Exercise, Physical Activity, and Exertion over the Business Cycle

Unemployed workers do not increase their recreational physical activity enough to make up for the physical activity that was demanded by their lost jobs.

In the United States, people get a substantial fraction of their exercise on the job, especially if those jobs are relatively physically demanding, for example in such sectors as construction, mining, and manufacturing. For the average individual between the ages of 25 and 55, work is responsible for about 26 percent of total daily physical activity. Among less-than-college educated males, work-related physical exertion rises to 33 percent of total daily physical activity.

In Exercise, Physical Activity, and Exertion over the Business Cycle (NBER Working Paper No. 17406), Gregory Colman and Dhaval Dave calculate that, on average, unemployed workers do not increase their recreational physical activity enough to make up for the physical activity that was demanded by their lost jobs. Their analysis is based on individual records from the American Time Use surveys between 2003 and 2010. These surveys consist of a detailed diary on time use from 4AM to 4PM on the interview day.

The authors use the standardized metabolic equivalent of task (MET) measures that are available for a wide variety of activities and occupations along with the information in each person's diary on the minutes spent performing specific activities like eating, sleeping, rollerblading, and work. They then compute total daily energy expenditure and compare the results before and after the most recent recession. They further control for various observed and unobserved factors, essentially comparing individuals' daily exertion within states over time as they were being affected by the business cycle, which is measured through the monthly state-specific employment rate.

Comparing population means before and after the recession which began in late 2007, the authors find that as unemployment rose, people did devote more time to recreational exercise, but that most of the time formerly spent at work was redirected towards housework, television watching, sleeping, and other relatively low MET activities. Their analysis confirms that while recreational exercise increased as a result of unemployment, total daily physical exertion declined. This effect was particularly significant because many of the largest layoffs were in physically active occupations like construction and manufacturing, affecting low-educated males who are particularly at risk of chronic and frequent bouts of unemployment, unhealthy behaviors, and poorer health outcomes.

--Linda Gorman