«Bruno Bosco2 Margherita Savona DEMS DSG University of Milan–Bicocca Piazza Ateneo Nuovo, 1 20127 Milan, Italy Abstract In this paper corruption is ...»
Starting with the seminal works of Rose-Ackerman (1975; 1978) a growing body of literature has studied corruption decisions as the result of optimal individual choices under uncertainty in the framework of crime, agency and rent-seeking theories. Public procurement and tax evasion have been the most natural initial case studies. Researchers have derived conditions for “optimal” bribe and penalty functions for private agents and public officers who may mutually gain from corrupt transactions. They have stressed the role of market competition among suppliers to the public sector in determining both corruption opportunities and rent appropriation and have also derived efficiency wage formulas designed to curb officers’ corruption by increasing the opportunity cost of risky corrupt activities. The conditions for efficient law enforcement, such as the judiciary activities and efficient penalty structure, have been extensively studied, too.
A second strand of literature has developed mainly after the publication of the indexes of (perceived) corruption from some international institutions. The publication of these indexes, such as the various versions of Transparency International Index, Corruption Perception Index and Word Bank Index, has lead economists to investigate what factors have a significant statistical influence on cross–countries differences in corruption. The fundamental legal structure of a country (e.g. civil vs. common law structure) as well as the strength of its political and administrative decentralization or the time length of its democratic organization seem to affect the probability that individuals of that country can be exposed to corrupted activities (La Porta et al, 1998; Treisman, 2000). The abundance/scarcity of raw materials (Ades et al, 1999) and the gender distribution of parliamentary seats and senior bureaucratic positions (Swamy et al., 2001) are other factors making corruption more/less likely.
Lawyers approach to the analysis of corruption can be understood from the very definition of corruption adopted by the Global Programme against Corruption run by the United Nations. Corruption is any “abuse of power for private gain”, which therefore covers corruption in both public and private sectors. Lawyers stress the need of monitoring and supervising the officers’ activity and to improve the enforcement of the law.
In spite of the fact that corruption is a many-faceted phenomenon for which a precise and comprehensive characterisation is difficult to formulate, yet during the last 40 years economists have attempted to offer a number of definitions of corruption to be used in the economic analysis of the illicit trade associated to the relationship between (corrupted) officials and (corrupting) privates. In the first of the two seminal papers quoted above, Rose-Ackerman (1975) deals with the relationship between market structure and the incidence of corrupt dealings in government contracting process. She defines the essential aspect of a corrupt behaviour as an illegal or unauthorized transfer of money or an in-kind substitute in favour of a person in a position of power acting as an agent for another individual or organization. The purpose of the bribe is to induce that person to place his/her own interests ahead of the objectives of the organization for which he/she works. Framing the concept of corruption in property rights terms, Shleifer and Vishny (1993) define government corruption as the sale by government officials of government property for personal gain, i.e. transforming illegally public into private property. Polisky and Shavell (2001) distinguish between two different kind of corrupting phenomena, namely the acceptance of a payment by an official in return for not reporting a violation and the threatening of an innocent individual in order to extort some money from him. In all these cases, as stressed by Rose-Ackerman (2006), corruption represents the illicit use of willingness to pay as a decision making criterion on the part of a private individual who makes a payment to a public official in return for actions that are against the interests of his/her principals. Hence, corruption occurs when the private search for economic advantages clashes with law and norms that condemn such behaviour (Rose-Ackerman and Søreide, 2011) and consists in acts in which the power of a public office is used for personal gain in a manner that contravenes the rules of the game (Jain, 2001).
Corruption has been seen for long time as an efficiency-enhancing practice that permits the market to defend itself from pre-existing burdensome government failures such as excess taxations or queues for services of various sorts (Lui, 1985). Consequently, bribes were considered as useful side-payments that improved bargaining outcomes and promoted overall efficiency. It was the progressive assimilation of the work of Myrdal (1968) what made clear to many that corruption ought to be seen as sand and not oil in the gears of economic systems because corruption negatively affects both GDP and GDP growth (Mauro, 1995; Brunetti et al, 1998); it deteriorates the investment climate and tends to reduce both domestic and foreign investment – perhaps with a few recent exemption (Helmy, 2013) – and ultimately distorts public expenditure decisions (Mauro, 1998). Moreover, interest rates are badly prejudiced by corruption in loans operations and countries perceived as more corrupt pay a higher risk premium when issuing bonds (Ciocchini et al, 2003) or encounter more difficulties when they introduce fiscal stimulus packages aimed at targeting future budget consolidations (Arin et al, 2011). There is also evidence (Lambsdorff, 2006) that bureaucratic corruption leads to misallocation of resources and reduces productivity and service quality, diminishes expenditure in education and distorts private sector activities by giving rise to shadow economy and tax evasion. Although appealing on intuitive grounds, the hypothesis that there is a negative effect of corruption on equality in resource distribution does not pass robust causal relationship statistical tests (Gupta et al, 2002).
Causes of corruption are the subject of both theoretical and empirical research.
Although many causes of corruption seems to be also consequences of corruption to such an extent that feedback loops may obscure the true causal relationship, still a list of possible critical conditions making corruption more likely can be compiled following Lambsdorff (2006). The list includes the size of the public sector, the quality of regulation, the degree of economic completion; the structure of government; the amount of decentralization; the impact of culture; the prevailing values and gender of the agents involved; the characters of some invariant features of a country, such as geography and history.
Lastly, we may recall that suggestions of anti-corruption reforms have emerged alongside the discussion of corruption causes. Accordingly, some suggested that the public sector should be downsized and privatization of SOEs further accomplished.
Other researches stress that regulation of economic activity and particularly public utilities should be simplified and made less discretionary. For others, officials’ salaries should increase because low salaries force public servants to supplement their income illicitly whereas high salaries are a premium that is lost if a public servant is caught and fired. Press freedom and independence are also indicated as forces conducive to a better anti-corruption environment. The judiciary, however, has received a special analytical attention only quite recently. Judiciary may deter corruption since the quality of the judiciary can make easier and quicker the repression of corruption. However, the judiciary sector too is open to bribery. Data discussed by Rose-Ackerman et al. (2012) show that judiciary, the only service having a pure public good nature among those they analysed, is the only sector in which corruption is higher in high income countries than in low income ones, as if a sort of Wagner Law for corruption was affecting the judiciary activity as countries become richer. In the same paper Rose-Ackerman et al.
(2012) also discuss the potential effectiveness of other measures designed to deter malfeasances such as: i) external monitoring and punishment; ii) transparency and bottom-up accountability initiatives; iii) civil service reforms (wage restructuring, personnel rotation and recruitment practices based on merit); iv) competitive service delivery, particularly for substitute services; v) international efforts to coordinated transnational anti-corruption activities; vi) reinforcement of legal services with the institution of National Courts and International Forums. Other policy measures, derived from the application of the tools of New Institutional Economics to the issue of corruption, are discussed by Lambsdorff (2002; 2007) who maintains that fighting corruption should focus less on individual moral attitudes or penalties and more on methods to destabilise corrupt relationships. The role of institutions and particularly the voting rules6 are also addressed explicitly or implicitly as sectors in which reforms might be introduced to reduce corruption (Potter et al., 2011).
Balancing uncertain gains and costs – given penalties – of two parties contemplating a corrupted bargaining has been the focus of the existing theoretical literature on corruption, which has progressively adapted this approach to the study of the different contexts in which corruption may emerge. Framing their researches into this analytical structure, many authors have discussed the properties the penalties should posses for effectively restrain the parties of a potential corrupted transaction to conclude the illicit deal. The main concern of the literature seems to regard the equilibrium relationship between briber and bribee given the probabilities of being discovered and the above mentioned penalties structure. Yet, as stressed in section 1, no general analysis of corruption has been undertaken with the purpose of deriving anti-corruption measures from the maximization of a properly defined social welfare function in which the wellbeing of the main actors (government, officers and privates) are included, compliance decisions are related to government regulation and compliance cost and the existence of asymmetric information on compliance costs between government and agents is properly taken into account. To conduct this kind of analysis we construct a model by assuming that regulation is costly for private subjects and that incompliance (when undiscovered) brings about a pay-off given by compliance cost avoidance. Since this provides the basis for a corruption bargaining we trust that using avoided costs as a measure of the gains pursed by a potential bribee allows to devise a sufficiently general analytical framework which may accommodate many of the specific cases of corruption discussed in the literature. Then, using the model discussed in this and in the following sections, we derive optimal punishment structure from a Principal-Agent scheme of the relationship between government and privates. Participation and incentive compatibility constraints are obtained from Nash equilibrium conditions emerging from the briberbribee bargaining and then incorporated into the government maximization problem.
The basic structure of the model is given by the following hypotheses.
The government sets some standards of conduct to be observed by firms or households operating in a society when they want to implement a project. Call these standards X and assume they can be cardinally measured (e.g. the number of pollution reducing filters in a production). Then the government invest resources to regulate privates’ behaviour by monitoring their compliance. Call these norms regulation and suppose they can be expressed as a variable r measurable in monetary terms. Assume the government invests the quantity r ≥ 0 of resources in regulation activities and obtains from this activity a return given by a function V(r) which we assume to be continuously increasing and strictly concave in r. V(r) is twice differentiable on [0, +∞) and satisfies the Inada conditions, V’(0) = +∞ and lim r ↑∞ V '(r ) = 0. Then, the benefit from standards and regulation-monitoring is V(r)X which is the gain of implementing X through r.
Specifically, the dimension of the voting districts seems to affect opportunities of corruption. Person et al. (2011) argue on empirical grounds that larger voting districts are associated with less corruption.