Pollution, Health, and Avoidance Behavior

It is important to account for avoidance behavior and measurement error in determining the full welfare effects of environmental quality.

Ozone is often described as having a deleterious effect on health, especially for children, the elderly, and those with existing respiratory illnesses. Yet the exact magnitude of that effect is difficult to estimate because individuals may select neighborhoods to live in based on the air quality there. For example, those with higher preferences for clean air or higher income may live in areas with better air quality, and these same individuals may make other investments in their continued good health. Furthermore, individuals may adjust their exposure in response to changes in pollution. Those most at risk of being negatively affected by pollution have the greatest incentive to adopt 'compensatory behaviors,' such as reducing physical activity on days with poor air quality.

In Pollution, Health, and Avoidance Behavior: Evidence from the Ports of Los Angeles (NBER Working Paper No. 14939), authors Enrico Moretti and Matthew Neidell propose a novel approach to estimating the biological effect of ozone on health. They isolate the short-term effect of ozone by using changes in pollution caused by boat arrivals and departures at the Ports of Los Angeles and Long Beach. Boat traffic significantly affects daily ozone levels in the Los Angeles area. But because boat traffic is unobserved by most residents, it generates an important source of variation in pollution that is difficult to avoid and cannot easily be offset by residents' compensatory behavior.

The authors draw two main conclusions from their analysis: first, the estimated effects of ozone on health are large. Second, it is important to account for avoidance behavior and measurement error in determining the full welfare effects of environmental quality. Simple correlations between ozone levels and health are significantly downward biased by unobserved avoidance behavior and/or measurement error.

The authors point out that behavioral responses to ozone levels may be substantial, especially in Southern California. Since ozone is greatly affected by weather conditions, it is highly predictable. Daily ozone forecasts that inform the public of dangers from episodic ozone conditions are widely available in Los Angeles through television and newspapers. This allows individuals to engage in protective behavior by spending less time outdoors to offset some of the adverse consequences from ozone exposure. If affected individuals compensate for changes in ambient ozone levels by reducing exposure, then estimates that do not account for these responses will understate the full welfare effects of ozone.

The authors' methodology avoids some of the measurement error problems that have plagued earlier studies. The most common approach for measuring exposure is to assign data from ambient air pollution monitors to the residential location of the individual using various interpolation techniques. Given the tremendous spatial variation in pollution within finely defined areas, this approach is likely to yield considerable measurement error. While several epidemiological field studies address this concern by using personal ambient monitors, these studies often involve very small samples that preclude the ability to obtain precise estimates of common outcomes of interest, such as hospitalizations. 

-- Lester Picker