Abe C. Dunn

Geographic analysis of consumption is often constrained by geographic borders such as counties, but economic agents often cross borders to consume. Using unique card transaction data, we estimate across-county spending flows between firms and consumers for every county in the United States and for 15 industries to provide a new consumption-link across counties that has not been previously studied. To demonstrate the importance of this consumption link, we reexamine the 2007–2009 Great Recession following the work of Mian and Sufi (2013) and Mian, Rao, and Sufi (2014), who demonstrate that counties with the greatest decline in housing net worth also had the largest declines in consumption and employment. We show that the effect of the housing wealth decline crosses borders to affect consumption and employment in a pattern consistent with our spending flows, even for the non-tradable sector. We find that not accounting for cross-border effects tends to understate the impact of local housing wealth shocks on employment and spending by 26 and 17 percent, respectively; it also misallocates where those effects are occurring, by about 11 percent for both spending and employment.

The administrative costs of providing health insurance in the US are very high, but their determinants are poorly understood. We advance the nascent literature in this field by developing new measures of billing complexity for physician care across insurers and over time, and by estimating them using a large sample of detailed insurance “remittance data” for the period 2013–15. We found dramatic variation across different types of insurance. Fee-for-service Medicaid is the most challenging type of insurer to bill, with a claim denial rate that is 17.8 percentage points higher than that for fee-for-service Medicare. The denial rate for Medicaid managed care was 6 percentage points higher than that for fee-for-service Medicare, while the rate for private insurance appeared similar to that of Medicare Advantage. Based on conservative assumptions, we estimated that the health care sector deals with $11 billion in challenged revenue annually, but this number could be as high as $54 billion. These costs have significant implications for analyses of health insurance reforms.

We examine the impact of Medicare Part D on mortality for the population over the age of 65. We identify the effects of the reform using variation in drug coverage across counties before the reform was implemented. Studying mortality rates immediately before and after the reform, we find that cardiovascular-related mortality drops significantly in those counties most affected by Part D. Estimates suggest that up to 26,000 more individuals were alive in mid-2007 because of the Part D implementation in 2006.