Since the end of the Great Recession, growth in health care spending has declined to historically low levels. There is disagreement over whether this decline was caused by falling incomes during the Great Recession (and therefore is likely to reverse once the recovery is complete) or whether the decline represents a structural change in the health sector (and therefore is more likely to endure). We exploit plausibly exogenous regulatory changes in the mortgage lending market to estimate causal effects of the financial boom and bust cycle on personal income in the health sector in a panel of U.S. counties. We find that counties that were exogenously more exposed to the financial crisis because of the regulatory reforms experienced a greater rise in the size of the health sector over the course of the boom and bust relative to control counties, with the differential persisting through the recovery. We also provide evidence that both the boom and the bust periods of the financial crisis increased mortality in treated counties compared to control counties.