Robustness test – system GMM (endogeneity)
| Dependent variable | ESG (β) | Robust SE | p-value | Lagged DV (β) | Robust SE | Hansen J (p) | AR(2) (p) | Instruments |
|---|---|---|---|---|---|---|---|---|
| ROA | 0.028 | 0.013 | 0.031 | 0.401 | 0.058 | 0.284 | 0.462 | 18 |
| ROE | 0.081 | 0.036 | 0.024 | 0.352 | 0.075 | 0.317 | 0.409 | 19 |
| Tobin's Q | −0.037 | 0.016 | 0.021 | 0.498 | 0.069 | 0.338 | 0.441 | 17 |
| Dependent variable | ESG ( | Robust SE | Lagged DV ( | Robust SE | Hansen J ( | AR(2) ( | Instruments | |
|---|---|---|---|---|---|---|---|---|
| ROA | 0.028 | 0.013 | 0.031 | 0.401 | 0.058 | 0.284 | 0.462 | 18 |
| ROE | 0.081 | 0.036 | 0.024 | 0.352 | 0.075 | 0.317 | 0.409 | 19 |
| Tobin's Q | −0.037 | 0.016 | 0.021 | 0.498 | 0.069 | 0.338 | 0.441 | 17 |
Note(s): Two-step System GMM with Windmeijer-corrected robust standard errors. ESG is treated as endogenous; lags t−2 and deeper are used as instruments. Instrument count is kept below the number of groups to limit instrument proliferation. Hansen J p-values above 0.05 and insignificant AR(2) tests indicate no evidence against instrument validity or second-order serial correlation
Sharing content requires targeting cookies to be enabled. Please update your cookie preferences to use this feature.