What would you be doing if you were not at work? Scrubbing the bathroom or watching the Olympics? Fixing the car or playing golf? Darning socks or doing crossword puzzles?
The easiest way to measure leisure is to take survey data on how many hours a week people spend at work and subtract. Since 1965, the number of hours the average American works for pay has not changed much. By this simple measure, then, leisure has stayed the same.
But are we really working as much as ever?
"All time away from work is not equal," said Erik Hurst, an economist at the Graduate School of Business at the University of Chicago. Some time off is just more work.
To put it in economic terms, we spend some time off the job in consumption (watching TV, hanging out with our friends, reading for pleasure) and some in production (cooking dinner, cleaning the house, doing household repairs). Some activities, such as sleeping and eating, fall somewhere in between, and others, including child care and gardening, combine pleasure and production.
The difference is not just that we enjoy some activities and dislike others. It is that we could, in theory, pay someone else to do the production for us. A cook or a restaurant can make dinner, but nobody else can play golf or watch TV for you.
That distinction can make a big difference in predicting how, for instance, people will respond to higher wages or lower taxes. Do they have to give up recreation to earn more money? Or are they trading one kind of work for another?
If they spend their off-hours cooking, Hurst suggested, "when they get richer, they can buy a microwave or order takeout."
That seems to be what has happened during the past few decades. Americans are not, in fact, working as much as they used to. They are just getting paid for more of the work they do. Using several different definitions of leisure, Hurst and Mark Aguiar, an economist at the Federal Reserve Bank of Boston, analyzed time-use surveys done from 1965 to 2003. Whether they defined leisure narrowly or broadly, they got a consistent result.
"Leisure time - measured in a variety of ways - has increased significantly between 1965 and 2003," they write in Measuring Trends in Leisure: The Allocation of Time Over Five Decades, a Boston Fed working paper. (The paper is available at www.bos.frb.org/economic/wp /index.htm.) Using the most restrictive definition, which includes only "entertainment/social activities/relaxing" and "active recreation," the economists found that leisure had increased 5.1 hours a week, holding demographics such as age constant. (Without that control, leisure has grown 4.6 hours.) Assuming a 40-hour work week, that is like adding six weeks of vacation - an enormous increase.
"I was surprised by the magnitude," Aguiar said, though the general trend agrees with earlier research.
The increase in leisure is particularly striking for women. During this period, they entered the paid labor force in large numbers but gained just as much leisure as men. The difference is in where the gains came from.
Ninety-seven percent of men ages 21 to 65 had jobs in 1965, compared with 87 percent in 2003. That drop accounts for about 60 percent of men's increase in leisure time.
By contrast, Hurst said, "for the women, the entire gain in their leisure time is coming from declines in nonmarket work. The time women spend on cooking and cleaning and laundry and other household maintenance has been dramatically declining over the last 40 years."
In 2003, women spent 11.1 fewer hours a week working at home than they did in 1965. The biggest drop, 6.2 hours a week, came in cooking and cleaning up after meals - not surprising, given the enormous growth in restaurant and takeout meals and the spread of microwave ovens.
More women working outside the home created more demand for such conveniences, which, in turn, enabled more women to work outside the home. By contrast, Hurst said, "A woman who was working full time in 1965 was also working full time at home, almost - 40 hours in the market, 20 or 25 hours at home."
Like income, the economists found, leisure time has become less equal across the population since 1965. While income has grown faster for highly educated people, leisure time has grown faster for the less-educated, in part because some have left the labor force.
"Low-educated men used to work a lot," Hurst said. "Now they don't work a lot."
Some of that new leisure time undoubtedly comes from involuntary unemployment, but, Aguiar noted, these men are not spending it doing chores.
"If they're the unemployed, I would have expected them to do much more home production to offset the fact that they don't have a salary," he said. "But they're taking a lot of it in leisure."
In new research, the economists are looking at how much leisure time people of the same age have had in different periods. This is one early finding: Middle-aged people have a lot more free time than they used to.
"The 40-year-olds in 1965 worked a lot, lot more than the 40-year-olds in 2003," Hurst said.
And, of course, one of the biggest increases in leisure time is deliberately missing from the working paper because it omits retirees. Longer life spans mean more retirement years.
"It used to be that you worked till 65 and died at 66. Now you work till 65 and die at 80," Hurst said. "The net increase in leisure is rather large."