I characterize the effects of empirically observed managerial incentives on long-run oligopolistic competition. When managers have a preference for smooth time-paths of profits – as revealed by the empirical literature on “income smoothing” – manager-led firms can sustain collusive agreements at lower discount factors. Capped bonus plans and incumbency rents with termination threats make collusion supportable at any discount factor, independent of contracts’ duration. When managers have these preferences/incentives and demand fluctuates, “price wars during booms” need not occur: the most collusive price may then be pro-cyclical.
Published in European Economic Review (July 2005), pages 235-254.
Leniency programs reduce sanctions for law violators that self-report. We focus on their ability to deter cartels and organized crime by increasing incentives to “cheat” on partners. Optimally designed “courageous” leniency programs reward the first party that reports with the fines paid by all other parties, and achieve the first best: complete and costless deterrence. “Moderate” leniency programs that only reduce or cancel sanctions may deter organized crime (a) by protecting an agent that defects from fines and from other agents’ punishment; and (b) by increasing the riskiness of crime/collusion, in the sense of Harsanyi and Selten (1988).
This paper analyzes the consequences of the size of an auction, in terms of the volume of product and the time lenght covered by the auctioned contract, on the incentive bidders have to collude. The paper shows that the recommendation to increase as much as possible the auction size in order to prevent bidders from colluding, found in several policy papers, is valid only in some exceptional circumstances. In many cases increasing the auction size may induce bidders to form a cartel that would not have been existed otherwise.
I describe a price game in which consumers face search costs and base their quantity decision on the expected price. Because of search costs, the choice of the firm they will buy from is described by a random process. I show that the expected equilibrium price is above the monopoly price. This result does not change if demand comes from a small share of perfectly informed consumers with zero search costs.
Published in The Journal of Industrial Economics, December 2003, Vol. LI, n. 4, pp. 427-32.
Economic evidence and proof of a collusive agreement. I investigate whether some economic observations may contribute to prove the existence of an illegal collusive agreement. The facts analyzed are: simultaneous increase in prices, price parallelism, market shares stability, price similarity among firms when firms have different costs, price similarity among firms and products when products have different costs, third degree price discrimination. I find that in many cases these observations per se do not allow to discriminate between a collusive and a competitive equilibrium and that further information on costs and demand is needed to obtain a legal proof.
Published in Economia e politica industriale.
Antitrust authorities often consider parallelism of firms’ strategies and market share stability as clues of illegal collusion in the form of concerted practices. In this paper I show that this inference may be contrary to some theoretical results. I develop a model of price competition with differentiated products in which demand and costs vary over time. If the market is perturbed by shocks on the supply or on the demand side which are common to all firms, then price parallelism will occur both in a competitive and in a collusive equilibrium and cannot signal which one occurred. If shocks are firm specific, price parallelism connotes a competitive market. The paper also shows that the competitive equilibrium is characterized by a higher market share stability than a collusive equilibrium.
Published in Rivista Italiana degli Economisti.
I show that as long as the stock market has perfect foresight, profits are distributed as dividends, and incentives are paid more than once or are deferred, stock-related compensation packages are strong incentives for managers to support tacit collusive agreements in repeated oligopolies. The stock market anticipates the losses from punishment phases and discounts them on stock prices, reducing managers’ short-run gains from any deviation. When deferred, stock-related incentives may remove all managers’ short-run gains from deviation, making collusion supportable at any discount factor. The results hold with managerial contracts of any length.
Published in RAND Journal of Economics 31/1 (Spring 2000), pages 22-42
We study the consequences of ‘leniency’ – reduced legal sanctions for wrongdoers who spontaneously self-report to law enforcers – on corruption, drug dealing, and other forms of sequential, bilateral, illegal trade. We find that when not properly designed, leniency may be highly counterproductive. In reality leniency is typically “moderate,” in the sense of only reducing, or at best cancelling the sanctions for the self-reporting party. Moderate leniency may greatly facilitate the enforcement of long-term illegal trade relations, and may even provide an effective enforcement mechanism for occasional (one-shot) illegal transactions, which would not be enforceable otherwise.
Following Bernheim and Whinston (1990), this paper addresses the effects of multimarket contact on firms’ ability to collude. Real world imperfections tend to make firms’ objective function strictly concave and market supergames “interdependent”: firms’ payoffs in each market depend on how they are doing in others. Then, multimarket contact always facilitates collusion. It may even make collusion sustainable in all markets when otherwise it would not be sustainable in any. The effects of conglomeration are discussed. “Multigame contact” is shown to facilitate cooperation in supergames other than oligopolies as long as agents’ objective function is submodular in material payoffs.
Published in Journal of Economic Theory 89/1 (November 1999), pages 127-139
