Notice & Comment

Cost-benefit Analysis as Policy and as Dialectics, by Shi-Ling Hsu

*This is the sixth post in a series on Michael Livermore and Richard Revesz’s new book, Reviving Rationality: Saving Cost-Benefit Analysis for the Sake of the Environment and Our Health. For other posts in the series, click here.

In Reviving Rationality, Livermore and Revesz (“L&R”) argue for a robust rehabilitation of cost-benefit analysis (“CBA”), after its cartoonish uses under the Trump Administration. It is difficult to argue for CBA from first principles alone, so L&R make out a case sounding in pragmatism. For example, L&R argue that CBA was critical in halting many Trump Administration initiatives, and was also helpful in advancing and preserving environmental rules from the Obama Administration. L&R also believe CBA to be appropriate for climate policy as a way of sorting and prioritizing the many ways in which government might reduce greenhouse gas emissions. Which fossil fuels should be phased out? How quickly? Making complex tradeoffs in climate policy is technically and politically challenging, and nothing else quite does the job that CBA does. For L&R, “rationality” is a notion that is derived from first principles (welfarism), but blended with some practical political and administrative considerations to make a normative case for CBA itself.

I fear that this pragmatic case for CBA, and perhaps more broadly for welfarism, falls on many deaf ears: a frighteningly large part of the Republican Party, and a substantial group of progressive Democrats. Turbo-charged by social media, appeals to values have become much more politically powerful than appeals to facts, reason, and analysis. Are welfarists tilting against windmills? Do people just prefer that government policy reflect their values rather than on efforts to maximize welfare?

The popularity of Donald Trump is illustrative. While there is economic grievance embedded in Trump’s appeal, the discontent is expressed more strongly in terms of lost values in wayward America: immigration restrictions were about fair play and rules abidance, climate policy was about rural supremacy, and even international trade was about fairness in competition. Now, one can argue quite credibly that what has caused discontented voters to support someone like Donald Trump and express these values so strongly is a deterioration in welfare that has undermined their communities and therefore threatened their values, embedded in those fading communities. But that does not solve the problem that these people seem willing to suffer welfare losses if it means their values are reflected in policy. Climate policy would increase welfare (or decrease welfare losses from damages due to climate change) for everyone, as would immigration reform and international trade, but are vigorously opposed by Trump supporters. Opposition to aspects of the Affordable Care Act were heaviest in groups that would have benefited most, including those heavily supporting Donald Trump. Darkly implicit in many of these values is xenophobia, unspoken but underlying. In any case the root problem is that CBA will not sway them, and any regulations designed to increase their welfare, justified by CBA, will only run headlong into a clash with their stated values. This is not limited to the United States: the United Kingdom vote in favor of Brexit also seemed to signal a willingness to suffer welfare loss in order to regain something less tangible, such as a measure of independence.

Perhaps the most serious objection to a welfarist, cost-benefit state, is that welfarism is just another value. Welfarism should stand alongside, and not above, other values such as equality, nondiscrimination, liberty, or freedom. Welfarism might even be subservient to some of these other values. Even welfarists acknowledge the importance of distributional issues, while they work to incorporate them into welfarist frameworks. President Biden has called for changes to CBA to account for distributional issues. Liberty and freedom are obviously fundamental to Americans, as well as other Western societies, as the Brexit vote may well indicate. So perhaps government isn’t even supposed to maximize welfare. It is supposed to reflect the values of a moral individual.

But this line of thinking then leaves unanswered the difficult and obvious question of what government is supposed to do about conflicts in values. What, indeed, is government to do to balance say, public health and safety against liberty and freedom? Personal responsibility against equality? Fairness against prosperity?

The pandemic has sharply highlighted this clash between welfarism and values. Mask and vaccination mandates are public health measures that clearly could be justified by any credible CBA. With full approval of the Pfizer vaccine, most people should be required to be vaccinated, with some limited exceptions. Many mandates, even vaccine mandates (such as for schoolchildren) have been uncontroverisally enacted, for public health crises that have done far, far less than kill the over 700,000 Americans that have died so far from COVID-19. And indeed, opposition to mandates have concentrated in those most likely to be helped: recent surges in cases and deaths in the United States are strongly concentrated in unvaccinated individuals, so a mandate could well have saved their lives.

Objections to mandates have largely taken the form of appeal to principles, such as “liberty” or “bodily integrity.” Those are universal values, to be sure. But how deeply held are they? In some cases, they are sincerely and deeply held. There are genuine religious objections to vaccination. But the lingering suspicion is that much objection is performative, political schadenfreude, incited by Facebook and by Republican politicians. There is surely a symmetrical suspicion on the objecting side.

But taking seriously those cases of objection that are sincerely and deeply held, and assuming that these legitimate objections cannot be distinguished from performative ones, the conundrum is that of choosing between majority coercion of a minority, and minority coercion of a majority. There are differences in the degree of oppression on both sides, and of different natures, but there would appear to be no way of reconciling or even comparing them. 

This is the kind of situation that cries out for cost-benefit analysis. In small scale problems, with small externalities, a deontological approach to policy might be more appropriate. However, these questions involving large populations in large effects implicate delicate and complex tradeoffs. Government decision making must find some way to weigh competing interests and values by aggregating preferences, to make sense of these sharp conflicts.

How could preferences for COVID policy be aggregated? In my favorable review of Retaking Rationality, I call for experimental economic analysis to try and measure the depth and sincerity of both (i) the aversion to mask-wearing or vaccination, and (ii) the perception of heightened risk due to proximity to unmasked or unvaccinated people. I propose the following hypothetical experiment: outside the entrance to a grocery store which does not require masks, approach people without a mask and offer them a mask and a payment of money, in exchange for the promise to wear the mask throughout their shopping trip. A variety of techniques familiar to economic researchers could measure variables relevant to the experiment, such as the amount of payment and the duration of the shopping trip, and demographics of shoppers.

This experiment would be a variation on contingent valuation, a method of eliciting values of non-market goods, but with an added advantage, unusual for economic research: instant verification that the shopper means what they say. The unmasked shopper who dons a mask for money demonstrates by their choice their actual personal cost for putting on a mask, not just their stated, possibly performative statement. A second, matching experiment would be less straightforward, but still feasible: asking masked shoppers how much they would be willing to pay to have everyone in the store wear a mask. This second experiment would not have advantage of instant verification of the first experiment, because the condition of payment cannot be guaranteed: it would be difficult to ensure that everyone in the store wears a mask. More engineering would be required to insulate the second experiment from performative responses.

The point of such experiments would be to express in dollar terms, exactly how much it is worth to for the mask-averse to exercise their rights to “liberty” or “bodily integrity,” and for those on the other side, who favor mask or vaccination mandates, how much it would be worth for them to get others to surrender their rights. The elicitation of valuation from a large, representative sample of respondents would allow aggregation of preferences, and a balancing using CBA. Such a CBA would serve as a formal weighing mechanism for choosing between majoritarian or minority interests.

Now, it is true that such a proposed COVID experiment would almost certainly be lambasted as the very kind of thing that Trump supporters have railed against, opaque techniques that suspiciously clash with their values. But to return to a question asked earlier, what is government to do? Is the future of public debate destined to be a vituperative, indeterminate, and ultimately violent clash of values? Whether conscious or not, the invocation of values these days divides, rather than resolves. Policy and political issues drip with grievance, and it seems that rhetorically, outcomes take on a winner-take-all nature, with no acknowledgement of competing values.

It would be far too much to assert that CBA is the salve for all that ails democracy and public debate. But at this frightening juncture in American history, a moment that seems like an inflection point in the history of democracy, there seem only to be bad options. If values-driven debate implicitly relies upon the political process as a means of resolution, it is hard to see how this ends well, at least in the United States. Americans and their elected officials seem to be hurtling towards a brutal political nihilism that no longer even pays lip service to the weighing of competing values. Whatever else is politically rotting in the United States, the invocation of values, the concoction of self-serving facts, and the dismissal of reason and analysis, seem to be aggravating factors. And while CBA is not a comprehensive answer, its approach of giving weight to competing values seems to be a step in the right direction. Reviving Rationality is a call to take that step.

Shi-Ling Hsu is the D’Alemberte Professor of Law at the Florida State University College of Law.