on the corrupt deal (e.g., disappointed bidders, taxpayers, consumers). The existence of losers, such as disappointed bidders, with a large stake in the outcome can facilitate efforts to limit corruption.
2.2. Anticorruption Laws and Credible Law Enforcement
A basic condition for corruption control is a viable legal framework that enforces the law without political favoritism or arbitrariness. The goal is both to deter those tempted to engage in corrupt acts and to educate the public to resist criminal conduct by officials. Tough laws are not sufficient. Many highly corrupt countries have exemplary formal statutes that have no real meaning because they are seldom enforced. A country serious about reform must have effective investigation and prosecution bodies and a well-functioning judicial system that is not itself corrupt. Because corruption is a two-sided offense, the law must specify the status of both those who make payments and those who receive them. If just one of the parties can be deterred, that is sufficient to prevent the deal from going through.
Designing an optimal deterrence strategy raises a seeming paradox. The more severe the penalties for corruption faced by officials, the lower the incidence of corruption, but the higher the bribes. If the risk of detection is high, officials must receive a high return in order to be willing to engage in bribery. One way around such a result is an expected penalty function that is an increasing function of the size of the bribe (Rose-Ackerman, 1978: 109-135). Conversely, if penalties on bribe payers have deterrent effects, this will lower the demand for corrupt services and the level of bribes at the same time.
An independent judiciary or some other kind of independent tribunal is a necessary condition for the use of law enforcement to check official malfeasance. This is a serious problem in many countries where the judicial system is backlogged and some judges are corrupt. Prosecutors, whether they are formally in the executive branch, as in the United States, or part of the judiciary, as in Italy, must be able to have the independence to pursue corruption allegations and need to be able to reward those who report on corrupt deals with lowered charges and penalties. Some countries have had success with independent anticorruption commissions or inspector generals reporting only to the chief executive or the parliament. These can be useful responses, but a single-minded focus on law enforcement is unlikely to be sufficient if the incentives for corruption are deeply imbedded in the structure of public programs and if law enforcement efforts can be diverted to harass political opponents.
2.3. The Civil Service
Many developing countries have very poorly paid civil servants. Although at independence most former colonies inherited civil service pay scales that exceeded private sector wages, this advantage has eroded over time. Wages relative to private sector wages have fallen in countries in transition in Eastern Europe and the former Soviet Union. The pattern varies across countries and over time. In some parts of the developing world public sector pay is so low that officials must supplement their pay with second jobs or payoffs. Some work suggests that there is a negative correlation between civil service wages (relative to private sector wages) and the level of corruption (Van Rijckeghem and Weder, 2001).
If officials are paid much less than people with similar training elsewhere in the economy, only those willing to accept bribes will be attracted to the public sector. Civil service pay should be set at least equal to equivalent positions in the private sector in order to make it possible to recruit based on merit and to permit those selected to serve without resorting to corruption. If the benefits under the control of officials are very valuable, however, parity may not be sufficient. Instead, civil service wages may need to be set above the going private sector wage with generous benefits, such as pensions, that will be received only if the worker retires in good order. This strategy, however, must be combined with an effective monitoring system. There must be a transparent, merit-based system of selecting civil servants or else people will pay the powerful to be allotted desirable government jobs.
Pay reform is necessary, but not sufficient. Penalties must be tied to the marginal benefits of accepting payoffs. In cases where corruption’s only efficiency cost stems from its illegality, the payments should be legalized. In the design of such systems, however, it is important to avoid giving monopoly power to bureaucrats that they can use to extract increased levels of rents.
2.4. Public Accountability
Corruption can be checked by structures that create independent sources of power and information inside and outside the government. Although not sufficiently taken by themselves, these options complement other reform strategies by reducing corrupt opportunities and increasing the risks of paying and accepting payoffs. There are several linked aspects in a system of public accountability over and above the checks provided by periodic democratic elections.
• Outsiders, such as ordinary citizens or the media, can obtain information about how the government is operating and have a way of expressing their displeasure about general policies. Nongovernmental organizations can organize easily and face few legal hurdles. They may even be subsidized.
• The structure of government includes guarantees that protect the individual against the state. Government actions may be checked by a specific Bill of Rights that limits state power, and individuals can appeal attempts to extort bribes. The legal system provides protection and perhaps rewards to individuals who come forward to "blow the whistle", on corrupt practices, but the state is also constrained by legal rules that protect the accused.
• Higher level governments and international organizations can use what leverage they have to constrain the behavior of individual governments.
• The threat of exit can be a powerful constraint on governments, reducing corrupt opportunities and limiting the scope for waste.
First, the private sector, particularly an independent media, can be an important check on the arbitrary exercise of power by government, but only if the government provides information, if the press is not controlled, and if people can organize into associations. Accountability to the public requires both that individuals can find out what the state is doing and that they can use this information to hold public actors accountable. Governments must publish budgets, revenue collections, statutes and rules, and the proceedings of legislative bodies. Financial data should be independently audited. Secret funds available to chief executives and top ministers are an invitation to corruption. Procurement regulations must keep the process open and fair. Scandals frequently occur because top officials overrule tender boards or because lower level officials operate without formal controls on their purchasing decisions.
Freedom of information acts in the United States and in a number of European countries are an important precondition for effective public oversight. These laws permit citizens to request information as members of the public without showing that their own personal situation will be affected.
Finding out what is happening is of little value however, unless people can use their knowledge to influence government. Individuals face a familiar free rider problem in seeking to control political and bureaucratic processes and to limit malfeasance. Information may be, in principle, available, but no one may have an incentive to look at it. Laws that make it easy to establish associations and nonprofits will help. For example, Transparency International has local chapters that carry out a range of activities including participation in Integrity Workshops, sometimes organized with the help of aid agencies. These workshops bring together concerned people from both the public and the private sectors to discuss the problem of corruption. Nonprofit organizations can carry out and publish public opinion surveys that reveal public attitudes toward government services. An alternative to NGO surveys of service users is the creation of "hot lines" so that citizens can complain directly to the government. The information from such complaint mechanisms will be less systematic than a survey and may well be self-serving, but hotlines provide a means of making a complaint without the necessity of establishing an organization. This method will only be successful, however, if those who complain can either do so anonymously or are not fearful of reprisals. Furthermore, if the complaints concern individuals, they must have a credible way of defending themselves against false accusations.
The second aspect of accountability is the way the government structure protects individuals against the state. The forms of administrative law and the protection they provide to individuals are of critical importance. If an official tries to extort a bribe from individuals or firms, do they have any recourse? Obviously, if the bribe is to be paid to permit illegal activities or to soften a legal regulation or tax assessment, the answer is no. Corruption of this type is unlikely to be revealed by the parties to the deal unless they have been arrested and are seeking to mitigate their punishment. However, those who face bribe demands as a condition for obtaining a legal benefit may not go along with the demand if they can appeal to an honest forum, such as an appeals board within the agency or the courts. In order to make appeals worthwhile, however, the processes must not only be honest, but also speedy and efficient.
The Ombudsman represents one route for citizen complaints. Many countries have established Ombudsmen to hear complaints of all kinds, not just those related to malfeasance. These offices can help increase the accountability of government agencies to ordinary citizens, but they are seldom a way to uncover large scale systemic corruption and most have no authority to initiate lawsuits.
Ombudsmen and other complaint mechanisms are insufficient if people are unwilling to complain. Reporting the peculations of others can be dangerous. Thus, governments should consider promulgating whistleblower statutes that protect and reward those in the public and the private sector who report malfeasance. However, whistleblower protection is obviously pointless unless the prosecutorial system follows up, the courts are incorruptible and relatively efficiently run, and the penalties are severe enough to deter potential offenders.
The third check on corruption can arise from intergovernmental relations. In a federal system, the national government can constrain the states, and the states, the localities. Similarly, institutions operating internationally may provide a check on national governments. This kind of leverage has problematic aspects since those who exercise it can make no straightforward claim to represent the interests of the affected citizens. There are two cases, however, in which such actions may be justified. First, corruption and waste frequently have cross-border consequences. Corrupt politicians or those engaged in legal joint ventures with private firms may try to use their political power to restrict commerce across state borders. Internationally, officials working in collaboration with corrupt business firms harm the prospects of honest businesses. Second, state and local governments may be under the control of narrow elites that use the apparatus of government for personal gain. Although both oversight from above and competition between jurisdictions for investment resources limit corrupt possibilities at the local level, they do not eliminate them. In fact, cross-country empirical work suggests that federal states are, on balance, more corrupt than unitary states suggesting that the negative effects outweigh the positive (Treisman, 2000).
Exit, the final constraint on corruption, has the advantage of not requiring a concerted organizational effort. In a world with many coequal governments, the corruption and ineffectiveness of government officials is limited by the ability of constituents and business firms to go elsewhere. Multinational firms trying to decide where to locate a manufacturing plant can limit bribe demands by locating several feasible sites. Residents of a village whose officials extract large payoffs for routine services can move elsewhere. The mobility of people and businesses clearly limits the ability of officials to extract payoffs for services to which one is entitled.
Mobility, however, is not always helpful. It will make it more difficult for an individual jurisdiction to control undesirable behavior. Suppose, e.g., that a city government has installed an honest police force that cracks down on illegal gambling. The gamblers may simply move to a friendly suburb that they can control and establish their business there. Several examples of this phenomena exist in United States urban areas. The ease with which funds can cross national borders, coming to rest in various "financial paradises" is another example of how multiple, competing jurisdictions can make control of corruption, fraud, and tax evasion more, not less, difficult. Thus inter-jurisdictional competition should be encouraged when it reduces the economic rents available for corrupt distribution and helps control waste but should be limited when it facilitates the illegal behavior that corruption often makes possible or requires.
A system of public accountability implies that once a law or regulation is put in place, individuals and groups both inside and outside government have the ability to find out how it is being administered, to complain, and to set in motion a legal or political enforcement process. To be a meaningful anticorruption check, however, knowledge must be combined with the existence of institutions that can take effective action both to promulgate new laws and to enforce existing ones.
3. Conclusions
Corruption has a moral dimension, but it can be understood and combated through the application of political-economic principles. A first step in the understanding of corruption is the documentation of the incentives for private gain built into political and bureaucratic processes. Next is an evaluation of the social costs when officials and private citizens succumb to these incentives. Part of the reform agenda involves explaining the social harm of corruption and trying to change a culture of tolerance both within government and in the citizenry and the business community (Rose-Ackerman, 2002). Moral suasion may work if backed up by concrete arguments for why corruption is harmful to society. Reformers do not simply point to corruption and appeal for people to change their behavior; rather they demonstrate that reducing corruption provides real gains, not just symbolic victories. The key point is to encourage people to look beyond the net gains from any particular corrupt deal to see how tolerance of corruption has negative systemic effects.
However, as Public Choice theory teaches, most people will not behave well simply because they are told that such actions are in the public interest. A change in behavior needs to be in their interest as well. A political-economic approach can go beyond documenting the costs of corruption to suggest ways to lower its incidence and impact. Although reforms in law enforcement and in internal monitoring are part of the story, the most important lessons of a political-economic approach are its recommendations to turn attention to the redesign of individual public programs, on the one hand, and to ways to increase government transparency and accountability on the other. That strategy both reduces the corrupt incentives facing bribe payers and recipients and facilities effective public oversight by the population.
