The “Science Charade” After 'Chevron'

What happens after the U.S. Supreme Court makes it harder for agencies to regulate? There are at least a couple of possibilities. Option One: an agency might just stop trying to regulate under that policy. Or Option Two: an agency might seek another path to achieve the same thing. The danger of Option Two may be one of the most important—but underappreciated—of the Court’s decision in Loper Bright, which overruled Chevron deference. My fear is that agencies will not simply give up but instead will lean into what Professor Wendy Wagner has dubbed “the science charade.”

Let’s start with some basics. Under Chevron, courts would defer to an agency’s reasonable interpretation of ambiguous statutory language. The idea was that because agencies are more politically accountable than courts and have a better technical grasp of how complex statutory schemes work, when a statute administered by an agency is ambiguous, courts should get out of the way and let the agency act so long as the agency’s resolution of the ambiguity is reasonable. Chevron presented legal and conceptual problems (including why ambiguity should favor the agency rather than regulated parties, who may be punished—sometimes even criminally—for violating the agency’s view of the statute), but also a practical one that goes to the heart of administrative incentives. Because agencies could expand their power by finding ambiguities, agency officials, often responding to political demands, would unsurprisingly stretch to find them so they could pursue aggressive policies that Congress never authorized.

In Loper Bright, the Court essentially said “enough.” Under our Constitution, the legislature makes the law, and courts ensure that the executive stays within the law as written by Congress. After Loper Bright, courts decide the meaning of statutes, even statutes with some ambiguity. As Justice Clarence Thomas has explained, Article III’s vesting of the “judicial power” in the judiciary “calls for that exercise of independent judgment,” but “Chevron deference precludes judges from exercising that judgment,” thereby “wrest[ing] from Courts the ultimate interpretative authority to ‘say what the law is,’  and hand[ing] it over to the Executive.”

Loper Bright thus should be a welcome development for purposes of respecting the separation of powers, especially if agencies accept the limits of their authority. But there is a danger: What if they don’t? What if the same political dynamic that prompted agencies to stretch statutes in the first place may also prompt agencies to find alternatives to Chevron? 

I have recently penned an article about one such alternative: the science charade. Wagner coined the term decades ago to explain an important dynamic within administrative law. As she observed, because judges often defer to agencies on questions of science, “the courts offer agencies strong and virtually inescapable incentives to conceal policy choices under the cover of scientific judgments and citations.” Rather than justifying the agency’s policy choice as a policy choice, agencies instead may dress-up their decisions as compelled by science.

To be sure, there are limits to the science charade. Agencies must engage in reasoned decision-making and justify their conclusions as not arbitrary or capricious. So if agencies push too hard, reviewing courts will sometimes catch on that a regulator’s policy choice has outrun its science. For example, I once worked on a case where the National Marine Fishery Service used a “model [that] assumed that salmonids would be exposed to lethal levels of the pesticides continuously for a 96-hour period,” but never explained “why the 96-hour exposure assumption accurately reflected real-world conditions.” The appellate court didn’t buy it—but the district court did. This illustrates how difficult it can be to persuade a court to second-guess an agency’s invocation of science. (I often wonder what would have happened had the Environmental Protection Agency itself not criticized the National Marine Fishery Service’s “unreasonable” assumption.)

The intuition driving Wagner’s theory, thus, is impossible to brush aside. To be clear, I do not claim that agencies do this all the time. When we discuss the administrative state, we often focus on unusual occurrences rather than on an agency’s more banal, bread-and-butter operations. But that does not mean we should not worry about incentives or ignore the risk that unthinkable behavior may become more thinkable if bad incentives are not curbed. Agencies are filled with people who want certain policies. Human nature being what it is, people sometimes respond to incentives. So if the best way to get a policy through is to drape a policy decision in as much science as an agency can credibly muster, shouldn’t we expect regulators sometimes to succumb to the science charade’s temptation?   

And that brings me to my thesis: Because agencies can no longer use Chevron to pursue policies that Congress has not allowed, their incentive to use the “science charade” should increase, again, at least at the margins. 

As I explain in my article, suppose Congress has authorized an agency to “regulate Chemical X if it harms the public health.” Suppose further that agency officials want to restrict Chemical X because it harms birds, but it is unclear whether it has negative health effects on people. Under Chevron, the agency might have argued that the statute is ambiguous as to whether its authority is limited to protecting human health, so it can use the statute to protect birds, too. Of course, such a strained reading may have worked even before Loper Bright, but now agencies know that this interpretation won’t fly. So instead, the agency may lean into the science charade. Because generalist judges may be more comfortable deferring to scientific analysis than to overt policymaking, agencies may deduce that they should not say “we care about birds,” but instead should overstate what the science says about the effects of Chemical X on human health. Using the science charade as a substitute for Chevron, may thus allow them to protect birds under the guise of protecting human health.

This increased incentive to rely on faux science should be alarming for at least two reasons. One, the statute books overflow with delegations that are triggered when certain facts about the world exist—facts that require scientific or technical (e.g., economics) judgments beyond the ordinary experience of judges. Agencies may thus stop scouring the U.S. Code for ambiguities and instead scour it for delegations that kick in if certain scientific findings are made. And two, there is a “boy who called wolf” danger. Good policy needs good science, but if agencies cannot be trusted, skeptical courts may erroneously reject agency conclusions that, in reality, are supported by good science.   

Unfortunately, there is no great solution to the science charade. The reason why the charade can work is that judges are not scientists, and even if they have some scientific or other technical training, no one can know everything about everything. Generalist judges are simply not equipped to understand all the technical issues the administrative state presents. Although there are downsides, the best answer might be greater procedural formality in the regulatory process—complete with more extensive cross-examination of agency experts to create a record that may be more understandable to judges. (Of course, the dynamic effect of that prospect may be to dissuade bad science from the get-go.) As I have explained elsewhere, increasing procedural rigor is not costless, which is one reason the administrative state has largely moved away from procedural devices such as cross-examination. But for certain categories of regulatory action, it might make sense to head off bad incentives. Of course, some may argue (presumably, Wagner herself) that such costs are not worth it. But especially given the heightened incentive caused by Chevron’s demise, I’m not so sanguine. 

Like most complex systems, the administrative state resists easy answers. It is important to think through incentives and unintended consequences. The Court’s decision to overrule Chevron deference addresses one incentive—the enticement to hunt for statutory language that agencies can claim is ambiguous. But it may have the unintended effect of strengthening the temptation to rely on the science charade. There is no silver-bullet solution; it is important to recognize why agencies act as they do and to create systems to best maximize the benefits of agency expertise while preventing its abuse.

Aaron L. Nielson is a senior fellow at the Civitas Institute and holds the Charles I. Francis Professorship in Law at the University of Texas at Austin School of Law.