October 30, 2003

#128: Factual Error

P. Krugman

We don't normally get too focused on Paul Krugman's chronic lack of fact-checking. We prefer instead to document and ridicule his devolvement from noted economist to political hack and point out the hoops of implausibility he regularly jumps through attempting to square his politics with sound economics. HOWEVER, in Too Low a Bar (10/24/03) he committed a factual gaff so atrocious we decided to point it out.

"If we want to improve the dismal prospects of job seekers � currently, 75 percent of those who lose jobs still haven't found new jobs when their unemployment benefits run out � the number of jobs must grow faster than the number of people who want to work."
The 75 percent figure is nothing less than preposterous! We checked with some labor economists who told us, off the top of their heads, it was more like 40 percent. So we looked it up. The Bureau of Labor Statistics (BLS) collects detailed data on labor force participation rates (that's basically what they do) and this includes exhaustion rates for unemployment insurance (UI). As shown in the top table in this BLS link, the correct number for the fraction of unemployed workers who exhaust their UI before finding a job is 43.6 percent in the most recent reporting period, the 2nd quarter of 2003. A related statistic shown further down on the same page indicates that the average worker on UI takes about 16.1 weeks to find a job. This is known as the "average duration" of unemployment. If Krugman's figure of 75 percent were correct, a rough estimate of average duration would be around 30 weeks, and UI only runs 26 weeks unless extended. There is no way he can defend this figure.

[The Truth Squad is a group of economists who have long marveled at the writings of Paul Krugman. The Squad Reports are synopses of their discussions. ]

Posted by John Weidner at October 30, 2003 6:42 AM
Weblog by John Weidner