Despite prosecutors’ difficulties in proving corporate bribery, nearly all enforcement actions end with a settlement at the pretrial stage. Compared to court proceedings, settlement-based enforcement provide prosecutors with flexibility to reward offenders’ self-reporting and cooperation, and reach quicker conclusions to complex cases. In this article, we explain, such enforcement needs regulation to minimize potentially harmful side-effects. When the difference between a court and settlement sanction exceeds a certain size, the alleged offender accepts a settlement regardless of actual responsibility of misconduct. For the prosecutor, the option of offering a lenient settlement means weaker incentives to ascertain the material facts of the case. Society receives less information about the blameworthy act, little opportunity to evaluate the sanction, and less reason to expect sanctions to deter bribery. We show why such consequences result in under-deterrence of bribery and weaker rule of law. The use of settlement may have a self-escalating effect because the enforcement mode can reduce the predictability of the law, while a defendant’s inclination to accept a settlement offer depends on the predictability of the law. Our results suggest that United Kingdom’s current escalation of enforcement of corporate bribery laws will lead to a mixture of settlements and court decisions, while in the United States firms will continue to negotiate settlements as if there were no opportunity to have their cases tested in court.