Time Allocation

BIBLIOGRAPHY

The international comparative time-use surveys organized by Alexander Szalai (1966, 1972) are usually seen as the first modern time-use measurements. These surveys were followed by an increasing number of time-use surveys in many countries. Initially, these data were used descriptively for the valuation of household work and for information about leisure activities, commuting, and travel behavior.

Table 1 gives the average shares of a day allocated to various activities for both males and females aged twenty to seventy-four over two years for three European countries and the United States. Depending on the country and year, market work occupies on average about 15 percent of all available time, as does household work. Tertiary time, about 45 percent, is mostly sleep and personal activities, while the remaining 25 percent is devoted to leisure.

Economic analysis of the rationale for time allocation was inspired by the pathbreaking work of a group of economists at the University of Chicago. They were led by Gary Becker (1965), who saw the household as a unit that produced utility giving commodities from market goods and time input from household members. For surveys of the first decades of time-use analysis, see F. Thomas Juster and Frank Stafford (1991) and N. Anders Klevmarken (1999).

Becker’s revised theory of choice assumes that a household derives utility from commodities, say Z_i i = 1, …, m, such as meals, a clean house, sleep, and going to a movie, which are produced by the household according to the household production functions,

The functions specify how much input of market time, T_i and goods, x_i, is needed to produce Z_i (The proportionality factors t_i and b_i are not necessarily constant.) The household is assumed to maximize utility subject to its budget and time constraints. These two constraints are not independent because time for consumption can be converted into money income by allocating more time to market work. The two constraints can be combined into one,

Total expenditures on market goods and household production time valued at the wage rate ω add up to the sum of nonlabor income V and total time T valued at the wage rate. The latter sum has been called full income . Substituting (1a, b) into the constraint (2), it becomes,

The expression in parenthesis is the full price of the i th commodity, and it is the sum of the prices of the goods and the time used per unit of the commodity i . The price of time input is the earnings forgone by using time to produce a unit commodity rather than to work in the market. If all t_i are zero, the model reduces to a conventional model of consumer choice. Becker’s model thus generalizes the conventional model by including the cost of time input to produce the utility-yielding commodities.

If the wage rate, the time cost, and the cost of market goods are fixed, it follows from the maximization of the utility function under the constraint (3) that marginal utility is proportional to the full price at maximum. We can now derive a number of predictions from this model: An increase in the wage rate will increase earnings forgone, and the full price of time-intensive commodities will increase more than that of good-intensive commodities. It follows from basic choice theory that the consumer will substitute away from time-intensive commodities and toward good-intensive commodities if the increase in the price of time is income compensated. At the same time, consumption time is freed for market work, which will increase. The effect of an uncompensated increase in the wage rate will depend on the relative size of the substitution and income effects.

If productivity of consumption time increases—that is, if t_i decreases—the relative price of time-intensive commodities will decrease, and consumers will substitute toward these commodities. For instance, when time-saving techniques are introduced into household work, we do more washing, cooking, and so on than we otherwise would have done. Whether or not we use less time in these activities will, as usual, depend on the relative size of the substitution and income effects, but if these commodities also become more good intensive, more market work is needed to generate the income required to buy the goods that go into these commodities.

Becker’s model is a good conceptual model for theorizing about time allocation, but it has weaknesses as a model for empirical work: One and only one good contributes to each commodity; the wage rate and the time and good productivities are not necessarily fixed and independent of the consumer’s choice; and the definition of a commodity is far from obvious. The model has been generalized to cope with these problems, but a more difficult problem is that the amount produced of each commodity is difficult if not impossible to observe. Without observations on the output from household production, it is in general not possible to identify preference parameters separately from the household production functions. All we can do is estimate mongrel time-use functions that depend both on preferences and household production technology.

In Becker’s model, the household is treated as a single decision-making unit, and there is no place for the separate decisions of the household members. Although Becker obtains some results from his model concerning the division of labor within a household—for instance, those who are more efficient at market activities use relatively less time at consumption activities—this issue is better analyzed within a bargaining model (see, for instance, the survey by Behrman [1997]).

Time allocation is not only an issue of how many hours of the twenty-four-hour day or of the 8, 760 hours of a year are spent on various activities; it is also an issue of when in a day or in a year various activities are done. One could also extend the domain of time allocation research by asking how much time is spent with whom? These issues have only recently attracted the interest of social scientists, but it is clear that when we do things and with whom is a matter of choice. These choices are sometimes constrained by laws, such as those regulating the opening hours of shops; by nature—for instance, we typically sleep during the night; or by social and religious conventions, such as those associated with Christmas and Easter. For recent contributions in this domain, see Daniel Hamermesh and Gerard Pfann (2005).

SEE ALSO Becker, Gary; Labor Force Participation; Leisure; Utility Function

BIBLIOGRAPHY

Becker, Gary S. 1965. A Theory of the Allocation of Time. Economic Journal 75 (299): 493–517.

Behrman, Jere R. 1997. Intrahousehold Distribution and the Family. In Handbook of Population and Family Economics, ed. Mark R. Rosenzweig and Oded Stark, 126–187. Amsterdam, NY: Elsevier.

Burda, Michael C., Daniel S. Hamermesh, and Philippe Weil. 2006. The Distribution of Total Work in the EU and the US. Unpublished manuscript. http://www.eco.utexas.edu/faculty/Hamermesh/BHW4.0.pdf.

Hamermesh, Daniel S., and Gerard A. Pfann, eds. 2005. The Economics of Time Use . Amsterdam, NY: Elsevier.

Juster, F. Thomas, and Frank P. Stafford. 1991. The Allocation of Time: Empirical Findings, Behavioral Models, and Problems of Measurement. Journal of Economic Literature 29 (2): 471–522.

Klevmarken, N. Anders. 1999. Microeconomic Analysis of Time Use Data: Did We Reach the Promised Land? In Time Use: Research, Data, and Policy, ed. Joachim Merz and Manfred Ehling, 423–456. Baden-Baden, Germany: NOMOS Verlagsgesellschaft.

Szalai, Alexander. 1966. Trends in Comparative Time Budget Research. American Behavioral Scientist 9 (9): 3–8.

Szalai, Alexander, ed. 1972. The Use of Time: Daily Activities of Urban and Suburban Populations in Twelve Countries . The Hague, Netherlands: Mouton.

N. Anders Klevmarken