In practice, hedonic price functions have proven difficult to estimate because the amenity of interest is typically not distributed randomly across locations. For example, locations with health risk due to air pollution tend also to be urban, industrial areas with particular labor market characteristics. When differences between locations are imperfectly measured and covary with health risk and housing prices, it becomes difficult to disentangle the price effects of health risks from the price effects of other locational amenities. The problem of omitted variables is compounded by an important sorting issue. Households move to locations endowed with amenities that match their preferences. When households near the amenity of interest are not representative of the population at large, it becomes difficult to interpret observed price differentials.

This paper measures the effect of health risk on housing values by exploiting a natural experiment that mitigates both econometric problems. The analysis focuses on an isolated county in Nevada where residents have recently experienced a severe increase in pediatric leukemia. Housing prices are compared before and after the increase with a nearby county acting as a control group. The variation in health risk over time makes it possible to control for unobserved differences across locations. In addition, because the leukemia cases were unanticipated there is no reason to expect sorting of households according to preferences prior to the increase. Finally, because many houses were sold repeatedly during the sample period it is possible to control for unobserved property- [sic]

The results provide a robust estimate of the MWTP [marginal willingness to pay] to avoid pediatric leukemia risk. Housing prices in the affected county closely followed trends in housing prices for the control county and the rest of the state of Nevada during the period leading up to the leukemia increase. Then, beginning when eight children were diagnosed in 2000, housing prices in the affected county declined significantly. Least-squares estimates indicate that houses sold during the period of maximum risk sold for 15.6 percent less than equivalent houses not affected by the leukemia increase. Fixed effects estimates indicate a 14-percent differential. The estimated MWTP to avoid pediatric leukemia risk is used to calculate the value of a statistical case of pediatric leukemia.

...In 2002 the budget for the Environmental Protection Agency included $3.2 billion for safe drinking water, $1.7 billion for waste management, and $598 million for clean air. A primary motivation for this spending is to protect households from cancer-causing substances and other environmental health risks. Reliable estimates of household valuations of these risks are imperative if programs are to be funded at cost-effective levels.

Original Source: a report by Lucas W. Davis, Department of Economics, University of Wisconsin,THE AMERICAN ECONOMIC REVIEW DECEMBER 2004, Volume: 94 Number: 5 Pages: 1693 -- 1704