In the last few years, law schools and law professors have given new attention to how questions of race can be interwoven into courses that are not explicitly about race. Much has been written about how to do so in both first-year and upper-level courses, and, from all reports, the law school classroom has meaningfully changed. My sense, though it is completely impressionistic and unscientific, is that the typical Administrative Law course may have changed less than many others. It seems fair to say, at least, that there has not developed a standard suite of topics that a professor wanting to integrate questions of race and racism might include. (Though for those interested, the 2020 Symposium on Racism in Administrative Law on the Notice & Comment blog is a very useful place to start.)
Daniel Farber’s Inequality and Regulation will be of enormous value to those looking for an entrée for discussing race and Administrative Law. Moreover, wholly apart from its relevance to the classroom, it is an important substantive contribution regarding the role of race, and of poverty, in regulatory policymaking. And it tackles these thorny topics in a highly readable fashion, with a minimum of jargon, obfuscation, and, relatively speaking, citations. (Were it in a student-edited Law Review, the editors would have been pretty grumpy about the above-the-line to below-the-line ratio. It may be one advantage of faculty-edited journals is a refreshing rejection of the citation addiction (or fetish).)
Farber sets the scene with a familiar descriptive account of the central role of cost-benefit analysis (CBA) in regulatory policymaking since the Reagan Administration, with some attention to the (in the grand scheme of things relatively minor) variations from president to president and Executive Order to Executive Order (EO). He then focuses in particular on the differing ways in which the EOs have treated unquantified benefits, equity, and distributional impacts. Since Bush II these have been acknowledged as legitimate considerations, receiving particular emphasis from Presidents Obama and Biden. But in practice they have received rather scant attention. Farber recounts a similar story with regard to Environmental Justice: formal attention, including in EO 12898 and Plan EJ 2014, with a much more modest actual policy impact.
Part III then addresses the role of economic inequality in regulatory policy. Farber joins others in rejecting the view held by “economic purists” that regulation should do nothing other than maximize net regulatory benefits, leaving issues of income distribution entirely to tax and spending programs. Indeed, he emphasizes that CBA as practiced by federal agencies reflects several equity-enhancing features. The most notable is the universal practice of quantifying benefits on the basis of an aggregate willingness to pay (WTP) figure. Given enormous differences in WTP based on age, race, and, most importantly, wealth, the value of a poor person’s life “should be” lower than that of rich person’s. But every agency uses a single number for the Value of a Statistical Life (VSL); all lives are of equal monetary value in a cost-benefit analysis, just like all lives are of equal value in a moral analysis. Farber attributes this approach to what he terms “harm egalitarianism”: “equal protection against equal harms.”
Having identified the core principle, Farber considers an alternative approach to accounting for economic inequality in regulatory policymaking. That would be to disaggregate WTP but, at the same time, more fully account for distributional impacts and the incidence of regulatory burdens. One method for doing would be “equity weighting,” i.e. assigning greater weight to dollars held by poorer people. Drawing on recent work by Daniel Hemel, among others, Farber rejects this approach. He does so primarily on practical grounds, but also because he is disconcerted by taking a powerful, but hidden, redistributive approach in regulatory analysis when the actual redistributive system (taxes and benefits) is not so aggressive.
So harm egalitarianism is the central concept of the article. Farber offers it as both descriptively accurate and normatively attractive. The principle of devoting equal resources to prevent equal harms is “implicit in current regulatory practice”; it “should play a central role in regulation”; and it is a requirement of “justice.” He would make it the dominant principle by which regulation takes account of equity and distributional concerns. Dominant, but not exclusive. Farber leaves the door open to a direct consideration of redistributive concerns, which might in some cases still be a “soft factor” in decision making.
Mentioning this possibility but not fully exploring it strikes me as a gap. Here is the problem. It is possible that a uniform VSL may harm poor people more than it helps them. It all depends on the incidence of costs and benefits. To the extent the costs of regulations using the uniform (and thus, for poor people, high) VSL are borne by poor people themselves, that compelled payment of more than the benefit is worth to them likely makes them worse off. As Cass Sunstein has put it, pithily: “Requiring poor people to buy Volvos is not the most sensible means of assisting them.” Farber’s redistributive “soft factor” suggests he is open to requiring “too much” safety but then subsidizing poor people’s purchase of that safety and/or taxing rich people to pay for it. But how, when, and how much that should happen are left unexplored.
The article then takes a doctrinal turn. Harm egalitarianism involves taking account of, and acting to avoid, the disparate impact (often racially disparate impact) of government policies that are facially neutral. Might doing so violate the equal protection clause? So far, the Supreme Court has not so held, and Farber concludes that such a result is imaginable but not likely. He also thinks most agencies most of the time have the statutory authority to take such concerns into account.
(Here let me diverge from the article briefly, like the workshop questioner who wants to discuss the article the presenter did not write. Farber does not investigate or assert the converse argument: that “equal protection against equal harms” might be not just permitted but required. At least three theories come to mind. First, it obviously resonates with equal protection principles. Indeed, as Farber’s use of the very term “equal protection” indicates, this constraint more clearly implicates equal protection than do many classifications subject to review under that clause. On the other hand, the intent requirement keeps us in rational-basis land; the “one step at a time” cases cut against grounding harm egalitarianism in the equal protection clause. Second, in relevant settings failure to take disparate impact on the poor and people of color would violate the Clinton Environmental Justice Executive Order. However, the EO, of course, is not judicially enforceable. Third, and most promising, there is a plausible argument that promulgating a regulation with a racially disparate negative impact would be arbitrary and capricious because the agency had failed to “consider the relevant factors.” What factors are “relevant” for purposes of arbitrary and capricious review is remarkably underexamined, but this seems like a potentially fruitful line of argument.)
Having defended agency consideration of the disparate impact of its own regulations, Farber acknowledges that it would raise different concerns if an agency designed regulations to remedy historical or structural injustices (an approach that would seem to go beyond “equal protection from equal harms”). Here, Farber anticipates more robust judicial and constitutional objections. Farber’s discussion is tentative. Such an action might be seen as, and swept aside with, affirmative action programs, which rest on a similar remedial theory. Yet the settings are different. Unlike, say, affirmative action in college admissions, this sort of agency action does not favor, or disadvantage, any individuals, is not zero sum, and would itself be facially neutral. The constitutional objection is “nonnegligible,” but would probably (and should) fail. Statutory objections might also arise. Their validity will depend on the statute, but clearly an agency would be on stronger statutory ground when levelling up protection of communities especially burdened by the regulated risks than it would if explicitly opting to reduce risk in communities of color rather than in white communities as a response to structural racism.
That last point leads into the final, and most concrete, section of the article, which discusses the ways in which risk modeling often already incorporates concerns over inequality and might do so more extensively. Here the “equivalent protection from equivalent harms” principle is operationalized. A risk assessment requires knowing not just the extent of exposure but the vulnerability of the exposed population. Vulnerability is a function of some factors having nothing to do with race and wealth, such as age. But it also turns on other exposures, poverty, poor health, and even, in some circumstances, race itself. All of these factors may make population A more vulnerable than population B, meaning that the risk (the harm) of an identical exposure is greater for the former than the latter. In such circumstances, “equivalent protection from equivalent harms” means prioritizing regulation to protect population A and/or regulating to ensure a lower exposure for population A than for population B. Current risk analyses do this to some extent, but the inputs, including in particular the size of the geographic units considered, could and should be more granular. With more reliable and fine-grained data on exposure and careful attention to differential exposures and vulnerability, regulatory analysis would avoid the potential legal pitfalls of explicit attention to economic and racial inequality while still achieving appropriately targeted protection for poor communities and communities of color.
