Elsevier

World Development

Volume 34, Issue 2, February 2006, Pages 349-358
World Development

Persistence of corruption: some theoretical perspectives

https://doi.org/10.1016/j.worlddev.2005.03.010Get rights and content

Summary

The paper examines how pervasive corruption can be persistent. Despite the presence of anti-corruption measures and incentives, corruption has a strong tendency to persist. In the first part, we discuss how a high level of corruption or a low level of compliance can be an equilibrium outcome. In such a case, corruption becomes the social norm. In the second part of the paper, we use ides from evolutionary game theory to discuss how corrupt behavior can be immune to interventions and it can sustain itself against different individual behavioral norms.

Introduction

Is corruption simply a manifestation of deviant behavior from the norm or is it the norm itself? Corruption is commonly defined as “behavior that deviates from formal duties because of private gains.” Hence, the very definition of corruption would suggest that corrupt acts are deviations from implicit or explicit behavioral norms (with or without legal and ethical connotations). However, the widespread nature of corruption in some societies indicates that corrupt behavior is the norm itself—despite the fact that it is inefficient and generally condemned. In this paper, we try to examine the various factors that contribute to the persistence of such behavior.

Historically, many societies have endured long periods of widespread corruption. Huang (1974) examines how corruption emerged in China under the Ming dynasty in the 14th century and continued to spread during the Qing dynasty through to the 19th century. There were many attempts including large scale salary reforms to curb corruption but corruption grew unabated. Similarly, Waquet (1991) discusses how corrupt practices were widespread during the 17th century in Florence despite the presence of fairly repressive anti-corruption laws during this period. Many contemporary societies are also beset with similar problems. In the early 1960s, the Government of India viewed corruption as a top priority and appointed a high-powered committee to look into corruption. The committee, chaired by Santhanam, made a detailed study and made a number of recommendations. But the anti-corruption measures over the last decades have not met with any success and corruption has grown.

It is important to bear in mind that we are concerned with persistence of widespread corruption and not just the practice of corruption. The phenomenon of corruption is certainly very old. References to bribery and the punishments for bribery can be found in many ancient sources like The Code of Hammurabi, King of Babylon (22nd century BC), The Eddict of Harmhab, king of Egypt (14th century BC), and Kautilya’s Arthasastra (14th century BC). Corruption is as old as the notion of kingdom itself and a corruption-free society is akin to an ideal state.1

There are various ways to approach the persistence issue. Some scholars would argue that corruption persists because of the functional characteristics associated with it. Merton’s (1968) fascinating study of American machine politics clearly sets out this intellectual tradition. Machine politics reached a peak at the turn of the century when corruption of various forms thrived. However, there was no systematic attempt to remove these because of the several latent functions they fulfilled. Even when there were serious attempts, as in the case of Waquet’s study of Florence, “the laws were intended not only to control corruption but also render it tolerable.” While there might be some merit in this view, we do not pursue it here.2

One could also argue that corruption persists because of inadequate initiatives and proper incentives. Bureaucratic corruption can be attributed to lack of sufficient political will and likewise, political corruption can be attributed to lack of adequate political competition. This is somewhat similar to the recent issue of corruption and competition. It is generally believed that greater competition in the form of privatization and deregulation would lead to substantially lower levels of corruption. But the recent experience of many countries (including some of the transition countries) shows that this may not indeed be the case. Corruption seems to be on the rise despite large scale privatization and deregulations. One can argue, along the same lines, that the reforms have not been proper or adequate and the right incentives are still not in place.3 There may be some merit in this view, since some societies or organizations have managed to control corruption in a significant way. But it does not provide adequate explanation and can be tautological in many instances.

In this paper, we consider two complementary approaches to the problem of corruption. First we show that pervasiveness of corruption contributes to its persistence in a significant way. When there are many corrupt individuals in the society, it becomes optimal to be corrupt despite the presence of anti-corruption policies and incentives. This way corrupt behavior becomes the equilibrium behavior or the social norm.

This view is not new. Economists have shown that endemic corruption can be viewed as an equilibrium outcome in models with multiple equilibria. It has been noted that different societies with relatively same levels of development, judicial machinery, and politico-legal structures can exhibit varying degrees of “illegal (pre) occupation” like corruption, tax evasion, and other regulatory non-compliance. The explanation for this observation is that different societies can get caught in different equilibria. At a general level, this multiplicity arises due to various forms of externality. For example, if people expect more people to be corrupt, then the expected cost of being corrupt would be less (the probability of apprehension might be low or even the social sanction against corruption could be low) leading to more people being corrupt.4 Like all models of multiple equilibria, these models cannot explain why some get caught in the bad equilibrium, but still aid to our understanding of the persistence of corruption in some societies.

We discuss a similar situation where both low compliance and high compliance can be equilibrium outcomes. The novelty of the analysis lies in the exact mechanism which gives rise to the multiplicity. Existing models of corruption tend to focus primarily on the cost and benefit of non-compliance, but we also look at the cost and benefit of compliance. It is often seen that in many of the corrupt societies, those who comply with the law or social standards often become victims of harassment, extortion, and alleged corrupt behavior. This means that the incentive to remain compliant declines leading to rise in the level of non-compliance. We are able to highlight the importance of several factors like the nature of information technology, the effectiveness of judicial redress, and compliance costs. We discuss how and when societies can avoid the low-compliance equilibrium. But guaranteeing these conditions become difficult when there is widespread corruption.

The second approach focuses on individual norms of behavior. The previous approach would suggest that people tend to be corrupt because there are many others who are corrupt or “being corrupt” is the social norm. This is different from saying that these individuals would be corrupt irrespective of the nature of the social surroundings. We wish to examine the extent to which corruption as an individual norm of behavior can sustain itself in the long run. We introduce some preliminary ideas from evolutionary game theory to discuss the persistence (and success) of these behavioral norms. It is shown that corrupt behavior can be self-sustaining and stable against other forms of behavior. Roughly speaking, this would mean that in the long run only corrupt people would survive in a population of corrupt and honest people. This is derived in a simple stylized model but it suggests that we should be asking the question “why don’t we see corrupt behavior everywhere?” rather than the question “why do we see so much corruption?” This suggests that societies have over time tried to curb corrupt behavior in various ways and many of them would have to do with development of individual and social values.

The plan of the paper is as follows. In Section 2 we discuss how low compliance can be an equilibrium outcome. We present a model which admits two equilibria for certain parameter values. Section 3 discusses the evolutionary model of corrupt behavior. Section 4 concludes with a few comments.

Section snippets

Corruption and social norms

We present a simple model of enforcement in this section. It is shown that corruption or non-compliant behavior can be the equilibrium outcome in some cases. In these situations, corruption is the norm rather than deviant behavior. For easy reading, we have omitted most of the technical aspects of the model. It is a general analysis of compliance—corruption being one such application. In the later subsection, we discuss how the ideas can be applied in an organizational context as well.

Corruption and evolutionary dynamics

In this section, we look at some of the ideas from evolutionary theories to see whether the persistence issue can be addressed in such a framework. There are at least three motivations for focusing on an evolutionary approach. First, recent work on evolutionary game theory has shown that evolutionary stability can address the issue of equilibrium selection to some extent. Since corruption happens in one such equilibrium, we can ask what kind of social dynamics would select such an outcome

Conclusion

The aim of the paper is rather modest. The paper discusses two different approaches to study the issue of persistence. In the first case, using a static framework, it is shown how pervasive corruption becomes the social norm. In the second case, we have shown how corrupt behavior by individuals can survive and succeed in the long run in an evolutionary setting.

The first half of the paper shows that it is possible to have a situation of low-compliance and pervasive corruption as an equilibrium

Acknowledgment

This paper was prepared for the symposium on “Re/Constructing Corruption” held on May 1, 2003 in University of East Anglia. I would like to thank Nick Duncan and Indranil Dutta for making me write on this topic. I would also like to acknowledge intellectual debt to Kaushik Basu and Dilip Mookherjee for many discussions on this and related topics. I have also benefited immensely from comments by two anonymous referees.

References (28)

  • Dutta, I., & Mishra, A. (2003). Corruption and competition in the presence of inequality and market imperfections....
  • R. Huang

    Taxation and government finance in sixteenth century Ming China

    (1974)
  • S. Huntington

    Political order in changing societies

    (1968)
  • Kandori, M. (1996). Evolutionary game theory in economics. University of Tokyo,...
  • Cited by (99)

    • Condoning corruption: Who votes for corrupt political parties?

      2023, Journal of Economic Behavior and Organization
    • Can inflation targeting reduce price information asymmetry and alleviate corruptive behavior? Evidence from developing countries

      2022, Economic Systems
      Citation Excerpt :

      Additionally, the persistence of corruption and the relevance of the historical information provided by its own process, and that of the other explanatory variables, make it suitable for introducing a dynamic modeling framework, described by the hypothesized data generating process provided in Eq. (1). Lui (1985, 1986), Mishra (2006), Accinelli and Carrera (2012), and Accinelli et al. (2014) provide a modeling framework of corruptive behavior where individuals create their strategies independently. However, as time passes, they compare their payoffs to the optimal ones, and thus adjust their respective strategies.

    • Exit or Voice? Corruption Perceptions and Emigration

      2024, Studies in Comparative International Development
    View all citing articles on Scopus
    View full text