How to Measure Underemployment?

Working Paper
August 2013

Unemployment rates in the United States have been slow to fall over the last couple of years despite a modest growth in the number of people employed. The increase in nonfarm payrolls has averaged 191,000 a month over the last year, but the number of unemployed has fallen only by about 81,000 a month. In addition, average weekly hours of all employees have risen (to 34.5 in June 2013 from 33.8 in June 2009). What explains rising hours and rising employment but slowly falling unemployment? Bell and Blanchflower suggest that one factor is the extent to which workers already employed prefer to work more hours to fill the increasing needs of their employers. As a result, as demand grows, employers may first offer more hours to existing workers before hiring new people. The authors have captured this "excess capacity" in the UK labor market using UK data. Their "underemployment index" reveals that the potential for existing workers to take advantage of the recovery by increasing hours rather than creating new jobs, at the expense of the unemployed, is significant. Similar data do not exist in the United States, however, but the authors recommend that the Bureau of Labor Statistics produce the data needed to construct an equivalent US index.