Job Conflict: How Do Hours Add Up?

Are Americans working more than they used to?

Yes, say two economists in a recent report for the Washington-based Economic Policy Institute.

No, say other academics who have conducted detailed use-of-time studies.

The confusion has generated a squabble in these scholarly circles, and neither side can fully explain the contradictory conclusions. But it appears the conflict centers on one point: There is more than one way to measure work.

Laura Leete-Guy, an assistant professor at Case Western Reserve University, and Juliet Schor, associate professor at Harvard University and author of a highly publicized book on the subject, argue that fully employed Americans worked an average of 138 additional hours each year between 1969 and 1989.

The figure, which includes both work in the marketplace and unpaid work at home, rises to 158 hours when commuting time and a decline in paid days off are included.

But University of Michigan economist Frank Stafford does not buy that calculation. “Basically, it’s totally wrong,” he said.

He cites Bureau of Labor Statistics figures showing that average hours at work per week amounted to 39.4 in 1989, virtually the same as the 39.5 figure in 1969.

Leete-Guy does not dispute those numbers but argues that weekly hours don’t tell the whole story, because many Americans are working more weeks per year. This is especially true for mothers, many of whom worked just part of the year in the past, perhaps taking summers off to be home with their children, but who now are working more weeks per year to make ends meet.

A second point of confusion is gender: As work hours for men have declined, they have risen dramatically for women, Leete-Guy and Schor report. They say women increased their paid employment by 276 hours from 1969 to 1989.

All agree that women are working more, but Stafford contends that when work hours for men and women are combined, there is no evidence of the dramatic increases reported in the study.

And then there is the question of whose data are more accurate. Stafford and University of Maryland sociologist John Robinson say the Census Bureau survey’s statistics used by Leete-Guy and Schor are unreliable. When people are asked how many hours they worked the previous week, they will often report a standard figure, such as 40 hours, rather than calculating carefully the exact amount of time worked.

But Leete-Guy said the time diaries used by Stafford and Robinson are also flawed. She said that the samples are small and that they were taken during business cycles that failed to reflect more recent increases in work hours.