Heterogeneous Effects of Brownfield Remediation by Neighborhood Value
| Variable | Dependent Variable: Log Sales Price | ||
|---|---|---|---|
| (1) | (2) | (3) | |
| BDI × I(Low-price neighborhood) | –0.0208*** (0.0024) | –0.0205*** (0.0024) | –0.0181*** (0.0037) |
| BDI × I(Medium-price neighborhood) | –0.0078** (0.0040) | –0.0024 (0.0037) | –0.0092* (0.0049) |
| BDI × I(High-price neighborhood) | –0.0010 (0.0019) | 0.0070** (0.0030) | 0.0115* (0.0059) |
| BDI total × I(Low-price neighborhood) | 0.0023 (0.0019) | ||
| BDI total × I(Medium-price neighborhood) | –0.0079** (0.0039) | ||
| BDI total × I(High-price neighborhood) | –0.0101*** (0.0034) | ||
| Census block group FE | Yes | Yes | No |
| Property FE | No | No | Yes |
| Observations | 43,787 | 43,787 | 17,506 |
| Adjusted R-squared | 0.724 | 0.724 | 0.860 |
Note: Each column shows a single regression. Each regression includes housing and neighborhood characteristics, year-month fixed effects, year-town fixed effects, and neighborhood presample mean price interacted with year-quarter fixed effects. In Columns 1 and 2, the sample is identical to Table 2. In Column 3, only properties transacting more than once are included. Neighborhoods (defined by census block groups) are sorted into three categories of low, medium, and high house prices based on average sales price for January 2000–September 2003. I(Low-price neighborhood), I(Medium-price neighborhood), and I(High-price neighborhood) are indicator variables based on that categorization.
↵*, **, *** Significance at 10%, 5% and 1%, respectively.