Forschungsgruppe ORCOS
> Zum Inhalt

Dynamics of Corruption

Corruption in some form or other occurs so frequently that it is hard to describe it as deviate behavior. Although the phenomenon has many dimensions including historical, political and social ones we restrict our considerations here to the economic aspects of corruption. What we have in mind are economically motivated illegal transactions between a member of an organization, e.g. a bureaucrat, and an agent outside of it. The bureaucrat provides an economic value, e.g. a local information or an unwarranted license, for the non-member in order to gain some economic rent.

Corruption is a perennial phenomenon, to some degree corruption is unavoidable, a necessary evil or - within certain limits - even useful. Public attitude requires a certain level of reported corruption before opposition is triggered. It is this threshold property as well as other nonlinearities which make corruption a phenomenon whose investigation is not only economically interesting but also worthwhile from a mathematical point of view. For an introduction into the field see Rose-Ackerman (1975, 1978), recent surveys on the economics of corruption are Andvig (1991) and Schleifer and Vishny (1993).

A major conclusion resulting from these economic investigations is that, given equal conditions, bribery and thus corruption increase with the amount of goods traded through the political process and through bureaucracies as opposed to being traded in the market. More concisely, corruption is positively correlated with government's ownership and interventions in the economy. The reason is that in a competitive market nobody holds monopoly power over certain property rights, allotments, and privileges.

The abundant literature on case studies of reported corruption in various regions of the world suggests several general questions. Mathematical models of corruption try to get answers to these open problems. Let us briefly summarize some of these questions: 

  • What is the impact of corruption on welfare? While high corruption levels reduce economic efficiency (delays, no decentralisation, unability to decide), 'light bribing' might have positive economic effects (e.g. increase of service speed).
  • Why are not more people corrupt? If corruption is economically advantageous and not too risky, why does the social norm 'to stay clean' still persist over time and space?
  • Why do we observe different levels of corruption in comparable situations?

To answer the second questionAndvig (1991) proposed to include reputation into the utility of corrupt agents. According to that 'people want to be rich and famous ' the latter not being redundant; see Dawid and Feichtinger (1996) for a dynamic setup of this problem. The third problem has been attacked by taking into consideration social interaction. By this we mean that the utility an individual receives from a given action depends on the choices of the others in that individual's reference group. It turns out that the same kind of agents within the same kind of economic and political system may through their interaction generate several levels of corruption (compare Andvig, 1991)

The present research project will try to give answers to all those questions in an intertemporal context. In what follows we briefly describe some related studies in the dynamics of corruption. Rinaldi et al (1998) investigate the dynamics of corruption at the top (i.e. by politicians). For this purpose a dynamic, politico-economic framework is developed and analyzed. A particular feature of this investigation is an analytical characterization of the system. This allows one to interpret these properties in terms of economic and institutional parameters. As an example, the framework is applied to simulate Italian history from 1948 to today.

Feichtinger and Wirl (1994) consider the widespread phenomenon of political corruption and analyzes some of its political, economic and dynamic properties. They assume an individually rational politician (homo oeconomicus) who derives benefits from popularity and from corruption (directly or indirectly). Of course, popularity will suffer as the evidence of corruption piles up. Given these dynamic reactions, the politician has to trade off between popularity and corruption within an intertemporal framework. This may lead to complex, in particular to cyclical and even unstable patterns. However, these complexities are restricted to those cases where popularity is less important ('dictatorship'), while the necessity of high public approval rates ('democracy') is sufficient for stable regimes.