Associate editor

It is a seemingly immutable law of modern Republican rhetoric that the word “regulation” can never appear unadorned by the essential adjective: “job-killing.”

As in nominee-in-waiting Mitt Romney, after winning the Illinois primary: “Day by day, job-killing regulation by job-killing regulation, bureaucrat by bureaucrat, this president is crushing the dream.”

Or House Speaker John Boehner (R-Ohio) denouncing “the president’s job-killing regulatory agenda” last month after the Environmental Protection Agency (EPA) proposed new limits on coal-fired power plants.

Or Rep. Michele Bachmann (R-Minn.), who said during her presidential campaign that the EPA should be renamed the “Job-Killing Organization of America.”

Hating regulation is an old argument, but the phrase is a relatively new trope. A Nexis search of articles from U.S. newspapers and news services shows that the words “job-killing regulations” appeared just a handful of times in 2007 — but several hundred times in 2011.

This inflated rhetoric is often accompanied by bad science — or, perhaps more precisely, inherently inexact science badly used. Opponents of a particular regulation tout inflated projections of the regulatory body count, more often than not financed by the affected industry. Ditto, by the way, for those on the other side.

For example, when the EPA last year issued rules to limit mercury and other power-plant emissions, the industry-backed American Coalition for Clean Coal Electricity estimated the regulations would trigger the loss of 1.44 million jobs.

At the same time, the Political Economy Research Institute at the University of Massachusetts Amherst concluded that the rules would instead create 1.46 million jobs through retrofitting old plants and switching to new sources of renewable energy.

The EPA itself came up with much more modest predictions — that the rules would create about 50,000 one-time jobs and another 9,000 additional jobs annually. All in the broader context of a rule that the agency estimated would deliver annual net benefits of between $166 billion and $407 billion from cleaner air, including avoiding as many as 51,000 premature deaths annually.

Lesson One: If you plug your cherry-picked assumptions into your preferred model, it’s easy to obtain the desired result. Lesson Two: Jobs are only part of the larger picture.

A new report from the Institute for Policy Integrity at the New York University School of Law attempts to bring some economic rationality to the regulatory discourse — however quixotic that might be in the current political environment, not to mention in a presidential election year.

The report is titled “The Regulatory Red Herring: The Role of Job Impact Analyses in Environmental Policy Debates.” Yet somewhat surprisingly, Michael Livermore, the institute’s executive director, does not oppose factoring job impact into the cost-benefit analysis. Rather, he argues for adopting a more sophisticated approach than the prevalent knuckleheaded assumption — my words, not his — that increased regulation inevitably results in fewer jobs.

If an employer’s costs increase as the result of a regulation, Livermore notes, that is another way of saying that the employer has to hire workers to, say, install new technology while other employers hire workers to produce the new equipment.

In a healthy economy, the cost of layoffs should be transitory, as workers quickly find new jobs. In an economy like the current one, the impact of such layoffs may be more persistent — but any new jobs created may be more significant since, in a soft labor market, otherwise unemployed workers may be hired.

Can these cross-cutting impacts be accurately measured in a dynamic economy? Perhaps more important for the current discourse, is it possible to have the jobs and regulation discussion without ignoring the inherent limitations of economic modeling?

“The jobs impact analysis is important and we should do it, but the way it’s discussed now is completely wrong,” Livermore told me.

First, he said, “we talk about the jobs impact on the one hand and the other impacts (such as health and safety improvements) on the other hand, and they’re treated as apples and oranges.” Instead, he said, “we need to integrate the jobs impact into the broader cost­benefit analysis.”

Second, Livermore said, is a failure among those doing the analyzing to disclose the assumptions and limitations of their models — and the willingness of politicians (and the media, for that matter) to treat the resulting figures as gospel rather than guesstimate.

“The real problem is the way they’re used in the political back and forth,” Livermore said. “They’re used as sledgehammers to beat up the other side.”

No surprise there. But a useful reminder at a time when the phrase job-killing has become mind-numbing.