Table 4.
Estimation of economies of scale: regression results.
| Model 1 Fixed-effects | Model 2 System GMM | Model 3 System GMM | Model 4 System GMM | |
|---|---|---|---|---|
| Log(AC)i,t-1 | 0.250** (0.072) | 0.230** (0.075) | 0.171* (0.064) | |
| Log(Scale) | −0.239*** (0.016) | −0.268*** (0.014) | −0.246*** (0.029) | −0.601** (0.192) |
| Contact quality | 0.241 (0.226) | −0.544 (0.500) | ||
| Log(Scale) x Contact quality | 0.482+ (0.260) | |||
| Constant | 4.879*** (0.108) | 3.772*** (0.266) | 3.575*** (0.250) | 4.318*** (0.449) |
| Time fixed effects | Yes | Yes | Yes | Yes |
| HB-HCT clusters fixed effects | Yes | Yes | Yes | Yes |
| R-squared | 0.657 | |||
| No. of groups | 22 | 22 | 22 | 22 |
| No. of time periods per group (average) | 17.8 | 14.1 | 13.7 | 13.7 |
| No. of observations | 391 | 310 | 302 | 302 |
| No. of instruments | 11 | 13 | 13 | |
| Arellano-Bond test for AR(2) (p-value) | 0.425 | 0.749 | 0.768 | |
| Hansen test of over-identification (p-value) | 0.105 | 0.145 | 0.569 |
Notes:+p < 0.1, ∗ p < 0.05, ∗∗ p < 0.01, ∗∗∗ p < 0.001. Standard errors in parentheses (cluster-robust in Model (1), and Windmeijer-corrected cluster-robust in Model (2)).