The fairness of sanctions: some implications for optimal enforcement policy

In this article we incorporate notions of the fairness of sanctionsinto the standard model of public enforcement. When both theprobability and magnitude of sanctions may be varied, the usualsolution involves a very high sanction and a relatively lowprobability of enforcement if individuals are risk neutral.When the issue of fairness is added to the analysis, the optimalsanction generally is not extremely high because such a sanctionwould be seen as unfair. The optimal probability of imposingsanctions may be higher than in the usual case (to offset thelower sanction) or lower than in the usual case (because thelower sanction reduces the effectiveness of enforcement).     

The conflict between notions of fairness and the Pareto principle

Most legal academics and policy makers believe that notionsof fairness should be accorded positive weight in evaluatinglegal policies. We explain, however, that ascribing importanceto any notion of fairness (other than one concerned solely withthe distribution of income) will sometimes lead to a conflictwith the Pareto principle. That is, to endorse a notion of fairnessis to endorse the view that it can be desirable to adopt a legalrule that will reduce the well-being of every person in society.     

Law versus Morality as Regulators of Conduct

It is evident that both law and morality serve to channel ourbehavior. Law accomplishes this primarily through the threatof sanctions if we disobey legal rules. Morality too involvesincentives: bad acts may result in guilt and disapprobation,and good acts may result in virtuous feelings and praise. Thesetwo very different avenues of effect on our actions are examinedin this article from an instrumental perspective. The analysisfocuses on various social costs associated with law and morality,and on their effectiveness, as determined by the magnitude andlikelihood of sanctions and by certain informational factors.After the relative character of law and of morality as meansof control of conduct is assessed, consideration is given totheir theoretically optimal domains—to where moralityalone would appear to be best to control behavior, to wheremorality and the law would likely be advantageous to employjointly, and to where solely the law would seem desirable toutilize. The observed pattern of use of morality and of lawis discussed, and it is tentatively suggested that the observedand the optimal patterns are in rough alignment with one another.     

Optimal Discretion in the Application of Rules

Discretion is examined as a feature of the design of rule-guidedsystems. That is, given that rules have to be administered bysome group of persons, called adjudicators, and given that theirgoals may be different from society's (or a relevant organization's),when is it socially desirable to allocate discretionary authorityto the adjudicators and, if so, to what extent? The answer reflectsa tradeoff between the informational advantage of discretion—thatadjudicators can act on information not included in rules—andthe disadvantage of discretion—that decisions may deviatefrom the desirable because adjudicators' objectives are differentfrom society's. The control of discretion through limitationof its scope, through decision-based payments to adjudicators,and through the appeals process, is also considered.     

On the Superiority of Corrective Taxes to Quantity Regulation

The traditional view of economists has been that correctivetaxes are superior to direct regulation of harmful externalitieswhen the state's information about control costs is incomplete.In recent years, however, many economists seem to have adopteda different view—that either corrective taxes or quantityregulation could be superior to the other. We emphasize thatone argument for this newer view, identified with Weitzman (1974),holds only if the state is constrained to use a fixed tax rate(a linear tax schedule) even when harm is nonlinear. But if—asseems more plausible—the state can impose a nonlineartax equal to the schedule of harm or can adjust the tax rateupon learning that it diverges from marginal harm, then correctivetaxes are superior to quantity regulation. Another argumentfavoring quantity regulation is that it gains appeal when thestate is uncertain about the harm caused by an externality.In this case, however, a corrective tax schedule (equal to theexpected harm schedule) is superior to quantity regulation.