Insight on U.S. Unemployment Trends

Tomorrow (or today, Irish time) sees the release of U.S. employment figures, which are expected to show a marginal increase in unemployment. Readers may be interested in visualising U.S. unemploymnent trends (by state, since 1990) using the new offering by "Google Public Data", as shown in the chart above. "Google Public Data" is discussed on the Google Research Blog here. The chart (above) shows Michigan, California, Louisiana, Vermont, and the U.S. average. (Misissippi is not shown though it follows a similar trend to Louisiana).

Of note, both Louisiana and California had above-average unemployment before 2000; but Michigan has fared relatively worse over the last decade. How did it go so wrong for Michigan when Louisiana (and Misissipi) and California previously suffered from the highest unemployment rates? This is a question that may see labour-economists swopping notes with trade-theorits and geographers.

Though Louisiana-State trends closely to Califorinia, Louisiana has a notable unemployment spike in 2005-2006, which corresponds to Hurricane Katrina (see some notes on Louisiana here). Despite an employment boost since Hurricane Katrina, Louisiana trends towards the national unemployment rate since summer 2008 (i.e. financial crisis). (Misissippi was also affected by Katrina but is omitted from the chart for reasons of parsimony).

Michigan's situation over the course of the last year is quite striking; the slope function of the state's unemployment increase is very steep. The outstanding (policy) question is: what is to be done about unemployment in Michigan; right now? From a research perspective, what can we learn from the 2005-06 unemployment shocks in Louisiana and Misissipi?