NCPA - National Center for Policy Analysis

Is the Unemployment Rate Accurate?

August 28, 2014

Many people see the unemployment rate as the main indicator of the health of the labor market, but David Leonhardt of the New York Times reports that the number may be questionable. According to a new study from researchers at Princeton University, the unemployment rate has become more and more inaccurate over the last 20 years.

What is responsible for the drop in accuracy? Partly, it is due to response rates, says Leonhardt. Americans have grown more unwilling to respond to surveys:

  • As landlines have become increasingly rare and caller ID has become more popular, fewer people are picking up the phone to respond.
  • Trust in institutions has dropped over the last few decades, writes Leonhardt. Americans are skeptical of survey questions and increasingly concerned about their privacy.
  • In 1997, the average telephone poll had a 36 percent response rate -- a number that had dropped to 9 percent in 2012. The Department of Labor has higher response rates for its monthly jobs survey (89 percent), though it has also fallen, down from 96 percent in the 1980s.

How does the government calculate unemployment rates?

  • It surveys groups of people for four months, takes an eight-month break, then surveys them again for another four months.
  • According to the study, people being interviewed in the later months of the survey are less likely to say that they have been looking for a job. That response would not count the person as "unemployed," because unemployment requires an active job search.
  • Yet, responses from older respondents are weighted more heavily than are those from new respondents, meaning that these respondents -- who are unemployed but report that they are not looking for a job -- are not counted as unemployed. Instead, they are considered outside of the labor force.

This brings the reported unemployment rate down. For example, the unemployment rate for the first half of 2014 was officially recorded as 6.5 percent. Among those being interviewed for the survey, those interviewed in the first month recorded a 7.5 percent unemployment rate, compared to a 6.1 percent unemployment rate among those in the last month of being interviewed.

Source: David Leonhardt, "A New Reason to Question the Official Unemployment Rate," New York Times, August 26, 2014. 


Browse more articles on Economic Issues