Antitrust authorities of many countries have been trying to establish appropriate competition policies based on economic analysis. Recently an anti-cartel policy called a "leniency program" has been introduced in many countries as an effective policy to dissolve cartels. In this paper, we studied several kinds of leniency programs through laboratory experiments. We experimentally controlled for two factors: 1) cartel size: the number of cartel members in a group, small (two-person) or large (seven-person), 2) schedule of reduced fine: the number of firms that are given reduced fines. The experimental results showed that (1) an increase in the number of cartel members in a group increased the number of cartels dissolved, (2) changing the coverage of reduced fine had no significant effect both in two-player case and in seven-player case.
