Introduction
Ian Steedman’s recent book on consumption and time has pointed out that, despite the early recognition by Gossen of the fact that consumption requires time, the theory of the consumer has almost entirely assumed away the issue. Steedman’s book has also developed a series of simple, elegant and insightful models to examine the implications of the fact that it takes time to consume, and to show how some basic implications of consumer theory have to be altered by taking account of this simple fact.
This chapter follows Steedman’s lead by pursuing some further implications of the fact that consumption takes time. It departs from Steedman in three important respects. First, unlike Steedman, who assumes that consumers fully take into account the constraints imposed by time in making their optimizing decisions, it recognizes the fact that consumers may not be completely “rational” in their choices, because they are not fully aware of the fact that it takes time to enjoy the pleasures of consumption, and they may therefore choose suboptimal consumption bundles. Second, while Steedman assumes that consumption generally requires time, it takes into account the fact that not all types of consumption take time in the same way (some goods may take up little or no time), and examines the implications of this fact for consumption decisions. Finally, while Steedman’s focus is on the behavioral implications of taking consumption time into account, this chapter focuses on its implications for the utility or happiness of the consumer.
A substantial body of evidence has been produced to show that despite significant increases in consumption and income, the level of experienced utility or happiness, as reported by the consumer, has in fact not increased in many rich countries, and the same appears to be generally true across countries if we concentrate on rich countries (see Easterlin, 1995, 2000, Frey and Stutzer, 2002, Layard, 2005). There has been some analysis of the implications of the fact that consumption takes time for happiness. The pioneering contribution is that of Linder (1970) who argues that economic growth leads to a scarcity of time, which gives consumers less time for consumption and thus makes them “harried”. Linder’s book contains some simple models of consumption behavior in which it is assumed that consumption takes time, but does not explicitly analyze the consumption-happiness relationship. Most of the discussion on consumption and happiness, however, relates to factors other than consumption time, such as the role of advertising, relative consumption and status, and consumer debt (see Schor, 1998, Frank, 1999).
The rest of this chapter proceeds as follows. First, I consider a one-commodity case to show that although consumption time can explain why increases in consumption may not make people happier, it requires a form of consumer irrationality. I also explore to what extent consumers can be irrational in this sense. Second, I consider a two-commodity case to argue that the shortage of time may imply that even fully rational consumers may be no happier when they consume more, because they purchase goods which, while requiring less time to consume, fail to increase happiness in the long run. Third, I discuss some other considerations relevant to the analysis of consumption time. This is followed by my conclusion.
A one-commodity case
My point of departure is the simplest model of consumer choice which takes time into account, the model of labor-leisure choice. The consumer chooses between labor and leisure to maximize his or her utility. Leisure yields utility directly, while labor results in income and hence consumption, which yields utility, and the consumer effectively chooses between leisure and consumption. In this approach no time is taken to consume.
To allow consumption to take time we assume that the utility obtained from consumption depends positively on the amount of time spent on it. Thus we assume
where the flow of consumption services, c, requires both consumer goods, cg, and time, which we assume to depend positively with the amount of leisure time, l. Otherwise we follow the standard model, so that
and
Utility depends on consumption and leisure, the total time available to the consumer is normalized to unity, income is assumed to depend positively on hours worked, n, and exogenously given productivity, A, and leisure time yields utility directly as well (even if it is not used for consumption).1
The behavior of the utility-maximizing consumer can be examined by considering his or her choice of leisure by assuming continuous and differentiable functions for (1), (2) and (4). Substituting from (1), (3) with equality, (4) and (5) into equation (2) we obtain
u=u(c (f (1-l, A), l), l). | |
The consumer’s optimal choice of leisure may be found by setting the derivative of utility with respect to leisure, l, to zero, or
du/dl=u1·[-c1·f1+c2]+u2=0 | |
where the partial derivative of a function, x(), with respect to its ith argument is denoted by xi. Economic expansion, as captured by an increase in A is seen, by the envelope theorem, to be given by
which is positive, so that technological change leads to an increase in utility. The effect on the consumption of goods, cg, of this expansion, however, is unclear. Thus, although cg may rise with technological change and utility, it need not do so. This decline in consumption with technological change is not a possibility in the simple case in which there is no consumption time, as long as consumption (which is the same as the consumption good) is normal, and if the marginal product of labor increases with technological change.
To show that the amount of the consumption good purchased may fall as a result of technological change when consumption takes time, consider the simple special case of equation (1) of the fixed coefficients type,2
where
is the amount of time required for a unit of consumption, and where the utility function exhibits standard properties of smooth substitution and normality of consumption and leisure. We also assume tha...
