EVENTS

Date:	Friday, January 27, 2006
Time:	1:00 PM -- 3:00 PM
Location:	Wohlstetter Conference Center, Twelfth Floor, AEI 1150 Seventeenth Street, N.W., Washington, D.C. 20036

January 2006

Leisure Time in America

Do Americans enjoy more or less leisure time in the twenty-first century than they did in decades past? Has the long-term rise of women’s employment come at the expense of time spent with family, friends, and in recreation? Do more educated persons enjoy more leisure time? What happened to leisure-time differences between more and less educated persons in recent decades as their wage differences expanded? At a January 27 AEI panel discussion, two leading scholars, Erik Hurst and Valerie Ramey, drew on their recent research to address these questions and discuss the evidence on trends in leisure time.

Erik Hurst
University of Chicago

Over the past forty years, we have seen men’s work time fall and women’s work time rise, and we conclude that leisure time goes the other way.

We must understand how households allocate their time away from the market. We value our leisure time, so people may be willing to earn less to have time off.

Looking at time diary data sets, we can track how households use their time in detail and see the breakdown of time that they spend brushing their teeth or making a sandwich. My analysis focuses on individuals aged twenty-one to sixty-five--those facing a work/leisure time trade-off. We control for education, age, and family structure, and so can view how a forty-year-old woman with children’s use of time has changed.

Looking at the data, we see confirmation of the conventional wisdom. Total work time is the same as it was in 1965, although men work six hours per week less and women five hours per week more. In non-market work, there is a total decline of four hours despite houses getting larger and yards bigger. We now have better washing machines, vacuum cleaners, and takeout food. If you add together market and non-market work, men and women’s work has declined in almost the exact same way (by eight hours), even though women have shifted from non-market to market work.

Leisure time has increased by between six and eight hours, depending on how you measure leisure time. As income has gone up over the past forty years, we could have improved consumption of leisure accordingly. It might therefore be the case that work itself is more leisurely, since we are now no longer working on the production line and can check ESPN at work.

The increase in leisure is larger for the less educated, although it is hard to draw well-being conclusions from leisure time data, since some involuntary unemployment is included in the data.

Valerie Ramey
University of California-San Diego

In 1930, John Maynard Keynes considered “The Economic Possibilities for Our Grandchildren,” and predicted that GDP per capita would be four to eight times higher within 100 years. He worried that productivity and leisure time would increase so dramatically that there would no longer be enough work to go around.

I shall attempt to give a longer-term evaluation of changes in leisure time by looking at Angus Maddison’s data on hours worked per worker. In 1900, the average worker worked 2,700 hours per year, compared to 1,500 hours in 2000. Yet, this leaves out much information regarding labor market participation, turnover, and unemployment.

By focusing only on the prime age population, we may miss important within-household transfers. In 1900, 36 percent of the population was aged up to fifteen, and that figure is now 22 percent. The population over sixty-five has gone up from 4 percent to 12 percent. It is important to include unpaid family workers, especially with farms at the beginning of the century. In 1900, the average annual hours of work was 1,000; in 2000 it was 850, with a peak for the Second World War and a trough for the depression in between.

High school enrollment was 12 percent in 1900 and is now in the upper 90 percentile. The average days attended per K-12 student has gone up from 100 to 160, and absenteeism has fallen since fewer leave to work on farms. Overall, time spent in school has risen compared with market work.

Gender, age, and employment status are important determinants of work time. The conventional wisdom holds that “diffusion of household utilities and appliances dramatically reduces the hours of household chores,” but there does not seem to be any trend between the hours of housework by non-employed women until 1965. Why did the diffusion of such appliances not reduce housework? How could washing machines not have reduced housework time? Appliances replaced immigrant labor. Also, appliances just led to increased household output. Washing machines meant you washed your clothes twice as often, and so household appliances were merely meeting new demands.

Home production hours are higher now on average, since there are fewer children to contribute to home productions. Average leisure per capita hours have remained constant between 2,700 and 2,800 hours over the past 100 years, except for peaks during the depression and a collapse during the Second World War.

Keynes’ prediction has not yet come true in the United States--although some claim that it has for Europe--but this may be the result of there not being so many children.

Steven J. Davis
AEI

These studies span many data sets, some of which are not always easy to put together. The pre-1960 data is very difficult to judge, since it precedes modern statistical sampling techniques.

Why has there been extra leisure for working age Americans since the mid-1960s? There are fewer children per person, although this effect is modest and may only account for one hour out of seven. On the other hand, the increase in the number of elderly may present a countervailing effect.

“Engines of liberation” such as disposable diapers, power tools, and washing machines may have had an impact, but as Valerie Ramey noted, these may have led to improved quality of output rather than reductions in work time by women. Greater reliance on market substitutes, such as more time being spent eating out, is consistent with household consumption expenditure data. We can now buy pre-washed salads and pre-cut chicken that reduce the individual time input required to prepare food. Time-sensitive innovations in retail distribution, such as self-service supermarkets, have provided greater convenience and variety. Online retailers like Amazon.com make it easier to find books and save that time.

There are now better opportunities for multitasking, which allow you to do business on the phone while driving or riding the subway. It is now possible to do productive work while traveling, and you can write a paper on your personal computer while riding the train. If you have a sick child at home, you can still do work, and this also leaves more time for leisure.

The rise in leisure time is stronger for less educated men. This may indicate “loafers, slackers, and bums”--people sitting on the dole. This is true of men as it is not for women.

How much is this extra leisure worth? A back-of-the-envelope calculation shows that five extra hours of leisure per week adds up to 260 per year. If you take average hourly earnings of $18.09, this means $4,700 dollars per person or $3,300 at a 30 percent marginal tax rate. With roughly 173 million working age Americans, this produces $570 billion dollars--equivalent to about 7 percent of aggregate personal consumption expenditures.

How does this value of extra leisure compare to other measures of economic gain? The value of extra leisure is about 12 percent as large as the measured increase in disposable income and is not measured by national income statistics. Usually, we only hear about rising real income, but we must also consider longer lives and more leisure per year for working age in addition to the extended period of possible retirement.

AEI research assistant Chris Pope prepared this summary.