The Persistent Employment Effects of the 2006-09 U.S. Housing Wealth Collapse

We show that the housing wealth collapse of 2006-09 had a persistent impact on employment across counties in the US. In particular, localities that had a larger loss in housing net-worth during that period had more depressed employment as late as 2016, without a commensurate population response. The use IV’s and controls to identify the causal impact of the wealth shock amplifies those results, leading to an estimate that a 10 percent change in housing net- worth between 2006 and 2009 causes a 4.5 percent decline in local employment by 2016, as compared with a 2006 baseline. We do not find a long-term causal impact of the shock on wages. Sectoral results indicate, however, that the results are unlikely to be purely a result of persistently low demand, since, contrary to the short-run effects, the effect over the longer horizon is less concentrated in the non-tradables sectors and is instead more prominent in the high-skilled services sector.