8. Ethics in Climate Economics: Balance or Bias?

As a philosophical discipline, ethics reflects on the moral beliefs of people and societies. The often-used term ‘equity’ can be understood in a narrow sense as addressing principles of fairness that apply to economic activities (often regarding distributive issues). Meanwhile, a huge body of literature on economics and ethics exists, pioneered by distinguished scholars such as Ian M.D. Little, Amartya Sen, Partha Dasgupta, John Broome and Peter Ulrich, among many others.

E.g., Streeten (1950), Anderson (1993), Beckerman (2011), Dutt and Wilber (2010), Hausman and McPherson (2006), Peil and Staveren (2009), Sen (1988), Putnam and Walsh (2012), Wallacher (2011), Dietz et al. (2009).

Little (1957) and Sen (1988) have provided seminal works on this relationship.

Welfarism is consequentialism in terms of individual utility, disregarding non-utility features.

Sen (1970, p. 21) states, interpreting Pareto’s approach, “the two following rules are used: (a) if everyone in the society is indifferent between two alternative social situations x and y, then the society should be indifferent, too; and (b) if at least one individual strictly prefers x to y, and every individual regards x to be at least as good as y, then the society should prefer x to y.” He goes on to point out with some literature examples that “the Pareto principle has been often taken to be free from value judgments” (p. 57).

As Little pointed out (‘Little Criterion’), there is only a Pareto improvement if (1) “expected social benefits are greater than expected social costs” and if (2) “the effects of the project on the distribution of income does not violate society’s distributional values” (Beckerman 2011, p. 54). Pareto optimality has “a built-in bias in favour of the status quo” (Beckerman 2011, p. 55). “In short, a society […] can be Pareto optimal but be perfectly disgusting” (Sen 1970, p. 22).

The two fundamental theorems of welfare economics based on Walras and Pareto are: (1) equilibrium of a (perfectly) competitive market is always Pareto efficient, and (2) any Pareto-efficient allocation in a competitive market is a market equilibrium (if indifference curves and isoquants are strictly convex). Since these theorems are very theoretical and presuppose perfectly competitive markets, they cannot play an important role in practice (Kleinewefers 2008, p. 104). Paretian theory has been used in political debates to argue for free, competitive markets until more differentiated economic theories, and new empirical findings, made it more or less obsolete – although it is still a very influential theory (Kleinewefers 2008, pp. 44; 193; Chap. 10).

For this paragraph, see particularly Kleinewefers (2008, pp. 45–47; Chap. 11).

For a criticism of these premises see Sen (1970), Touffut (2009) or Arrow et al. (2002).

An example of a social choice is a majority decision. Although social choice theory developed out of welfare economics, it is still a research field of its own (Kleinewefers 2008, p. 56; Sen 1970, p. vii). For an introduction see Arrow et al. (2002), especially Part 3.

Giersch (1993, p. 10) and Kleinewefers (2008, p. 53). Sen (1986) provides an overview of the literature on Arrow’s theorem.

As Giersch (1993) explains, a societal welfare function of the original Bergson type is possible despite Arrow’s theorem, because it does not refer – as Arrow suggested with his ordinal utilitarianism – to ordinal utility in interpersonal comparison. In contrast, it demands a general weighing of ordinally ranked preferences, which does not necessarily mean “utility.” The interesting question is, then: are interpersonal comparisons of utility necessary for determining social welfare? Obviously, this is not an economic but an ethical question.

The neoclassical approach is sometimes used synonymously with economics itself (Dutt and Wilber 2010, p. 10). Two different definitions are common. The first one “relates to method:” neoclassics “analyzes the economy by examining the behavior of individual, self-interested, optimizing agents and how they interact with each other. The second is a narrower definition related to how the economy operates which, in addition to the assumption made in the first definition, assumes that the economic agents interact with each other in smoothly functioning markets in which all resources are being fully utilized and in which there are no distortions such as imperfect competition” (Dutt and Wilber 2010, p. 10). Dutt and Wilber (2010, pp. 10–12) identify four main characteristics of neoclassical economics: (i) the rationality of maximising individual utility; (ii) competitive markets, which work perfectly, or where market distortions have to be removed; (iii) under certain conditions these things lead to efficiency; (iv) but do not (necessarily) lead to fair distribution since there is a trade-off between fairness and efficiency. The first definition above does not say “that free markets will result in efficient outcomes” (Dutt and Wilber 2010, p. 12), but it also implicitly values efficiency; therefore, policies that “move the economy closer to the perfectly competitive distortion-free situation” are preferred. The second definition above prefers laissez-faire policies, apart from distributional issues in order to create fairness. The appeal of neoclassical theory – after surviving so many attacks – is that it provides a methodological tool to answer almost every political question.

An interesting recent report provided by the Stiglitz-Sen-Fitoussi commission on the measurement of economic performance and social progress states that there are three major current strands regarding welfare measurement: (1) happiness research, (2) the capability approach and (3) gross domestic product (GDP), or resourcism. See http://​www.​stiglitz-sen-fitoussi.​fr/​en/​index.​htm, accessed 2 Mar 2015.

Many publications discuss the concept of homo economicus, its diverse assumptions, strengths and limitations. See, e.g., Sen (1999); Kirchgässner (2000); Rodriguez-Sickert (2009); Davis (2009); Wallacher (2011); Bowles and Polania-Reyes (2012). The different logical axioms and assumptions of the rationality theory are explained, e.g., by Endres and Martiensen (2007).

See, e.g., Kleinewefers (2008, p. 278); Dewey (1927, 1988).

Referring to revealed preferences is typical for newer welfare economics (Sect. 8.1.2): “The modern theory […] disowns its Benthamite origins and cannot be properly understood if these trappings of its childhood are not entirely discarded […] A rational individual is only said to behave as though he were satisfying preferences or maximising a utility function and nothing is claimed at all about the internal mental processes that may have led him to do so. A utility function, in the modern sense, is nothing more than a mathematically tractable means of expressing the fact that an individual’s choice behaviour is consistent” (Binmore 1974, pp. 50f).

Oral presentation by Dieter Birnbacher, Bucharest, Nov. 2011; Fullbrook (2009, p. 125).

See, e.g., Endres and Martiensen (2007) for a precise and comprehensive definition of the ‘(free) market.’ A briefer explanation could be the following: “In a market economy goods are produced for, distributed by and subject to contractual forms of exchange in which money and property rights are transferred between agents” (O’Neill 2009, p. 317).

For value-laden judgements related to the idea of economic globalisation and globally free markets see DeMartino (2009) and Müller and Wallacher (2005).

For further reading and literature on “trust” see Nooteboom (2009).

However, the three IAMs introduced in Chap. 7 are “social planner” models and do not model markets as such.

With W = objective function, t = time, C(t) = consumption of goods and services within a period of time, L(t) = labour within a period of time (labour is equivalent to population, at least in the ReMIND variant, Bauer et al. 2011) and ρ = pure time preference (see Sect. 8.4).

With α = elasticity of marginal utility of consumption and L(t) = population, proportional to labour inputs. The model variant, RICE (region-based Dice), uses a similar approach.

There is much literature critically discussing GDP (even though GDP, as a flow variable, originally was not developed to provide information on welfare, see Brümmerhoff 2007, p. 276). Collections of arguments against GDP include Kleinewefers (2008, pp. 245f), Nordhaus and Tobin (1973), Beckerman (2011, Chap. 9) and Dutt and Wilber (2010, Chap. 9).

The model comparison study by Edenhofer et al. (2010, p. 30) is an example of a case where at least some basic, ethically relevant limitations are made transparent.

Many more examples are discussed in the literature. Kleinewefers (Table 3.1, 2008, p. 59) provides a helpful overview of normative judgements found in different approaches to welfare economics. Box 2.1 concerning welfare functions in the Stern Review (Stern 2007, p. 30) is also helpful. Here, Stern acknowledges that his approach is “not always consistent with ethical perspectives based on rights and freedoms. But the approach has the virtue of clarity and simplicity, making it easy to test the sensitivity of the policy choice that emerges to the value judgements made. It is fairly standard in the economics of applied policy problems and allows for a consistent treatment of aggregation within and across generations and for uncertainty.” For a broader discussion on the ethical aspects of neoclassical consumer choice theory, see Hodgson (2001). Sen (1982, p. 328) offers a critique on welfarism – i.e., on the idea that individual utilities are ethically decisive for evaluating social states – as well as a critique on Pareto’s welfare economics (Sen 1970). The ethical anthropocentrism in welfare economics has rarely been criticised yet.

See Schneider (1997) for one of the best articles on this topic, as well as Schneider (2005), Sanstad and Greening (1998), Broome (1992), DeCanio (2003), Nelson (2008), Füssel (2007), Ackerman et al. (2009), Stanton (2011), Weyant (2009), Heal (2009), Hope (2005), Hof et al. (2008), Sluijs (2002), Tavoni and Tol (2010) or Edenhofer et al. (2010).

See Stanton (2011) for a detailed technical explanation of Negishi weights in IAMs as well as a critique on them.

Dasgupta (2007); Weitzman (2007); Nordhaus (2007); Dietz et al. (2009); etc.

Similar arguments can be found in Kleinewefers (2008, p. 225), but he argues that the social discount rate should not be too far away from the market rate of interest, otherwise there are incentives for sub-optimal investments reducing economic growth. Without growth it might be more difficult for future people to overcome their societal problems (Beckerman 2011, p. 206).

Montuschi (2014, pp. 130–134) clearly points out several normative-ethical issues of Stern’s discounting. For an ethical discussion of discounting, see Caney (2009); Meyer (2010); and Kowarsch and Gösele (2012). Some scholars have proposed alternatives, such as a variety of discount rates, e.g. a higher one for Africa to overcome poverty (Beckerman 2011, p. 206).

Interestingly, also the concept of ‘development’ has shifted its meaning from mere economic growth to more emphasis on distributional aspects and well-being, as well as environmental issues (Dutt and Wilber 2010, p. 207).

This is particularly true if IAMs use damage functions. Moreover, see the interesting project at http://​www.​teebweb.​org/​ or the “Copenhagen Consensus” exercise at http://​www.​copenhagenconsen​sus.​com/​ (both accessed 3 Mar 2015) as examples of such an integration and evaluation – without however suggesting that these are best practices.

The discussion about cost-benefit analysis versus cost-effectiveness analysis in IAMs (Sect. 7.​2.​2) perfectly illustrates this issue. See Chang (1997) for the question of (in)commensurability of plural values (values require a common numeraire to be commensurable).

A more fundamental normative implication of IAMs is the assumed need for governmental institutions, e.g., to regulate market failure and distortions. Some radical libertarians might not agree with such an active role of the state. Beckerman argues, “although market failure may constitute a prima facie case for public intervention, government failure is equally notorious in many situations. The normative significance of market outcomes in some cases may well be weak, but the public sector is also likely to be inefficient in some broad sense and the choices made by public decision-makers in many situations may be just as irrational as those made by ordinary citizens” (Beckerman 2011, p. 49).

Myrdal (1898–1987), a Swedish economist and Nobel Laureate, already identified crucial economic concepts as thick ethical concepts: economic productivity, equilibrium, balance, rationality, optimality, utility, development, efficiency, etc. (Dutt and Wilber 2010, p. 21). One could add the concepts of scarcity and goods (Robbins 1945). Even the widely used concept of CO₂ equivalents is based on value judgements implied in the typical economic methods of (intertemporal) valuation that are used to define the CO₂ equivalents.

Millner (2013) draws a similar conclusion. Note that in the present chapter, I did not analyse IAM-based literature as a whole (rather, I performed spot checks) and therefore cannot justifiably state that there is a severe bias in the literature as a whole. But, having discussed some highly influential IAM-based works, there is actually some bias and some lack of transparency.

Economics usually claims to follow the sequence of “effectiveness – efficiency – justice” when developing policy options. Economics allegedly provides answers to the first two steps and society then decides about issues related to ethics and justice. The first two steps, however, already imply a lot of ethical issues, as seen above.