Quo vadis : Does economic theory need a sustainability makeover?

Introduction

Professor Marglin (2013) aptly synthesises a number of crucial themes relating to the challenge of achieving (and maintaining) truly sustainable economic outcomes globally. The urgency evident in his discussion is eminently appropriate and, shockingly, unshared by far too many in both rich and poor countries in the world. Grave dangers to the survival of human civilisation loom in our future of possibilities, and stark and fundamental changes are required in response.

One such domain of change argued for by Prof. Marglin is economics itself. It is a commonplace in the sustainability literature (particularly in the case of advocates of strong sustainability) that economists place excessive weight on the efficacy of market mechanisms, and accord undue significance to material considerations in relation to the theoretical conception of human welfare. Furthermore, it is regularly maintained that the values of actual humans are shaped (indeed, warped) by the experience of personal development within market (capitalist) economies so that human agents come to resemble the agents of theory to a greater degree than they would under some set of ‘ideal’ alternative conditions. ¹

Prof. Marglin makes these points explicitly, calling for an economics that places a higher value on discretionary time relative to consumption, a change in the values of human agents that privileges non-material sources of welfare relative to material sources and an economics that eschews policy implications in which unlimited economic growth is seen as both desirable and achievable.

Obviously, these themes are interdependent, part and parcel of the overall vision argued for by Prof. Marglin, namely a world without throughput growth, supported by a bespoke economics which is itself a reflection of a suitable set of values held by the population at large.

This article aims to discuss the challenges involved in coming up with the kind of new economics that Prof. Marglin deems necessary and, by extension, the challenges involved in ushering the new world of which he is in favour. In doing so, I am not in any way expressing disagreement with his fundamental premise – the need to show absolute respect for the ecological constraints we face as a species. Rather, I wish to explore some of the ways in which, in the arguments about welfare put forward by advocates of the steady-state economy, aspiration runs ahead of appreciation of the fundamental challenges involved, leading to (false) accusations on occasion that it is only a lack of political courage that stands in the way of salvation.

Economic theory, consumption and welfare

The difficulties facing the claim that economics puts excessive weight, in welfare terms, on consumption (whether directly material or necessitating material input for delivery in the case of services) are both conceptual/theoretical and empirical in nature.

The first source of challenge is the antiquity of the consumption–welfare link. The ultimate source of the subjective theory of value in the Western tradition is Aristotle’s treatment of justice in exchange in Book V of the Nicomachean Ethics. Aristotle ponders the question of what it is that is equalised in the exchange of heterogeneous goods between men of unequal status, such equalisation being required to achieve the ‘proportional reciprocation’ that justice demands. Aristotle identifies demand (arising from need) as both the basis for the exchange to take place and the mechanism through which the equalisation is achieved. Hence, at the dawn of economic theorising, the importance of consumption as a contributor to welfare played a central role as the basis for an indispensible social activity.

Aquinas cites Aristotle – and is supported in this interpretation by Schumpeter (1954) – in his account of the Scholastic argument that the just price must cover the costs of production. He refers to Aristotle’s claim in Ethics V.4 that the ‘arts’ would be destroyed if the quantity and quality of the contributions of producers and consumers did not correspond sufficiently. Here, the appeal to the importance of consumption is implicit – the consequential weight of the diminution of production and exchange is sufficient in Aquinas’ eyes to justify the moral imperative associated with the just price. Nonetheless, the importance of the link between consumption and welfare is attested to strongly again. Aquinas’ contribution is crucial in the development of Western thought. The cost-of-production thread in Scholastic analyses of the just price leads on to the objective theories of value that emerge in the 17th century, and forms one aspect of Smith’s reconciliation, in Wealth of Nations, between cost-of-production theories and the identification of the market price with the just price.

Myint (1948) argues persuasively that the Classical economists relied upon an implicit ‘physical’ conception of welfare that assumed a strict proportionality between economic welfare and physical output. This conception underpins the labour and other objective theories of value and renders the economic problem as ‘the struggle between man and nature in which the success of man is to be measured by the quantity of his net physical product’ (p. 229). This approach survives in Keynesian and Post Keynesian economics via Marshall’s (1920) conception of economics as ‘the study of mankind in the ordinary business of life’, focusing on ‘the attainment and the use of the material prerequisites of well-being’ (p. 1). Following the analytical shift toward subjectivism across the board in the 1870s, welfare in neoclassical theory was increasingly characterised in terms of satisfaction of preferences, with an increase in welfare ultimately now defined as an increase in the size of the feasible consumption set.

Hence, there is a common thread in the development of economics over more than two millennia (despite the fundamental transformation therein from a topic area in ethics, to a form of applied social theory, to a pseudo-scientific formalist apparatus) which links consumption strongly to welfare. Of course, this is in no way a justification of the proposition itself. Economics could be wrong. The point is that it would have to be root and branch wrong, and always have been. Prof. Marglin refers to ‘the central notion of conventional economics … that is no limit to our wants, no limit to the satisfactions from consuming more, more, and ever more’ (para 24). There is a pervasive tendency among those who echo these sentiments in the wider literature to construe this characteristic of economic theory as some kind of perverse ideological overhang of the Marginal Revolution, and to buttress such claims with hackneyed and uninformed diatribes on homo economicus, and other tired tropes of the often insufficiently scholarly ‘heterodox’ literature.

The implication of the need for a new economics from this perspective is that what is required is the correction of an aberration, a deviation from a (presumed identifiable) true path. This serves the purpose of making the task seem easier than it is, and thereby, conveniently provides the claimants with a practical excuse for their failure to provide a detailed plan of action for achieving it. The fact is that if economics is wrong in attaching major importance to consumption as a driver of welfare then all economics has always been wrong, and what is required is a complete knock-down and rebuild. Shying away from that reality while doing no more than calling for a new economics based on a different conception of welfare in relation to consumption is ultimately untenable, and lacking in intellectual honesty. The ‘fixes’, if required, in the realm of our ideas are no simpler or easier than in the domain of our interactions with the ecosystem.

Human welfare and consumption

Underpinning claims about the mistaken emphasis on consumption in economics are claims about what the link between consumption and human welfare actually is and, to the extent any such link in practice is undesirable, the need to change human behaviour accordingly. This need Prof. Marglin refers to explicitly. The point is significant, as it bears on the question of the welfare costs of adjustment and, consequently, the nature of the argument in favour of the transition to the economic system favoured by those in the strong sustainability camp. Furthermore, it speaks to the applicability of an economics based on a conception of welfare in which consumption plays a crucial role. The question of the appropriate conception of human welfare in relation to consumption is obviously a fundamental one, the ultimate resolution of which is beyond the scope of the present piece. However, some historical observations are pertinent.

Archaeology as a discipline largely originates from the analysis of funerary goods. The interment of goods along with corpses is a pervasive human practice dating back to the Palaeolithic and beyond. The fact that the living would seek to provide the dead with material artefacts indicates a belief in the role of consumption in welfare that extends even beyond life itself, and a role, given the spiritual significance of funeral rites, that is of central importance in cultural terms.

The increasing opulence of grave goods and other votive offerings from the Neolithic onwards provides evidence of processes of intensifying social stratification along status lines, but also an indication of the importance of the social relations of consumption for signalling status and other distinctions, the apogee being the funerary practices of the New Kingdom in Egypt. ² The important social role of consumption is attested to (both positively and negatively) in numerous literary sources as well. Macintyre (1966) observes that Aristotle holds that the ‘great-souled man’ (μϵγαλoψυχία) engages by necessity in conspicuous consumption in order to signal his superior status and to demonstrate his independence through the accumulation of beautiful, yet useless, things. Cicero derides the ‘piscanarii’ (Ad Atticum, I.19, I.20, II.9) of the late Republic, caught up in a positional goods arms race involving ever larger and more elaborate fishponds on their estates. Tuchman (1978) reports the humorous obsession in the 14th-century French court with poulaines, elaborate curling shoes that grew ever longer until even walking was impaired unless the tips of the shoes were fastened to the knees with cord. In King Lear (Act 2, Scene 4), Shakespeare has Lear answer his daughters’ campaign to reduce the size of his retinue with

O, reason not the need: our basest beggars

Are in the poorest thing superfluous:

Allow not nature more than nature needs,

Man’s life’s as cheap as beast’s: thou art a lady;

If only to go warm were gorgeous,

Why, nature needs not what thou gorgeous wear’st,

Which scarcely keeps thee warm.

What these examples represent is evidence of awareness, across millennia, of the role that consumption plays in human welfare via social channels. Once again, it need not be viewed purely positively; nonetheless, it has not escaped attention historically that humans value consumption over and above its contribution to physical well-being, and that consumption activity functions in part in the creation of personal identity.

Furthermore, these observations cannot be attributed to the effect of ‘consumerism’ as alluded to by Prof. Marglin. The usual interpretations of that term involve the ‘[inculcation of] values, beliefs and ways of being’ (p. 152) that is held to arise within capitalist economies subject to ‘market-driven growth’. The thrust of such accounts is that preferences and conceptions of the good are warped via interaction with capitalist institutions, leading to an excessive attachment to consumer goods. This position is not necessarily easy to square with the readily available evidence that humans have behaved in such a way for a very long time in a diverse range of social contexts, nor with the apparent human taste for novelty and the (perhaps) concomitant fact that it is difficult to point to examples of human populations who have obtained the capacity to enlarge their consumption sets in terms of both quantity and diversity, and yet have chosen not to exploit that ability. ³

Again, none of this supports the conclusion that it is appropriate or desirable for human agents to act as if there is a definite positive association between consumption and welfare. It does suggest, however, that they typically do, and so the ‘distorted values’ argument is therefore called into question. The second strand of the attack on the consumption–welfare link (as exemplified by Prof. Marglin) is to claim that even if human agents do believe that consumption and welfare are positively associated, and act accordingly, they are mistaken. This claim is predicated on the so-called Easterlin paradox, the assertion that subjectively reported well-being is insensitive to increasing income (Easterlin, 1974, 1995; Easterlin et al., 2010).

From the initial findings that average scores on happiness surveys did not vary systematically with aggregate income across countries, nor across time for individual countries, Easterlin and his followers in the happiness economics literature inferred that either there was a satiation point for happiness at a level of income below the average in the currently richest countries, or that human agents were more influenced by relative valuation (i.e. comparison to the average or to some other reference norm) than by evaluation of their absolute position (Clark et al., 2008).

For the strong sustainability advocates, the implications of the Easterlin paradox are clear: much economic growth in the developed countries has in fact been, ‘uneconomic’ in that it has not added to welfare – agents do not appear to feel that they are better off. The corollary, of course, is that any reductions in income and consumption in the rich countries necessitated by the move to a sustainable scale of human activity will not involve a loss of welfare. Indeed, Easterlin (2004: 33) holds that the substitution of non-pecuniary uses of time for working time (a necessary condition for consumption) would typically lead to enhanced welfare.

How much weight should be put on the Easterlin paradox? The conceptual foundations have certainly always been open to challenge. Easterlin (1974) recognises the problems associated with cultural specificity of the meaning(s?) of ‘happiness’ (pp. 93–99), the possibility that respondents are unable to reliably assess and report on their overall emotional state (differentiating it from how they feel in the moment in which they are asked), and the potential influence of conformism to expectations and framing effects triggered by the wording of survey questions. He concludes, however, that any bias arising in the results is not sufficient to threaten the validity of his conclusions. The basis for this assertion, however, remains unclear – it does not seem unreasonable to suspect that the issues identified could contaminate the data to a potentially serious degree. ⁴

Many of the surveys on which happiness research is based ask respondents to rate their happiness against a given scale. These can be verbal, in the tradition of Gurin et al. (1960) – respondents are asked to match their current emotional state against a set of descriptors such as ‘not very happy’, ‘somewhat happy’, ‘fairly happy’, ‘very happy’ and so on – or mapped to a numerical scale. The widely used Cantril Self-Anchoring Scale (Cantril, 1965) requires respondents to situate their current assessment of their life satisfaction on a scale running from the worst conceivable to the best conceivable, and then use a graphical device (the ‘ladder of life’) to assign a number between 0 (worst) and 10 (best) to represent their current degree of satisfaction. Country averages of Cantril scores feature prominently in the happiness economics literature.

Prima facie, it seems reasonable to expect that the fixed nature of the evaluative scales would lead to the appearance of some degree of satiation effects in the results. At one end, if ‘very happy’ is the most positive option available then individuals placing themselves in that category have nowhere to go but down subsequently, regardless of what happens to their material standard of living and indeed to what their actual state of happiness/well-being is. ‘Very happy’ becomes an open-ended catch-all category for high levels of happiness.

Respondents might show a tendency to avoid reporting extremes for a number of reasons. The choice between ‘fairly happy’ and ‘very happy’ might be difficult to make due to confusion over the appropriate boundary between ‘fairly’ and ‘very’ – this might be related to language game aspects of the use of intensifiers, a la Wittgenstein (1953): we might be able to confidently answer whether we are ‘fairly sure’ or ‘very sure’ of something when asked in ordinary conversation, but when put on the spot to answer a survey question we might find deciding whether we are ‘fairly’ or ‘very’ happy as challenging as coming up with a specific definition for ‘table’, or describing precisely what is meant by arriving somewhere ‘precisely’ at six o’clock. Cognitive challenges of this kind might prompt a larger proportion of respondents to consistently choose ‘fairly’ over ‘very’ leading to inbuilt inertia in the average scores, even at different points in time and with differing respondents.

Respondents might be reluctant to choose the highest rating for their happiness for precisely the reason that it leaves them no prospect for improvement, and might represent an undesirable failure of imagination on their part. This is of particular relevance to the Cantril measure. It is not unreasonable to expect that, at any time, only a minority of respondents would identify their circumstances as the best conceivably possible; it would no doubt be a somewhat depressing conclusion that one had nothing left to look forward to! The same argument applies to the step 9 on the ‘ladder’ – it is not hard to see why few respondents would, even if they were genuinely happy, choose to describe themselves as very nearly at the best possible level of life satisfaction. Similar reasoning applies at the lower end of the scale. Presumably, it would require true abject misery to override a natural human tendency to shy away from characterising their life as the worst possible, or even only a step away from it, if only because of superstitious fears of ‘tempting fate’. ⁵

This reluctance to utilise the extremes of the scale would naturally induce inertia in the average score, even if respondents over time were, in fact, getting happier and would not, if possible, choose to revert to their earlier lifestyles if given the option. Furthermore, those respondents living in dynamic economies, experiencing growth and innovation throughout their lives, could be expected to revise their assessment of what constitutes the best possible life. If the endpoint of the scale is, in effect, subject to upward revision, then even steady improvement in well-being might not close the gap between current circumstances and the optimum optimorum. Nor is there a reason to doubt that generations separated by decades would have very different conceptions of what is possible, based on experience, and where a cultural predilection to optimism prevails the resultant expectation of future improvement might well produce similar Cantril averages despite increases in material abundance which have made people more happy in an absolute sense rather than relative to their imagination.

The empirical bases of the results have come under sustained attack in recent years. Blanchflower and Oswald (2004) demonstrate that for the UK and the US, income is (statistically) significantly positively associated with happiness measures, while Graham and Pettinato (2002) reach similar conclusions for a range of emerging economies. Clark et al. (2005) find robust evidence of a positive relationship between changes in real income and changes in happiness in 12 countries using panel data methods to control for unobservable individual fixed effects. Di Tella et al. (2003) control for country fixed effects and find evidence of co-movement of happiness with gross domestic product (GDP), economic growth and inflation. Stevenson and Wolfers (2013) provide the latest comprehensive critique of the Easterlin paradox. They find that, utilising a range of international data sets on reported well-being and sufficiently long time-series on these measures and incomes, the relationship between income and subjective assessments of well-being is roughly log-linear across countries and within. This is inconsistent with the relative income strand of the argument. Furthermore, there is no evidence of a satiation point within the observed range of incomes. The inference invited is that the original, apparently paradoxical, Easterlin results were a statistical artefact of the low power of the tests arising from small sample size. In the light of these results, and other similar critiques, it is not unreasonable to say that the very existence of the Easterlin paradox is open to serious question.

A new economics?

The implications of this for arguments like that of Prof. Marglin are significant. Without these results to rely upon, the proposition that prospective welfare losses arising from the move to a sustainable path are largely illusory, a case of false consciousness with regard to the true empirical relationship between income/consumption and happiness, is nugatory. ⁶ Without the free lunch implied by the Easterlin paradox, the arguments in favour of moving to a sustainable path have to be recast purely in intergenerational justice terms – less income and consumption for those in the rich countries will mean that they will be worse off. It is not that they have already surpassed ‘enough’, in a sense that they would recognise subjectively, without realising it; there actually is no ‘enough’ and the proposition is that they have too much, from an objective point of view, and that their obligations to future generations, whether they recognise them or not, compel them to be made to have less.

This is a more demanding case to prosecute, not only politically but also philosophically. Intergenerational justice presents a potentially intractable challenge to virtually all approaches to the question of distributive justice. Consequentialist theories founder on the profound uncertainty attaching to future consequences (positive and negative) of current actions, especially when extending out (as in the case of climate change) over several centuries. Preference-based theories are skewered by the fact that future people don’t exist, and therefore don’t have preferences to take into account. Furthermore, we have no credible method for determining what those preferences might be. Ball (1985) argues that intergenerational justice, qua a concept framed in terms of the preferences of future agents, is actually incoherent for precisely that reason.

Deontological approaches to intergenerational justice must confront the problem that the notion of obligations across generations is undermined by the impossibility of reciprocity between the existent and the non-existent, and the concomitant difficulty of justifying one-way obligations involving potential welfare losses in the present. For an example of the denial of the existence of obligations to future generations, we need to look no further than the words of Prof. Marglin (1963) himself in a famous contribution to the debate over the social discount rate (no doubt prior to a subsequent Damascene conversion!):

I want the government’s social welfare function to reflect only the preferences of present individuals. Whatever else democratic theory may or may not imply, I consider it axiomatic that a democratic government reflects only the preferences of the individuals who make up the body politic … if after being made aware of future needs, present individuals wish to save on behalf of future generations, … each can adequately express his sympathy with the claims of future generations in his unilateral saving vs consumption decisions. But, if after being made aware of future needs, present individuals remain indifferent to the claims of future individuals, then, it seems to me, a democratic view of the state does not countenance governmental intervention on behalf of future generations. (pp. 97–98)

Contractarian theories a la Rawls (1971) struggle because of the problem of deciding whether all possible generations are to be included in the hypothetical bargain that establishes the principles of justice. Clearly, the number of possible generations is boundless and this makes it attractive to limit the parties to the bargain in generational terms. However, this has the effect of attenuating the incentives of the parties to the bargain to formulate the appropriate principles of justice in the absence of strong motivational assumptions. Paden (1996) provides a useful discussion of Rawls’ travails arising from the ‘present time of entry’ interpretation of the original position, that is, the requirement that the parties to the bargain are from, and know that they are from, the same generation. Of course, all contractarian approaches suffer from the problem of justification: why should I agree to, and act in accordance with, an inter-temporal allocation of resources that I perceive to be contrary to my interests when I am told that the reason why it is just is that in a purely hypothetical bargaining situation it would have been agreed to (and if my generation were to have been included in the deliberations then I would have agreed to this allocation)?

Any economic argument in support of obligation claims with respect to future generations, involving as it does welfare losses for those in rich countries, would have to rely on a welfare economics that does not yet exist, one based on something other than a subjective theory of value. In that sense, Prof. Marglin is broadly correct but has perhaps (along with advocates of strong sustainability) rather underestimated the task at hand. As noted, the link between consumption and welfare that is an obstacle for arguments for a move to zero-growth paths is in the DNA of the economics we have now. Furthermore, the historical evidence, and the empirical evidence associated with the rejection of the Easterlin paradox, suggests that the link is not without a good deal of validity so far as human beings go. Coming up with a new economics from scratch that recasts valuation in a way that both suits the normative arguments for strong sustainability and does not deviate seriously from what is empirically credible in terms of human behaviour will not be easy, and it is fair to say that despite the frequency of the calls to action on this front, little progress has been made by the advocates of change.

This perhaps explains that the need for a new economics is not felt as keenly by some of those who share Prof. Marglin’s overall vision. For example, Herman Daly (1977, 1991), somewhat of a godfather of the strong sustainability movement, argues strongly for a zero-growth future and advocates policies for achievement of that aim that rely heavily on the operation of markets as conventionally understood. ⁷ Hence, Daly argues for an extension of cap-and-trade systems to all basic resources, with quota allocations to be determined by auction. This is a thoroughly conventional approach, consistent with the standard maxim that ‘people respond to incentives’, that recognises the power of relative price change to drive agent behaviour in desired directions. The power to rapidly affect a multitude of decisions within a decentralised framework is obviously a historically observable advantage of market mechanisms, and one which, therefore should not be discarded lightly.

Daly addresses the income side of market exchanges by advocating changes to the mix of taxation, away from income, employment and value-added in favour of an array of Pigovian taxes on inputs (particularly where depletion is involved), and uses of the ecosystem’s assimilative capacity. Once again, the simultaneous freeing up of income and ratcheting up of the prices of ecologically damaging activities represents a thoroughly conventional method of discouraging undesirable activities and directing expenditure into superior alternatives.

Somewhat more controversially, Daly seeks to address the question of income inequality (and, no doubt, to provide a hard ceiling on consumption demand consistent with the achievement and maintenance of an optimal macroeconomic scale) by enforcing a personal income minimum and maximum consistent with a specified maximum range of inequality. The limits would be enforced via confiscatory taxation of income in excess of the maximum, and tax credits, or a basic income scheme (Christensen, 2008), to ensure that no individual falls below the minimum. Despite obvious questions about the extent to which such proposals are consistent with established liberal-democratic principles, it is clear again in this case that a proposal to limit incomes functions very conventionally as a part of a suite of measures intended to usher in a steady-state non-growing economy.

Daly’s challenges to established principles go further in that he is a noted advocate of transferable birth permits (with substantial financial penalties for non-permit births) as a means of population control. Whatever might be said about such a proposal, it is hard to conceive of a more striking sign of faith in the power of the market to drive human behaviour!

The nub of course is that Daly’s policy proposals reflect the very economics that for Prof. Marglin has passed its use-by date. Belief in the efficacy of systematic relative price change and limitation to personal incomes as drivers of both reduced economic activity and significant changes in its composition presupposes agents motivated by utility maximisation, and predictably responsive to the engineered variations in the constraints they face in terms of the volume and the distribution of their consumption activity. In short, they will only ‘work’ if economics ‘works’ – rejecting economics would mean rejecting these policies as ways of reducing throughput and overall human environmental impact. ⁸

Prof. Marglin is correct to identify the critical nature of the threshold we face. I would contend that indulging ourselves in the idea of a new economics just a corrective tweak away from what we have now, and concocting arguments about free lunches in welfare terms based on unsubstantiated representations of human behaviour, is likely to prove to be more of a roadblock to addressing the imminent challenges than anything else. It would be more fruitful to devote our efforts to what we understand currently regarding the management of human behaviour in the aggregate, and to develop more ingenious applications of that knowledge in the form of conventional (micro) economic policies. Education to influence values with regard to sustainability and conservation should continue apace, certainly, but history shows that values change as economic environments change, and manipulation of economic policy instruments is the most effective way to make that happen. Attempting to reinvent the wheel, without even a firm grasp of what a wheel is, is not.