| Stacked first differences | |||
|---|---|---|---|
| 1991–2011 | 1991–2007 | ||
| Without instrumentation | With instrumentation | Stacked periods 1991–1999 and 1999–2007 | |
| 100×Annual Δ in US exposure to Chinese imports | −0.81***b (0.16) | −1.30 (0.41) | −1.24 (0.37) |
| 1 {1991–1999} | −0.08 (0.36) | 0.05 (0.36) | 0.04 (0.36) |
| 1 {1999–2011} | −3.79*** (0.33) | −3.46*** (0.33) | |
| 1 {1999–2007} | −2.58*** (0.38) | ||
| Estimation method | OLS | 2SLS | 2SLS |
Abbreviations: OLS, ordinary least squares; 2SLS, two-stage least squares.
aN = 784 (392 four-digit manufacturing industries over two periods (1991–1999 and 1999–2011 or 1999–2007). Employment changes are computed in the County Business Patterns and are expressed as 100×annual log changes. Observations are weighted by 1991 employment. Standard errors in parentheses are clustered on 135 three-digit industries. Table adapted from table 3 in Acemoglu et al. (2016).
*p < 0.10, **p < 0.05, ***p < 0.01.