Fiddling while rome burns: 64 dead, 741 sick, and Cass Sunstein’s dangerous love affair with cost-benefit analysis

Former (de)regulatory czar Cass Sunstein is back, full of advice on how to run the government from his perch as a Harvard law professor.  In a “View” column for Bloomberg News entitled “Left and Right Are Both Wrong About Regulation,” Sunstein urges his former allies and enemies to redouble their efforts to “look back” at old rules. He claims that forcing agencies to rummage through their closets in search of bad rules has already saved “billions of dollars,” although the only tangible example he offers is the recent Federal Aviation Administration (FAA) decision to allow people to use electronics on airplanes—popular, to be sure, but probably not such a plus for the economy.  Sunstein is deaf to any perspective on the regulatory state other than his deeply held prejudice that it is over-regulating and must be choke-chained through the zealous application of cost-benefit analysis.  As he did when he held the tight-collared short leashes of the regulatory agencies from his corner office at the White House, he ignores the many recent public health crises that tougher rules would have prevented.

Consider, for example, the 2012 meningitis outbreak that sickened 741 and killed 64 people in 20 states.  In the early fall of that year, people began to die from virulent infections after receiving spinal injections of methylprednisolone, a steroid drug used to relieve back and shoulder pain.  Suspicious doctors discovered that the injections originated at the New England Compounding Center (NECC) in Framingham, Massachusetts.  The company, which had been in trouble with federal and state regulators repeatedly for more than a decade, is sadly representative of serious problems within this industry.  

When federal and state inspectors inspected 31 “high-risk” compounders in 18 states last April, 28 got Form 483’s—FDA-speak for bad conduct reports.  All were engaged in abuses from mixing drugs in “clean rooms” contaminated by mold to getting the composition of medications wrong.  Yet compounders are regulated by state pharmacy boards that are ineffective.   Companies are not required to register with the federal government, and the Food and Drug Administration’s (FDA) authority to prevent them from selling adulterated drugs is hamstrung by recent court decisions.  As she withstood blistering condemnation of House Republicans, FDA Commissioner Margaret Hamburg begged her congressional overseers to give her agency the tools needed to police the industry.

Compounders provide 40 percent of intravenous medications used in hospitals, up from 16 percent just a decade ago.  If NECC is not a rogue company but rather a typical example of how fast and loose practices allowed this industry to grow by leaps and bounds, we’re in big trouble.  Yet, when Congress reared up on its tiny hind legs to address this crisis, it passed a shamefully weak law that would let compounders choose whether they wanted to register with the FDA and be regulated, or whether they preferred to do business as usual.  The magical thinking behind this approach is that market forces will compel reputable companies to register. 

Sunstein never says a word about such episodes.  Instead, he urges the President and Congress to assign agencies like the FDA to double-down on elaborate calculations that purportedly measure whether the benefits of protecting public health might be outweighed by the costs of imposing such requirements on industry.  He admits that these analyses are often wrong, citing with approval a 2006 study by economist Winston Harrington that agencies make mistakes overestimating benefits a third of the time and underestimating them as often.  Sunstein dismisses that damning indictment, writing, “existing studies cover only a very small fraction of existing regulations. We need to know much more.”  Persevere, writes the former administrator of the White House Office of Information and Regulatory Affairs (OIRA), and more number crunching will “enlist evidence to cut through the competing dogmas.” 

We’re perplexed why Sunstein believes that more mistakes will lead to fewer mistakes.   Taken at face value, Harrington’s results suggest that cost-benefit analysis is not much better than pure chance.  The large amounts of agencies’ limited resources diverted toward producing these cost-benefit analyses—no one knows quite how much because, ironically, the enterprise has never been subject to a cost-benefit analysis—will further cripple their effort to do their jobs.  These elaborate calculations add years to the process for issuing rules.  Delay costs lives at the same time that it saves industry money.  If we are really going to spend more time giving economists free rein to conduct these analyses, they should have a better predictive value than mere chance.

In his earlier, pre-government life as an eclectic academic, Sunstein was more realistic about the shortcomings of cost-benefit analysis.  In a 2002 law review article, The Arithmetic of Arsenic, Sunstein presented a veritable how-to manual for manipulating both sides of the cost-benefit equation in order to produce nearly any result an interested party wishes.  He wrote: “We can see how creative citizens and lawyers, representing water systems or environmentalists, might be able to mount reasonable challenges to EPA’s decision, regardless (almost) of the content of those decisions.”  Of course, Sunstein does not come right out and endorse misuse of cost-benefit analysis.  But his cynicism then is not consistent with the argument he makes now that the methodology is the only and best means for ensuring unbiased, rational regulatory decision-making.   

Abandoning rigid cost-benefit analysis, where the monetary value of every cost and every benefit are calculated down to the last misleading dollar, doesn’t mean bringing the state of nature back to Jamestown levels or ignoring the burden on regulated industries of new regulatory controls.  Rather, being more realistic about the numbers we can crunch and the numbers that are—at best—guesswork would liberate agencies to pursue the missions Congress assigned them decades ago.  The gist of the problem with cost-benefit analysis is that it gives short shrift to public health, worker and consumer safety, and the environment by shoving off the table benefits that cannot be quantified in any reasonable way and should not be considered in an ethical society. 

So, for example, in their number-crunching frenzy, Sunstein and his fellow cost-benefit enthusiasts value the loss of an IQ point because a child is exposed to lead paint or lead in drinking water at somewhere between $1,500-$8,400.  This number is then discounted at the rate of seven percent annually to reflect that although children might be poisoned today, the damage won’t affect their earning power until they reach the age of majority.  Never mind the affliction of going through all of one’s life with a brain diminished from what it should be.

As it happens, most of the major regulatory statutes (for example, the Clean Air Act, the Clean Water Act, the Food, Drug, and Cosmetics Act, and the Occupational Safety and Health Act) establish standards for whether and how to regulate.  None mandate rigid cost-benefit analysis.  In fact, the cost-benefit requirement that Sunstein's old office applies often runs roughshod over those statutory provisions.  The statutes recognize that the life-and-death decisions regulators make—let’s not kid ourselves, that’s what they are—should not come down to whether companies that make very dangerous products should devote some of their profits to protecting people.  Instead, regulatory decisions ought to reflect the standards and values set forth in the statutes, which were adopted, after all, to prevent just such threats.

In the end, cost-benefit analysis seems best suited for satisfying the intellectual musings of someone safely ensconced in an Ivory Tower, high above and far removed from the very real dangers that agencies such as the FDA are tasked with addressing. We’d all be a lot safer if agencies were able to go about their business unfettered by the fruitless search for more information geared to support various dogmas.  Fungal meningitis from tainted drugs, after all, should be unacceptable to anyone, at any cost.