Table 4.
Robustness tests and treatment of endogeneity problems.
| （1） Substitution of explanatory variables | （2） Shrinkage processing | (3) Excluding municipalities | （4） Differential GMM | （5） System GMM | （6） Tool variables | |
|---|---|---|---|---|---|---|
| Digitalshangzhi | −0.023*** (0.003) | |||||
| Digital | −0.355*** (0.076) | −0.114* (0.067) | −0.045*** (0.017) | −0.010** (0.004) | −1.260*** (0.260) | |
| L.Ce | 0.834*** (0.017) | 0.905*** (0.016) | ||||
| Control variables | Yes | Yes | Yes | Yes | Yes | Yes |
| Individual fixed effects | Yes | Yes | Yes | Yes | Yes | Yes |
| Year fixed effects | Yes | Yes | Yes | Yes | Yes | Yes |
| LM Statistical quantities | 29.739 [0.000] | |||||
| Cragg-Donald Wald F | 32.430 {16.38} | |||||
| AR（2） | 0.392 | 0.405 | ||||
| Hansen | 1.000 | 1.000 | ||||
| N | 300 | 300 | 260 | 240 | 270 | 300 |
Note: ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively; values in parentheses are standard errors, values in square brackets are P-values, and values in curly brackets are critical values for the Stock-Yogo test at the 10% level. AR(2) and Hansen values are P-values for the test.