WHENEVER THE GOVERNMENT ATTEMPTS TO CONTROL UNDESIRABLE BEHAVIOR, IT MUST CHOOSE A SANCTION TO BE IMPOSED AS WELL AS AN AGENT TO DO THE ENFORCING. FREQUENTLY, A PARTICULAR SANCTION CAN BE ENFORCED UNDER A VARIETY OF INSTITUTIONAL ARRANGEMENTS. FOR EXAMPLE, SOME ANTITRUST PENALTIES IN THE U.S. ARE ENFORCED BOTH BY PUBLIC AGENCIES (E.G., THE JUSTICE DEPARTMENT AND THE FEDERAL TRADE COMMISSION) AND BY PRIVATE PARTIES, THE VICTIMS OF THE VIOLATIONS, AND THE VICTIMS' LAWYERS. THE MODEL PROPOSED ASSUMES THAT INDIVIDUALS HAVE THE SAME WEALTH AND ARE RISK NEUTRAL. EACH INDIVIDUAL FACES THE SAME PROBABILITY DISTRIBUTION OF PRIVATE GAIN FROM ENGAGING IN AN ACTIVITY WHICH IMPOSES DAMAGES ON OTHERS. ALSO, EACH INDIVIDUAL IS EQUALLY LIKELY TO BE THE VICTIM OF ANOTHER'S DAMAGES. AN INDIVIDUAL WILL ENGAGE IN THE ACTIVITY IF HIS REALIZED PRIVATE GAIN EXCEEDS THE EXPECTED FINE. FROM THE MODEL, GENERAL CONCLUSIONS ARE DRAWN REGARDING ENFORCEMENT METHODS. FIRST, REGARDLESS OF RELATIVE ENFORCEMENT COSTS, PRIVATE (COMPETITIVE OR MONOPOLISTIC) ENFORCEMENT LEADS IN A WIDE RANGE OF CIRCUMSTANCES TO LESS ENFORCEMENT THAN PUBLIC ENFORCEMENT, RATHER THAN MORE. SECOND, PUBLIC ENFORCEMENT IS SOCIALLY PREFERABLE TO PRIVATE ENFORCEMENT IN MANY CIRCUMSTANCES EVEN WHEN PUBLIC ENFORCEMENT IS MUCH COSTLIER. THIRD, DEPENDING ON RELATIVE ENFORCEMENT COSTS, MONOPOLISTIC ENFORCEMENT MAY RESULT IN MORE OR LESS ENFORCEMENT THAN COMPETITIVE ENFORCEMENT. THE FINAL CONCLUSION, APPLICABLE WHEN PRIVATE ENFORCEMENT IS CHEAPER THAN PUBLIC ENFORCEMENT, IS THAT REGULATING PRIVATE ENFORCERS BY PAYING THEM SOMETHING DIFFERENT THAN THE FINE FOR EACH VIOLATOR DETECTED CAN ACHIEVE THE SOCIALLY MOST PREFERRED OUTCOME IN THE COMPETITIVE CASE BUT GENERALLY NOT IN THE MONOPOLISTIC CASE. WAYS TO MAKE THE MODEL MORE REALISTIC WITHOUT AFFECTING THE BASIC RESULTS ARE ALSO DISCUSSED. AN EXAMPLE FULLY UTILIZING THE MODEL IS PRESENTED. FORMULAS, GRAPHS, AND FOOTNOTES ARE INCLUDED, AND PROOFS OF SOME STATEMENTS IN THE TEXT ARE APPENDED. (AUTHOR ABSTRACT MODIFIED--PRG)