October 17, 2007 — Minding the Gap

A recent study on trends in leisure time suggests income isn't the only way to gauge inequalities in economic welfare
By Kevin Bryan

Since the late 1970s, income inequality has substantially increased in the United States. This gap between the "haves" and "have-nots" has become a major political issue in the upcoming presidential election. However, a recent paper by economists Mark Aguiar at the University of Rochester and Erik Hurst at the University of Chicago suggests that the gap in total welfare hasn't been widening quite as quickly.

Income is only a rough measure of the welfare of consumers. Recent research has shown that much of the growth in the income gap has been driven by increased inequality over the life of each individual, rather than increased inequality between individuals.

Imagine an accountant earning $50,000 a year decides to become a real estate agent. In the good years, his salary goes up to $75,000. When the housing market gets rough, though, it drops to $25,000. If there is the same number of good and bad years, he can simply save $25,000 during flush times and maintain his consumption at $50,000 every year, just as he did when he was an accountant. However, income data after his job change will show that inequality has increased.

This problem could be solved by measuring inequality in consumption. However, consumption data are much less readily available than income data, which are widely available from tax returns, unemployment surveys, and other sources. Consumption is more difficult to measure — everyone knows his income, but few can accurately state how much they spent on food in the past year.

A second problem with using income inequality as a proxy for welfare inequality is that income is not the sole contributor to welfare. Leisure is also a factor.

If a company pays each worker $50,000 plus two weeks of vacation, an individual who agrees to make $45,000 with four weeks vacation will cause a company-wide rise in income inequality. From a welfare perspective, however, inequality may not have grown at all since that individual, by virtue of his choice, must prefer the two extra weeks of vacation to the lost $5,000 in income.

Therefore, if patterns of leisure have changed across the economy, this must be accounted for when evaluating inequalities in economic welfare. That is what Aguiar and Hurst did in their August 2007 study.

The economists aggregated a series of surveys collected by a number of different sources since the 1960s. In each survey, individuals are asked to record how much time they spend each day performing tasks such as cooking, sleeping, and working. Next, these data were combined with another set of surveys that asked people how much they enjoy various tasks. Finally, tasks were categorized as market work (time spent on the job), nonmarket work (such as time spent cooking and cleaning), child care, and leisure (such as time spent playing with your kids or watching TV).

Aguiar and Hurst found that the average leisure time for all individuals has risen nearly eight hours per week since 1965, primarily due to a fall in nonmarket work. The average, however, says nothing about the distribution of this new leisure. Have the rich been gaining substantially more leisure while the poor toil, or have the poor been working fewer hours while the doctors and lawyers put in 12-hour workdays? This is a very difficult question to answer given the available data, but clues exist.

For example, men without a college degree have gained an average of six hours per week more leisure since 1965 than men who were college graduates. Since education level and income are closely correlated, this strongly suggests that lower-income individuals have gained more leisure than high-income individuals. This may, to some extent, offset the greater income inequality between the two groups. Women have also gained leisure since 1965, with the decrease in nonmarket work more than compensating for the increase in time spent at the office.

Valerie Ramey, an economist at the University of California, San Diego who specializes in time use, has criticized Aguiar and Hurst's methodology. In particular, she takes issue with how they aggregate different time use surveys. Still, her research also shows that leisure among the less educated has increased faster than leisure among the more educated.

Ramey suspects that inequality findings hold up when inequality is measured by income. "Thus, the data support the idea that increased leisure among the poor counteracts increased income inequality to some extent," she notes.

The next question is why this discrepancy in leisure time has occurred. Is it the result of workers becoming discouraged about the prospect of finding a job, and therefore giving up? Or, is the leisure gap the result of trends such as early retirement and more flexible labor schedules? Current data cannot tell us the answer.

Measuring inequality in welfare, rather than in income, is a difficult task. Further research on consumption and leisure over the next few years may help policymakers and the public better understand the changing nature of inequality.