by Richard L. Revesz and Michael A. Livermore. New York: Oxford University Press, 2008. 262pp. Cloth. $34.95/£18.99. ISBN: 9780195368574.

Reviewed by Thomas W. Church, Department of Political Science, University at Albany, State University of New York. Email: TChurch [at]


Richard Revesz, Dean of the NYU Law School, and Michael Livermore, a recent graduate of the same school, have written a book aimed at convincing environmentalists and others of a “proregulatory sentiment” that cost-benefit analysis (CBA) is not their enemy but a potential friend, that they “need not lose their souls in order to embrace cost-benefit analysis. They only need to be reminded that reason is often on their side as well” (p.19).

The book is thus an upbeat appraisal of the promise of CBA for those concerned with protecting the environment and public health:

This book argues that cost-benefit analysis, properly conducted, can improve environmental and public health policy. Cost-benefit analysis – the translation of human lives and acres of forest into the language of dollars and cents – can seem harsh and impersonal. But such an approach is also necessary to improve the quality of decisions that regulators make. Saving the most lives, and best protecting the quality of our environment and our health – in short exercising our compassion most effectively – requires us to step back and use our best analytic tools. (p.3)

This theme is echoed in the book’s title, “Retaking Rationality.” The clear implication is that CBA is simply another term for reasoned thinking – “coolly calculating, rational, disinterested” (p.19) – and it must be recaptured from the antiregulatory groups that have improperly biased its practice over the past two decades. These biases in CBA methodology arose, according to the authors, because progressives chose to sit on the sidelines when important governmental determinations were made regarding how CBA was to be practiced in agency decision making. If the book’s title leaves any doubt about either the intended audience or the political message, it is dispelled early on with a call to arms more reminiscent of the Communist Manifesto than an academic treatise on an esoteric analytical technique:

Proregulatory groups must shake off their torpor. Their opposition to cost-benefit analysis, even if it was understandable at the outset, has become very counterproductive. Their position is now hindering their goals more than it is helping them. They must recognize that cost-benefit analysis can – with work – become a goad as well as a brake on government. As soon as they reach this realization, cost-benefit analysis will cease to be inherently antiregulatory and will become a tool that is exactly as good as we can make it (pp.44-45).

The irreverent analogy to a work that brings heartburn to most economists suggests my overall predisposition [*849] toward CBA, which I should probably state at the outset. I am a political scientist. Though I have had some training in both economics and law, I do not share either discipline’s distaste for, and distrust of politics, nor their attraction to decision-making processes in government that attempt to eliminate administrative judgment and discretion in favor of purportedly “objective” or “rational” methods that – at least in practice – are frequently neither. This is not the place for an extended critique of cost-benefit analysis, either in theory or in practice. (Those looking for a spirited critique can find it in Ackerman and Heinzerling 2004). Indeed, I am something of an agnostic on the subject, and suspect that some version of CBA has a useful place among the tools used by policy analysts and administrators to evaluate alternate approaches to policy issues. However, I believe that the largely unqualified embrace by the authors of this book of “properly conducted” CBA as the primary source of guidance for would-be regulators, and their implicit identification of the narrow economic rationality embodied in CBA with “reasoned thinking,” may prove to be bad advice for their intended “proregulatory” audience and for the future of effective health, safety, and environmental regulation.

The book is divided into three sections. The first, “Decisions Are Made by Those Who Show Up,” sets out the errors made by environmentalists and others supportive of governmental regulation when they failed to participate in decisions concerning the mechanics of how cost-benefit analyses would be conducted in government. This strategic mistake left the field to opponents of regulation, and led to a number of antiregulatory biases introduced into the way CBA is now conducted. The final section sets out in broad outline the authors’ program for “Instituting Regulatory Rationality” through better use of CBA.

While the first and last sections outline the authors’ arguments for the embrace of CBA by groups that have tended to shun the technique in the past, the analytical meat of RETAKING RATIONALITY is found in the central section of the book, which discusses eight purported errors or “fallacies” in the way CBA is either currently used in regulatory decision making, or which seem to be on the horizon. These errors “amount to a virtual Berlin Wall blocking good regulations” (p.145). While these criticisms of the conduct of CBA are not new, the discussion is cogent and informative. It is also novel in that it comes from such obvious partisans of the technique.

Three of the fallacies speak to what might be termed technical issues regarding the conduct of CBA; they raise empirical questions that presumably could be addressed by correcting analytical errors or by further research into empirical assumptions:

• Ancillary positive benefits from regulation are seldom taken into consideration in CBA, while predicted negative consequences of regulation always find their way into the analysis.

• The “health-wealth tradeoff,” based on the assumption that the costs of regulation decrease overall societal wealth and thereby diminish human health, is improperly used to decrease the expected benefits of regulation; the [*850] authors argue that this analysis is empirically incorrect because it confuses the correlation of wealth and health with causality.

• Costs of regulation are frequently overstated, not only because it is in the interest of the regulated parties to make predicted costs as high as possible, but also because CBA infrequently considers the potential for innovation.

The remaining issues cannot be addressed empirically, because they pose fundamental moral, ethical, and political choices that lie at the heart of CBA, and which are also illustrative – at least in my view – of the underlying subjectivity and indeterminacy of the entire enterprise.

• CBA “discounts” future benefits, even the benefits accrued by saving human lives in the future, in the same way economists discount the value of an income stream in the future; the assumption is that future benefits are less valuable than those obtained immediately, and are therefore “worth” less. While this piece of economic orthodoxy makes sense when applied to purely financial decisions, it operates more problematically when applied to long-term environmental problems – particularly those that may not have major impact on ourselves, but on future generations. The decision to devalue benefits and harms to future generations in the service of reducing costs to the present generation obviously involves ethical and political considerations, and a strong dose of subjectivity in setting a discount rate, rather than empirical or technical issues.

• Since CBA requires benefits to be valued in monetary terms, even human life must have a price tag. Revesz and Livermore consider two aspects of this process, currently gaining sway among CBA practitioners, to be “fallacies” in need of correction. These involve placing differential values on human lives, based on either age or on the “quality of life.” In the former case, older peoples’ lives are valued at a lower price than younger people, based on the predictable fact that they have fewer years left to live. In the latter case, an increasingly popular CBA technique values lives differently based on quality of life being lived. For example, an individual who loses use of his limbs, or suffers from serious lung disease, or painful arthritis is said to have a lower quality of life, and therefore to benefit less from life than an individual in good health. In both cases, the diminished value of the lives being saved lowers the monetised benefits of a proposed regulatory action and thus justifies expenditure of fewer societal resources. The moral and political implications involved in deciding whether, and the inherent subjectivity in determining how, to make such choices are obvious.

• CBA cannot deal with benefits or “intrinsic values” that have no ascertainable market price. While conceptually and statistically problematic, techniques exist for determining an extrapolated price for some goods that have no direct market price (such as human lives, or risks to human lives) if there are prices for related goods in the market (such as the increment workers are paid for employment in highly risky jobs, or what consumers are willing to pay for safer products). Based on such studies, [*851] for example, EPA has set the monetary value of a human life (or “statistical life”) at $6.3 million (p.70). But no relevant market exists for “natural resource values” such as the preservation of wilderness areas or endangered species. How does one monetise the benefit of maintaining a pristine Alaska National Wildlife Refuge, or protecting the polar bear from extinction? The usual method is to conduct what are termed “stated preference” or “contingent valuation” surveys in which individuals are asked how much they would be willing to pay for preserving a species or wilderness area they may never see. These surveys are accompanied by a host of conceptual and technical problems, leaving some analysts (typically those in the “antiregulation” camp) to argue that natural resource values that cannot be extrapolated from the market should simply be ignored. Revesz and Livermore regard this as another “fallacy,” though their criticism seems based less on the strength of existing valuation methodologies than on the fact that in the absence of such methods, preservation of many intrinsic natural values will not show up on the benefits side of CBA at all.

There is a final problem with CBA that the authors discuss, but suggest that no solution exists within the confines of the analysis: the problem of fairness which, in the environmental arena, is subsumed in the concern with “environmental justice” or, more pungently, “environmental racism.” All CBA calculations of benefits are attempts to approximate the price that would emerge in a free market; the best approximation of such a value is typically “willingness to pay” or WTP. The equity problem here is that poorer people are necessarily less willing to pay for environmental or public health benefits because their incomes are consumed by food, shelter, and other essentials; they are less willing because they are less able to pay. Basing environmental and public health policy on WTP, then, would necessarily concentrate environmental and health risks and damages on those segments of the population least able to pay for their amelioration. Of course, this is typically the result of the play of the political and economic system, but when considering public policy from the position of a concerned scholar, it is surely problematic to label this arguably unjust and unfair allocation of harms to be the result of “coolly calculating, rational, disinterested” reason.

Revesz and Livermore’s solution to this problem is to undertake “distributional analysis:”

Distributional analysis is not an easy undertaking, but it is a necessary corollary to cost-benefit analysis. Cost-benefit analysis, on its own terms, excludes concern for the distribution of the benefits and burdens of regulations. This omission is acceptable only if a separate effort is undertaken to account for these effects. (p.182)

So, we learn in the final pages that CBA is unacceptable without distributional analysis. Of course, it goes without saying that a determination of who should benefit from, and who should pay for a governmental policy is the very definition of a political decision. Yet the authors devote only two paragraphs to this issue and do not indicate what a distributional analysis might look like, how its results would be integrated into [*852] the results of CBA, what weight it should be given in regulatory decisions.

While the authors make a cursory nod (on the last page of the book) to the fact that cost-benefit analysis is “not a panacea” (p.190), this caveat is drowned in their unbridled enthusiasm for a new, purified version of the tool as the sine qua non of rational regulatory decision making. There is no serious discussion of how CBA could accommodate concerns over equity and fairness, or how it might fit with other modes of analysis, other less mechanistic – more “subjective” or even “political” – decision-making processes.

The authors’ overselling the promise of CBA, and their explicit identification of the technique with broader notions of administrative rationality, is especially troubling because legislators and the courts may buy into the view that cost-benefit analysis defines reasonable decision making in a regulatory context, that if an agency does not follow the dictates of economic rationality embodied in CBA, it is acting arbitrarily. Arbitrary administrative behavior, of course, is subject to reversal by judges, especially federal judges, who increasingly distrust the exercise of discretion by bureaucratic agencies and are ever more ready to substitute their own notions of reasonableness for those of agency officials. It is not clear that regulatory policy making would be improved if administrative decisions were routinely second guessed by federal judges attempting to determine whether proposed regulations square with their understanding of the dictates of rational choice economics.

Cost-benefit analysis as depicted by the authors of RETAKING RATIONALITY promises to remove (or at least substantially reduce) agency discretion, always subject to influence by agency clientele or other political pressure, and take us to a landscape of impartial analysis and rational, disinterested decision making: “By providing a more accurate assessment of the real costs and benefits of a decision, formalized cost-benefit analysis reveals the distortions of politics – the backroom deals and special interest politics – for what they are” (p.12). As such, CBA, at least as viewed by those who fail to appreciate the transcendently political and moral assumptions underlying its application, can be seen as providing a clear, apolitical metric by which to evaluate regulatory decisions. But, as we have seen, critical assumptions in the very guts of CBA – such as differential valuation of human lives, discounting the interests of future generations, or inability to consider the societal distribution of the benefits and costs of regulation – are highly charged moral and political choices that will frequently have a profound impact on the calculation of costs and benefits that emerge from the allegedly disinterested analysis. As a result, as even its most enthusiastic supporters admit, the results of CBA are frequently “indeterminate:” reasonable and conscientious analysts may produce widely divergent assessments of the costs and benefits of a proposed governmental action. For example, Cass Sunstein – a more measured supporter of CBA than the authors of RETAKING RATIONALITY – found that defensible estimates of the societal benefits of EPA’s proposed regulation of arsenic in drinking water ranged from $13 million – well below [*853] regulatory cost – to a figure more than 260 times greater – $3.4 billion – substantially above the cost of the regulation (Sunstein 2002, p.177).

Judge Richard Posner, a vocal proponent of CBA, states the following in the “Advanced Praise” blurb printed on the back cover of RETAKING RATIONALITY: “It is true that in noncommercial settings, cost-benefit analysis often cannot yield definitive conclusions without the analyst’s adopting assumptions that may be politically charged. But the charge can be, as the authors show, liberal rather than conservative.” In this context, irrespective of one’s political leanings, it is difficult to accept Revesz and Livermore’s depiction of cost-benefit analysis as “coolly calculating, rational, disinterested.” And it is similarly unclear that committed environmentalists should enthusiastically embrace a technique that promises the removal of political judgment from agency decision making while reinserting it under cover of seemingly technical issues such as “discounting” and “quality of life” calculations.

The authors readily admit that CBA has not in the past been a friend of those supportive of governmental regulation to promote human health and the environment. In light of the underlying subjectivity of the enterprise, the technical problems inherent in the calculation of public benefits, CBA’s frequent indeterminacy and intrinsic inability to consider basic notions of fairness, it is not clear that environmentalists and other supporters of regulation would be well served by casting their lot with CBA, even if it were biased in a proregulation direction as the authors propose. In my judgment, the critical issue for those concerned with effective regulatory policy should be how to integrate the findings of cost-benefit analysis and other quantitative analytical techniques such as risk analysis, into a more inclusive decision-making framework that does not substitute form for substance, and which insures that the value choices made by administrators are explicit and defensible in broader terms than those embodied in economic models.

Ackerman, Frank and Lisa Heinzerling. 2004. PRICELESS: ON KNOWING THE PRICE OF EVERYTHING AND THE VALUE OF NOTHING. New York: The New Press.

Sunstein, Cass R. 2002. RISK & REASON: SAFETY, LAW, AND THE ENVIRONMENT. Cambridge: Cambridge University Press.

© Copyright 2008 by the author, Thomas W. Church.