November 03, 2015
Hey, better late than never. It was good to see two columns reporting on new data indicating that the Current Population Survey (CPS), the main survey used to measure poverty rates, as well as employment and unemployment, seriously undercounts the number of poor people due to undercoverage in its sample. It’s an important point and deserves attention.
We thought so too, which is why John Schmitt was writing about the issue almost a decade ago for CEPR. Schmitt noticed a large gap between employment rates as shown in the CPS and the 2000 Census long-form. The latter was lower with the largest gap for the groups with the lowest coverage rate in the CPS. (Coverage rates in the Census are close to 99 percent due to extensive outreach efforts.) In the case of young African American men the gap was close to 8.0 percentage points.
Anyhow, this is an important issue and it is good to see it get some attention. Of course it would have been better if it got some attention a decade ago.
Comments