Table 4.
IV Estimation of the LC impact on bilateral manufacture trade flows by subsample, 1995–2014
| (1) | (2) | (3) | (4) | (5) | (6) | |
|---|---|---|---|---|---|---|
| Variables | ln mijt | ln mijt | ln mijt | ln mijt | ln mijt | ln mijt |
| LC | -0.282 | 0.144 | 0.266 | -1.043 | 1.087** | 0.957* |
| (0.236) | (1.866) | (2.935) | (1.385) | (0.496) | (0.525) | |
| ln (Distanceij) | -1.311*** | -0.972*** | -1.106*** | -1.732*** | -1.443*** | -1.434*** |
| (0.043) | (0.138) | (0.177) | (0.145) | (0.098) | (0.099) | |
| Contiguityij | 0.026 | 1.367*** | -0.0971 | 0.615 | 0.704 | 0.730 |
| (0.133) | (0.261) | (0.191) | (0.540) | (0.573) | (0.576) | |
| Common Languageij | 0.576*** | 0.213 | 0.319 | 0.963*** | 0.615*** | 0.619*** |
| (0.088) | (0.366) | (0.238) | (0.240) | (0.119) | (0.120) | |
| Past Colonialij | 1.046*** | 0.506 | 0.959*** | 0.133 | 1.155*** | 1.153*** |
| (0.121) | (0.828) | (0.176) | (0.380) | (0.177) | (0.178) | |
| Common Colonialij | 0.986*** | 1.046** | 1.448* | 0.440 | 0.443** | 0.449** |
| (0.133) | (0.442) | (0.773) | (0.365) | (0.208) | (0.208) | |
| Depth of PTAijt | 0.403*** | 0.566 | 0.403 | 1.355* | 1.225*** | 1.207*** |
| (0.139) | (0.445) | (0.300) | (0.700) | (0.335) | (0.342) | |
| Sample | All | South to South | North to North | North to South | South to North | South to North |
| Observations | 48,008 | 3,572 | 15,679 | 6,474 | 21,297 | 21,297 |
| Cragg-Donald Wald F stat a/ | 4,423 | 47 | 21 | 317 | 6,083 | 3,430 |
| importer—year Fixed Effects (it) | Yes | Yes | Yes | Yes | Yes | Yes |
| exporter—year Fixed Effects (jt) | Yes | Yes | Yes | Yes | Yes | Yes |
IV estimates. Standard errors in parentheses are clustered at the dyad level; *p < 0.1; **p < 0.05; ***p < 0.01. First-stage estimations are reported in Table A3
a/ The Stock and Yogo (2005) weak identification test critical value at 10% maximal IV size (for Cragg-Donald F statistic and i.i.d. errors) is 16.38
Source: authors' computation