New York University Public Law and Legal Theory Working Papers

Document Type



Northwestern University Law Review, Vol. 105, p. 1581, 2011


Whenever legislators and regulators adopt a regulatory change, they face an important question: how should existing actors be required to respond to the new law? For example, if regulators drastically reduce the level of emissions that new plants are permitted to discharge, they might grandfather existing plants, gradually phasing in the new regulation and granting existing polluters a certain period of time—perhaps even a long or undefined period—to incorporate the mandated changes. Alternatively, regulators could insist that existing plants immediately comply with this new regulation, even if some existing plants would have to close their doors. While the dominant view in the academic literature has been that transition relief is undesirable, this view has recently come under attack. Steven Shavell has argued that when actors have made significant investments in order to comply with an existing law, it might be socially optimal to grandfather those actors, as opposed to requiring them to comply with the new law. In this Article, we show that the dominant view has not paid sufficient attention to the issue on which Shavell focused: that the costs of retrofitting existing plants to come to compliance with the new standard can be a great deal higher than that compliance costs for a new plant. But we also show that Shavell has paid insufficient attention to the issue that concerned the dominant view: the desirability of having existing plants anticipate future changes in the legal standard. We therefore craft a distinctive approach to evaluating the desirability of grandfathering. This Article also disputes the argument that has been advanced by prominent scholars that transition relief is desirable because it can reduce wasteful lobbying expenses. They overlook that transition relief results in additional, and significant, lobbying by actors seeking to extend such relief beyond its expiration date. Finally, we argue that the approach to setting regulatory standards and transition rules that is endorsed in the academic literature is flawed because it assumes that regulators should first pick the standard that is optimal for new plants and then choose the best transition rule in light of that standard. Since transition relief often impedes new actors from entering the regulated activity, however, there may be very few new actors to actually meet the more stringent standards. Thus, it may be preferable to adopt a less stringent standard for new actors and a less generous transition policy for existing actors. Both the new rule and the transition rule need to be optimized simultaneously. First optimizing the new rule and then picking the best transition rule in light of the new rule—the universal approach of the academic literature—leads to undesirable results.

Date of Authorship for this Version