Leniency programs are widely used as instruments of competition policy. They are supposed to serve two purposes: in the short-run to facilitate the detection of cartels and thereby to reduce cost of legal enforcement, and in the long-run to deter firms from antitrust abuse. In our empirical study of the 1996 EU Leniency Program, we find strong evidence that the program provides incentives to reveal information on criminal activities in the sense that agencies are better informed about the cartel conduct than they would be absent the program. Despite of that, investigation and prosecution cost reductions seem to be rather modest.On deterrence, we neither find that the leniency program stabilizes cartels through facilitating punishing strategies, nor do we find that cartels are destabilized because defecting from a cartel becomes less costly. The sharp rise in the number of convicted cartels after introducing the program may be attributed to a rise of cartel activity caused by an accelerated market integration in the EU in the beginning of the 1990s.Journal of Economic Literature Classification: K21, K41, L4

This paper extends the interval Hotelling model with quadratic transport costs to the n-player case. For a large set of locations including potential equilibrium configurations, we show for n > 2 that firms neither maximize differentiation-as in the duopoly model-nor minimize differentiation-as in the multi-firm game with linear transport cost. Subgame perfect equilibria for games with up to nine players are characterized by a U-shaped price structure and interior corner firm locations. Results are driven by an asymmetry between firms. Interior firms are weaker competitors than their rivals at the corners. Increasing the number of firms shifts even more power to the corner firms. As a result, there is too much differentiation from the social perspective if n 3, while adding firms leads to a level of differentiation in equilibrium below the social optimum.

* This version of the article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the publisher's final version AKA Version of Record.

R&D cooperation, Endogenous coalition formation, Optimal coalitions, Antitrust, Patent thicket, K11, L41, M2,