It seems to be easy to make arguments one way or another about the effect of regulation on jobs. What does the evidence say? Those seeking an answer would do well to look at a recent book on the subject by Coglianese, Finkel, and Carrigan. Although the book is broader in scope, it provides a careful survey of the empirical literature. I’m going to forgo the opportunity to editorialize and just report what they found.
A leading study by Michael Greenstone found that areas classified as non-attainment by EPA had lower employment than other areas, amounting to about 3.4% of manufacturing jobs. The study design didn’t allow conclusions about whether existing jobs were lost or whether job growth just switched to the other states. A later study by Walker found that overall employment in the most regulated sectors fell by 15% relative to areas with less regulation. Two earlier studies had found no statistically significant effects.
The literature review concludes, based on these and other studies:
These major studies indicate that the relationship between regulations and jobs is far less pronounced than typically portrayed in political debate. The research has generated at most only tepid or mixed support for the proposition that regulation kills jobs. Although the results vary between positive and negative, statistically significant and insignificant, the studies do fairly consistently demonstrate that any effects of regulation are at most modest relative to the overall size of the labor market. [pp.8-9]
Another chapter in the book suggests that the impact on employment may vary between industries, with higher impacts in concentrated industries and lower impact on industries with growing demand. [p. 66] The book also includes a study the focuses specifically on power-sector regulations. The study found that a carbon price of $15/ton would cause on overall reduction of 0.1% in manufacturing, with losses of 1% in the most energy-intensive industries. [pp. 82-84]
On a percentage basis, these are very small fractions of the labor market. That doesn’t mean that the job losses are insignificant to the people who suffer them. I remain sympathetic to efforts to better take those costs into account, although it’s not clear to me whether that should be done in the regulatory process or through making better provision for the unemployed generally.