Introduction

Ideally, scholarship can improve not only understanding but also policy-making: contributing “knowledge for change”. Faced by the environmental challenges of today, this is something we badly need. In this chapter, I assess the contribution of economics. Can it help us to better understand consumption behaviour; and to change it?

Economics today is a highly formalised system of thought that places great emphasis on analytical rigour, often at the expense of empirical realism. It is not hard to show that the standard textbook economic theory of consumer behaviour is severely impoverished when viewed from a ‘real life’ perspective. I do not deny this. But this claim may lead one to ignore the fact that the few explanatory factors that this theory does include—most notably incomes and prices—do explain a good deal of how consumers behave. Furthermore, these simple explanatory factors have the advantage that they link directly to potential instruments of policy—taxes and subsidies—which can, if suitably applied, very effectively modify consumption behaviour.

In recent decades, other disciplines—such as sociology, anthropology, geography and management studies—have developed new theories and perspectives on consumption. Contributors to this literature have often—and I believe rightly—claimed that economics has had an unduly powerful influence. For example, Wilhite (2005: 1): “the theories and models of economics have dominated the study of both energy consumption and ways to reduce it”. Economic theory has also been criticised or dismissed. A recent example may be cited from a book co-authored by Elizabeth Shove, one of the most influential scholars in the field of consumption studies. Conceptualising Demand (Rinkinen et al., 2021) poses fundamental questions about our understanding of this phenomenon. It is claimed, citing Callon & Muniesa (2005), that “economics operates in a realm of ‘pure calculation’ developing and working with models and variables that constitute a kind of parallel universe at one remove from the complexities of actual economic exchange, and from the historically situated practices involved” (Rinkinen et al., 2021: 17). More specifically, “consumption is taken to be an expression of individual choice, based on the ambition of satisfying needs and meeting desires and wants, limited or enabled by a willingness or ability to pay” (Rinkinen et al., 2021: 18). The authors also argue that economists are in error for not “considering the possibility that forms of supply might be implicated in making and not simply meeting demand” (17). This latter point has been made by others, including economists such as Fine, (e.g. 2013) as discussed below. While such criticism is largely well founded, it is also sometimes exaggerated or over-simplified; for example Middlemiss (2018) in an otherwise useful introductory textbook on sustainable consumption, presents the contribution of economics under the misleading title ‘people are selfish’: an unduly narrow interpretation of ‘homo economicus’.

In this chapter, I begin by presenting the standard economic theory of consumer behaviour. I then show that orthodox economists have long been aware of the limitations of this theory, and that these are not simply attributable to the adoption of the homo economicus model of human behaviour but relate also to the discipline’s preferred methodology. I then demonstrate how some recent advances that have been accepted by the mainstream, notably in behavioural economics, have enhanced our understanding of consumer behaviour. Next I note the significant contribution made by the Marxist economist Ben Fine, who takes account of the significance of production. Finally, I assess to what extent these different theories are relevant for policy-making. Here, I argue that, despite its limitations, standard economic theory can be helpful in devising simple—and highly effective—instruments of policy to modify or reduce consumption. I conclude with some brief reflections on the potential of interdisciplinary or multidisciplinary research on consumption.

Standard Economic Theory

According to standard economic theory, a consumer’s demand for a specified good over a given time period is determined by the consumer’s income and by prices: of the good, and of competing and complementary goods. Also included may be the consumer’s expectations about future prices. Advertising, the consumer’s taste for good X, and ‘other relevant factors’ complete the model.

Formally, that is expressed in algebraic terms, the model is as follows. A consumer’s demand for a good X over a given time period T is determined by dX=f (PX, PR, y, E, T, A, Z), where

  • PX is the price of the good

  • PR are the prices of competing and complementary goods

  • Y is the consumer’s income

  • E is the consumer’s expectations about future prices

  • T is the consumer’s taste for good X

  • A is advertising

  • Z is tastes and ‘other relevant factors’

Why are these particular factors chosen in constructing this model of consumption? And why are many others effectively excluded—simply lumped together as tastes and ‘other relevant factors’? The answer, I shall argue, is a combination of the methodology favoured by economists (building models) and the economic conception of the individual as ‘homo economicus’: rational, self-interested, well-informed, autonomous, maximising. And these two factors are interlinked: the economist’s rather special conception of human behaviour is well suited to the construction of the models used by orthodox economists. Econometrics has for several decades, and to an increasing extent, been the favoured approach in economics. It implies formalisation, in mathematical (algebraic) notation. And it also implies quantification—a preference for factors that can be measured. Using econometrics as the basis for establishing causality, or so its defenders argue, justifies the economist’s claim to greater analytical rigour than other social sciences. But these two requirements—algebraic formalisation and quantification—severely limit the extent to which standard economic models can explain actual consumption behaviour. As one economist expressed it, half a century ago:

The theory of consumer behaviour in deterministic situations as set out by, say, Debreu (1959, 1960) or Uzawa (1960) is a thing of great aesthetic beauty, a jewel set in a glass case. … it now stands as an example of how to extract the minimum of results from the minimum of assumptions. (Lancaster, 1966: 132)

Certainly prices and incomes are significant factors in determining consumption behaviour. And these have the advantage not only that they can be measured but also that data may be readily available from existing sources. But the standard model fails to take account of many other important determinants of consumption behaviour. (The case of advertising is somewhat anomalous. It is included in the model, as having a significant effect on consumption behaviour, although according to the assumption of homo economicus it should presumably not be—since if consumers were indeed autonomous in their preferences then there should surely be no place for advertising. The ‘solution’ is to assert that advertising simply provides information.)

Despite these limitations, it is nevertheless true that the model has considerable explanatory power. Empirical evidence shows that prices and incomes are indeed very important determinants of consumer behaviour, at least in the short term. If the price of one type of bottled water rises while the price of others remains the same, then the demand for the former almost certainly does fall. And a fall in income will lead to a fall in consumption of most goods. But the model does not explain the effect of the other factors, which economics is apparently not able to deal with. Here, other social sciences come into play: sociology, anthropology, psychology, marketing and so on. In brief, the standard economic theory of consumer behaviour is impoverished. As noted earlier, referring to Rinkinen et al. (2021), critics from other disciplines have often, and with justification, pointed out these limitations. But they have not perhaps acknowledged that the severe shortcomings of the theory have been recognised by economists themselves for many decades, as I shall demonstrate.

Criticisms Within Orthodox Economics

As noted, the conception of homo economicus on which the model is based draws on a number of assumptions that are questionable: for example, that an individual taking a consumption decision is autonomous, rational and the possessor of perfect information. These—clearly unrealistic—assumptions are found necessary in order for the model to be applied. But these limitations have not escaped criticism within the discipline itself. This applies most notably to the factor entitled ‘tastes’ in the model. An early example of such criticism is Leibenstein’s, 1950 article ‘Bandwagon, Snob, and Veblen Effects in the Theory of Consumers’ Demand’Footnote 1 where he shows how consumers are far from autonomous in their choices.

The desire of some consumers to be “in style”, the attempts by others to attain exclusiveness, and the phenomena of “conspicuous consumption”, have as yet not been incorporated into the current theory of consumers’ demand. My purpose, in this paper, is to take a step or two in that direction. (Leibenstein, 1950: 183)

A less technical version of the same criticism is to be found in Galbraith (1958) The Affluent Society. And others have contributed to the same argument, such as Frank (1985) Choosing the Right Pond: Human behavior and the quest for status. Some have noted that Adam Smith himself made a similar point (Wisman, 2019). One of the most famous contributors to this debate was Gary Becker, who won the Nobel Prize in Economics in 1992.Footnote 2 He is notorious for carrying the economic model of selfish, rational, maximising behaviour to extremes—applying it to human behaviour regarding education, discrimination, crime, marriage and divorce and childbearing. But he also recognised the weaknesses in standard consumption theory, as it became increasingly formalised: “As greater rigor permeated the theory of consumer demand, variables like distinction, a good name, or benevolence were pushed further and further out of sight” (Becker, 1974: 254). He was particularly critical regarding ‘tastes’. Economists, he wrote, “have no useful theory of the formation of tastes; nor can they rely on a well-developed theory of tastes from any other discipline in the social sciences, since none exists” (Becker, 1976: 133). The latter claim is certainly overstated, revealing an ignorance of—if not contempt for—other social sciences.

In 1980, Angus Deaton (another winner of the Nobel Prize for economics) and John Muellbauer published Economics and Consumer Behaviour, which became a classic text for students of economics. They assert that: “it seems unrealistic to suppose that preferences are exogenous, God-given and unchangeable. Rather they are socially inherited and conditioned and are governed by the conventions of technology and social institutions” (Deaton & Muellbauer, 1980: 330).

This was 40 years ago. But a more recent survey confirms that the situation still obtains, despite the fact that much new research has been undertaken:

Hence, while standard economics largely abide by their assumption of stable preferences, an array of alternative approaches is now available to account for changing tastes. Some of these approaches are old and have been discussed in the literature for many decades while others are younger. However, all approaches have in common that they, in some cases surprisingly, have not made it to standard microeconomics textbooks. (Jacobs, 2016: 121).

Jacobs lists a number of factors which economists have sought to study, including incomplete information and bounded rationality, habit formation, interdependent preferences, culture and institutions, and he concludes that “it is clear that preferences are unstable from the viewpoint of the standard model in microeconomics since behavior is influenced by more factors than just prices and income” (Jacobs, 2016: 142). He further notes that “empirical and experimental studies have been concerned with a pile of phenomena that may be explained by interdependent preferences” (Jacobs, 2016: 142). The question is whether these can be handled using the methods favoured by orthodox economics. (It is worth remarking that an additional attraction of the standard economic model is its elegance; and not least the symmetry between the theory of demand and the theory of supply, which so neatly combine to create a holistic conception of the market system. To put it simply, the theory of consumer behaviour is mirrored by a theory of the behaviour of firms: each firm seeks to maximise profit—rather than utility—and in doing so acts rationally, and autonomously, on the basis of perfect information).

Some examples from mainstream economic literature will provide insight into why it is difficult to develop a more realistic model of consumer behaviour within this tradition. Take the case of the last-named phenomenon—interdependent preferences: the fact that a consumer’s tastes are influenced by the behaviour of others. One of the few economists who has attempted to find a resolution to this issue is Kapteyn, who noted that: “For a very long time the study of inter-dependent preferences has remained at the fringe of the economics profession” (Kapteyn, 2000). There are several reasons for this, but, as Kapteyn notes, “perhaps the most fundamental one is just lack of adequate data” (op.cit).

There is no shortage of research in social sciences other than economics to demonstrate that preferences are not independent. There are interesting parallels here with theories of voting behaviour in political science, where so-called neighbourhood effects have been found to have some explanatory power. See, for example, Johnston et al. (2005). But the evidence is generally not in the form of statistics suitable for econometric testing. Goldsmith-Pinkham and Imbens (2011) study what they call ‘peer effects’. These “are hypothesized to arise through networks that are formed by individuals making choices to establish links”. The authors do manage to express this hypothesis in formal (algebraic) terms—but their model clearly cannot be tested empirically.

Another article, on the closely related phenomenon of conspicuous consumption, does—unusually—go some way to expressing this in terms that allow statistically based empirical testing. The study analyses how different very broad classes of expenditure vary with increasing expenditure, contrasting “those commodities whose ownership or consumption is highly visible to the community” with those commodities “whose consumption is usually screened from public view” (Basmann et al., 1988: 533).Footnote 3 The empirical results from this econometric analysis do provide some support for the theory. But the contribution that it makes to our understanding of consumer behaviour is—by comparison with research in other social sciences—rather trivial.

In brief, the problem is not that economists actually believe that tastes are autonomous. Rather, I suggest, it is the dominance of econometrics as method,Footnote 4 This involves not only constructing a model that can be expressed in terms of measurable factors, but also—in order to test its validity—being able to obtain the necessary quantitative data. These impose extremely severe constraints on the model’s usefulness.

Rationality—another assumed quality of homo economicus—has also been challenged. Here, there have been some interesting developments in economics in recent decades based on theories of so-called bounded rationality, to which I now turn.

Bounded Rationality

Standard economic theory has recently been supplemented by the sub-discipline of behavioural economics, which draws on psychology. Richard Thaler was awarded the 2017 Nobel Prize in Economics for his contributions to this development. In the text by the Nobel committee justifying their choice, they begin by noting that economists make a ‘fruitful simplification’ of assuming that agents are perfectly rational ‘in order to build useful models’.

Nevertheless, economists and psychologists have documented systematic deviations from the rational behavior assumed in standard neoclassical economics. Incorporating insights from psychology into traditional economic analysis has spawned the field of behavioral economics, a flourishing area of research with significant impact on many subfields of economics. (Royal Swedish Academy of Sciences, 2017: 2)

Thaler’s work built on that of two previous Nobel Laureates. One was Herbert Simon, from whom he “adopted the idea of bounded rationality in its broadest sense but not Simon’s view of decision-making as a ‘satisficing’ process” (Earl, 2018: 107). The second was Daniel Kahneman, and his work with another psychologist, Amos Tversky (Tversky & Kahneman, 1974).

According to (Earl, 2018), Thaler exposed the empirical shortcomings of rational choice theory and developed the Kahneman and Tversky perspective to make sense of a wide range of anomalies, as well as “developing a theory of ‘mental accounting’ and using it to understand the ways that consumers respond to different kinds of pricing strategies” (Earl, 2018: 107). Thaler further contributed to the development of a new policy instrument—’nudging’—which I shall discuss below.

Interestingly, Thaler himself appears not to regard his contribution as a rejection of the standard theory of consumer behaviour. Thus he writes:

I do not base my critique of the economic theory of the consumer on an attack of the assumptions. I agree with Friedman and Savage that positive theories should be evaluated on the basis of their ability to predict behavior.Footnote 5 In my judgment, for the classes of problems discussed in this paper, economic theory fails this test. What I have argued in this paper is that the orthodox economic model of consumer behavior is, in essence, a model of robot-like experts. As such, it does a poor job of predicting the behavior of the average consumer. (Thaler, 1980: 57–58)

Economists’ interest in consumer behaviour has, in recent years, been associated with increasing concern for the environment, and calls for ‘sustainable consumption’. An article by another famous economist, Partha Dasgupta, is relevant here. He writes: “Our aim is to show that building links between sociological and behavioural economic approaches to the study of consumer behaviour can lead to significant and surprising implications for conventional economic analysis and policy prescriptions, especially with respect to environmental policy” (Dasgupta et al., 2016: 191). He too (Dasgupta, 2014) identifies several limitations of conventional economics of consumer preferences. First, habits: that a consumer’s past consumption of a good may increase preference for that good in the future. Second, competitive or conspicuous consumption, as discussed by Veblen; where consumers seek to distinguish themselves from others. Third, where consumption of some good signals that the consumer relates to members of a social group. This last, he notes, is different from conspicuous consumption: people are here not wanting to consume more, but to consume in a similar way. While the first two limitations have already been referred to above, the third is somewhat novel. It relates to the issue of interdependent preferences, but Dasgupta introduces the suggestion that consumption behaviour may here be motivated by solidarity: a shared concern for sustainability.

Thus, in summary, it is apparent that some orthodox economists have long been seeking to improve upon standard consumption theory. They have sometimes been inspired by new directions within the discipline, such as experimental economics or behavioural economics, and by other disciplines, notably psychology. But, as Jacobs (2016) has noted, these developments have not been included in microeconomics textbooks.

Heterodox Economics: Systems of Provision

I will now consider what so-called heterodox economists have contributed to the study of consumption behaviour. The term ‘heterodox economics’ covers a number of different approaches, such as institutional, evolutionary, feminist, post-Keynesian, ecological and Marxian. Although to varying extents critical of standard economic models and homo economicus, few of these have developed alternative theories of consumer behaviour. A major exception is the Marxian economist Ben Fine, who emphasises the importance of production. This, of course, is what Marx himself does in his analysis of the capitalist system. Before describing Fine’s contribution, I shall therefore devote a few words to what Marx himself had to say about consumption.

His critique of capitalism was based on the emphasis he placed on social relations; so too, his critique of what he called ‘bourgeois economics’ and the categories of economics which treated as ‘natural’ that which was in reality dependent on the shared beliefs and practices of the people. But his critique extends only to the sphere of production. This emphasis on production is evident in most of his writings, but the precise relationship obtaining between production, consumption, exchange (and distribution) is spelled out most clearly, and in most detail, in the draft Introduction to the Grundrisse, in which, Marx lays stress on “the primacy of production”: “Individuals producing in society—hence socially determined individual production—is, of course, the point of departure” (Marx, 1970: 6). He does not ignore exchange and consumption, but nevertheless treats production as dominant. And when it came to consumption, his analysis was as asocial as that of modern neo-classical economists. (McNeill, 2021)Footnote 6 “Marx excluded use-value (or as it would now be called, ‘utility’) from the field of investigation of political economy on the ground that it does not directly embody a social relation” (Sweezy, quoted in Rosdolsky, 1980: 73). Some writers, most notably Agnes Heller, have sought to argue against this view, despite asserting that, according to Marx, “Use value expresses the natural relation between things and men” (Heller, 1974: 35). The evidence is strong that Marx’s emphasis on production appears to have blinded him to the social relations of consumption.

Ben Fine, a more recent critic of ‘bourgeois economics’, has, however, developed an alternative theory of consumption—the ‘Systems of Provision’ (SoP) approach—which sees production as determinant of our consumption choices—of food, clothing, shelter and so on. In ‘Consumption Matters’ (Fine, 2013), he criticises mainstream economics for being trapped by its own methodology: “Advances in economics relevant to consumer behaviour tend to be confined within and adapted to the needs of the parent discipline” (Fine, 2013: 53) His approach includes insights from other disciplines, although he is wary of what he sees as the excesses of cultural studies. In his critique of standard theory, Fine noted that “the literature to an enormous extent, sets aside the role of public provision in consumption. … This is an extraordinary omission given the extent of the state in determining what and how we consume” (Fine, 1994: 396). The SoP approach was initially applied to the food and clothing sectors, in the 1990s, (Fine & Leopold, 1993) then to housing and water supply, and has been further developed since (Bayliss & Fine, 2020). Fine refers to what he calls systems, for example ‘the meat system’, ‘the sugar system’, ‘the strawberry system’, each of which can be understood as a complex web of structures, agents, processes and relations. His approach, he claims, recognises the significance of both material and social factors. Thus, for example, the demand for household durables is linked to changing patterns of female employment.

The SoP approach focuses on the household rather than the individual. “In this light, household consumption is not the sum of individual behavioural patterns, each consciously motivated and evaluated by the actor. Instead, household consumption is a whole set of … social practices carried out by applying sets of rules and shared norms. They are also connected to production and distribution systems (technological and infrastructure networks) that enable certain lifestyles that connect consumers to one another” (OECD, 2002: 8).

Fine describes this approach as studying “vertically organized structures and processes that comprise the economic and social relations through which particular commodities are produced, distributed, advertised, retailed, and consumed within a material culture” (Fine, 1994: 395). This contrasts with the standard ‘horizontal’ approach which generalises across fields such as energy, food, transport, housing and clothing. Each of these, he argues, exists and functions differently (Fine, 1994: 395). Thus, in contrast to orthodox economics, SOP is much more empirically based, although Fine claims that his approach, though context and sector specific, “is not without theoretical implications” (Bayliss et al., 2013: 39).

To summarise my argument so far: both orthodox and heterodox economics have—despite their limitations—contributed significantly to our understanding of consumption behaviour. I will now, more briefly, consider the issue of how to change it.

Policy: Changing Consumption Behaviour

If a theory of consumer behaviour is to be useful for the purposes of policy-making, it should enhance our understanding of consumption as a phenomenon, and thus provide insights into how consumption behaviour might be modified. But this is of little practical use unless there also exist appropriate and feasible policy instruments; measures which can be taken to modify people’s behaviour, based on these insights. The need for governments to change people’s consumption behaviour has become of increased importance in recent years as public concern for the environment and people’s health has risen higher on the political agenda. Calls for sustainable production and consumption were included in the Sustainable Development Goals agreed by all nations of the world in 2015, but there has been resistance to some of the measures proposed to promote this. Clearly it is not sufficient to identify instruments of policy that are technically feasible; these must also be politically acceptable.

The point is well illustrated by the first factor in the standard economic model of consumer demand, namely income. It is certainly possible for a government to reduce people’s consumption—simply by reducing their disposable incomes through direct taxation. But this would be extremely unpopular; a guarantee that the politician or party proposing it would lose support. To cite a rather sardonic comment by the columnist George Monbiot “Our problem is that no-one ever rioted for austerity. People tend to take to the streets because they want to consume more, not less. Given a choice between a new set of matching tableware and the survival of humanity, I suspect that most people would choose the tableware” (Guardian Weekly January 2004). In the early years of the twenty-first century there were some calls, in the richer countries of the world, for moderating consumption. But when the financial crisis hit, in 2008, with its promise of slowing or even reversing economic growth, there was an immediate call for people to consume more—almost as an act of loyal citizenship. In short: reducing incomes would certainly be a very effective instrument for reducing consumption—but it is in political terms unfeasible.

What about prices? Here also there exist potentially effective policy instruments, namely taxes and subsidies. And these have a great advantage over measures to reduce income: they need not reduce total consumption, but only modify it. Taxes and subsidies can be used to change people’s behaviour in such a way as to bring about improvements in terms of health or environment. Such measures are, indeed, in common use in most countries—for example with taxes on cigarettes and alcohol. And they work: to an extent which can be quantified. The so-called price elasticity of demand is a measure of the extent to which a change in price affects total demand; and is thus a good indication of the impact of taxes. In high-income countries (HICs) recent studies estimate that for cigarettes the figure is around −0.4, implying that a 10% increase in price reduces overall consumption by 4%. For low and middle-income countries (LMICs) estimates range from −0.2 to −0.8 (Chaloupka et al., 2019: 89). The figure for alcohol averages about −0.64 in both HICs and LMICs (op.cit: 190) For sugar-sweetened beverages the figure for HICs is −1.2, and about the same in—at least some—LMICs (Mexico, Ecuador and Chile) (op.cit: 191). Thus, such measures are effective. But here, also, politics is a major constraining factor. For example, there is often strong resistance to increasing taxes on petroleum; to do so has led to riots in the streets, putting governments at risk.

In brief, to modify prices—through taxes or subsidies—is undoubtedly a very effective method of modifying consumer behaviour. There are, it is true, political constraints on how far this is feasible; but there can be no doubt that this economic instrument is a very powerful one.

As noted above, orthodox economic theory has now been supplemented by behavioural economics. Here too, there exists a policy instrument that corresponds to the theory, namely ‘nudging’. According to this approach, (Thaler and Sunstein, 2008) beneficial changes in behaviour can be achieved by minimally invasive policies that nudge people to make the right decisions for themselves. This approach emphasises the use of ‘choice architecture’, that is, the design of the environment where choices take place. A simple example is, in a supermarket, to locate carrot sticks in place of chocolates next to the check-out counter. Other examples include the strategic use of background music, lighting, smells and product placement (Carolan, 2018: 144). The latest development is so-called hyper-nudging which makes use of information about consumers available from Big Data to “make it possible for automatic enforcement to take place dynamically, with both the standard and its execution being continuously updated and refined within a networked environment that enables real-time data feeds which, crucially, can be used to personalise algorithmic outputs.” (Yeung, 2017: 122).

There is now considerable experience with ‘nudging’. In ‘A Review and Taxonomy of Choice Architecture Techniques’ Münscher et al. (2016) distinguish between three different types, influencing either ‘decision information’, or ‘decision structure’, or providing ‘decision assistance’. Decision information “target(s) the presentation of decision-relevant information without altering the options themselves, for example, by (re)arranging existing information or changing its presentation”. Decision structure works “by modifying the available options in the decision situation, including their range or composition, the default option, or the effort required for selecting an option and the consequences of selecting it”. And ‘decision assistance’: helps people “to follow through with their intentions. For example, choice architects can foster deliberate commitment or take measures to remind people of preferred behavioral options.”

Nudging is a more subtle measure than taxes or subsidies. For this reason it may generally be more politically acceptable. Even when people are well aware that they are being ‘nudged’ they may nevertheless support such a measure. There has, however, been some resistance to the use of this instrument, both with regard to specific instances of its application and to the philosophy of ‘libertarian paternalism’ with which it is associated.Footnote 7 It has also been questioned whether its effects may decline in the longer term. In his books, Sunstein (2008, 2014, 2016) has responded to criticism that nudges diminish autonomy, threaten dignity, violate liberties or reduce welfare, and argued that ‘choice architecture’ is inevitable, and that some form of paternalism cannot be avoided. This new instrument—nudging—supplements the existing arsenal of (non-economic) policy instruments, notably laws and regulations, and information campaigns, by seeking to change what might be called the ‘soft’ context in which consumption decisions are taken. Another type of instrument changes the ‘hard’ context: a person’s material surroundings. For example, in relation to consumer demand for mobility, the provision of roads and other transport infrastructure. Non-economic instruments such as these derive their theoretical justification from theories from non-economic social sciences, such as practice theory. But it may well be the case that a proposed policy instrument relies on insights from both economics and other social sciences. For example, one approach for reducing energy consumption—that has been suggested, and tested—involves making consumers more aware of the costs of the energy they consume; and perhaps also how this varies over time (Wilhite & Ling, 1995). This instrument is thus based, in part, on the economic theory that the price of a good or service is one factor determining demand. Elsewhere, also, Wilhite acknowledges that economics has a role to play. Thus Wilhite (2008: 5) asserts that, in economics, “consumption is stripped from everyday practice and actors from their social interactions” but continues “this is not to say that price, income and other economic considerations are not important to consumption; in Kerala, middle class access to capital and income are one part of the explanation for growing consumption.”

Conclusion

Despite its manifest shortcomings, economic theory has much to contribute to our understanding of consumption. In addition to standard textbook theory, I have described also two other theories of consumer behaviour, based on behavioural economics and a ‘systems of provision’ approach. Both of these involve moving beyond the bounds of orthodox economics as narrowly, and typically, defined (although the former is now accepted into the fold). This is surely the way to go—if one’s ambition is not methodological purity but rather a better understanding of the real world. However one may define one’s terms—interdisciplinarity, multidisciplinarity and so on—it is clear that an adequate understanding of consumer behaviour must draw on disciplines beyond economics. And the same applies when one moves from theory to policy. An adequate array of policy instruments for effectively changing consumer behaviour must also draw on disciplines beyond economics. This is not a call for a single, unified theory, but rather a call for eclecticism in both theory and policy. Economics has a valuable role to play, along with what many economists—perhaps disparagingly—refer to as ‘other social sciences’.