It’s the end of an era. The GOP, in the person of Donald Trump, has fallen out of love with subjecting regulations to cost-benefit analysis.
Our story begins in December 1980, when President Jimmy Carter signed (over the objection of four Cabinet agencies) a law creating the Office of Information and Regulatory Affairs, or OIRA. The purpose was to establish within the White House budget office a clearinghouse to, in Carter’s words, “regulate the regulators” by reviewing and sometimes jettisoning paperwork requirements that regulatory agencies imposed on the public.
Two months later, Carter’s successor, Ronald Reagan, greatly expanded OIRA’s mandate by giving it review power over the regulations themselves. All major new regulations (defined as any that imposed a cost to industry of $100 million or more) would henceforth be submitted to OIRA for cost-benefit review. If the benefits didn’t outweigh the costs, the regulation would be modified or scrapped.
Liberals complained that cost-benefit analysis biased OIRA against protecting health and safety because a regulation’s benefit to society, being widespread, was harder to quantify than its narrow cost to specific businesses. In addition, because calculations of regulatory cost relied on industry input, these were highly exaggerated.
For Republicans, though, regulatory cost-benefit analysis became a kind of religion. So confident were they that it was a reliable regulation killer that in the 1994 midterm elections they pledged, as part of Newt Gingrich’s Contract With America, to make it a statutory requirement. After Republicans retook the House of Representatives in 1995, the House passed a bill that fulfilled that pledge—and rigged cost-benefit methodology even further in industry’s direction. The legislation died in the Senate, in large part because President Bill Clinton had already signed in 1993 an executive order that reaffirmed Reagan’s cost-benefit requirement for major rules. By then cost-benefit analysis was, among social science wonks across the ideological spectrum, too respectable to discard. Whatever its flaws, cost-benefit analysis was rational, it was market-friendly (these were the “Washington consensus” years), and, with a little effort, its pro-business bias could be reduced.
Lo and behold, that started to happen during the presidency of Barack Obama. Obama’s OIRA administrator, Cass Sunstein, broadened cost-benefit’s scope to include, for instance, the calculation of benefits to people living outside the United States, and to assign a (human) cost to the release into the atmosphere of carbon ($21.40 per metric ton). You can guess what happened next. “As economists got better at measuring the benefits of regulation,” Stuart Shapiro, a onetime OIRA analyst and now professor of public policy at Rutgers, observes in The Regulatory Review, “benefit-cost analysis began to be seen as a tool that supported more stringent regulation of the economy.”
How committed, really, are Republicans to cost-benefit analysis? In 2019, Berkeley law professor Daniel Farber, author of the classic 1986 New Republic essay “The Case Against Brilliance,” found a way to measure this, and if it wasn’t brilliant, it was certainly shrewd. In 2017, a Republican Congress had used expedited procedures under the Congressional Review Act, or CRA, to eliminate 14 Obama-era regulations. To what extent, Farber asked, did Congress rely on cost-benefit analysis in selecting regulations to kill off?
The pool of regulations eligible to block was limited to those passed at the tail end of Obama’s presidency, so Farber looked only at those. What he found was that the regulations Congress targeted through proposed CRA resolutions corresponded not at all with OIRA cost-benefit analyses. Indeed, fully one-third of the regulations that members of Congress sought to eliminate had no OIRA cost-benefit analysis at all, because their economic impact was too slight. Among major rules eligible for elimination, most never even got considered, even though they came with cost-benefit analyses. And among those regulations that were considered, minor ones were repealed at twice the rate of major ones.
If not cost-benefit analysis, what determined which rules Congress chose to eliminate? You can probably guess. The deciding factor was whether a given regulation addressed a hot-button issue such as guns or contraception, or whether it affected some key Republican constituency such as the oil business. Even as the Supreme Court was preparing to push regulatory decision-making onto Congress, Congress was demonstrating that its approach followed no logic except crude political gain.
During his first term, Trump imposed a limit on the regulatory cost of new regulations. This appalled even President George W. Bush’s notoriously pro-business OIRA administrator John Graham, who was a fervent cost-benefit advocate. Prior to Trump, observed a highly critical 2020 article in the journal Regulation that Graham co-authored, “there was no annual cap on the additional cost burdens an agency can impose.” Trump created a cap and set it, of course, at zero. “Thus has cost-benefit analysis,” wrote Georgetown law professor Lisa Heinzerling, “mutated, by executive directive, into cost-nothing analysis.” During his second term, Trump issued an executive order setting the cost of new regulations (with apologies due Elvis Costello) at “significantly less than zero.”
In 2023, President Joe Biden’s OIRA administrator, Richard Revasz, updated the agency’s analytic process for the first time in 20 years. When calculating a proposed rule’s future benefit, Revasz explained last February in The New York Times, OIRA applies a discount rate fixed to the long-term, inflation-adjusted return on government debt. The higher the discount rate, the lower the calculated benefit. Based on bond yields over the previous 30 years, Revasz set OIRA’s discount rate at 2 percent, leaving room for future adjustments based on new data.
But when the Trump administration came back in, it raised the discount rate from 2 percent to as high as 7 percent, based on some vague notion that societal well-being will be higher in 30 years. Will it be? That’s an interesting but unanswerable question. On one hand, Trump won’t be around, and that’s a definite plus. On the other hand, I think everybody over 35 can agree that societal well-being in December 1995, despite its many deficits, was a damned sight better than societal well-being in December 2025.
Trump also jettisoned from cost-benefit review the social cost of carbon (and also methane, and other greenhouse gases) because it was “marked by logical deficiencies, a poor basis in empirical science, politicization, and the absence of a foundation in legislation.” Global warming is a hoax!
In October, Jeffrey Bossert Clark, acting OIRA administrator, issued a memo leveling yet another blow at cost-benefit analysis. Instead of taking 90 days to evaluate some proposed deregulatory action, OIRA will now take 28 days—and 14 if a given rule is “factually unlawful.” That’s not even close to enough time. “If high-quality analysis suggests that deregulation is costly, and it often will,” concludes Shapiro, “the [Clark] memo effectively instructs OIRA to turn a blind eye.”
(If I may be permitted a digression: Can something be “unfactually unlawful”? Under Trump, I suppose the answer is yes. Such a doctrine would perhaps justify Trump’s serial unsuccessful attempts to indict Letitia James and James Comey.)
Increasingly, it’s left to others to calculate the true cost of Trump’s policies. For instance, Trump is promising a $2,000 rebate on his tariffs for low- and middle-income Americans. But according to the nonprofit (and very centrist) Tax Foundation, even if Trump were to distribute every penny of tariff revenue, the result would be a rebate well below $2,000—and that’s assuming a means test with a hard cutoff at $100,000. Partly that’s because total tariff revenue must be discounted by about 25 percent to take into account tariffs’ shrinkage of the tax base. “A better way to provide relief from the burden of tariffs,” the Tax Foundation concluded, “would be to eliminate the tariffs.”
The decline and fall of cost-benefit analysis is in step with the decline of other conservative values under Trump: fiscal conservatism, a united West, free trade, tight money at the Fed, and of course that old standby, dignified leadership at the top. It’s another reminder that Trumpism has almost nothing to do with what Republicans purport to stand for (excepting tax cuts for the rich and deregulation, and even Mitt Romney is having second thoughts about the former). Liz Cheney was right: When the Trump nightmare is over conservatives will need to start a new party because the old one will be too compromised by Trump. It will have no more legitimacy than Vichy after the Allied liberation of Paris. Instead of trying to revive neoconservatism, David Brooks should try to revive the Whigs. Maybe Cheney, who disappeared from view after departing Congress, will come out of hiding to help.
