The term "job-killing regulations" has been used so often that in some circles it seems to have become a single word. Many Republicans believe that regulations are job-killing by definition.
Their view appears to fit with common sense. Suppose the Environmental Protection Agency imposes significant new burdens on businesses. The cost of production will increase, eventually raising prices for consumers and thus decreasing demand.
As a result, employment should decline. Businesses will let workers go, or at least hesitate before hiring new ones.
Many Democrats reject this view. Some of them believe that regulations increase employment. For example, regulations often require companies to hire people to build and install new equipment.
This view also appears to fit with common sense. To the extent that regulations mandate new hiring, they will increase employment, at least in the short run.
In light of these competing possibilities, it isn't possible to predict, in the abstract, whether regulations will kill jobs or create jobs. That's an empirical question. The latest evidence, outlined in an important new book, "Does Regulation Kill Jobs?" suggests that the polar positions are wrong.
Some research finds that regulations have a negative effect on employment -- but that effect is pretty modest. Two of the book's editors, Cary Coglianese and Christopher Carrigan, explain that "the empirical work suggests that regulation plays relatively little role in affecting the aggregate number of jobs in the United States."
An influential study, conducted by the Resources for the Future economist Richard Morgenstern and his colleagues, explored the effects of environmental regulations on several industries. On average, they find that an additional $1 million in regulation-induced spending produced a net decrease of just 1.5 jobs.
In some industries, such as petroleum and plastics, they find that regulation actually had a (small) positive impact on employment. Compliance with environmental regulation required new hiring, and consumer demand wasn't much decreased by the new costs.
By contrast, Massachusetts Institute of Technology economist Michael Greenstone finds that in its first 15 years, the Clean Air Act produced a loss of 590,000 jobs in heavily regulated industries. That is a pretty big figure.
But as Greenstone emphasizes, the law didn't have a large impact on total employment in those industries. Nor did his study explore whether the Clean Air Act created jobs as well, or whether those who lost their jobs found other positions. For these reasons, it doesn't provide a complete picture of the employment effects of the law.
Some research suggests that if we focus on how regulations affect overall productivity, or on their impact on foreign direct investment, we will find some real warning signs about the potentially harmful effects of expensive requirements.
But to date, there is little direct evidence that regulations have produced significant job losses in the United States (whether the Affordable Care Act will provide such evidence is, of course, hotly disputed, and remains to be seen).
But it would be a big mistake to conclude that public officials should ignore the risk that regulations will cost jobs. In a period of high and sustained unemployment, they should be giving a lot of attention to that risk.
A reasonable approach, followed in recent years by the Obama administration (in which I served), is for agencies to accompany their required cost-benefit analysis with a separate treatment of employment effects, good or bad.
If a regulation would eliminate a lot of jobs, regulators should at least reconsider it, or explore how to rewrite it so that it doesn't have that consequence.
A more ambitious approach would try to incorporate job losses directly into cost-benefit analysis. In an important essay, economist W. Reed Walker of the University of California at Berkeley explores what happens to people who lose their jobs as a result of environmental regulation.
He finds that workers in regulated sectors lose a lot, largely as a consequence of sustained unemployment or lower earnings in their subsequent job. The wage loss is about 20 percent, on average, of what workers would have otherwise earned.
Extrapolating from evidence of this kind, University of Chicago law professors Jonathan Masur and Eric Posner contend that $100,000 is a reasonable monetary estimate for the cost of each job loss. Use of that figure could tip the cost-benefit analysis against some regulations.
A lot more remains to be learned, but two things are clear. First, there isn't much direct evidence to date that regulation has caused significant job losses in the U.S. Second, some regulations do cost jobs, and they create real and sometimes long-term human hardship as a result.
In deciding whether and how to proceed, regulators should take account of that hardship -- and try to minimize it.
Cass R. Sunstein, the Robert Walmsley University professor at Harvard Law School, is a Bloomberg View columnist. He is a former administrator of the White House Office of Information and Regulatory Affairs, the co-author of "Nudge" and author of "Conspiracy Theories and Other Dangerous Ideas," forthcoming later this month. Follow him on Twitter CassSunstein.