Cost Savings from Allowance Trading in
the 1990 Clean Air Act: Estimates from a
Nathaniel O. Keohane
Title IV of the Clean Air Act of 1990 represented a major innovation in environmental policy in the United States. Before 1990, clean air regulations for electric power plants in the United States were exclusively prescriptive in nature—requiring individual units to meet strict performance standards or even install particular abatement technologies. Title IV took a much different tack: It instituted a market for sulfur dioxide (SO2) emissions. Utilities affected by the regulation were allocated a certain number of allowances, with each allowance representing one ton of SO2 emissions. Utilities that reduced their emissions could sell their excess allowances to other utilities or bank them for future use.
Economists have long asserted the virtues of such market-based environmental policies over traditional prescriptive or command-and-control regulation, such as performance or technology standards. In theory, market-based instruments are cost-effective: That is, they can achieve a given level of abatement for the least possible total cost.1 Title IV offers the first major test of how a market-based instrument can perform in practice.
Overall, this grand policy experiment (Stavins 1999) has been overwhelmingly successful to date. After a slow start, allowance trading has been vigorous. The utilities' ability to bank allowances, meanwhile, led to deeper than expected cuts in emissions during Phase I with concomitant environmental benefits.2 Surprisingly, however, a consensus has yet to emerge on the actual cost savings realized from the allowance market. Expectations of cost savings certainly ran high before the program began. Studies done in the early 1990s by the General Accounting Office and by the Environmental Protection Agency (EPA) anticipated cost savings