Time for Things Labor, Leisure, and the Rise of Mass Consumption

22 Time for Things

1.3. The Structural Version of the Puzzle

Given the questionable degree to which the trend toward increasing consumption, as opposed to reducing work, seems explicable in straightforward cost-benefit terms, a further question arises, this time on a structural level. If the bias toward consumption under industrial capitalism is not explained by utility optimization, how is it that levels of aggregate efective demand under industrial capitalism have, nonetheless, generally been suficient to ward of structural crisis? The question is particularly relevant in the case of Fordist mass production. The Fordist order that fully emerged in the middle part of the twentieth century employed a set of regulatory policies to manage efective demand, primarily by broadening the distribution of purchasing power. But for those policies to work, a propensity to use higher incomes to increase consumption, rather than work less, had to be in place.

The question has Marxian and Keynesian versions, as well as a hybrid version represented by the social structures of accumulation (SSA) paradigm in American political economy. According to the Marxian version, most systematically developed by Paul Sweezy (1942) (and later on also by Michel Aglietta [(1977) 2000]), once mass production is in place, capitalism faces the problem of how to generate a volume of consumption commensurate with its output. Within the Marxian framework the problem is complicated by the division of the economy into Department I, which produces the means of production, and Department II, which produces consumption goods (Marx [1893] 1993). While it is possible for a while for shortfalls in efective demand in Department II to be made up for by increasing demand in Department I, most Marxists hold that in the long run the two sectors of the economy must grow together.23 In order for balanced expansion of the economy to be possible, real wages must therefore increase at a rate suficient to generate levels of aggregate efective demand for consumer goods capable of supporting the growth of Department I. Yet there are pressures on capitalists to reduce their wage bills as much as possible, and this leads to a tendency for imbalance between the two parts of the economy. Underconsumption in Department II then leads to overaccumulation in Department I.24 This was especially a problem in the early decades of the twentieth century. Those decades, according to Michel Aglietta, were characterized by a contradiction between the new system of mass production and the tendency for wages to be pushed down under what he describes as the “competitive regime of regulation” that held sway in the United States and elsewhere in the capitalist world up until the 1930s. Without measures in place to create a mass market for consumer goods, industrial capitalism in the era of mass production would have faced an ongoing crisis of underconsumption.25 For Aglietta the answer to this threat was to increase hourly real wages while introducing policies to encourage the masses to spend those wages on goods (hence sustaining aggregate efective demand) by way of the creation and promotion of what he calls the “working class norm of consumption.” In his view: “For the first time in history, Fordism created a norm of working-class consumption in which individual ownership of commodities governed the concrete practices of consumption” (Aglietta [1977] 2000, p. 158 [emphasis added]). This norm allowed for the development of a “mode of consumption” that complemented the Fordist “regime of accumulation” and consequently made possible the stable pattern of economic growth in the postwar decades ([1977] 2000). Developing Aglietta’s theory, Robert Boyer places even more emphasis on the centrality of mass consumption to the Fordist order. Boyer (1988, p. 11) characterizes Fordism as “a shift in consumption norms, mainly as a result of nominal wages rising to keep pace with anticipated gains in productivity.” Fordism is a system that “combines the development of consumption norms and production norms” (Boyer and Saillard [1995] 2002, p. 39). Yet it is not clear what processes underlay the alignment of mass consumption and mass production under Fordism—why it was that productivity was channeled into ever increasing efective consumer demand rather than less work time. Aglietta’s only argument bearing on that question is rather mechanistic. He suggests that the mode of consumption in the era of mass production was in large part a functional adaptation to changes in the labor process associated with Fordism. As the labor process intensified, so workers, in an efort to reconstitute themselves after their exertions on the job, increased their consumption. A new “mode of consumption” thus emerged, based on private commodity acquisition: “Individual commodity consumption is the form of consumption that permits the most efective recuperation from physical and nervous fatigue in a compact space of time within the day, and at a single place, the home” (Aglietta [1977] 2000, p. 159).26 From the perspective of the puzzle of work time and consumption under Fordism, however, Aglietta’s explanation leads us in a circle. According to Aglietta, increasing mass consumption was required because the labor process became more intense, while the willingness of labor to acquiesce in a high output / high income regime ensured that the demands of work remained unrelenting. Yet if labor had pushed more resolutely for increases in productivity to be turned into fewer work hours, then the demands of work would have been lessened, and consequently, less consumption would have been required to reproduce labor power.27 Furthermore, arguably the most efective way for workers to restore themselves after their depletion by the labor process is rest and recreation— in other words, leisure time. If that is the case, then we are returned to the question of why increasing productivity was turned into more income rather than less work at constant income.

The Puzzle

23

In a spirit similar to the regulation theorists, members of the US school of radical political economy have envisaged particular “social structures of accumulation” as securing suficient aggregate demand in the Fordist era (Gordon, Edwards, and Reich 1982). According to these theorists, the liberal regime of accumulation that preceded the Fordist era enabled capitalists to change the distribution of income in their favor, which resulted in insuficient aggregate demand resulting in overcapacity. In response, a set of institutions developed, the purpose of which was to ensure a more even distribution of purchasing power. The SSA approach allows for more contingency than does regulation theory, giving an important role for institutional and cultural developments that are exogenous to the logic of capitalist production. Yet both SSA theory and regulation theory share a tendency to take the sources of aggregate demand as more or less given. The underlying assumption is that once the right set of supporting institutions was set up, the virtuous cycle of increasing productivity, higher real incomes, and increasing aggregate demand, which defined capitalism during its Golden Age, simply fell into place—and continued until thwarted by structural contradictions located primarily in the sphere of production. The question of why labor came to accept a compact in which wage income increased while work time remained relatively constant is not given much consideration. The demand for goods over free time is left unexamined, with the distribution of purchasing power being the only variable adduced to explain the coordination of aggregate demand with mass production. For example, Thomas Weisskopf, Samuel Bowles, and David Gordon, prominent theorists working within the SSA tradition, argue that underconsumption (inadequate aggregate demand) is caused by the excessive power of capital over labor, which enables capitalists to drive down real wages (Wiesskopf et al. 1985).28 On this view, it is real wage levels that determine whether the demand-side cogs of industrial capitalism are suficiently well oiled for the growth machine to grind on. For capitalism to be stable, the power of capital to suppress the wages of labor must therefore be circumscribed by things like state-sponsored collective bargaining rights and minimum wages. The implicit supposition is that aggregate demand, when backed by suficiently high wage rates, will rise to levels capable of absorbing aggregate output.29 Yet the fact that this has tended to happen depends on a pervasive materialism, expressed in a tendency for increasing productivity and higher wage rates to be turned into commodities rather than free time. The conditions that give rise to this materialism, which channels productivity improvements into higher demand, are not given much consideration by the regulation and SSA schools. Despite their sophistication, both the regulation and SSA approaches assume that once appropriate distributive and regulative measures are in place, demand and supply will fall into equilibrium or, at least, will be suficiently aligned to facilitate the ongoing accumulation of surplus value.

24

Time for Things

The quasi-Keyensian version of the problem of mass consumption, as developed by Michael Piore and Charles Sabel (1986), focuses on the contribution of investment to aggregate efective demand rather than directly on consumption.30 At the same time, because predictions about the future state of consumer demand have tremendous influence on investment, consumption remains a crucial variable in determining the stability of industrial capitalism. Piore and Sabel point out that an economy organized around mass production involves very substantial fixed capital investments in specialized machinery and argue that this presents challenges to the stability of the system. High fixed capital investments of the sort involved in mass production mean that in order to be profitable, factories must be run at or near full capacity. This makes the mass production economy very inflexible, limited in its ability to cope with sudden downturns in demand. If measures are not in place to ensure that aggregate demand is sustained at levels suficient to support full capacity production, there will be a risk of underinvestment. Thus, Piore and Sabel interpret the Depression in broadly Keynesian fashion, as caused by a loss of nerve on the part of investors about consumer demand. This led to a self-fulfilling prophecy, as falling investment led to increasing unemployment, which had the efect of depressing aggregate demand, which in turn made the outlook for further investment even grimmer, prompting further declines in investment and more job losses, and so on, in a vicious cycle.31

Levels of aggregate efective demand commensurate with the growth in capacity of industrial production were secured by a series of institutional developments, stretching from the late nineteenth century to the postwar period. On what Piore and Sabel call the microeconomic side, the most important of these was the emergence of the corporation. Also important at the level of the firm was the development in the 1920s of welfare capitalism, which increased the purchasing power of wage earners.32 On the macroeconomic side, demand was stabilized by a new, government-backed industrial relations regime, by various components of the welfare state, including minimum wages, unemployment insurance, and Keynesian fiscal and monetary policy. Of these developments, the most significant was the system of collective bargaining that emerged in the years following World War II. This system reflected a compact between labor and capital, one mediated by the state. Norms of collective bargaining in the United States in the postwar period were set by the landmark 1948 agreement between the United Automobile Workers union and General Motors, which established the principle that wage levels should be reset annually according to the formula change in productivity plus change in the consumer price index. Real wages, and therefore mass purchasing power, thus became linked to the increasing productivity of the economy. This allowed the virtuous cycle of increasing returns to scale and increasing mass consumption to take root, ushering in the postwar era of economic expansion. Importantly, Piore and Sabel note that the success of the US model led to its emulation across the advanced industrial world, as various nations developed their own versions of Fordist regulation—the American model became a kind of paradigm for the regulation of capitalist economies in the postwar era. Indeed, the United States works well as a case study for Fordism precisely because it served, at least to some degree, as a model followed by other countries in that era of capitalism.

The Puzzle

25

Thus, in diferent ways Marxian theorists of underconsumption, those working within the regulation school, SSA theorists, and quasi-Keynesians such as Piore and Sabel see the management of aggregate demand as a central problem for twentieth-century industrial capitalism. To coordinate mass production and consumption (while ensuring adequate investment and profit realization) capitalism had to be regulated by an ultimately quite unwieldy set of institutions, policies, and practices. But once more, all of these theorists take it as given that ensuring mass purchasing power was a suficient condition for mass consumption. The solution to the problem of aggregate demand was to widen the distribution of spending power, primarily by increasing wages. As long as enough money could be channeled to consumers, and as long as investors remained confident that demand would continue to be robust, all would be well. On the Marxian model, profits could be realized, while for the Keynesians, this would ensure that capitalists would feel confident enough about the future prospects of the economy to invest their money, thereby securing the economy against the danger of falling into a liquidity trap. Yet, as noted, policies devised to increase the purchasing power of the masses, in order to stimulate the economy, crucially presuppose that increasing income will add to aggregate demand, rather than being used to reduce work hours. Measures like income-based tax credits, or industrial policy aimed at bolstering the power of unions so that they can increase their wage rates, only work to stimulate the economy if the extra money put in people’s hands has the efect of increasing demand rather than reducing work hours at constant levels of spending. However, it is not clear just why an increase in purchasing power per unit of work-time should lead to increased aggregate consumption rather than to decreased work hours. Indeed, it was not always the case that increasing hourly income was automatically used to increase consumption. There is, as noted in the introduction, historical evidence that wage earners in the later nineteenth century used increasing hourly rates to cut their work hours rather than to increase consumption (Dora Costa 2000a). Had that pattern continued—had increasing productivity been channeled into reducing work time at fixed levels of consumption—then the positive feedback loop between increasing returns to scale in production, increasing real wage income, and ever-increasing consumption would not have been possible. The Fordist economy would then quickly have run into demandside dificulties of the sort that its regulatory apparatus was supposed to prevent.33 For capitalism in the era of mass production to be sustainable, a stable pattern in which aggregate demand increases with increases in productivity and output was required. In order for aggregate demand to at least roughly match the output of mass production, it was not enough simply to raise labor’s real hourly wages, for increasing wage rates could, in theory, be used to reduce work time. Nor, for the same reason, would government measures to provide supplemental income have been suficient to stimulate the required level of aggregate efective demand. Rather, the coordination of demand and supply under industrial capitalism ultimately depended on an inclination on the part of the population to make use of increases in hourly wages to increase consumption rather than to reduce work time. A fundamental condition of possibility for the stability of the industrial capitalist order was, therefore, a general preference among wage earners for the accumulation of consumer commodities over more free time—or at the very least an indiference between the two alternatives, such that wage earners were amenable to an arrangement that privileged purchasing power over free time.34

26

Time for Things

The Puzzle

27

The propensity to increase consumption rather than reduce work time is, however, not a fact of nature. Nor, at the same time, is it plausible that it appeared in 1945 ex nihilo. In this book, I explain the dynamic consumer demand that underpinned the postwar period of stable expansion of industrial capitalism as a consequence of a set of social, cultural, and institutional developments that took place roughly over the first half of the century. The most important of these—to some degree underlying all the others—was the increasing extent to which wage labor became construed as a potentially fair form of economic exchange, based on the commensurability of labor and its wage. Seeing wage labor as a potentially fair exchange of commensurable substances, afected how use value was perceived in goods and activities, imparting unto consumer commodities the characteristics of the free activity given up (through work) in exchange for the wage. The balance established between mass production and mass consumption under advanced industrial capitalism was thus made possible by the ways in which the normative ground on which wage-labor exchange became legitimate was linked to the meaning of consumer commodities.

1. Possible Explanations

In order to clear the ground for this alternative, exchange-focused account of mass consumption, in this section I briefly consider the main currents of social scientific thought about consumption and assess the degree to which they might ofer a satisfactory solution to the questions under consideration: Why, under conditions of afluence, does consumer accumulation nonetheless continue unabated? Why has increasing productivity tended to be used to that end rather than to reduce work time? The currents of theory considered are neoclassical economics, behavioral economics (an increasingly influential heterodox current within economics), status theories, Marxian theories, and cultural approaches.

28

Time for Things

2.1. The Utility Maximization Explanation

Because of the prominence of economic explanations for consumer and worker behavior, it makes sense to begin with a more detailed consideration of the economic approach.35 In the introduction, and at the beginning of this chapter, I raised some serious objections to the economic explanation for the pattern of consumption and work that developed under advanced industrial capitalism. There are, however, some responses economists could make to my criticisms.

First though, an overview of the approach might be helpful. In neoclassical economics, human behavior tends to be regarded simply as an expression of preferences, with individuals following the principle of utility maximization, posited as a very general feature underlying and organizing human action.36 The commitment to this rational, optimizing picture of action is sometimes expressed in startlingly strong terms. “The view that consumers maximize utility is not merely a law of economics, it is a law of logic itself,” comments Paul Samuelson (1976, quoted in Ainslie 2001, p. 8), presenting an axiomatic assumption in the modern economic theory of consumer behavior in a standard textbook of modern neoclassical economics. Although the content of preferences and sources of utility are usually taken to be exogenous in economic theory, once a utility function is given, the assumption is that economic action proceeds in a rational and predictable manner.37 The economic theory of the consumer assumes that economic agents seek to optimize utility, and economists expend much creative energy demonstrating that any given instance of consumer behavior accords with this maxim. Consumer practices, from this perspective, are then most parsimoniously explained by the utility they produce for consumers. If workers, following an increase in wage rates, choose to increase their income rather than reduce their time at work, this simply indicates that the added utility for them of more consumption must outweigh the added utility of increasing their free time.38

Although the neoclassical model of human agents involves many questionable assumptions, it would be hard to deny that modern consumer practices are at least in part satisfactorily accounted for in broadly utilitarian terms. Marx was surely correct to note that for things to become commodities they must be potentially useful. That certainly seems right as we venture further down a hierarchy of needs, to those that are more basic, such as that posited by Maslow (1943). Yet, as we have seen, the utility maximizing approach, when presented as a full and satisfactory account of the pattern of work and consumption under industrial capitalism, leaves certain puzzles. First, according to economic theory itself, the marginal utility of consumption falls not just for a given product—for example, for the nth ice cream consumed—but also for the set of all things consumed.39 If, therefore, consumers are simple maximizers, as the neoclassical model would have it, beyond a point, as productivity and hourly wages go up, we should expect the rate of consumption to begin to level of and, in place of further increases in consumption, one of two possible alternatives. Increasing productivity should either be converted into more free time (as opposed to increased output), or else the rate of saving should increase.40 Yet actual patterns of behavior under industrial capitalism do not conform to that expectation.41 Although there is variation across diferent parts of the advanced capitalist world, over the course of the twentieth century, and especially in its second half, increasing productivity in the afluent world has tended to be used to increase consumption levels.

The Puzzle

29

Moreover, as noted earlier, even if levels of consumption are not at the point at which, for reasons of satiation, utility drops of steeply at the margin, if ever more commodities are accumulated at fixed or only very gradually decreasing amounts of work time, the amount of time available to make use of each commodity will decline—asymptotically approaching zero. This existential constraint on consumption, which limits the degree to which commodities can be made use of, becomes particularly salient when seen through the lens of modern consumer theory, as developed by Kelvin Lancaster (1966a, 1966b) and Gary Becker (1981). Becker and Lancaster maintain, quite sensibly, that goods are not themselves sources of utility but rather should be thought of as tools used to facilitate practices that generate utility. Consumer commodities are thus equivalent to capital equipment, by means of which the labor of consumption—the activity of making use of consumer goods—produces utility as a final output. For example, roller skates are capital equipment that facilitate the production of the utility generated by the activity of roller-skating. Becker and Lancaster thus hold that an important distinction can be made between the potential usefulness of an object and the utility realized through that object being put to use. The utility of an object only gets realized in use. From this perspective, to explain the apparent preference for higher income and more consumption in terms of maximizing utility entails that commodity objects are, on average, put to use to a degree suficient to realize an amount of utility at least equal to the disutility of the work endured in order to obtain them. However, as we have seen, what is puzzling about the trade-of between commodities and free time in afluent industrial societies is that ever more objects are accumulated, while the time available for their use decreases or remains constant. If the use value of objects is realized in activity and activity demands time, then this pattern does not make much sense. From the point of view of modern consumer theory, at least some aspects of the pattern of consumption under modern capitalism therefore seem somewhat hard to explain.

30

Time for Things

One possible objection to the reasoning I follow here is that it does not take into account that commodities can vary greatly in the amount of utility they produce per unit of time spent using them. Becker and Lancaster could argue that as new products emerge, the activities they facilitate become more eficient at generating utility so that the same amount of utility can be generated in less time. As goods become more eficient in this way, consumers might well have good reason to seek to upgrade their set of commodities, progressively retiring products as better ones become available. If higher income makes objects yielding more utility per unit of time-in-use afordable, then it might make sense for people to use increases in their hourly wages to raise their incomes (in order to acquire those objects) rather than reduce work time at constant income.42 But while this point might be valid on a formal level and might, indeed, account for some accumulative consumer practices, it is unconvincing as a comprehensive explanation for the general pattern of accumulative consumption in afluent capitalist societies. It certainly does not account for the accumulation of multiple, functionally equivalent items or for the acquisition of items that seem unlikely, given constraints of time, ever to be made use of enough to justify their expense. Moreover, if it was the case that the utility-generating productivity of consumer goods increased at a rate suficient to ofset the decline in the time available for making use of them, surely that should be reflected in a positive correlation between increasing income and subjective well-being. If increasing spending at fixed amounts of free time increases the amount of satisfaction experienced in that time, then as income increases, so should happiness. However, as we have seen, the evidence indicates that this is not necessarily the case in afluent societies.

Thus, the neoclassical theory of the consumer seems unable to provide a good explanation for accumulative consumerism, at the expense of increasing free time, under conditions of afluence. The notion that modern consumption is fully accounted for in terms of the realized utility of the things acquired stands in some tension with some general patterns in modern capitalism, particularly in its US incarnation.

The Puzzle

31