It’s critically important to have high-quality cost-benefit assessments for proposed regulations, especially at the Environmental Protection Agency, which issues some of the federal government’s most expensive rules. But a new report from the Government Accountability Office finds that the EPA has employed some suspicious math in assessing several recent regulations aimed at improving air and water quality.

For example, the EPA claims it examined the potential effects on employment as it drafted regulations for industrial boilers, commercial incinerators, and waste-water dischargers — but, the GAO found, the EPA was reaching these estimates by using 20-year-old labor-market research that examined only four industrial sectors.

“Without improvements in its estimates,” the GAO reported, “EPA’s [regulatory-impact analyses] may be limited in their usefulness for helping decision makers and the public understand these important effects.”

That’s a tactful way of saying the EPA’s assessments were completely outdated and practically useless. But don’t expect anything to change soon: “EPA officials said they are exploring new approaches for analyzing these effects but were uncertain about when such results would be available,” the GAO report noted.

Keep in mind, too, that once these rules are passed, they’re nearly impossible to roll back, a problem that has plagued the United States for decades. As early as 1999, the GAO cautioned that of 101 “economically significant regulations issued by the EPA from 1981 through 1998, only five were the subject of retrospective studies.” So if the EPA’s initial cost-benefit calculations were off the mark, it was unlikely that anyone would ever find out, much less reverse the regulations.

The new GAO report, issued Monday, focuses on only a few shortcomings in how the EPA determines whether the benefits of regulation outweigh the costs. But flaws abound.

As I wrote earlier this week, recent proposed regulations, which would require power plants to cut carbon emissions by 30 percent or more between 2005 and 2030, have the potential to greatly harm industries concentrated in certain states, as well as the national economy.

Given that risk, the EPA should be able to solidly explain why such restrictive regulations are worthwhile. But as the Cato Institute recently noted, the agency forgot to include one very important calculation in the information they released about the proposed rules: whether or not they will actually affect climate change. “There’s really no reason to go after carbon emissions unless you think they cause climate change,” Chip Knappenberger, assistant director for Cato’s Center for the Study of Science, tells me. The impact on climate change is key. But the EPA hasn’t publicized any finding on that supposed link.

Knappenberger and his colleague Patrick J. Michaels crunched the numbers using an EPA-developed climate-model emulator. They found that the regulations would somewhat affect the climate — by eighteen-thousandths of a degree Celsius by 2100.

“We’re not even sure how to put such a small number into practical terms, because, basically, the number is so small as to be undetectable,” Knappenberg and Michaels wrote when they released their findings. “Which, no doubt, is why it’s not included in the EPA Fact Sheets. It is not too small, however, that it shouldn’t play a huge role in every and all discussions of the new regulations.”

Omitting that statistic is purposefully misleading. The EPA must know about that minuscule number — after all, it’s part of the agency’s Social Cost of Carbon calculation, which itself involves some dubious math: It attempts to assess the per-ton dollar cost of emissions based on the impact of hypothetical future events, such as hurricanes and flooding, supposedly caused by climate change. That highly subjective calculation is then used to justify the purported savings for reducing emissions.

The EPA’s questionable math adds up, but not in the way the Agency thinks. By the Office of Management and Budget’s estimates, the EPA alone accounted for between 43 and 53 percent of the total costs of major federal regulations between 2001 and 2011, carrying a price tag in the billions. During Lisa Jackson’s tenure alone, the agency issued 20 major regulations, each carrying a cost of at least $100 million, which together imposed a total initial cost of $7 billion on the American economy as industries struggled to begin meeting the EPA’s standards. The EPA owes it to taxpayers to conduct more honest, accurate assessments before issuing new regulations. For an agency that purports to be motivated by science and data, such misleading and sloppy work is inexcusable.

— Jillian Kay Melchior writes for National Review as a Thomas L. Rhodes Fellow for the Franklin Center. She is also a senior fellow at the Independent Women’s Forum.