Dan Hamermesh, an economist at the Royal Holloway University of London, isn’t afraid to take on controversial topics. His past research includes estimating how much it pays to be beautiful—”Why Attractive People Are More Successful”—and he made headlines last year when he quit a tenured professorship at the University of Texas after it allowed students to carry concealed firearms.
His latest paper (pdf), co-authored with Katie Genadek and Michael Burda, is sure to grab attention. In it, the researchers measure whether workers from different racial and ethnic backgrounds spend more or less time working when on the job, wading into a loaded debate about stereotypes that still persist in the US and elsewhere. When asked what motivated the study of such a sensitive topic, Hamermesh, typical of economists driven by pure intellectual curiosity, said it hadn’t occurred to him that it might be so controversial.
In an earlier paper, Hamermesh and his co-authors measured how many hours people actually worked while on the job. They noticed the variable for race and ethnicity stood out as a small but statistically significant factor. They were intrigued enough to dig further into the data to find out why. Hamermesh was surprised by the “vitriolic” emails he received the day after his latest research was posted online.
The economists did a careful study using the American Time Use Survey, a long-running poll that asks detailed questions about how people spend their time, including what exactly they do when they are at work. The economists estimated that, after controlling for demographics, work hours, and detailed occupation and industry characteristics, African-American, Asian-American, and non-black Hispanic men said they put in about 1% fewer hours of work while on the job than non-Hispanic white men. (The same pattern held for women, but the difference was even smaller.) Non-work on the job includes breaks, socializing, and all the things we do that are not explicitly working while on the clock.
It is important to put these findings in context: namely, that 1% is a very small difference in work hours, the equivalent of just a few minutes per day. Meanwhile, studies have shown that African Americans, for example, tend to be paid 10-15% less than whites for similar jobs, which is way out of proportion to the minor difference in working time. The researchers also couldn’t observe differences in productivity—it is possible that if employees take more breaks, they make up for it by being more productive while working.
The authors tried to identify a reason for the small difference in working time observed between groups. Neither segregation (workplaces dominated by certain groups) nor race-specific job protections (laws that allow workers to sue for discrimination if they are fired) could explain the gap. The economists figured one reason could be that lower pay and promotion prospects might reduce minority workers’ motivation, but they dismissed the idea because the difference in non-working time also showed up among the self-employed. They speculate that it instead comes down to marginal preferences about how to spend the work day, and what drives those preferences remains a mystery.
Hamermesh points out all Americans work longer days and take fewer holidays than their counterparts in other developed economies, and aren’t markedly more productive as a result. What’s an extra few minutes of downtime per day in the grand scheme of things? “People who work less should be applauded,” he says. “Americans already work too much.”