The leniency program in Korea and its effectiveness

In this article, we examine the determinants that induce cartel members to apply for leniency in the case of Korea, using data on detected cartel cases covered from 2005 to April 2012. Factors such as fines before deduction, past years' total fines, total market share of participating firms, number of firms involved, and more are used to test which variables increase the probability that a firm will apply for leniency. We also analyze whether leniency-invoked cases have higher fines before deduction, and whether those cases account for shorter durations of investigation, using the simultaneous equations model with endogenous switching. Our main findings show that cases with a small number of cartelists during the years 2008 and 2009 indicate a higher probability of leniency applications. In addition, fines before leniency were shown to be larger and the lengths of investigation shorter for leniency-applied cases from the estimated sample selection model.