Table 2. Main analysis of the effect of academic freedom (AF) on innovation quantity (patent applications).
| Model | (1) | (2) | (3) | (4) | (5) | |||||
|---|---|---|---|---|---|---|---|---|---|---|
| Dependent variable | Patent applicationst+2 (ln) | Patent applicationst+2 (ln) | Patent applicationst+2 (ln) | Patent applicationst+2 (ln) | Patent applicationst+2 (ln) | |||||
| Statistic | Coeff. | (SE) | Coeff. | (SE). | Coeff. | (SE) | Coeff. | (SE) | Coeff. | (SE) |
| Indepdent variables | ||||||||||
| AF (z) | 0.345 | (0.032)*** | ||||||||
| AF (IV: Democracyt-5, z) | 0.477 | (0.060)*** | ||||||||
| AF (IV: Democracy stockt-10, z) | 0.442 | (0.057)*** | ||||||||
| AF (IV: AFt-5, z) | 0.330 | (0.043)*** | ||||||||
| AF (IV: AF stockt-10, z) | 0.315 | (0.042)*** | ||||||||
| Controls | ||||||||||
| Year/country dummies | Yes | Yes | Yes | Yes | Yes | |||||
| Countries (observations) | 157 | (12,392) | 157 | (11,809) | 157 | (11,257) | 157 | (11,569) | 156 | (10,753) |
| R2 (R2 adjusted) | 0.811 | (0.807) | 0.815 | (0.811) | 0.818 | (0.814) | 0.817 | (0.812) | 0.820 | (0.816) |
| MOP weak instrument test (F) | - | 2,395.13 | (τ < 0.05) | 4,114.46 | (τ < 0.05) | 5,946.18 | (τ < 0.05) | 11,895.90 | (τ < 0.05) | |
Notes: The estimates in Model (1) are obtained from a pooled OLS regression. The estimates in Models (2)–(5) come from 2SLS regressions. All variables are described in Table 1. Robust standard errors in parentheses. The MOP weak instrument is based on Montiel Olea and Pflueger (2013), as implemented in STATA’s weakivtest command. High F-values indicate a strong instrument.
* p < 0.10
** p < 0.05
*** p < 0.01.