Valuing the Benefits of Superfund Site Remediation: Three Approaches to Measuring Localized Externalities

We apply three complementary approaches designed to identify the localized effects of Superfund site remediation under the CERCLA, examining data at the level of (i) the census tract (paying attention to within tract heterogeneity), (ii) the census block, and (iii) individual house transaction. Our analysis of the within-tract housing value distribution detects statistically and economically significant appreciation in the lower tails resulting from hazardous waste cleanup; deletion of a site raises tract-level housing values by 18.2% at the 10th percentile, 15.4% at the median, and 11.4% at the 60th percentile. These tract results are confirmed by (i) house transaction data that show cheaper houses within each tract are more likely to be exposed to waste sites within one kilometer, explaining their greater appreciation from site cleanup, and (ii) high-resolution census block data that show greater appreciation among blocks lying closer to the cleaned sites. House-level repeat-sales data confirm results from our national level census analysis by showing that deletion raises housing values relative to proposal in specific markets, such as northern New Jersey, but they also uncover a great heterogeneity in the effects of remediation across markets, with no statistical effects from deletion relative to proposal detected in Los Angeles metro, southwestern Connecticut or Boston metro.