Working Paper: NBER ID: w3008
Authors: Louis Kaplow
Abstract: Even when society would wish to deter all acts of some type, such as tax evasion and many common crimes, the benefits from deterrence often will be insufficient to justify the expenditures on enforcement that would be required to deter everyone. If some individuals are not deterred, however, they will bear risk when fines are employed as a sanction. As a result, it may be optimal to reduce total risk-bearing costs by reducing the number of individuals who bear any risk. This can be accomplished by increasing enforcement above the level that would be justified considering only the benefits of deterrence and the direct costs of enforcement. Another possibility is that it may be optimal reduce the risk borne by those who act, by employing fines below the maximum feasible level. This latter result constitutes an instance in which the well-known implication of Becker's analysis that it is optimal to employ extreme sanctions for all offenses is invalid.
Keywords: fines; deterrence; risk aversion; enforcement; optimal policy
JEL Codes: K42
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Risk aversion of individuals (D11) | Optimal fine levels (H21) |
| Optimal fine levels (H21) | Risk-bearing costs (G22) |
| Enforcement policies (K42) | Number of individuals bearing risk (G52) |
| Total harm caused by acts (H84) | Enforcement policies (K42) |
| Risk-bearing costs (G22) | Welfare (I38) |