WPS6596 Policy Research Working Paper 6596 When Competition Corrupts A Theoretical Analysis of Market Structure and the Incidence of Corruption Kaushik Basu Tamara McGavock Boyang Zhang The World Bank Development Economics Vice Presidency September 2013 Policy Research Working Paper 6596 Abstract The paper develops a simple model to demonstrate that, maybe because they pay a bribe to avoid installing the paradoxically, greater competition may exacerbate the environmental safeguards required by law—such that problem of corruption. Market participants engaging honest players are driven out of the market when the in corrupt practices enjoy lower production costs— market becomes sufficiently competitive. This paper is a product of the Development Economics Vice Presidency. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The authors may be contacted at tjm276@cornell.edu. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team When Competition Corrupts: A Theoretical Analysis of Market Structure and the Incidence of Corruption Kaushik Basu, Tamara McGavock and Boyang Zhang September 7, 2013 Abstract The paper develops a simple model to demonstrate that, paradoxically, greater competition may exacerbate the problem of corruption. Market participants engaging in corrupt practices enjoy lower production costs–maybe because they pay a bribe to avoid installing the environmental safeguards required by law–such that honest players are driven out of the market when the market becomes sufficiently competitive. Keywords: Market Competition, Corruption, Industrial Organization. JEL Classi�cation Numbers: D73, D4, L1, O1 1 1 Motivation As economists, we normally associate competition with healthy and desirable market outcomes. This is so in terms of standard measures that economists use, such as efficiency and consumer surplus: competition enhances these. Similar claims are also often made for some social indicators. For example, Becker (1957) argues that discrimination of different kinds, including racial, falls as market competition increases.1 In this paper we argue that, paradoxically, corruption will tend to flourish under competition and may be less pervasive when competition is less �erce. Some prominent papers in the literature have previously analyzed the link between cor- ruption and market structure (see, eg., Rose-Ackerman, 1975; Ades and Di Tella, 1999; Yoo, 2013) but the present paper provides a very different approach. In our analysis, corruption itself generates rents arising from lower costs, which implies that standard competition leads to higher, not lower, incentives for corruption. In this manner, our paper shares a loose connection with Tirole (1986) in the sense that higher degrees of competition force “agents� to collude with a “supervisor� in the environment of Tirole’s paper. In contrast to the tra- ditional argument made by Acemoglu and Verdier (2000), we posit the role of government in mitigating the undesirable effects of competition on corrupt practices. 2 Model with Two Types The standard economic model of corruption and crime is entirely amoral: the fact that an action entails corruption means nothing to the individuals apart from its economic returns, which they consider in isolation when deciding whether or not to choose an action (Becker, 1968). Other work goes further and attributes this amorality also to the enforcers of the law, by assuming that police and magistrates will enforce the law or choose to take bribes depending only on their personal returns from such actions (Basu, Bhattacharya and Mishra, 1 One of us has elsewhere (Basu, 2010, Ch.5) argued that this may not be the case. There are strong theoretical explanations of how discrimination and racial prejudice can survive competition. 2 1992). In this paper we break from that tradition and assume that while there may be amoral people to whom breaking the law is appealing if and only if it is economically pro�table, there are also people who �nd corruption distasteful, with some even refraining from being corrupt no matter what the returns to being corrupt are. In this section, we consider the polar case in which there are two types of people: the honest and the corrupt. The corrupt are like the standard human being in the models of economists (make what you will of the economics profession) and the honest are those who will, given a choice, never choose the corrupt option (see Guha and Guha (2011) for a similar setting).2 Following this, we will speak of �rms as being corrupt or honest, meaning that �rms are managed by corrupt entrepreneurs or by honest entrepreneurs. Firms choose to enter a certain goods industry if they will earn positive pro�ts.3 Suppose the number of �rms in the industry is exogenously constrained to n. If more �rms desire to enter the industry than there are spaces available, �rms who apply for entrance will be randomly selected to enter. Demand in the industry is linear, given by: P = a − bQ (1) Once in the industry, the n �rms have uniform �xed marginal cost c (c < a) and engage in Cournot competition. Denoting �rm i’s output by qi and using a star to denote Cournot equilibrium, we obtain (from the �rst order condition) the following condition for each i: ∗ ∗ a−b j =i qj −c qi = (2) 2b This means that the Cournot equilibrium price P (n) and aggregate quantity Q(n) will be functions of the number of �rms: 2 This is in contrast to Myerson (2004) where honest behavior emerges in equilibrium from people who are fundamentally amoral and hence are fully corruptible. 3 Note that this assumes that the outside option is equivalent for honest and corrupt managers. In practice, this may not be the case, but we assume this for simplicity. 3 n a−c Q(n) = (3) (n + 1) b a + nc P (n) = (4) n+1 We will, in what follows, denote each �rm’s Cournot equilibrium output by q (n). Clearly, a−c q (n) = b(n+1) . Now assume that before starting operation, a �rm is supposed to incur a �xed cost L to put in some environmental safeguards, but it is possible for the �rm to avoid this by paying a bribe l. We assume of course that l < L and without loss of generality, we will assume l = 0. Firms thus incur �xed cost hL, where h = 1 for honest �rms and h = 0 for corrupt �rms. Proposition 1 If the industry is sufficiently competitive (i.e. the number of �rms in the industry is above a certain threshold level n∗ ) then all �rms in the industry will be corrupt. Proof. Suppose there are n �rms in the industry. Then pro�t earned by an honest �rm is given by: P (n)q (n) − cq (n) − L. (5) Using the de�nition of P (n) and q (n) from above, it is easy to see that the pro�t earned by an honest �rm will be zero if: a−c n= √ −1 (6) Lb Denote the value of n that satis�es equation (6) by n∗ . It follows that if n > n∗ then the industry will be populated only by corrupt �rms. Suppose next that the number of �rms in the industry is not exogenously given. Instead there is free entry of �rms. If the total number of corrupt people in the economy is greater than n∗ , it is obvious that in equilibrium with free entry, all �rms in the industry will be 4 corrupt. The argument is simple. As long as pro�ts earned by corrupt �rms are positive, corrupt �rms will continue to enter. As this continues and the number of �rms in the industry increases above n∗ , only the corrupt �rms will remain in the industry. Hence the next corollary: Corollary 1 In a competitive industry with free entry, corruption will be endemic. 3 Generalization It may appear at �rst sight that our result is driven by the binary nature of the attitude people have toward bribery. The result that a critical point exists beyond which corruption suddenly increases in a competitive industry seems especially to be a consequence of the assumption in the previous section of there being only two kinds of people, the honest and the corrupt. Interestingly, this is not so. Even with different kinds of behavior, the threshold effect will be present. So let us move to the more general case by assuming, as indeed is real, that people vary in the degree of their attitudes toward corruption and honesty (which, in the present paper is equated with the aversion to corruption), ranging from incorruptible to the completely amoral. We shall here denote a �rm’s (meaning its entrepreneur’s) propensity for honesty by h ∈ [0, ∞), which denotes the psychological cost to a person of being dishonest. If h = ∞, such a person will not be corrupt no matter how large the reward from corruption. Note that in this paper we are not distinguishing among different kinds of corruption. This is harmless here since the only corruption that occurs in this model is bribery to evade environmental regulations. We shall use φ(h) to denote the fraction of individuals whose propensity for honesty is less than or equal to h. Let us denote a �rm’s pro�t ignoring the ˜ and call this the variable pro�t. If there are n �rms in the industry, then: �xed cost by π a + nc a−c a−c ˜ (n) ≡ π −c n+1 b(n + 1) b(n + 1) 5 (a − c)2 ˜ (n) = or π (7) b(n + 1)2 Clearly, a �rm of type h will choose to enter the industry if and only if: ˜ (n) ≥ min{h, L} π (8) If (8) holds and h < L, this �rm will enter the industry and be corrupt. If (8) holds but h ≥ L, it will enter the industry but not be corrupt. We used an arbitrary tie breaker rule for h = L, but that is innocuous. It is now easy to describe the relationship between the degree of competition and the (a−c)2 ˜ (n) increases toward incidence of corruption. Note that as n goes to 1, π 4b . Starting ˜ (n) declines monotonically. De�ne from n equal to one and increasing n, it is obvious that π ˜ to be such that π n n) = L. If the number of �rms allowed in the industry n is less than n ˜ (˜ ˜, all �rms will want to enter the industry and a random selection of �rms will manage to do so. The fraction of �rms in the industry that pay a bribe will be given by φ(L). ˜ , then π If n > n ˜ (˜ ˜ (n) < π n) = L. Now only those �rms whose h is less than L will �nd it worthwhile to enter the industry. But if h < L, then it pays to give a bribe. Hence, the incidence of bribery will be total. In other words, all �rms in the industry will be paying a bribe. Hence our fundamental result is unchanged: there is a critical level of competitiveness such that once an industry becomes competitive beyond this level, bribery and corruption will be ubiquitous. It is interesting to note that even if individual propensity toward corruption varies �nely from the incorruptible to the totally amoral, the incidence of corruption rises abruptly beyond a certain level of competitiveness in the industry. 4 Observations We believe our paper points to an important reason why corruption is so rampant in some industries. India’s large system of ration shops has a high incidence of corruption (in 6 this case it takes the form of cheap grain received from the government and meant to be sold to poor households at a special low price instead being sold illegally at the higher market price). A little calculation shows that for a fully honest ration shop owner, the pro�t margin is so low that it is not worthwhile for a person to run a ration shop unless he or she is prepared to be corrupt (Khera, 2011). It is this idea that drives our theoretical model. The model may give the impression that corruption is inevitable in competitive industries. That is, however, not true; our model was developed for an environment in which bribery is worthwhile for amoral individuals. The way to root out corruption is to make sure that this is not the case in society. If, for instance, we choose a police force by selecting innately honest people or by having punishment rules which make it not worthwhile for an officer to take a bribe, then even in competitive industries there will be negligible corruption. Further it is possible for societies to develop self-enforcement norms whereby bad behavior is collectively punished by all and where people punish bad behavior because not to punish bad behavior is itself bad behavior (Akerlof, 1976; Basu, 1986). Finally, by moving away from the patently unrealistic textbook assumption that all hu- man beings are amoral and by allowing for the fact that people incur psychological cost for being corrupt, the paper draws attention to the role of the distribution of morality in a society. References Acemoglu, D. and Verdier, T. (2000). “The Choice Between Market Failures and Corrup- tion.� American Economic Review: 90, 194-211. Ades, A. and Di Tella, R. (1999). “Rents, Competition, and Corruption.� American Eco- nomic Review: 89, 982-993. Akerlof, G. (1976). “The Economics of Caste and of the Rat Race and Other Woeful Tales.� Quarterly Journal of Economics 90(4), 599-617. 7 Basu, K. (1986). “One Kind of Power.� Oxford Economic Papers 38(2), 259-282. Basu, K. (2010). Beyond the Invisible Hand: Groundwork for a New Economics. Princeton University Press. Basu, K., Bhattacharya, S., and Mishra, A. (1992). “Notes on Bribery and the Control of Corruption.� Journal of Public Economics: 38, 349-359. Becker, G. (1968). “Crime and Punishment: An Economic Approach.� The Journal of Po- litical Economy: 76(2), 169-217. Becker, G. (1957). The Economics of Discrimination. The University of Chicago Press. Guha, A. and Guha, B. (2011). “The Persistence of Goodness.� Journal of Institutional and Theoretical Economics: 168, 432-443. Khera, R. (2011). “India’s Public Distribution System: Utilisation and Impact.� Journal of Development Studies 47(7), 1038-1060. Lambsdorff, J. (2006). “Consequences and Causes of Corruption: What do we Know from a Cross-section of Countries?� in International Handbook on the Economics of Corruption. Susan Rose-Ackerman (ed.), Edward Eglar Publishing. Myerson, R. (2004). “Justice, Institutions, and Multiple Equilibria.� Chicago Journal of International Law 5(1), 91. Rose-Ackerman, S. (1975). “The Economics of Corruption.� Journal of Public Economics: 4, 187-203. Tirole, J. (1986). “Hierarchies and Bureaucracies: On the Role of Collusion in Organiza- tions.� Journal of Law, Economics, and Organization: 2, 181-214. Yoo, S.H. (2013). “Competition, Corruption, and Institutional Design.� Mimeo: Korea Uni- versity, Seoul. 8