The effects of robots and Chinese imports on manufacturing and non-manufacturing employment
. | Long difference, OLS . | |||||||
---|---|---|---|---|---|---|---|---|
. | (1) . | (2) . | (3) . | (4) . | (5) . | (6) . | (7) . | (8) . |
. | Denmark . | Finland . | Germany . | Italy . | Norway . | Spain . | Sweden . | United Kingdom . |
Panel A. Manufacturing employment | ||||||||
Exposure to robots (instrument) | −0.097 | −0.592 | −0.060 | −0.320*** | −0.506* | −0.206** | −0.233 | −0.238*** |
(0.523) | (0.701) | (0.051) | (0.085) | (0.302) | (0.076) | (0.178) | (0.080) | |
Exposure to Chinese imports (instrument) | −0.017*** | 0.025** | −0.011 | −0.104** | 0.002 | −0.049*** | 0.008 | −0.216*** |
(0.004) | (0.010) | (0.013) | (0.041) | (0.006) | (0.013) | (0.013) | (0.044) | |
R-squared | 0.486 | 0.557 | 0.357 | 0.886 | 0.340 | 0.976 | 0.448 | 0.604 |
Panel B. Non-manufacturing employment | ||||||||
Exposure to robots (instrument) | 0.625 | 0.446 | −0.043 | −0.225 | −1.391** | 0.279** | −0.075 | −0.144 |
(0.762) | (0.366) | (0.032) | (0.141) | (0.607) | (0.131) | (0.119) | (0.186) | |
Exposure to Chinese imports (instrument) | 0.014 | −0.001 | −0.049*** | 0.132 | −0.008 | 0.084*** | −0.006 | −0.032 |
(0.015) | (0.006) | (0.019) | (0.095) | (0.010) | (0.029) | (0.011) | (0.085) | |
R-squared | 0.353 | 0.848 | 0.334 | 0.715 | 0.416 | 0.906 | 0.583 | 0.365 |
Panel C. The effect of interaction between robots and Chinese imports on manufacturing employment | ||||||||
Exposure to robots (instrument) | −0.038 | −0.605 | −0.138* | −0.187 | −1.859*** | −0.156 | −0.235 | −0.248* |
(0.482) | (0.689) | (0.082) | (0.209) | (0.605) | (0.117) | (0.205) | (0.150) | |
Exposure to Chinese imports (instrument) | −0.009 | 0.001 | −0.020 | −0.098** | −0.006 | −0.041 | 0.008 | −0.217*** |
(0.008) | (0.014) | (0.014) | (0.046) | (0.006) | (0.028) | (0.015) | (0.044) | |
Exposure to robots × Chinese imports | −0.120 | 0.087*** | 0.012* | −0.046 | 0.824** | −0.024 | 0.003 | 0.011 |
(0.076) | (0.032) | (0.007) | (0.078) | (0.363) | (0.062) | (0.346) | (0.120) | |
R-squared | 0.495 | 0.597 | 0.378 | 0.887 | 0.458 | 0.976 | 0.448 | 0.604 |
Observations | 99 | 70 | 319 | 110 | 74 | 49 | 100 | 352 |
Regional FE | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
Baseline controls | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
. | Long difference, OLS . | |||||||
---|---|---|---|---|---|---|---|---|
. | (1) . | (2) . | (3) . | (4) . | (5) . | (6) . | (7) . | (8) . |
. | Denmark . | Finland . | Germany . | Italy . | Norway . | Spain . | Sweden . | United Kingdom . |
Panel A. Manufacturing employment | ||||||||
Exposure to robots (instrument) | −0.097 | −0.592 | −0.060 | −0.320*** | −0.506* | −0.206** | −0.233 | −0.238*** |
(0.523) | (0.701) | (0.051) | (0.085) | (0.302) | (0.076) | (0.178) | (0.080) | |
Exposure to Chinese imports (instrument) | −0.017*** | 0.025** | −0.011 | −0.104** | 0.002 | −0.049*** | 0.008 | −0.216*** |
(0.004) | (0.010) | (0.013) | (0.041) | (0.006) | (0.013) | (0.013) | (0.044) | |
R-squared | 0.486 | 0.557 | 0.357 | 0.886 | 0.340 | 0.976 | 0.448 | 0.604 |
Panel B. Non-manufacturing employment | ||||||||
Exposure to robots (instrument) | 0.625 | 0.446 | −0.043 | −0.225 | −1.391** | 0.279** | −0.075 | −0.144 |
(0.762) | (0.366) | (0.032) | (0.141) | (0.607) | (0.131) | (0.119) | (0.186) | |
Exposure to Chinese imports (instrument) | 0.014 | −0.001 | −0.049*** | 0.132 | −0.008 | 0.084*** | −0.006 | −0.032 |
(0.015) | (0.006) | (0.019) | (0.095) | (0.010) | (0.029) | (0.011) | (0.085) | |
R-squared | 0.353 | 0.848 | 0.334 | 0.715 | 0.416 | 0.906 | 0.583 | 0.365 |
Panel C. The effect of interaction between robots and Chinese imports on manufacturing employment | ||||||||
Exposure to robots (instrument) | −0.038 | −0.605 | −0.138* | −0.187 | −1.859*** | −0.156 | −0.235 | −0.248* |
(0.482) | (0.689) | (0.082) | (0.209) | (0.605) | (0.117) | (0.205) | (0.150) | |
Exposure to Chinese imports (instrument) | −0.009 | 0.001 | −0.020 | −0.098** | −0.006 | −0.041 | 0.008 | −0.217*** |
(0.008) | (0.014) | (0.014) | (0.046) | (0.006) | (0.028) | (0.015) | (0.044) | |
Exposure to robots × Chinese imports | −0.120 | 0.087*** | 0.012* | −0.046 | 0.824** | −0.024 | 0.003 | 0.011 |
(0.076) | (0.032) | (0.007) | (0.078) | (0.363) | (0.062) | (0.346) | (0.120) | |
R-squared | 0.495 | 0.597 | 0.378 | 0.887 | 0.458 | 0.976 | 0.448 | 0.604 |
Observations | 99 | 70 | 319 | 110 | 74 | 49 | 100 | 352 |
Regional FE | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
Baseline controls | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
This table presents OLS estimates of the impact of the exposure to robots and Chinese import competition on manufacturing employment in panel A as well as non-manufacturing employment in panel B. Panel C presents OLS estimates of the impact of the exposure to robots with the interaction between robots and Chinese imports. The outcome variables are the long difference in manufacturing or non-manufacturing employment-to-population ratio. Regressions are weighted by population in the start-of-period. The list of covariates is documented in Appendix A. The missing geographic values in Germany are due to confidentiality-related data limitations. Statistical significance based on robust standard errors (reported in parentheses) is denoted by ***P < 0.01, **P < 0.05, and *P < 0.10.
The effects of robots and Chinese imports on manufacturing and non-manufacturing employment
. | Long difference, OLS . | |||||||
---|---|---|---|---|---|---|---|---|
. | (1) . | (2) . | (3) . | (4) . | (5) . | (6) . | (7) . | (8) . |
. | Denmark . | Finland . | Germany . | Italy . | Norway . | Spain . | Sweden . | United Kingdom . |
Panel A. Manufacturing employment | ||||||||
Exposure to robots (instrument) | −0.097 | −0.592 | −0.060 | −0.320*** | −0.506* | −0.206** | −0.233 | −0.238*** |
(0.523) | (0.701) | (0.051) | (0.085) | (0.302) | (0.076) | (0.178) | (0.080) | |
Exposure to Chinese imports (instrument) | −0.017*** | 0.025** | −0.011 | −0.104** | 0.002 | −0.049*** | 0.008 | −0.216*** |
(0.004) | (0.010) | (0.013) | (0.041) | (0.006) | (0.013) | (0.013) | (0.044) | |
R-squared | 0.486 | 0.557 | 0.357 | 0.886 | 0.340 | 0.976 | 0.448 | 0.604 |
Panel B. Non-manufacturing employment | ||||||||
Exposure to robots (instrument) | 0.625 | 0.446 | −0.043 | −0.225 | −1.391** | 0.279** | −0.075 | −0.144 |
(0.762) | (0.366) | (0.032) | (0.141) | (0.607) | (0.131) | (0.119) | (0.186) | |
Exposure to Chinese imports (instrument) | 0.014 | −0.001 | −0.049*** | 0.132 | −0.008 | 0.084*** | −0.006 | −0.032 |
(0.015) | (0.006) | (0.019) | (0.095) | (0.010) | (0.029) | (0.011) | (0.085) | |
R-squared | 0.353 | 0.848 | 0.334 | 0.715 | 0.416 | 0.906 | 0.583 | 0.365 |
Panel C. The effect of interaction between robots and Chinese imports on manufacturing employment | ||||||||
Exposure to robots (instrument) | −0.038 | −0.605 | −0.138* | −0.187 | −1.859*** | −0.156 | −0.235 | −0.248* |
(0.482) | (0.689) | (0.082) | (0.209) | (0.605) | (0.117) | (0.205) | (0.150) | |
Exposure to Chinese imports (instrument) | −0.009 | 0.001 | −0.020 | −0.098** | −0.006 | −0.041 | 0.008 | −0.217*** |
(0.008) | (0.014) | (0.014) | (0.046) | (0.006) | (0.028) | (0.015) | (0.044) | |
Exposure to robots × Chinese imports | −0.120 | 0.087*** | 0.012* | −0.046 | 0.824** | −0.024 | 0.003 | 0.011 |
(0.076) | (0.032) | (0.007) | (0.078) | (0.363) | (0.062) | (0.346) | (0.120) | |
R-squared | 0.495 | 0.597 | 0.378 | 0.887 | 0.458 | 0.976 | 0.448 | 0.604 |
Observations | 99 | 70 | 319 | 110 | 74 | 49 | 100 | 352 |
Regional FE | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
Baseline controls | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
. | Long difference, OLS . | |||||||
---|---|---|---|---|---|---|---|---|
. | (1) . | (2) . | (3) . | (4) . | (5) . | (6) . | (7) . | (8) . |
. | Denmark . | Finland . | Germany . | Italy . | Norway . | Spain . | Sweden . | United Kingdom . |
Panel A. Manufacturing employment | ||||||||
Exposure to robots (instrument) | −0.097 | −0.592 | −0.060 | −0.320*** | −0.506* | −0.206** | −0.233 | −0.238*** |
(0.523) | (0.701) | (0.051) | (0.085) | (0.302) | (0.076) | (0.178) | (0.080) | |
Exposure to Chinese imports (instrument) | −0.017*** | 0.025** | −0.011 | −0.104** | 0.002 | −0.049*** | 0.008 | −0.216*** |
(0.004) | (0.010) | (0.013) | (0.041) | (0.006) | (0.013) | (0.013) | (0.044) | |
R-squared | 0.486 | 0.557 | 0.357 | 0.886 | 0.340 | 0.976 | 0.448 | 0.604 |
Panel B. Non-manufacturing employment | ||||||||
Exposure to robots (instrument) | 0.625 | 0.446 | −0.043 | −0.225 | −1.391** | 0.279** | −0.075 | −0.144 |
(0.762) | (0.366) | (0.032) | (0.141) | (0.607) | (0.131) | (0.119) | (0.186) | |
Exposure to Chinese imports (instrument) | 0.014 | −0.001 | −0.049*** | 0.132 | −0.008 | 0.084*** | −0.006 | −0.032 |
(0.015) | (0.006) | (0.019) | (0.095) | (0.010) | (0.029) | (0.011) | (0.085) | |
R-squared | 0.353 | 0.848 | 0.334 | 0.715 | 0.416 | 0.906 | 0.583 | 0.365 |
Panel C. The effect of interaction between robots and Chinese imports on manufacturing employment | ||||||||
Exposure to robots (instrument) | −0.038 | −0.605 | −0.138* | −0.187 | −1.859*** | −0.156 | −0.235 | −0.248* |
(0.482) | (0.689) | (0.082) | (0.209) | (0.605) | (0.117) | (0.205) | (0.150) | |
Exposure to Chinese imports (instrument) | −0.009 | 0.001 | −0.020 | −0.098** | −0.006 | −0.041 | 0.008 | −0.217*** |
(0.008) | (0.014) | (0.014) | (0.046) | (0.006) | (0.028) | (0.015) | (0.044) | |
Exposure to robots × Chinese imports | −0.120 | 0.087*** | 0.012* | −0.046 | 0.824** | −0.024 | 0.003 | 0.011 |
(0.076) | (0.032) | (0.007) | (0.078) | (0.363) | (0.062) | (0.346) | (0.120) | |
R-squared | 0.495 | 0.597 | 0.378 | 0.887 | 0.458 | 0.976 | 0.448 | 0.604 |
Observations | 99 | 70 | 319 | 110 | 74 | 49 | 100 | 352 |
Regional FE | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
Baseline controls | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
This table presents OLS estimates of the impact of the exposure to robots and Chinese import competition on manufacturing employment in panel A as well as non-manufacturing employment in panel B. Panel C presents OLS estimates of the impact of the exposure to robots with the interaction between robots and Chinese imports. The outcome variables are the long difference in manufacturing or non-manufacturing employment-to-population ratio. Regressions are weighted by population in the start-of-period. The list of covariates is documented in Appendix A. The missing geographic values in Germany are due to confidentiality-related data limitations. Statistical significance based on robust standard errors (reported in parentheses) is denoted by ***P < 0.01, **P < 0.05, and *P < 0.10.
This PDF is available to Subscribers Only
View Article Abstract & Purchase OptionsFor full access to this pdf, sign in to an existing account, or purchase an annual subscription.