Cleaner at a Cost

At a recent Energy Policy Institute of Chicago Workshop Series Event, Professor Chad Syverson of the Chicago Booth School of Business presented his paper “The Effects of Environmental Regulation on the Competitiveness of U.S. Manufacturing,” co-authored with Michael Greenstone and John List.

The paper quantifies the impact of the 1970 Clean Air Act and its amendments on total factor productivity (TFP) in the manufacturing sector, providing quantitative support to previous qualitative claims that economic productivity is affected by the tightening of environmental standards. The authors find that there has been a statistically significant decline in TFP due to Clean Air Act regulations, conservatively estimated at 2.6 percent, or $12.9 billion dollars annually.

The 1978 Clean Air Act Amendments (CAAA) established county-level standards for particular pollutants such as ozone, carbon monoxide, total suspended particulates, and sulfur dioxide. Based on a “bubble” approach, the CAAA require that the aggregate pollution within one county, or bubble, does not exceed certain limits, meaning that polluting plants may operate but must be offset by cleaner plants to lower the collective emissions. The Environmental Protection Agency (EPA) then gives counties “attainment” or “nonattainment” designations based on the collective emissions of the plants within its bubble.

Using CAAA panel data and the U.S. Census Bureau’s Annual Survey of Manufacturers (ASM), Syverson and his co-authors exploit the variation between the two designations by measuring the changes to the respective total factor productivities from each of these counties annually. The study takes advantage of the fact that the plants within a county receiving a nonattainment designation for a particular year are subject to strict abatement standards. In adapting to the new standards, a plant’s TFP (or productivity) is negatively affected.

In total, 3,141 U.S. counties were included in the study, spanning the 28-year period from 1972 to 2000.

In order to remove geographical trends in variation of TFP levels, Syverson and his colleagues compared non-emitting plants to emitting plants in the same county. Since only those plants emitting the target pollutants were subject to abatement requirements, the non-emitting plants served as controls, allowing the researchers to distinguish between changes in TFP not related to the EPA designations.

Moreover, the authors identified three sources of potential bias. First, their data was not able to capture the TFP lost by plants that exited the market due to the abatement regulations. Second, the data used revenue to proxy for output, which can lead to an undercounting of output losses when an internalization of abatement costs pushes prices up, causing gross revenues to appear greater. Finally, those plants not subject to abatement regulations were not included in the data, placing a downward bias on TFP losses.

When accounting for these biases, the estimated loss in TFP due to the CAAA increases to 4.8 percent, or $24.4 billion dollars annually.

However, Syverson and his authors do not assert that the CAAA has no benefits. In fact, as Syverson points out, there are multiple studies that show that the monetary benefits of the CAAA outweigh the costs.