Table 5.
Estimation results.
| PSP | PGP | |
|---|---|---|
| Density of PSP | – | 0.371*** |
| (0.038) | ||
| Density of PGP | 0.679*** | – |
| (0.118) | ||
| Density of CSP | −0.036 | −0.060 |
| (0.167) | (0.091) | |
| Density of CGP | −0.376 | −0.396*** |
| (0.254) | (0.127) | |
| Density of private hospital bedsa | −0.017 | 0.003 |
| (0.077) | (0.025) | |
| Density of public hospital bedsa | 0.070*** | −0.022** |
| (0.026) | (0.010) | |
| Average incomea | 0.340 | −0.181 |
| (0.327) | (0.191) | |
| Educationa | 1.656** | 0.342 |
| (0.591) | (0.255) | |
| Living areaa | 0.010 | −0.011 |
| (0.028) | (0.012) | |
| Observations | 847 | 847 |
| R2 | 0.880 | 0.787 |
| Overidentification: χ2(13) | 13.953 | 10.727 |
| Endogenous regressor | PGP | PSP |
PSP, non-contract specialist; PGP, non-contract general practitioner; CSP, contract specialist; CGP, contract general practitioner. Bootstrapped standard errors in parentheses (50 replications).
a
Variables are time-invariant.
* Significance at 10 percent level.
**
Significance at 5 percent level.
***
Significance at 1 percent level.