Sorting for Life: Housing Market Estimates for Seniors’ VSL
AbstractLongevity varies across the US (Chetty et al., 2016) and some of this variation is believed to be caused by features of local environmental quality that affect seniors’ health (Finkelstein et al., 2018). This means that seniors can purchase statistical life extension by moving to places that offer health-enhancing amenities such as mild climates, clean air, low crime, and good health care. Our study uses revealed preference methods to translate these decisions into measures for the value of statistical life (VSL) among Americans ages 65 to 100.
We extend the spatial model of Roback (1982) of compensating differentials to incorporate health and longevity. First, we adapt methods from Finkelstein et al. (2018) to estimate how residential locations affect mortality at different ages (65-69, 70-74,..., 90+). Then we use hedonic methods to estimate the housing market ”price” of life extension, which we convert to VSL. The main identification challenge is that amenities may simultaneously affect the quantity and quality of life. Failing to distinguish these effects can bias VSL estimates upward. However, controlling for amenities can bias VSL estimates downward. We acknowledge these biases and estimate bounds. Our bounds for 65-69 year-olds ($4.1 to $8.4 million) include conventional VSL measures used by the EPA and academic literature. However, our upper bounds at older ages are less than half of conventional measures; e.g. $0.7 to $1.5 million for ages 85-89. Thus, these bounds are informative.
The VSL for seniors is perhaps the most impactful and least understood statistic for evaluating environmental regulations. Seniors comprise about 80 percent of deaths avoided by the Clean Air Act (EPA, 1999) but these mortality reductions are typically monetized using wage-hedonic evidence on workers who are decades younger and healthier. Our study advances knowledge by providing the first housing market evidence on seniors’ VSL.