We extend Becker's model of crime by allowing for collusion between an
arresting officer and the criminal at the expense of the police depar
tment. The weakening in the deterrent power of a given set of criminal
sanctions that results is explored in a perfect information setting.
Applying a solution concept developed in related work by Cadot we deri
ve optimal policies for each group, We show in particular that the Wel
l-established results from the Becker model do not all carry through,
In particular we find that the maximal fine may no longer be optimal.
(C) 1997 by Elsevier Science Inc.