On Balance: What Should OIRA Do about Equity, Justice, Dignity and Moral Responsibility?
On his first day in office, President Biden issued a memorandum titled “Modernizing Regulatory Review,” directing the director of the Office of Management and Budget to produce a set of recommendations for how to improve regulatory review.
The main purpose of the memorandum is summed up in the following quote:
This directive confronts the OMB director with a dilemma, which is whether to recommend monetizing the impact of regulations on all of these policy goals, or instead to recommend accounting for some of them in some other way, whether quantitative or qualitative.
Consider: to the extent that regulatory review promotes any kind of regulation, it is by demonstrating that said regulation has benefits that more successfully “justify” its costs than alternative regulations. Thus, when calling for ways that regulatory review can “promote” certain policy goals, what President Biden is calling for is a more comprehensive and effective way of accounting for all of the listed policy impacts, so that they can all be taken into consideration in the determination of whether costs are justified. We already have a way of accounting for effects on public health and safety, economic growth, social welfare, and at least some important components of environmental stewardship and the interests of future generations, which is to monetize them using willingness to pay (WTP). What is called for, then, is a way to account for racial justice, human dignity, equity, and whatever components of environmental stewardship and the interests of future generations are not accounted for in the measures of aggregate WTP we currently use. (In the latter cases I’m largely thinking of the moral duty to care for the environment and for future generations, some parts of which may not be captured by conventional methodologies.)
An obvious choice would be to simply add these things to the list of policy impacts we monetize using WTP. It’s easy: just use contingent valuation, which can be applied to anything, including racial justice, environmental stewardship, human dignity, equity, and the interests of future generations. In an unpublished study, for example, I measured aggregate willingness to pay for, among other things, the rights of transgendered individuals to serve in the military, using an online survey. (Not surprisingly, WTP was negative for a non-trivial minority of respondents.) WTP is how regulatory impact analysis accounts for the positive and negative impacts of policies, whenever possible, because the federal government implicitly requires monetization of all policy impacts, to the extent possible. Circular A-4 states that, “'Opportunity cost' is the appropriate concept for valuing both benefits and costs. The principle of 'willingness-to-pay' (WTP) captures the notion of opportunity cost by measuring what individuals are willing to forgo to enjoy a particular benefit." And elsewhere, "You should develop quantitative estimates and convert them to dollar amounts if possible." President Biden is apparently extending the list of costs and benefits that must be included in regulatory impact assessments, and Circular A-4 makes it implicitly clear that all costs and benefits should be monetized, if possible. It is possible to convert rights and dignity into dollar amounts. Circular A-4 says that you must. This approach has been endorsed in some benefit-cost analysis (BCA) literature, explicitly by Zerbe, Bauman and Finkle (2007) and implicitly by Cass Sunstein (2018, chapter 6).
I think monetizing racial justice, human dignity, equity, and moral responsibilities is a terrible idea. There are reasons we don’t do so, and they are good reasons. One argument is that such impacts are incommensurable with the kinds of things that are commonly measured using WTP. They can’t be measured on the same cardinal scale “without doing violence to our considered judgments about how these goods are best characterized.” (Sunstein 1994.) Aggregate WTP for things that are incommensurable is like adding height and weight. The units of measure have no meaning. We recognize things that are incommensurable in this way because they are things for which we consider that “a monetary offer is entirely inappropriate.” (Sunstein 1994.) I remain hopeful that we live in a society in which it is still considered inappropriate to exchange money for racial justice, human dignity, equity, or the moral duty to protect future generations or the environment. Boardman et al (2020) make essentially the same point, saying that inclusion of distributional concerns in BCA obscures the tradeoffs between what are essentially different kinds of value.
I have proposed another reason for not monetizing impacts of this sort. (Acland 2021.) When we monetize risk of injury, or days of fishing, or consumer products, and use aggregate WTP to account for their value to society in RIA, we are saying that it doesn’t matter why people value them (because WTP does not tell us that), and that their value to society is nothing more than their value to the individuals who express WTP for them (because including WTP for them in BCA is the only way we account for them). For impacts of this kind, it is enough to know that individuals value them, and how much, in dollar terms. One way to frame this is that these are impacts that we think should be governed by “consumer sovereignty,” in the sense that we feel that it is acceptable to make decisions about them purely on the basis of individuals’ preferences, in the way that markets do. I refer to these as “sovereignty warranting” impacts, and they include almost everything that it is common practice to monetize in benefit-cost analysis.
I define sovereignty warranting impacts as those goods for which, “i) it should not matter to society why individuals value them, and ii) there are no other reasons they should matter to society, over and above the fact that they matter to the individuals who receive them.” (Acland 2021.) Human dignity, racial justice, and equity clearly do not satisfy this definition. A simple test is to ask oneself, “if WTP for racial justice benefits were outweighed by the cost of implementation, is it possible that we would consider the costs justified by the racial justice benefits, if those benefits seemed significant enough.” Clearly the answer is yes. Racial justice is a moral value that matters for its own sake. To adequately weigh the pros and cons of policies that affect racial justice, policy makers need to know more than just the aggregate WTP for the increases or decreases in racial justice effected by said policies. Inter alia, they need to weigh moral and ethical considerations in light of their own beliefs about what is wise, what makes for a good society.
The argument might be made that BCA should nonetheless include WTP for non-sovereignty warranting policy impacts, because how much individuals value such impacts is important for decision makers to know, even if it is not all they need to know. I argue against this, on the grounds that it leaves the decision maker unable to determine which kinds of value, what reasons for caring, have already been weighed in the balance using aggregate WTP, and which ones he or she needs to add to his or her considerations above and beyond aggregate WTP. Suppose people are willing to pay for a reduction in income inequality. (I have measured this.) WTP does not tell us if it is because inequality makes them feel inferior to their fellow citizens, or because they have altruism towards those at the bottom of the distribution. Knowing which of these quite different kinds of value have been accounted for in BCA is crucial for a decision maker. If WTP is based primarily on altruism, the decision maker might rightly feel that little additional consideration need be given to altruism. But if WTP is primarily based on positional concerns, the decision maker might feel that they themselves have a responsibility to additionally weigh altruism, because it matters to society, even if it isn’t why people have WTP for income compression. Far better would be to inform decision makers about the effect of policies on equity, dignity, and racial justice in some other way, qualitative or quantitative (but not monetized), and let decision makers weigh those considerations against the aggregate WTP for sovereignty warranting impacts. Describe the effects qualitatively. Include quantitative data when possible. If citizens’ sentiments seem relevant, include the results of opinion surveys or focus groups. But not WTP.
A crucial point to make is that BCA is not synonymous with regulatory impact assessment. It is true, as I explained above, that Circular A-4 implicitly requires at least an attempt at monetization of all cost and benefits. But it is equally clear that – at the same time, and in contradiction to that implication – the intent of the authors was for RIAs to include additional information about the pros and cons of a policy, including, most saliently, its distributional impacts, as Robinson and Zeckhauser (2016) document. My belief is that the authors of the circular were simply not considering the possibility that one could monetize distributional impacts, as, in fact, one can. A careful reading makes it clear that what they had in mind was essentially a “multi-goal” analysis that included net benefits as measured by BCA, and other factors, which they did not expect would ever be monetized, even if they could be. And, although the executive orders that require regulatory impact assessment for federal regulations call for choosing regulations that “maximize net benefits,” they simultaneously call for choosing regulations whose benefits “justify” the costs. If the BCA component of a regulation nets out negative, but the policy has desirable distributional effects (some of which might have implications for racial justice and human dignity), it is understood that those desirable distributional effects can be treated as a benefit to society that could justify the net cost in BCA terms. True also of environmental stewardship and moral responsibility to future generations.
Up until now, OMB has not made any clear statement about whether dignity, justice, equity and moral responsibility should be monetized. As I stated earlier, Circular A-4 implies that all costs and benefits should be monetized if possible. But it is equally clear that the authors were not thinking about equity, justice, dignity and moral responsibility. The examples of “non-monetizable” benefits cited in the circular are things like the performance of off-road vehicles and increases in water quality and fish populations. As explained to me by a former chief economist at a federal agency, the real goal was to get analysts to monetize and include such things as the shadow price of time, consumer utility, and the value of displaced actions. There is no mention of monetizing impacts such as equity, dignity and racial justice, and there is no clear guidance on how to account for such impacts in regulatory impact assessment.
But Biden is forcing the issue, because he is going beyond the requirement embedded in EO 12866 and re-emphasized in EO 13563, that regulators should “assess” or “take into account” all costs and benefits, and adding the requirement that there be “concrete suggestions” for how this be done. My concrete suggestion is that OMB explicitly list those impacts that should not be monetized, and it should include equity, human dignity, racial justice, and the moral duty to care for the environment and for future generations. Then, some alternative concrete suggestion can be made about how to account for these impacts – quantitatively when possible – so that they can be weighed against the impacts that are included in the BCA component of a regulatory impact assessment, and it should be made explicit that regulatory impact assessment is intended to be a multi-goal analysis, and not just an attempt to tally all pros and cons in the same unit of measure.