Douglas A. Kysar*

Abstract:  Economic cost-benefit analysis aims to evaluate regulatory proposals by identifying, monetizing, and comparing the proposals’ expected positive and negative consequences. The methodology has been received critically in the area of environmental, health, and safety regulation, where scientific uncertainty, difficulties of valuation, and uncommonly long time horizons are said to render cost-benefit analysis especially problematic. This Essay reviews such criticisms through a discussion of the use of cost-benefit analysis in the particular context of climate change policymaking. In this context, generic criticisms of cost-benefit analysis in the environmental, health, and safety area become even more pronounced, raising significant doubt about the methodology’s philosophical and practical appropriateness as a guide for climate change.

In a remarkably clear, candid, and courageous new book, James Gustave Speth documents both the continued deterioration of the global environment and the failure of the international community to respond forcefully and effectively to that deterioration.1 As the current dean of the Yale School of Forestry and Environmental Studies, a founder of the World Resources Institute and the Natural Resources Defense Council, an advisor on environmental issues to Presidents Carter and Clinton, and a former chief executive officer of the United Nations Development Programme, Speth brings considerable experience and authority to the account. His views reflect the seasoned gravitas of one who has been intimately involved with the modern environmental movement since its inception and who, despite having played a leading role in many of its successes and accomplishments, [*PG556]knows well the movement’s failures and limitations.2 Accordingly, when Speth concludes that “[t]he current system of international efforts to help the environment simply isn’t working,”3 we should listen carefully both to his diagnosis and to his prescription for change.

As Speth notes, most of the underlying causes of unsustainable and inequitable development patterns are readily identifiable,4 as are the policy measures necessary to address them.5 What is missing, in Speth’s view, is not adequate knowledge of our problems and their potential solutions, but rather a common environmental consciousness that regards other nations, other generations, and other life forms with sufficient awareness and concern to vault environmental solutions to the forefront of the global agenda. Accordingly, Speth believes that an entirely new model of global environmental governance is required, one in which international obligations flow not from the empty platitudes of treaties that lack political and institutional support,6 but rather from a fundamentally transformed culture that views environmental sustainability and global equity as moral imperatives to be implemented through dynamic local, regional, and nongovernmental initiatives, as well as through a well-funded and well-respected World Environmental Organization.7

[*PG557] This Essay supports Speth’s case through a discussion of climate change, cultural transformation, and comprehensive rationality.8 Narrowly, it critiques the use of one aspirant to comprehensive rationality, economic cost-benefit analysis (CBA), in the context of climate change policymaking. More broadly, it describes a manner of evaluating human activities and policy choices that impedes, in deep and subtle ways, the type of cultural transformation that Speth regards as necessary to the continued vitality and development of humanity and its environment. As will be explained, “the two central ideas of environmental ethics” that Speth argues must be embraced as part of our re-envisioning of ourselves and our relationship to the natural world—namely, “the protection for their own sake of the living communities that evolved here with us,” and “our trusteeship of the earth’s natural wealth and beauty for generations to come”9—both are denied or discounted by the methodological assumptions of CBA. It follows, then, that if one agrees with the central argument of Speth’s new book, one should be troubled by the use of CBA in climate change policymaking.

To be clear, this Essay does not challenge efforts to identify the expected consequences of climate change or of proposals to mitigate or adapt to climate change. There are obvious and sound reasons for wanting to know, with as much precision as is feasible at a given moment, what the likely positive and negative effects of engaging in a particular course of action are in response to an impending problem. Nor does this Essay oppose the implementation of particular climate change policy goals through the use of economic instruments such as tradable permits or pollution taxes.10 There are, again, obvious and sound reasons for wanting to achieve a specified goal with as little social cost as possible. If economic policy instruments truly are more cost-effective than traditional regulatory tools, as their proponents argue, then there is little reason in principle to oppose them. What this Essay disputes instead is the use of CBA to determine the content of climate change policy goals, rather than merely the method by which those goals are implemented. It is in this sense that CBA threatens to impede [*PG558]the transformation of culture that Speth and others regard as vital to the future of the planet.

Part I of this Essay offers a very brief introduction to the use of CBA in climate change policymaking, utilizing controversial environmental commentator Bjørn Lomborg’s study as an illustrative example of the methodology and the type of policy conclusions that it is capable of generating. Part II then discusses four key limitations of the use of CBA in the climate change context: (1) the inconsistency between CBA’s problem-state assumptions and the actual level and nature of uncertainty surrounding climate change; (2) the inability of the valuation methodologies utilized by CBA practitioners to consider appropriately the anticipated losses of human life and environmental resources; (3) the impropriety of using exponential discounting within CBA to resolve decisions regarding the intertemporal distribution of costs and benefits; and (4) the inherent orientation of CBA toward the status quo, or what David Driesen would call its static, rather than dynamic, efficiency orientation.11 Together, these limitations suggest that CBA is an unacceptably crude device for guiding policy choices in the context of a massively complex and morally imbued problem such as global climate change.

It bears emphasizing at the outset that most of the critiques levied in this Essay are not original. Rather, they come from careful thinkers who have examined and challenged the use of CBA in environmental, health, and safety decisionmaking more generally, including Elizabeth Anderson,12 Derek Parfit,13 Mark Sagoff,14 and Lisa Heinzerling.15 Their points are worth restating here, however, for at least two reasons: first, because they apply with particular force in the climate change context and, second, because no matter how thoughtful and well-reasoned the critiques may have been when first offered, [*PG559]they seem not to have slowed the drive toward a “cost-benefit state.”16 Indeed, with international action on climate change currently being thwarted by the United States at least in part due to arguments premised on CBA,17 one sensibly might ask whether we are headed toward not just a cost-benefit state, but a cost-benefit globe. If so, then as this Essay will explain, Speth’s hope for the emergence and assertion of a “new consciousness”18 may face legal and intellectual impediments that are even stronger than he fears.

I.  Cost-Benefit Analysis of Climate Change Policies:
A Brief Example

Climate change implicates the economic, social, cultural, and religious practices of all people in all regions of the world. Broadly speaking, it challenges both the established patterns of living, moving, eating, and consuming that prevail in the developed world and the aspirations of the developing world to achieve the same.19 More specifically, climate change emphasizes the need to shift to a decarbonized energy infrastructure at some point in time well before humanity otherwise would tap and combust the last remaining oil and coal reserves. Following the most recent suite of consensus scientific and economic reports from the Intergovernmental Panel on Climate Change (IPCC),20 few seriously question this need. Debate instead focuses primarily on issues of timing and the degree and nature of government involvement that is desirable in order to bring about the requisite transformation.

In Speth’s view, the necessary changes can occur on a global level only with the active leadership of the United States: “Not only are we Americans a big source of problems (as with climate change), but we are also essential to cooperative solutions. . . . The world needs a United States that leads by example and diplomacy, with generosity and compassion.”21 Unfortunately, with the exception of its appropriately lauded role in the effort to protect and restore the earth’s ozone layer, Speth believes that the United States has “dragged its feet on [*PG560]just about every international environmental treaty that has been suggested.”22 In the climate change context, the most dramatic illustration of such recalcitrance occurred when the U.S. Senate passed, by a vote of 95–0, the Byrd-Hagel Resolution, which voiced objection to any climate treaty that did not include binding emissions targets for developing nations on the same schedule as developed nations.23 Faced with such domestic resistance, the Clinton Administration accomplished little in the way of actual progress toward reaching the emissions targets set by the international community in the Kyoto Protocol.24 More recently, President George W. Bush’s decisive withdrawal of the United States from the Kyoto Protocol seemed to deride the very notion of international environmental lawmaking.25

In deciding to abstain in this manner from multilateral discussions concerning what many regard as the most important policy issue facing the world today, U.S. officials may have taken comfort from various economic analyses that purport to reveal the Kyoto Protocol to be a welfare-decreasing policy measure.26 Lomborg, for instance, has relied on climate changes studies by Yale economist William Nordhaus27 to conclude that “the total and long-term damages from [*PG561]emitting an extra ton of carbon today is the equivalent of $7.5[0].”28 Weighed against the costs of cutting emissions, this figure implies that a “4 percent cut in 1995 [emissions levels] is the optimal carbon reduction for the globe,”29 a level far more modest than the Kyoto Protocol’s aggregate cut of 5.2 percent below 1990 emissions levels.30 Indeed, based on Nordhaus’s calculations, Lomborg concludes that, “[d]espite our intuition that we naturally need to do something drastic about . . . global warming, economic analyses clearly show that it will be far more expensive to cut CO2 emission radically than to pay the costs of adaptation to the increased temperatures.”31

Lomborg’s language requires unpacking. At bottom, he is arguing that the economic benefits of continued fossil-fuel combustion more than outweigh the human, physical, agricultural, and ecological costs that would be averted by restricting global greenhouse gas emissions to the levels called for by the Kyoto Protocol. Viewed in the aggregate, Lomborg asserts that the present value of all harms expected to be caused by the current level of emissions is $4820 billion.32 Initially, this may seem to be a staggering sum requiring swift and decisive action, but Lomborg also claims that the optimal policy of emissions reduction will cost $4575 billion to implement, thereby resulting in a net gain of only $245 billion.33 Any more stringent policy, moreover, will result in a net loss to society, as emissions reductions will have been undertaken at a cost greater than the harm that they avert.34 Thus, Lomborg argues that the world will be best off if we restrict emissions merely to the level that is implied by adoption of the ideal carbon tax of $7.50 per ton.35

Lomborg’s study, which depends principally on work by Nordhaus, is only one example of CBA in the climate change context. As one [*PG562]might expect, results from other studies vary, sometimes considerably. For instance, estimates of the direct benefits of reducing greenhouse gas emissions levels range from as little as $5 to as much as $125 per ton in 1990 U.S. dollars.36 Conversely, estimates of the price necessary in order to achieve the emissions targets of the Kyoto Protocol range from $20 to $400 per ton of carbon.37 Among the causes of such wide variation are the base level assumptions included in the models regarding population, economic growth, and anticipated future emissions levels—all matters about which researchers legitimately may disagree.38 In addition, although Lomborg’s figure includes only direct benefits, increasing attention also is being paid to possible ancillary benefits produced by climate change policies such as the incidental reduction of other air pollutants through emissions abatement, including sulfur dioxide, nitrogen oxides, volatile organic compounds, and particulates.39 Current estimates of such benefits range up to an additional $20 per ton, in 1990 U.S. dollars.40

In short, although CBA of climate change is widely conducted and at times influential, it nevertheless is capable of considerably divergent results even according to its own established practices. Moreover, as the next Part will explain, close inspection of the established practices of CBA gives rise to several more fundamental sources of concern regarding the propriety of its use in the climate change context.

II.  Four Limitations

The normative premise of CBA is that its use will lead society to maximize the efficiency with which it devotes resources to policy goals. When Lomborg claims to have identified the “optimal carbon reduction for the globe,” he is claiming to have considered all pertinent consequences of emissions-induced climate change and to have identified the level of emissions abatement that is “just right” for all concerned.41 In other words, he is claiming to have found not just an answer, but the answer to the climate change problem. Viewed in this [*PG563]light, CBA obviously constitutes a form of rationality in that it seeks to derive conclusions from the application of formal logical rules. Moreover, it aspires to be a form of comprehensive rationality in that it aims to countenance all consequences of climate change within a single rubric that is both consistent and complete.42 In theory, therefore, the results of CBA should afford no basis for serious objection or dissent, as all important considerations will have been identified, tabulated, and weighed in a manner that leads to the identification of a first-best social outcome. There will be particular winners and losers, of course, as costs and benefits are not necessarily distributed equitably by policy interventions that are designed to maximize efficiency. But, according to proponents of CBA, any “disinterested observer” will conclude that CBA has led to the “right answer” in terms of aggregate social welfare.43 Part II explains why this is not the case.

A.  Uncertainty

The CBA methodology presumes by its very nature the availability of good data and understanding regarding the magnitude of physical and economic effects of various climate change scenarios, as well as the probability that those scenarios will occur. Yet uncertainty pervades, we might even say defines, the climate change problem.44 We are unsure how much of observed warming is attributable to greenhouse gas emissions. We do not know with certainty what the size of the human population will become over the next century. We do not know what the ref[*PG564]erence case of economic growth will be. We do not understand and therefore cannot model how sulfate particles, water vapor, and clouds interact in the atmosphere to mitigate or to enhance warming effects. We do not know with any degree of precision how temperature increases will impact agriculture, forests, vector-related diseases, heat-related deaths, flood zones, coastlines, storm intensity levels, freshwater supplies, or species extinctions. Perhaps most troubling, we cannot begin to pinpoint the likelihood of catastrophic climate-related events such as the disintegration of the West Antarctic Ice Sheet, the shutdown of thermohaline circulation in the Atlantic, or the release of frozen methane deposits. In short, we do not know with anything other than an anemic level of confidence what will be the consequences of our atmospheric experiment.45

By providing a semblance of order and exactitude where none exists, therefore, the results of CBA implicitly obscure the severity of climate change uncertainties. Consider as a simple example Lomborg’s treatment of the human health toll threatened by climate change.46 To calculate potential climate-induced losses of human life, Lomborg relies upon Nordhaus’s calculation of human health impacts from climate change which in turn relies upon a study that is restricted to climate-related diseases.47 The IPCC meanwhile reports with high confidence that climate change also will result in an increase in heat-related deaths and illnesses due to more frequent heat waves, an increase in drowning, intestinal, and respiratory diseases due to more frequent flooding, and an increase in malnutrition due to reduced crop yields in certain regions of the world.48 Nevertheless, because these health effects have not been estimated with sufficient precision to calculate a damages function, Lomborg’s CBA simply pretends that they do not exist.

The uncertainties of climate change, moreover, go to our very ability to identify and understand categories of costs and benefits, not just our ability to measure them and to predict their likelihood. In this regard, CBA seems especially ill-equipped to address the threat of potentially catastrophic consequences of climate change. Although several scenarios have been discussed in the scientific literature in which climate change spins dramatically and irreversibly out of control, at present our understanding of such scenarios is limited simply to assessing their scientific plausibility, rather than estimating their empirical likelihood.49 Despite this limited knowledge, a specially-appointed committee of the National Research Council recently did conclude that “greenhouse warming and other human alterations of the earth system may increase the possibility of large, abrupt, and unwelcome regional or global climatic events.”50 Such abrupt climatic events could include “changes in coupled modes of atmospheric-ocean behavior, the occurrence of droughts, and the vigor of thermohaline circulation in the North Atlantic.”51 According to the committee, each of these events could be accompanied by ecological and socioeconomic effects of enormous impact.52 In sum, climate change poses real, but as yet non-quantifiable, catastrophic threats that must be considered as part of any complete policymaking process.

Substantially for this reason, a variety of international interests and voices have converged around the precautionary principle as an appropriate response to climate change, particularly with respect to its potentially catastrophic but currently inestimable risks.53 Although susceptible to numerous interpretations, the precautionary principle often is taken to hold “that if it is known that an action may cause profound and irreversible environmental damage which permanently reduces the welfare of future generations, but the probability of such damage is not known, then it is inequitable to act as if the probability is known.”54 In other words, it is inequitable to maintain nominal adherence to the notion of comprehensive rationality when the preconditions for its effective operation are absent. Instead, society should establish “safe minimum standards . . . for protecting Earth’s life-[*PG566][*PG565]support systems in the face of virtually inevitable unpleasant surprises.”55 In the climate change context, for instance, the safe minimum standards imperative would seem to require acting to restrict greenhouse gas concentrations in the atmosphere at least to a level that would eliminate the plausible threat of catastrophic scenarios.

Critics of the precautionary principle, on the other hand, argue that the unrecognized costs of behaving according to its dictates may sometimes turn out to be more significant than the harm sought to be avoided.56 Behaving with precaution, in other words, entails its own costs, both direct (e.g., compliance costs) and indirect (e.g., opportunity costs, iatrogenic risks) that must be considered alongside the cost that is being guarded against through precautionary action. For this reason, critics argue in favor of a decisionmaking framework that maintains fidelity to the notion of assessing consequences comprehensively, rather than resorting to a conservative position, such as the precautionary principle, that may turn out in individual cases to be suboptimal.57 Of course, in a decisionmaking environment of true uncertainty, in which outcome probabilities cannot be assigned to the possible consequences of an action or event, it is by definition unknowable whether the consequences of regulating (e.g., averting potentially devastating climate-induced events in the future) will outweigh the consequences of not regulating (e.g., foregoing the benefits of continued levels of greenhouse gas emissions).58 Comprehensive rationality therefore cannot be attained. Instead, a more pragmatic approach is required in order for decisionmaking to continue to function at all when true uncertainty is present.59

In raising these objections, critics of the precautionary principle seem to have assumed that the only meaningful way in which to think about and react to potential hazards is to compare their likelihood and magnitude against the cost of averting them; in other words, to evaluate the consequences of action or inaction.60 From this perspective, resort to the precautionary principle appears to be an abandonment of the quest to maximize some overall desideratum solely in light of expected consequences. But, of course, this brand of consequentialism does not exhaust the universe of respectable and widely-held moral theories nor, indeed, is it necessarily the best-suited theory to govern decisionmaking under true uncertainty. Non-consequentialist theories, for instance, might begin with the existence of an inalienable right held by each generation to the benefit of a minimally stable climate and a corresponding duty on the part of each preceding generation not to act in a manner that jeopardizes the baseline level of climate stability. On this account, the existence of a plausible worst-case scenario in which climate stability is severely and irreversibly disrupted would in itself constitute a sufficient reason for restricting greenhouse emissions at least to a level in which the worst case scenario no longer seemed plausible—a course of action that seems precisely to mirror the advice of the precautionary principle.

Seen from this light, the precautionary principle is neither “incoherent”61 nor “indeterminate,”62 as critics claim, but rather reflects a moral theory that is capable of rational and internally consistent application even under conditions of true uncertainty. CBA, in contrast, must revert under true uncertainty to devices such as (1) the minimax principle, which shares more conceptual ground with the precautionary principle than with the maximizing aspirations of CBA;63 (2) Monte Carlo procedures and other statistical methods for generating hypothetical distributions of unknown probabilities, which can help to identify those policy prescriptions that would prevail under various possible states of the world, but which cannot eliminate underlying uncertain[*PG568][*PG567]ties;64 or (3) a Bayesian approach, which attempts to get the consequentialist ball rolling simply by assigning, more or less arbitrarily, a prior subjective belief to the unknown probability.65 Because none of these approaches truly overcomes the problem of uncertainty, none is able to fulfill CBA’s normative promise that its outputs ensure the maximization of aggregate welfare. Instead, decisionmakers are left to rely on a conviction that their methodological devices provide better outcomes than less formulaic approaches to human judgment, even though at their heart the devices contain a similar but much less transparent leap of faith.

Significantly, as Daniel Farber notes in a provocative recent discussion of complexity theory, there are important reasons to suspect that some environmental risks may not conform at all to the well-behaved patterns presumed by statistical uncertainty procedures such as those frequently used in conjunction with CBA.66 As Farber describes, “[t]he problem is not simply that we are in the early stages of scientific investigation but also that many environmental issues involve complex dynamic systems with nonlinear properties.”67 Under such conditions, scientists cannot and should not rule out the possibility that the uncertain system will behave according to what are known as power law distributions, in which “there is no [meaningful] ‘average’ event,” but rather “simply many small ones, a few larger ones, and occasionally extremely large ones.”68 Significantly, our current understanding suggests that climate change may represent just such a chaotic system. Statistical uncertainty procedures, in contrast, typically presume a world that is incapable of the “nasty surprises” entailed by power law distributions.69 Thus, rather than adopting the type of precautionary approach that is counseled by the implications of complexity theory,70 CBA instead seems to adopt a process of highly sophisticated, but potentially groundless, statistical roulette.

[*PG569] Nor do alternative methods of dealing with true uncertainty in the CBA framework overcome these limitations. Somewhat in line with the Bayesian approach, for instance, Nordhaus’s model includes a cost estimate of the risks of catastrophic events that was derived using “a survey of experts” who were asked to respond to the following question:

Some people are concerned about a low-probability, high consequence output of climate change. Assume by “high consequence” we mean a 25 percent loss in global income indefinitely, which is approximately the loss in output during the Great Depression . . . . What is the probability of such a high-consequence outcome . . . if the warming is 3 degrees C in 2090 . . . ?71

Using the probability estimate generated by responses to this survey, Nordhaus calculates the global “WTP [willingness-to-pay] to avoid catastrophic risk”72 by multiplying the likelihood of a catastrophe by its dollar equivalent—made tractable by the fact that Nordhaus has defined catastrophe as a percentage loss of global income—and adjusting the result upward to reflect society’s general aversion to risk.73 In this manner, Nordhaus concludes that the world currently is willing to sacrifice only one percent of global GDP in order to avoid the consequences of a catastrophic global warming event.74 As noted above,75 this amount, combined with estimates of other harmful consequences of climate change, translates into only a very modest optimal reduction of global carbon emissions—far less than would be required under a non-consequentialist, precautionary approach that sought to eliminate the possibility of catastrophic climate-change scenarios altogether.

Suppose, though, that one of the generic “high consequence” scenarios described by Nordhaus actually eventuates, in which case human suffering, death, and other adverse consequences deemed equivalent to “a 25 percent loss in global income indefinitely” befalls future generations.76 Will members of those future generations believe the decisionmaking process that led us to pose this threat was adequately protective of their interests? Is it sufficient that some number of experts [*PG570]will have assigned a purely subjective probability figure to currently inestimable but scientifically plausible worst case scenarios? Is it sufficient that the global willingness-to-pay to avoid such scenarios will have been calculated in a manner that avoids asking any human directly how much they would be willing to sacrifice in order to eliminate a risk of devastating magnitude on future generations? What are the reasons for assuming that direct human responses to such a question would be less reliable than a method of decisionmaking, such as CBA, that studiously avoids addressing matters of right and responsibility? What are the reasons for treating catastrophic losses indistinguishably from an agglomeration of trivial ones, albeit with a slight upward adjustment for some decontextualized notion of risk aversion?

In light of questions such as these, CBA proponents frequently emphasize that the results of their analyses are intended merely to provide one input into the overall decisionmaking process and that policymakers therefore remain free to give the results of CBA as much or as little weight as they desire. This reassurance is unsatisfactory, however, for it ignores the cognitive lure of comprehensive rationality. CBA offers the appearance, if not the reality, of certitude and objectivity. As Laurence Tribe observed over thirty years ago, “[s]uch ‘end-result’ theories have great appeal, for the notion of maximizing some desired end may seem the very essence of rationality.”77 Moreover, by generating estimates of the “optimal carbon reduction for the globe,” CBA proponents like Lomborg quite clearly imply that alternative, more stringent reduction policies are suboptimal and therefore require significant justification by their proponents.78 Nevermind that the supposedly optimal figure has failed to account for a host of possible consequences about which we currently lack sufficient understanding to model, including some consequences that may jeopardize the very ability of human civilizations to continue to exist.

B.  Valuation

As noted already, CBA seeks to do more than simply generate a systematic list of the pros and cons of climate change or of efforts to mitigate or adapt to climate change. In addition, CBA seeks to provide a common metric for them, such that policymakers need only run a spreadsheet in order to determine the optimal policy intervention into the carbon economy. Obviously, then, the policymaker and her con[*PG571]stituent public must have a great deal of faith in the valuation methodologies used in order for CBA to generate democratically acceptable results. For good reason, however, CBA practitioners have been unable to come up with widely agreeable methods for determining the monetary value of non-marketed goods such as the Great Barrier Reef, a child’s life, or an endangered species. This Section will address only a few of the philosophically and methodologically troublesome aspects of the valuation exercises typically utilized in conjunction with regulatory CBA. These problems, however, cut to the very core of CBA’s appropriateness for use in policy contexts such as climate change.79

Recall Speth’s hope that humanity will resolve to ensure “the protection for their own sake of the living communities that evolved here with us.”80 If such a commitment does develop as part of Speth’s hoped-for cultural transformation, it will reflect the judgment that other life forms are capable of holding interests of their own and are, therefore, worthy of respect and care over and above the level that might be merited solely according to an assessment of their value to humans. CBA, on the other hand, is premised on a liberal market conception in which only humans—and more specifically only individual humans—are capable of holding interests. For that reason, the value of non-human life forms is acknowledged only to the extent that humans value them. This distinction is highly significant in light of our growing understanding that climate change threatens the rapid extinction of literally thousands of non-human species.81 In order for such a massive potential loss of biodiversity to register as a cost in the view of the CBA practitioner, humans presently must demonstrate some credible commitment to the preservation of species.

One method of demonstrating such a commitment is through the use of contingent valuation surveys in which researchers attempt to elicit the amount that individuals would be willing to pay to support public goods such as the preservation of an endangered species, assuming that such goods were capable of being traded through market [*PG572]transactions.82 Using these hypothetical market instruments, researchers hope to derive an accurate estimate of the individual’s “true” valuation of the endangered species. As Ilana Ritov and Daniel Kahneman note, however, the estimates derived from such surveys are better understood as expressing an attitude regarding “the perceived severity” of the need to preserve the public good, rather than actually being “motivated by the good that is to be acquired.”83 That is, individuals often interpret the contingent valuation instrument as a strength-of-opinion survey, rather than a market transaction of the form that researchers hope to simulate. This account helps to explain, for instance, why respondents sometimes do not vary their stated willingness-to-pay as the purported amount of the public good being preserved is changed, even by orders of magnitude.84 It also helps to explain why significant numbers of respondents refuse to participate in the surveys at all or state willingness-to-pay amounts that researchers deem unrealistically large.85 In essence, respondents are attempting to make a political statement through the market medium of the survey instrument—a medium, incidentally, that is being forced on them by the experimental researcher.

Notably, contingent valuation practitioners typically dismiss responses that do not conform to the researcher’s expectations of what a well-behaved preference ordering for public goods should resemble.86 Dismissal is necessary because such “protest votes” reflect views that the CBA methodology simply cannot countenance, such as the beliefs that not everything valuable is necessarily monetizable, that explicitly trading-off sacred categories in CBA may unwittingly diminish the sacredness of those categories, and that some judgments should be made collectively through the democratic process, rather than individually through decentralized market transactions. Because the CBA methodology is premised on a contrary set of foundational assumptions, researchers must sanitize the responses that they gather from contingent valuation surveys in order to preserve the methodology’s nominal coherence. As Frank Ackerman and Lisa Heinzerling describe, however, much is lost in the process of sanitization: “By narrowly constraining the conditions under which responses in contingent valuation surveys will be counted, the researchers have imposed a new literacy test for participation, demanding that voters who want their votes to be counted frame their answers in an artificially quantified vocabulary.”87

Contingent valuation surveys represent what is known as a stated preference methodology. That is, researchers ask individuals to state the amount that they would be willing to pay to preserve a public good, but do not actually require individuals to make good on their promised contributions. Revealed preference methodologies, on the other hand, attempt to infer an implicit valuation of some public good through the actual market choices of individuals.88 Environmental Protection Agency economists, for instance, recently examined a parent’s willingness to pay premium prices for organic baby food as an indirect measure of the parent’s valuation of her infant’s life—a value that the economists argue can then be used by regulators to set government health and safety standards for children more generally.89 Such methodologies are thought to produce especially realistic valuations because, unlike hypothetical survey instruments or political referenda, the choices examined will have occurred under the disciplinary constraints of actual markets with real resources at stake. Moreover, such methodologies may offer the only credible method of deriving willingness-to-pay measures for saving human lives, given the seemingly in[*PG574][*PG573]surmountable problems facing contingent valuation studies that attempt to depict human lives as being explicitly subject to market trading.90

In part for these reasons, when evaluating environmental, health, and safety regulations, U.S. federal agencies typically use a value of life that has been empirically derived through observations of labor markets.91 The central premise of this methodology is that, other things being equal, workers will demand a higher wage—or a “compensating wage differential”—in order to accept a higher level of occupational fatality risk. Large data sets have been analyzed with this assumption in mind and a consensus view among CBA practitioners seems to have developed that the implicit value of life revealed by wage premiums is somewhere between $3 and $7 million.92 Such figures routinely are used to compute the monetized benefits of proposed regulations that aim to save human lives, affording in the process a ready basis for comparison against the purported economic costs of regulation. As noted above,93 these figures are said to provide an especially reliable basis for setting public safety standards because they represent the actual tradeoffs that individuals make between money and life when determining their own tolerances for risk.

Even on CBA’s own terms, however, the reliability of the wage-risk premium studies depends critically on the tightness of fit between subject labor markets and the competitive market ideal. At least two of the most important assumptions of the market ideal—risk awareness and free mobility among laborers—lack a strong empirical basis when one examines actual market operations.94 According to one study, for instance, the implicit value of life revealed by wage-risk interactions appears to be several million dollars higher for union workers than for non-union workers, holding constant other significant variables.95 Union membership, in other words, seems to be a strong determinant of the size of the compensating wage differential—a wage-risk effect that more closely captures bargaining power than willingness-to-pay and that, therefore, suggests a less normatively attractive basis for policymaking than has been supposed by CBA practitioners. By failing to control for the influence of labor market structure in this manner, previous wage-risk premium studies may have generated misleading results—an unsurprising outcome given the sheer number of economic factors that influence wages other than fatality rates and that therefore threaten to confound study results.

A recent study in the Journal of Risk and Insurance attempts to get around these kinds of data problems by focusing on changes in industry accident and fatality rates, as well as wage rates, over a period of seven years.96 The researchers reasoned that, if the compensating wage differential thesis is correct—that is, if labor markets really are characterized by sufficient risk information and bargaining equity to reliably correlate compensation with occupational risk exposure— then we should expect to observe a relationship between accident rates and wages over time, after controlling for other important variables.97 The researchers in this study, however, observed no such relationship, despite using a Bureau of Labor Statistics data set that covered the most consistent and significant change in occupational injury and fatality rates in the last quarter century.98 Reflecting on the various methodological problems associated with the wage premium literature, the researchers concluded by observing: “[W]e are frankly surprised that estimates of the value of a life based on occupational fatalities and wages have ever been used.”99

Many of the data problems found in the wage-risk premium literature are caused by the fact that the United States has managed to drastically reduce its occupational hazard levels over the course of the past century—an accomplishment, incidentally, that occurred through New Deal and other protective legislation that did not depend on CBA for its justification.100 For this reason, the few remaining labor segments of [*PG576][*PG575]the economy that exhibit an occupational mortality rate high enough to even plausibly support the wage-risk premium methodology tend to be segments populated by those with the least social, economic, and political capital.101 CBA practitioners seek to extrapolate from the circumstances of these less well-off individuals a value of life that will apply across the entire range of risk contexts that confront government policymakers. No allowance is made for the possibility that better-off individuals might wish to devote resources to public protection at levels higher than male, blue-collar workers are able to devote in their private workplace struggles. Thus, through the wage-risk valuation methodology, CBA practitioners enshrine into public policy aspects of our economy that many might regard as regrettable side-effects of market capitalism, rather than desiderata in their own right.

Such problems are in many respects magnified in the global policymaking context, where potential distortions due to divergent resource endowments are likely to be even more pronounced. Nevertheless, many analyses of the costs and benefits of climate change do in fact distinguish between the value of lives on a country by country or a region by region basis. One prominent analyst, for instance, first divided countries into high income, middle income, and low income categories, and then assumed a value of human life of $1,500,000, $300,000, and $100,000, respectively, for the three categories.102 Although purportedly a value-neutral approach based only on varying willingness-to-pay calculations for the three regions, this method of valuation nevertheless entails a willingness to invest resources to protect Americans at a level fifteen times higher than sub-Saharan Africans.103 One might think that such an approach has it exactly backwards. After all, the majority of human-induced greenhouse gas concentrations currently in the atmosphere are traceable to the industrial activities of the developed world, while the majority of premature deaths and other adverse health effects expected from climate change threaten to strike the [*PG577]developing world.104 Speaking in terms of moral responsibility, then, rather than merely in terms of credits and debits in a greenhouse ledger stripped of human context, one naturally might think that the interests of the developing world should count at a rate greater than those of the developed world.105 One would, however, have reached a conclusion opposite to the approach often implicitly taken by CBA.106

Each of the valuation approaches considered in this Section assumes that the proper measure of a good’s worth is the amount that individuals would be willing to pay in order to preserve the good in a well-functioning market. A different approach altogether would be to assume that valuations are more reliably captured through society’s willingness to act collectively in order to preserve the threatened good, including, for instance, through protective legislation. As Sagoff observes, unlike the comprehensive rationality of CBA, “[t]his kind of rationality depends on the virtues of collective problem solving; it considers the reasonableness of ends in relation to the values they embody and the sacrifices we must make to achieve them.”107 To be sure, a monetary value will be implied by the level of resources chosen to be committed to a good’s preservation, but this value will not have driven the initial selection of public policy. Rather, the value will be merely an ancillary effect of a policy choice that instead was premised on social values, explicitly discussed and mediated through a democratic decisionmaking process.

In the climate change context, therefore, an entirely different approach to policymaking is to simply identify the expected costs of implementing a given proposal, so that observers can have a clear sense of the amount that must be given up in order to achieve a policy goal that, at least in part, is to be considered and evaluated through the decisionmaking process itself. As Lomborg notes, economic studies suggest that implementation of the Kyoto Protocol would entail an annual cost [*PG578]of somewhere between two and four percent of developed nations’ gross domestic product (GDP).108 Armed with this knowledge, decisionmakers and their constituent publics need only ask themselves whether incremental progress toward stabilizing greenhouse gas emissions is worth foregoing two to four percent of GDP. The benefits of the Kyoto Protocol on this approach are not given by a contrived monetization of the averted adverse effects of climate change, but rather by the amount society collectively is willing to sacrifice in order to purchase a margin of safety under the Protocol’s dictates. Monetization may or may not help to inform such a decision, but it cannot replace it.

C.  Discounting

Given the atmospheric persistence of greenhouse gases, relevant time frames for climate change CBA stretch into the hundreds of years. As a result, many of the benefits of climate change mitigation policies consist of deaths and illnesses that otherwise would afflict generations far into the future. Faced with these time horizons, researchers must articulate some method of comparing the impacts of atmospheric warming across time, even if it is the simple method of assuming that a life today is “worth” the same amount as a life tomorrow.109 Practitioners of CBA typically attempt to normalize the intertemporal distribution of costs and benefits through the use of an exponential discount rate, which is the logical converse of compounding interest.110 Accordingly, lives that will be lost in the future due to increased carbon loading of the atmosphere typically are not valued by CBA at the going rate for lives today. Instead, they are discounted to a present value, which often results in numbers that are vanishingly small: one million lives discounted over 145 years at ten percent are “worth” less than a single life today.111

The moral basis for discounting in this manner is rarely articulated. Sometimes it is argued that discounting future harms is appro[*PG579]priate because future generations will be richer and therefore better able to handle the harmful consequences of climate change.112 But this is not an argument that we should discount something because it will occur in the future; it is an argument that we should discount something because it will happen to better off individuals.113 As a matter of distributive justice, we may agree with such an argument, but it requires a much different implementation device than numerical discounting. Moreover, if we use the standard income-dependent or income-influenced economic methods of valuing life,114 then an anticipated rise in future incomes suggests that future lives are more worth saving than present lives, not less.115 Therefore, rather than discount future benefits of climate change mitigation, on this view, we should magnify them. The fact that this result is counterintuitive suggests that the normative justification for discounting has been insufficiently established.

A stronger defense of discounting invokes the economic concept of opportunity costs. By discounting future lives to a present value, the argument goes, governments ensure that climate change mitigation policies produce a “return on investment” at least as high as the discount rate that is being utilized. In that manner, regulatory expenditures can be avoided that might have been invested more profitably in other social projects or in capital markets. Strangely enough, therefore, it seems that future generations will be better off if we discount the value of their lives, because we will leave them with a resource base that has taken advantage of the best available investment opportunities. Along these lines, Lomborg concludes that “a sensible investment with a good yield will leave our descendants and future generations of poor people with far greater resources, and this is probably a far better way of looking after their interests than investing in low-yielding greenhouse gas reductions.”116

[*PG580] This line of reasoning appears compelling until one recalls that the benefits of climate change mitigation consist at least partially of human lives that might otherwise have been lost in the future. The normative case for discounting—that it will leave future generations with a more valuable stock of resources—is in considerable tension with the fact that some members of those future generations are sacrificed in order to make the very alternative investments that are supposed to inure to their benefit. Obviously, if one takes seriously the moral injunction that human lives should not be used instrumentally without the consent of the sacrificed,117 then discounting demands still further justification. All the CBA apologist demonstrates by appealing to opportunity costs is the fact that a life lost in the future may be compensated for at lower cost than a life lost today. The decision actually to sacrifice the life—and thereby to bring about a situation in which compensation becomes relevant—remains an entirely separate, and philosophically more problematic, matter.

Consider a thought experiment to help elucidate this point. Suppose that our society was to set up a trust fund from which equal payments would be made to surviving family members or representatives on account of the death of any individual member of the society. In light of the differing life expectancies that society members would have, the managers of our trust fund quite properly could earmark a lesser amount of money for younger members than for older members. Likewise, they could set aside an even lesser amount for the not-yet-born. In both cases, the later-expected death of the individual would allow the lesser amount set aside to grow through alternative investment until payment eventually became due. Notice, though, that the more or less arbitrary fact of being born later in time does not support a judgment that the future life actually is worth less than the current life, at least not in any sense that would justify the knowing sacrifice of one life over the other.

To see why, suppose that some unforeseen societal emergency actually did necessitate the knowing sacrifice of a given number of lives. [*PG581]Would we agree that the youngest members of society automatically should be chosen simply because our trust fund accounting appears to show that their lives have the lowest current monetary equivalent? Suppose further that we could respond to the societal emergency by sacrificing a given number of the not-yet-born, perhaps by storing hazardous waste in a container that will leak in a hundred years and cause a predictable number of deaths. Are we clear that the time value of money by itself would justify taking those future lives over the currently living? Suppose finally that we could respond to the emergency either by sacrificing a given number of the not-yet-born or by lowering our current standard of living by two to four percentage points of GDP.118 Now, do we not think that our trust fund accounting is entirely beside the point?

The situation is even worse than this, for not only are we currently choosing to sacrifice future lives in order to preserve two to four percentage points of GDP, but we also are failing to ensure that resources actually are transferred to those future generations whose members we are sacrificing. To give the opportunity cost argument any practical significance, we would need some reliable intergenerational transfer mechanism to ensure that the resource base actually is expanded for the benefit of surviving members of future generations. As it is, however, we have inadequate means of knowing whether the resources intended to make future generations better off are being conserved as durable capital or whether instead they are being consumed for the fleeting pleasure of the currently living. National income measures such as GDP are notoriously bereft of information about non-marketed goods such as environmental and social capital. GDP simply measures the absolute flow of commodities through the economy over a given time period; it makes no effort to measure the stock of resources from which that flow ultimately originates or to otherwise determine whether the flow is sustainable.119 The CBA discounter may take comfort in the knowledge that, in some purely hypothetical sense, future generations may be better off for having been discounted. Future generations themselves may not feel as sanguine.120

[*PG582] On a more technical level, it also bears noting that our very decision whether or not to save natural resources for future generations influences the rate of return for capital which, in turn, forms the basis of the strongest argument that CBA practitioners have for discounting—the opportunity cost argument. But, of course, if the savings rate for natural capital in part determines the rate of return for all capital— if, in other words, the decision whether to conserve natural resources influences the size of the opportunity cost that supposedly determines whether or not to conserve—then invoking opportunity costs as a justification for discounting is, at bottom, nothing more than conceptual bootstrapping. After all, how does Lomborg know that climate change mitigation is a “low-yielding” investment?121 He knows this because he has implicitly compared it—through discounting—to a rate of return for capital that takes as given the fossil fuel economy with its laissez faire climate change policy. Put differently, the status quo of continued greenhouse emissions appears to be a better investment than abatement in part because the benefits of abatement have been discounted at a rate that assumes abatement will not occur. Such circularity may be tolerable for decisions of modest practical impact—in which the ultimate outcome might not be affected by the specification of a different reference case—but it hardly seems appropriate for addressing an issue with the enormity of climate change.

In part for this reason, some CBA proponents attempt to unhinge the discount rate entirely from the rate of return for capital, relying for public policymaking instead on the use of a “‘social’ discount rate” that is set according to the rate at which individuals seem to prefer the present to the future.122 At this point, the normative justification for discounting no longer depends on the opportunity cost argument, which at least purports to take account of the interests of future generations, but rather now is based on the bare assertion that “people are impatient,”123 and tend to prefer present consumption to future consumption of an identical item. CBA practitioners seek to replicate this impatience in public policymaking by discounting with similar haste the value of future benefits that will occur from proposed activities such as climate change mitigation. In that manner, [*PG583]public resources will be devoted to the future only at the level that individuals seem to prefer when delaying their own personal gratification. Even proponents of CBA sometimes seem sheepish when offering this defense of discounting. Famed economist Robert Solow, for instance, argued that although as an ethical matter “no generation ‘should’ be favored over any other,” discounting nevertheless is appropriate as a “concession to human weakness.”124

Others argue that the social discount rate should be set according to openly normative judgments about intergenerational equity.125 They recognize that this “human weakness” justification for discounting has strength, if at all, only in the intragenerational context, in which the effects of the public policy decision will be felt by the same individuals from whom the analyst is deriving the rate of time preference.126 These analysts typically argue that future benefits should be discounted at some positive rate because otherwise present generations perpetually will be forced to forego present benefits in order to achieve the same level of benefits later at a lower present cost. That is, “if invested, our resources are expected to grow at [a market] rate, so that if we forego spending and invest the money instead, we can save more lives in the future with the amount foregone today.”127 And, of course, the same logic will apply tomorrow, such that current consumption seemingly will be postponed indefinitely whenever CBA is forced to treat the interests of future generations pari passu with the interests of the present generation.

The problem with this argument is that it merely demonstrates that the present generation might prefer to avoid unduly sacrificing for future generations; it does not explain why discounting is the [*PG584]proper method for implementing such a preference.128 Indeed, upon reflection, the choice of discount rate often seems downright trivial in comparison to the underlying distributive judgment that the discounting methodology is attempting, however awkwardly, to subsume. Consider, for example, the following question: “At what rate should the benefits of future timber harvesting be discounted when using CBA to determine the appropriate current level of harvesting from a particular forest?” If one’s response is some variation of, “at whatever rate results in a sustainably managed forest,” then one is in agreement with the view that distributive judgments precede the choice of discount rate. To be sure, the example uses a context in which the distributive judgment is relatively tractable and even appears to admit of an “optimal” outcome: Renewable resources should be harvested at a level that ensures maximum sustainable yield.129 Nevertheless, the same conceptual framework underlies all issues of intergenerational resource distribution, including the intertemporal distribution of our atmosphere’s capacity to absorb persistent greenhouse gases. Discounting cannot resolve these issues; it can only obscure them.

In a recent article, Dan Farber seeks to elide this difficulty by assuming that “we [already] have determined by some means—whether cost-benefit analysis or pure intuition—how much of our current resources we are willing to sacrifice to obtain future environmental benefits.”130 With that assumption in place, Farber then demonstrates that the benefits of an environmental investment inevitably will decline over time and thus, he argues, “discounting environmental benefits is not only ethically permissible but often unavoidable.”131 Farber only has shown, however, that a discount rate is implied by any social decision involving a finite resource that must be distributed over an infinite time horizon. He has not shown that discounting is the proper method for resolving the initial question of how much of the resource to distribute to future generations. Indeed, that is the very decision that Farber treats as exogenous to his account. The [*PG585]problem, of course, is that it is also the decision that proponents of CBA seek to resolve in part through the use of discounting.132

To sum up, a third limitation of the use of CBA in the climate change context is that some components of the disutility caused by environmental harms simply cannot be priced and incorporated into cost-benefit calculation without first asking a threshold question. This threshold question concerns the intergenerational distribution of natural resources, biodiversity, and other environmental goods, including, critically, the carbon-absorbing potential of the earth and its atmosphere. Just as each distribution of resources among the presently living gives rise to a unique efficient allocation of resources, each distribution of resources among different generations also entails a separate efficient equilibrium.133 Accordingly, the question of how natural resources should be distributed across time remains fundamentally an ethical question—one that will imply a discount rate, but not be determined by it. Only by directly confronting the ethical question in that manner can we hope to honor the second of Gus Speth’s two central environmental obligations, our duty of “trusteeship of the earth’s natural wealth and beauty for generations to come.”134

D.  Vision

As we have seen, CBA attempts to subsume a variety of important ethical and practical judgments within the rubric of a purportedly straightforward efficiency-maximization exercise. Decisions about how to behave in the face of true uncertainty concerning potentially devastating consequences of human action are masked by statistical uncertainty procedures that, despite their technical sophistication, may depend on flatly unwarranted assumptions about the operations of the physical world.135 Society’s level of commitment to the preservation of species, ecosystems, and human lives is divined from real or contrived individual market decisions in a process that, at best, is beset by a variety of methodological problems and, at worst, fundamentally misconstrues the nature of the very public goods problem that demands collective judgment.136 Finally, difficult and unavoidable choices about how to distribute potentially massive costs and benefits among generations are elided through an elaborate mathematical fiction, the discount rate.137 At bottom, each of these features of CBA exhibits a common flaw, observed long ago by Laurence Tribe: “[Q]uantitative decision-making techniques . . . reduce[] entire problems to terms that misstate their underlying structure, typically collapsing into the task of maximizing some simple quantity an enterprise whose ordering principle is not one of maximization at all.”138

Indeed, this tendency to obscure difficult judgments through mischaracterization or misspecification seems to follow from the very structure of the CBA project. As intimated above,139 CBA practitioners typically treat individual preference formation as a process that is entirely exogenous to their analysis. Thus, the many internal compromises, hesitancies, and regrets that characterize individual behavior and choice never register a blip on the practitioner’s radar. Instead, whatever preferences individuals seem to reveal through their market behavior are taken to be the best measure of true “wants” or “desires” and, therefore, also are taken exclusively to provide the valuation inputs that in critical part determine the policy outputs of CBA. In practice, this means that CBA proponents have no means of evaluating the desirability of different market or policy outcomes, other than to examine whether or not the given set of fixed individual preferences are being satisfied in the most efficient manner possible. Under this approach, then, preferences may be maximized (assuming away the conceptual and methodological problems already discussed), but they cannot be scrutinized.

There is a nearly tautological aspect to this approach—what is good for the individual is what she chooses and what she chooses is what is good for her—and a deeper problem as well. The deeper problem is that the neoclassical economic project, as exemplified by CBA, excludes from consideration the very feature that many philosophers identify as uniquely constitutive of humanity. That is, what distinguishes us from other animals and makes us distinctly human is not the ability to satisfy our goals, but the ability to reason and deliberate about the content of those goals. Indeed, the very project of life [*PG587][*PG586]might be said to consist of shaping, revising, and reflecting on one’s goals or, put differently, on what one wants to want.140 A final concern with CBA, then, is that the methodology seems ill-suited to grapple with this central project of life.

In very simplistic terms, CBA asks whether diverting resources from current patterns of production and consumption toward climate change mitigation would produce a net enhancement of social welfare. The benefits of mitigation—avoided human deaths, preservation of ecosystems, survival of other species, and so on—are compared to the opportunity cost of whatever utility would have been provided by the foregone combustion of fossil fuels. The reference case for defining and measuring utility in this process remains unequivocally focused on the status quo pattern of production and consumption. No allowance therefore is made for the possibility of individuals to adapt their preferences in light of changed circumstances, to acknowledge the moral responsibility created by climate change, to accept as being well and just any newly imposed constraints on their harmful activities, or simply to get on with the project of life by deriving utility in new but not necessarily inferior ways. In short, no allowance is made for the possibility of individuals to grow.141

The Intergovernmental Panel on Climate Change (IPCC) in contrast utilizes a scenario-based approach in its important consensus reports regarding the causes, consequences, and policy implications of climate change.142 That is, it develops a set of richly described future courses of development that humanity might take in order to consider the ecological and socioeconomic ramifications of each. As the IPCC notes, this approach,

assists in the understanding of possible future developments of complex systems. Some systems, those that are well understood and for which complete information is available, can be modeled with some certainty . . . and their future states predicted. However, many physical and social systems are poorly understood, and information on the relevant variables is so incomplete that they can be appreciated only through intuition and are best communicated by images and stories.143

Thus, unlike a system of comprehensive rationality, which by its nature must hold some important variables constant in order to avoid generating inconsistent results, the IPCC’s multiple scenarios approach enables inspection of the full range of determinants of the climate change problem. Moreover, each of the scenarios is based on scientifically credible assumptions and is designed to include parameters that are internally consistent within individual scenarios.144 Thus, at least theoretically, each scenario represents a future that actually is possible for humanity to achieve, given emergence of the requisite social, economic, and political forces.

The IPCC’s scenario-based approach is preferable to CBA in the climate change context precisely because it provides a foundation for discussing what we want to want. As Tribe observed,

the whole point of personal or social choice in many situations is not to implement a given system of values in the light of the perceived facts, but rather to define, and sometimes deliberately to reshape, the values—and hence the identity— of the individual or community that is engaged in the process of choosing.145

Questions of that nature depend not on cost-benefit balancing, but rather on the will of the global community to conceive of and to realize a transformation of culture toward some shared ideal. At present, like most important aspects of the climate change problem, it is uncertain whether such a transformation can occur in the face of CBA’s contrary but analytically powerful affirmation of the status quo. As Speth argues, however, “[w]e must hope that the answer is yes, for contrary to the [*PG589][*PG588]conventional perspective, it is business as usual that is utopian, whereas creating a new consciousness is a pragmatic necessity.”146


Red Sky at Morning begins with a recounting of Gus Speth’s experience serving on President Carter’s Council on Environmental Quality and helping to draft one of the first significant reports to identify the problem of climate change for a broad political and public audience.147 In his foreword to the report, issued in 1981, Speth urged his readers to consider the problem one of immediate and serious global import, not only for the sake of the presently living, but also for the sake of other generations and other forms of life:

One imperative we share is to protect the integrity of our fragile craft and the security of its passengers for the duration of our voyage. With our limited knowledge of its workings, we should not experiment with its great systems in a way that imposes unknown and potentially large risks on future generations. In particular, we cannot presume that, in order to decide whether to proceed with the carbon dioxide experiment, we can accurately assess the long-term costs and benefits of unprecedented changes in global climate.

Whatever the consequences of the carbon dioxide experiment for humanity over the long term, our duty to exercise a conserving and protecting restraint extends as well to the community of life—animal and plant—that evolved here with us. There are limits beyond which we should not go in disrupting or changing this community of life, which, after all, we did not create. Although our dominion over the earth may be nearly absolute, our right to exercise it is not.148

This passage, and the failure of the global community to heed its words over the past two decades, illustrate well the pitfalls of aspiring to comprehensive rationality within a context as vast and polycentric as global climate change policymaking. CBA, by its very design, is ill-equipped to grapple with deep scientific uncertainty, to reliably value human life and respect the existence of other beings, to assess and [*PG590]honor the needs and rights of future generations, or to contemplate, discuss, and pursue as-yet-unrealized ways and modes of living. Without these capabilities, CBA offers only meager assistance to climate change policymaking.

?? ??