K. Sabeel Rahman
In October 1984, two-year-old Joy Griffith of Miami poked her head between the seat and the footrest of her grandfather’s brown recliner. The footrest folded in on her, restricting her breathing. Minutes later, when Griffith was discovered, according to contemporary news reports, she had suffered irreversible brain damage.
The case drew national attention. Eight other children had died in similar incidents in recliner chairs, a study found, and manufacturers offered to modify the design. “We recognized it as a responsibility of the industry,” the vice president of the American Furniture Manufacturers Association said, according to The Washington Post. Costs, he said, “really didn’t matter.”
But the Consumer Product Safety Commission (CPSC), the federal agency regulating hazardous retail goods, disagreed.
“It is our recommendation that nothing be done beyond mentioning [the problem] in safety alerts,” Warren Prunella, assistant to the director of the CPSC’s economic analysis staff, wrote in an internal memo. Even then, they should be wary of going too far, Prunella cautioned, since “the psychic costs associated with the anxiety that accompanies the release of information on household hazards is to be considered against any accompanying benefits.”
Consumer advocates were outraged, and even manufacturers were taken aback. They went ahead with their planned modification, and against the CPSC’s recommendation, voluntarily installed a device between the leg-rest and seat to prevent similar injuries.
To critics, Griffith’s case became emblematic of a craze for cost-benefit analysis (CBA), used not to help government deliver better services and regulations, but to block almost any state intrusion into markets.
President Jimmy Carter had placed a greater emphasis on cost-effectiveness since 1979, when, facing stubborn inflation, he declared war on government programs that couldn’t prove their worth. In response, Al Gore, then a Democratic congressman from Tennessee, called cost effectiveness an “attempt to cripple” government functions. “I remain to be convinced that any kind of ethereal academic analysis can adequately value the benefits of lives saved by regulations,” Gore said.
But the next decades saw a bipartisan embrace of CBA, first under Ronald Reagan, and then in successive Democratic administrations. Cass Sunstein, administrator of Obama’s Office of Information and Regulatory Affairs (OIRA), made his career as the sober face of blocking regulations, speaking up for their often invisible costs.
Still today, many proponents of CBA are happy to agree with Gore that it is intended as a check on government action—a way to not just improve regulation, but to keep it from occurring in the first place. It presumes that government action introduces risk and uncertainty, and the burden should be on the meddling regulator to prove his usefulness.
“OMB is to the executive branch what the Senate is to the legislative—it is a forced cooling function on the passions of the moment to help ensure we all get more thoughtful policy, policy that fully considers both the benefits and costs of changing the status quo,” DJ Gribbin, former infrastructure adviser to President Trump, told the Prospect. “Agencies are advocates and OMB’s job is to say ‘no.’”
THE BIDEN ADMINISTRATION HAS BROKEN with that consensus, pursuing a series of changes to government budgeting, purchasing, and other agency decision-making processes that attempt to reclaim cost-benefit analysis from economists broadly skeptical of state action.
The latest is the uncharismatically named Circular A-4, the guidance document of the Office of Management and Budget (OMB) on CBA for regulations. Originally issued in 2003, the circular was revised for the first time this month.
This is the document that tells federal agencies how to calculate regulatory costs. Though its impact is indirect, it is consulted widely, not only by federal agencies but by economists across the government, who determine what kinds of factors to use in deriving the value of a project.
“The new [circular] is much more accepting of the idea that the government can play a useful, potentially wide-ranging role,” Noah Kaufman, an economist who worked on the new circular while at the Biden Council of Economic Advisers, told the Prospect. In the updated version, he said, “the presumption seems to have shifted from ‘Do not regulate unless you must’ to ‘How can regulations improve society?’”
“The new [circular] is much more accepting of the idea that the government can play a useful, potentially wide-ranging role.”
Biden’s April executive order on modernizing regulatory review, which called for revising the circular, redefines a “significant regulatory action” as one expected to have a $200 million annual economic impact, up from $100 million.
The updated circular expands the geographic scope of analysis to include international effects of a regulation, including the effects on noncitizens living abroad. It also provides more guidance on distributional weighting, which adjusts for the fact that a dollar spent in a poor neighborhood is worth more than one spent in a rich neighborhood.
Distributional weighting is not new, but actually applying it as a “non-trivial component” of regulation has proved, according to OIRA Administrator Richard Revesz shortly before his appointment, “a failure.” Proponents say that the renewed emphasis on distributional weighting is not an ideological case for prioritizing the poor—it just corrects for the declining marginal utility of income for the well-off. MIT economist David Autor, in a joint comment on the regulations, endorsed the guidance, writing that “most economists support the view that a $1 transfer to the poor generates more social surplus than a $1 transfer to the rich.”
PERHAPS MOST SIGNIFICANTLY, the circular updates the discount rate for federal regulation. This is how the government takes market information—specifically, inflation-adjusted Treasury yields—and gives it implicit moral weight, by using it to decide how much value to assign the future costs and benefits of regulation and investment.
The discount rate is a lot like the private sector “hurdle rate,” which determines the minimum expected returns for an investment to be worth pursuing. Just as a mortgage becomes more affordable when interest rates fall from 7 percent to 2 percent, a regulation—cutting down on microplastics, redesigning food packaging—looks more worthwhile when the discount rate falls. To have a high discount rate, then, is quite literally to discount the future.
Economists and business typically argue for a high discount rate. The future is far off, and betting on it is uncertain. And anyway, they argue, future people will be richer and more able to bear the costs of regulation.
Non-economists have tended to argue that the government should devote more resources to public goods with long-term benefits, like education, infrastructure, preventing pollution, and investing in a greener economy. (Exceptions include Tyler Cowen of George Mason University, who argues for a lower discount rate here.)
Discount rates have lately broken into popular discourse through the work of William Nordhaus, who won the 2018 Nobel Prize in economics. Nordhaus argues that since the energy transition is expensive, we should assign it a higher discount rate, embarking on costlier decarbonization efforts in a richer future. That gradualism has been critiqued on multiple grounds, including by economists who point out that the energy transition can itself be a motor of economic growth, and by those who point out that uncertainties, tipping points, and exponential costs of future climate change call for all-out crisis-fighting today.
Whatever the merits of these critiques, it’s not easy to drill down on a society’s desire to invest in future generations, often at the expense of present ones, and assign it a single rate.
The 2003 circular accomplished this by using two discount rates: a 3 percent “social cost of time preference” to describe risk-free investment in U.S. Treasuries, and an “opportunity cost of capital” rate of 7 percent. The updated circular discards the 7 percent metric, and sticks with the same methodology for deriving the social cost of time preference, using the 30-year average of the rate of return on ten-year Treasuries, minus inflation. Updated with current data, this spits out a discount rate of 2 percent.
The drop from a 3 to 2 percent discount rate, said K. Sabeel Rahman, a Cornell law professor who until recently ran OIRA as associate administrator, “basically suggests that we have been vastly undercounting future benefits and costs, to the tune of hundreds of billions of dollars.”
A purely technical update conveniently made future costs and benefits look more important at a time when rising interest rates are pointing in the opposite direction.
A Wall Street Journal editorial suggested that the lower discount rate was a political move, adopted to “make hoped-for benefits that might occur years in the future weigh more heavily against the present regulatory costs.”
But the OMB did not have to tilt the scales to produce a lower discount rate. The past 40 years of low interest rates meant that when it updated the data, the high-interest rates of the 1980s fell out of the frame, and the same method coughed out a lower discount rate.
A purely technical update, then, conveniently made future costs and benefits look more important at a time when rising interest rates are pointing in the opposite direction, and when the Biden administration has ambitions to rapidly build out infrastructure for the energy transition.
That happy coincidence was not lost on business groups opposed to a low discount rate. A coalition including the U.S. Chamber of Commerce and several oil and gas groups wrote during the public comment period that the sample should not be “skewed by the historical anomaly of hyper-low interest rates.”
Do Americans’ investment choices offer meaningful information about how much they value future generations? The “social cost of time preference” is derived from market data, using revealed preference to infer the discount rate from Treasury yields.
It’s a convenient rule of thumb, but also quite a profound epistemological claim, implying that a democracy can derive thick notions of how much it collectively values the future from the behavior of market actors who are not being asked to think explicitly about that question.
Some have challenged this “descriptive” approach. Kaufman has written that instead of retreating to empirical analysis for long-term decisions that involve ethical questions, it would be better to “confront them head on.”
But in a document explaining its decision-making, OMB wrote that it “continues to believe that it is generally important for agencies to produce analysis that enables policymakers ‘to base resource allocation on the tradeoffs that society actually makes.’”
THERE ARE BROADLY THREE REASONS progressives have been skeptical of cost-benefit analysis. First, there is a kind of Victorian nose-holding sentiment that putting a price on items like human life is in poor taste.
Second, there is a concern with uncertainty. Setting market values for the costs and benefits of complex assets can descend into nonsense territory. Even the near-term hazards of climate change, for example, come with huge error bars. Prunella, the CPSC champion of cost-effectiveness, strained credulity when he proposed to factor the “psychic costs” of warning labels to consumers into his assessment of a regulation’s effect.
Third, there is a more acquired skepticism toward CBA, and toward the way economic analysis writ large has been used in practice. Over the past 40 years, progressives argue, economic models have not been neutral, but have been used to thwart progressive agendas. That frustration has led some to suggest that in order to pass good policy, CBA should be sidestepped altogether.
The fight over CBA—how much to use it, whether it is neutral—has a long history.
Thorne Auchter, a Florida building contractor who became Reagan’s first director of the Occupational Safety and Health Administration (OSHA), vigorously fought a Supreme Court case upholding occupational safety and health laws protecting textile workers from brown lung, which had been challenged on the basis that they did not pass CBA.
Congress had never required OSHA to weigh costs and benefits, Justice William Brennan wrote in the decision. “Congress was fully aware that the Act would impose real and substantial costs of compliance on industry, and believed that such costs were part of the cost of doing business.”
Undeterred, Auchter attempted to revise the rule to achieve benefits at lower cost. His broader efforts to establish the agency’s neutrality must have seemed, to non-initiates in economic reasoning, extraordinarily tendentious.
Auchter drew outrage, for example, when he ordered the destruction of 100,000 brochures about cotton dust, a cause of brown lung disease. Auchter argued that the brochure’s cover, which depicted Louis Harrell, a North Carolina textile worker killed by brown lung disease, was “offensive” and “obviously favorable” to labor.
Progressives point to this sort of skewed reasoning as evidence that economic tools, and institutions charged with CBA, have long been deployed, asymmetrically, to constrain their agendas. To proponents of CBA, this sounds like an admission of defeat: Feeling that they are at a disadvantage in winning the political debate that their favored causes are beneficial, it is said that progressives are trying to build those preferences into the regulatory process itself.
But for better or worse, the updated circular does nothing so dramatic as taking moral issue with CBA, or downgrading its importance. It’s a document written by economists that updates the procedure for evaluating regulations with new data, and more recent consensus in the academic literature—which has departed from an earlier, hard-line use of economic reasoning.
“A lot of times what ends up happening is, in public discourse, we will take a monetary value where we have one, because it’s easy to talk about, and then where we don’t have a monetary value, we’ll overlook that issue,” Rahman said. “The point of the new A-4 is, quantify and monetize where you can, but you shouldn’t ignore impacts just because you don’t have a number. You still have to analyze it rigorously.”
