09.11.2016

Decision theory and the doom scenario of climate catastrophe

slide1Frost (1920)[1] Fire or Ice? In many Western countries, climate policy has become one of the most expensive policy areas, and the Paris Agreement suggests that the amounts of money spent on fighting climate change will increase dramatically. There are widely diverging views on the right measures to address climate change, however. Virtually everybody believes that the climate changes over time, and that humanity has some influence on the climate. Despite massive climate alarmism and a constant stream of doom scenarios, fewer of us believe that we heading towards climate catastrophe. How citizens will see this issue, will likely have an important influence on future climate policy-making. And how ambitious climate policies will be, is, in part, a function of citizens’ belief in climate catastrophe. Pascal’s Wager So, people have to decide whether to believe in climate catastrophe. Assuming they want to make a rational decision, can decision theory help them decide whether to support ambitious climate policy to avert catastrophe? The decision model developed by Blaise Pascal (1623-1662) would appear to offer an interesting approach. The model is known as Pascal’s Wager, and intends to answer the question whether a rational human being should believe in God. In Pascal’s view, this is a question nobody can avoid.[2] Pascal arrives at the conclusion that, in their own self-interest, people should believe in God. His reasoning proceed as follows:

  • There are two possibilities: either God exists or He does not exist. Under each of these two scenarios, a person can decide to believe in God or not believe in God.
  • Once the outcomes for each are compared, a rational person should believe in God. This is so because, if God does not actually exist, such a person will have only a finite loss while on Earth (the loss of the enjoyment of some pleasures, luxuries, etc.), but, if God does exist, he will receive infinite benefits in Heaven (the benefits of Heaven are enjoyed for an infinite period of time), and will avoid infinite losses (eternity in Hell).
Decision Matrix In Pascal’s Wager, the decision matrix looks like this:[3]   slide2Thus, we can now see that a rational person would choose to believe in God, because, if God exists, belief in God produces infinite gain (and disbelief produces infinite loss), and, if he does not exist, belief produces only a finite loss (and disbelief a finite gain). Note that Pascal assumes complete uncertainty about God’s existence. He, of course, also assumes that God’s existence implies the existence of Heaven and Hell, and that belief or disbelief has consequences in terms of where a person will spend his time in eternity (Heaven or Hell) and what that implies. The Dismal Theorem Atheists would reject Pascal’s assumptions about the possible existence of God, Heaven and Hell, and about the consequences of belief versus disbelief. Remarkably, however, Pascal’s logic has been applied by scholars to the issue of climate change. In 2009, a professor of economics at Harvard University published a mathematical version of Pascal’s Wager to argue that rational people should belief in climate catastrophe.[4] He posited a ‘very strong form of a “generalized precautionary principle” for situations of potentially unlimited downside exposure’ such as climate change in which cost-benefit analysis cannot be applied. On the basis of this principle, also known as the ‘Dismal Theorem,’[5] a rational agent should be willing to spend an theoretically infinite amount now to avert climate catastrophe in the future. At a more pragmatic level, the Harvard professor has suggested that people should be “a lot more open to at least considering undertaking serious mitigation measures (including, perhaps, geoengineering in the case of climate change).”[6] Anybody who disagrees would bear the burden of proof and persuasion. Criticism of the Dismal Theorem In line with the expectations of rational people, the Dismal Theorem was severely criticized. One of the critics was, surprise, a Yale economics professor.[7] The first critics, however, were two climate economists, one of whom first articulated a more limited version of a dismal theorem. They argued that the Dismal Theorem renders the attempts to reach consensus on the social cost and central tendencies of climate change obsolete, as all estimates are infinitely too small.[8] They also point out that the Dismal Theorem assumes that policy makers are risk averse, rather than risk-neutral. Rather than spend huge amounts on preventing climate catastrophe, they suggest that we could invest in developing the scientific understanding of the fundamental processes that might produce catastrophic impacts. More generally, they posit a potential for policy learning and mid-course corrections in the management of climate change. Further, they argue that there is no reason not to apply the generalized precautionary principle inherent in the Dismal Theorem to all social issues for which there is potential for catastrophe (in other words, virtually all big policy issues), but, if we do that, we will be unable to set priorities for distributing the planet’s finite resources in the social interest. ‘The economic tradeoffs would simply be undefined.’[9] The criticism of the Yale economist builds on this analysis. Recognizing that uncertainty is ubiquitous, he observes that the Dismal Theorem would hold not only in climate change but also in the areas of ‘biotechnology, strangelets, runaway computer systems, nuclear proliferation, rogue weeds and bugs, nanotechnology, emerging tropical diseases, alien invaders, asteroids, enslavement by advanced robots, and so on.’ Consequently, the Dismal Theorem would cause us to ‘dissolve in a sea of anxiety at the prospect of the infinity of infinitely bad outcomes.’ Emphasizing the step-wise, iterative policy-making process, he concludes confidently that ‘[a]s long as policy is not shut down, the world economy can avoid catastrophic outcomes.’[10] Pascal’s Wager Compared To the Dismal Theorem Pascal’s decision theory involves the same assumption as the Dismal Theorem: the loss resulting from an uncertain event in the future is infinite. In Pascal’s theory, the infinite loss is living in Hell, in the Dismal Theorem, it is climate catastrophe. But there are important differences between the two models. In Pascal’s Wager, the decision to be made (‘believe in God’) is a decentralized, private matter that is costless to the collectivity, except, possibly, for the time a person needs to spend to think the problem through. Each individual person needs to make the decision whether to believe or not believe, and each person’s decision is independent of the decisions of others. In the Dismal Theorem, however, the decision (‘prevent climate catastrophe’) by necessity is collective and requires enormous spending by society. Thus, unlike Pascal’s Wager, the Dismal Theorem involves a decision about the allocation of scare resources, which raises all of the issues typically associated with collective decision-making influenced by special interests. Further, in Pascal’s Wager, no evidence can be obtained about the existence or non-existence of God; although Pascal allows ‘the faces of the cards’ to be seen through the ‘scripture and the rest,’ this would not provide information about the probability of God’s existence. In the Dismal Theorem, to the contrary, there are all sorts of actions one could take that provide information on the probability of climate catastrophe. Promoting unbiased, disinterested climate science would be good start. Stimulating innovation in energy technology would be another useful action. Neither of these actions currently seem to have priority in policy-making, however. Proponents of the Dismal Theorem could insist that, no matter how good the science is and how fantastic innovation might be, there always is some residual uncertainty, however small, of an infinite loss. That is not a fact, it is a belief. It is the power of imagination; or, its rationalized version, ‘possibilistic’ (as opposed to probabilistic) risk assessment.[11] So, is there risk of climate catastrophe at some point in time? Pascal answered for the advocates of the Dismal Theorem: ‘by faith we know [its] existence.’ What Should A Disbeliever Do? Proponents of the Dismal Theorem invoke the possibility of infinite losses due to climate catastrophe to persuade opponents of ambitious climate policies, even if they may not suffer these losses during their life time. Apparently, this threat is deemed a sufficient ground for collective decisions to allocate massive resources to the fight against climate change. Logically, this fight should then be extended to combat also the natural causes of climate change that might result in catastrophe. Disbelievers can be conveniently disregarded as long as they do not prevent the adoption of the necessary policies. Pascal has other advice to disbelievers: ‘[L]earn your inability to believe, since reason brings you to this, and yet you cannot believe. Endeavour then to convince yourself, not by increase of proofs of God, but by the abatement of your passions (…), which are your stumbling-blocks.’ Similarly, in the area of climate policy, moralists and philosophers have argued that mankind in the West should temper its consumption, and give up amenities, such as traveling by airplane and eating beef, all in an effort to save the planet. Indirectly, proponents of the Dismal Theorem do the same by drawing resources away from all these other endeavors. The Centrality of Belief Proponents of the Dismal Theorem do not need to persuade everybody; they just need to persuade those in power who can make the relevant policy decisions. Of course, this still is a tall order. Any effective climate policy would need to be global in nature. In the absence of an effective world government, it is unlikely that countries around the world will adopt the far-reaching measures that would be necessary to eliminate or even substantially reduce the risk of climate catastrophe. Given the size of the investments involved and the interdependence of the policy measures adopted by nation states, ‘free riding’ and taking the lead in a ‘race to the bottom’ (or, maybe, lagging behind in some sort of ‘race to the top’) will likely be the dominant strategies. Is climate change then the ultimate instantiation of the ‘tragedy of the commons’? No, not quite. The folly of the Dismal Theorem is that, by necessity, it has to claim that it is a uniform theory of knowledge governing all human affairs. An erroneous decision by the whole world to spend all or most of its resources on fighting climate change, to the detriment of other worthwhile causes (including other risk management programs), would be the ultimate ‘tragedy of science-based collectivism:’ in its quest for supreme rationality, it destroys rationality and with it humanity. A comparison of Pascal’s Wager and the Dismal Theorem teaches us that in both cases the key is belief. While Pascal attempted to demonstrate the rationality of belief, not of the existence of God, the Dismal Theorem tries to establish the rationality of the belief in the possibility of climate catastrophe, not of climate change. To those who insist that we need evidence of the existence of God or of climate change before we believe in either, both theories are unsatisfactory. Can Decision Theory Help? This essay posed the question whether decision theory can help a rational person to decide whether to believe in climate catastrophe? The Dismal Theorem, of course, is only one decision theory, and it a simplistic one for that. There are much richer models of cost-benefit analysis that are more helpful to decision-making in the real world. Even these models, however, merely help to organize and generate relevant information; as all areas of science, they can inform, but should not prescribe policy decisions. The folly of the Dismal Theorem is not only that it attempts to do exactly that. It also, by necessity, has to claim that it is a uniform theory of knowledge governing all human affairs. Obviously, it fails in this regard. Both Pascal’s Wager and the Dismal Theorem appeal to faith, not doubt. Only those who believe in the necessity of forming a belief will buy these theories. Pascal believed that belief in God is good for human beings. Proponents of ambitious climate policies believe that a collective belief in climate catastrophe is necessary for the survival of humanity. Rational people may have very different views about the likelihood of climate catastrophe or the existence of God. Some might find it convenient to refer to Pascal’s Wager or the Dismal Theorem. In the final analysis, however, both theories cannot avoid the question Sagan posed in his parable of the Dragon: ‘what’s the difference between an invisible, incorporeal, floating dragon who spits heatless fire and no dragon at all?’[12] In the case of Pascal’s Wager, it is not a problem that the answer to Sagan’s question is faith, because religious belief is a personal matter. In the case of the Dismal Theorem, however, it would be irresponsible to spend huge amounts of society’s resources on the basis of faith in the doom scenario of climate catastrophe. ***** [1] Robert Frost, “Fire and Ice”, in: A Group of Poems by Robert Frost, Harper’s Magazine , December 1920, p. 67. [2] PASCAL’S PENSÉES, Part III, http://www.gutenberg.org/files/18269/18269-h/18269-h.htm [3] Cf. Pascal’s Wager About God, http://www.iep.utm.edu/pasc-wag/ . Pascal’ Wager, https://en.wikipedia.org/wiki/Pascal%27s_Wager Alan Hájek, Pascal’s Wager, Stanford Encyclopedia of Philosophy (November 6, 2012), http://plato.stanford.edu/entries/pascal-wager/index.html [4] Martin L. Weitzman, On Modeling and Interpreting the Economics of Catastrophic Climate Change, Review of Economics and Statistics, February 2009, 91(1), pp. 1–19. [5] The Dismal Theorem holds that if a non-zero probability of catastrophic climate change exists and marginal utility goes to infinity as consumption goes to zero, in the absence of prior information that rules out the upper tail of climate risks, rational agents would be willing to spent an infinite amount of resources to ensure a positive consumption level in the future. See Ross McKitrick, CHEERING UP THE DISMAL THEOREM, University of Guelph, Working Paper 1205, 2009, http://www.rossmckitrick.com/uploads/4/8/0/8/4808045/dismal.pdf [6] Martin L. Weitzman, On Modeling and Interpreting the Economics of Catastrophic Climate Change, Review of Economics and Statistics, February 2009, 91(1), p. 18. [7] William D. Nordhaus, The economics of tail events with an application to climate change. Review of Environmental Economics and Policy, 2011: 5(2):240–57. William D. Nordhaus, An Analysis of the Dismal Theorem (January 16, 2009). [8] Gary Yohe & Richard S.J. Tol, Precaution and a Dismal Theorem: Implications for Climate Policy and Climate Research, September 2007, DOI: 10.1002/9781118467381.ch7 [9] Gary Yohe & Richard S.J. Tol, Precaution and a Dismal Theorem: Implications for Climate Policy and Climate Research, September 2007, DOI: 10.1002/9781118467381.ch7 [10] William D. Nordhaus, An Analysis of the Dismal Theorem (January 16, 2009). For a rebuttal, see Martin L. Weitzman, Fat-Tailed Uncertainty in the Economics of Catastrophic Climate Change, Review of Environmental Economics and Policy, volume 5, issue 2, summer 2011, pp. 275–292. [11] Frank Furedi, PRECAUTIONARY CULTURE AND THE RISE OF POSSIBILISTIC RISK ASSESSMENT, Erasmus Law Review, 2009, pp. 197-220. [12] Carl Sagan, The Dragon In My Garage, http://www.godlessgeeks.com/LINKS/Dragon.htm   This article appeared on the Climate Etc. website at https://judithcurry.com/2016/09/11/decision-theory-and-the-doom-scenario-of-climate-catastrophe/]]>

Join our effort to build a greener future.