HARRINGTON, Winston

On the accuracy of regulatory cost estimates - 2000

This study compares ex ante estimates of the direct cost of individual regulations to ex post assesments of the same regulations. For total costs the results support convetional wisdom, namely that the costs of regulations tend to be overestimated. This is true for 14 of the 28 rules in the data set discussed, while for only 3 rules were the ex ante estimates too low. For unit costs, however, the story is quite different. At least for EPA and OSHA rules,unit cost estimates are often accurate, and even when they are not, overestimation of abatement costs occurs as often as underestimation. In contrast, for those rules that use economic incentives, unit costs are consistently overestimated. The difference between the total-cost and the unit-cost results is caused by frequent eros in estimates of the effects of individual rules, which suggests, in turn, that the rule`s benefits may also be overestimated. The quantity erros are driven both by difficulties in determining the baseline and by incomplete compliance. In cases of unit-cost overestimation, unanticipated technological innovation appears to be an important factor - especially for economic incentive rules, although procedural and methodological explanations may also apply