3.3 Trade and Specialization
Finally, we estimate the impact of trade on specialization (h), equation (6). Table 7 reports
the estimation results. Column (1) is the bivariate OLS regression of trade openness on the
Herfindahl index, while column (2) controls for log per capita PPP-adjusted GDP from Penn
World Tables. The coefficient on trade is significant at the one percent level. Since trade
openness is likely endogenous to diversification, columns (3) and (4) repeat the exercise
instrumenting for trade using natural openness from Frankel and Romer (1999). Results
are unchanged, and the magnitude of the coefficient is not affected dramatically. In order
to probe further into this finding, we also analyze more directly how the export patterns
are related to industrial specialization. We construct the Herfindahl index of export shares
in a manner identical to our index of production concentration. The results are presented
in column (5). The coefficient on trade openness decreases by about one third, but remains
significant at the one percent level. The coefficient on the Herfindahl of export shares is
highly significant as well.
We illustrate these results in Figure 4, which presents partial correlations between trade
openness and the Herfindahl index of sector shares for the available countries, once per
capita income has been netted out. It is clear that there is a positive relationship between
trade and specialization. The effect of trade openness and export concentration on the
specialization of production is sizeable. A one standard deviation change in log trade
openness is associated with a change in the log Herfindahl of production equivalent to about
0.53 of a standard deviation. A one standard deviation change in export specialization is
associated with a change in the log Herfindahl of production of roughly 0.65 standard
deviations.
Appendix Table A7 presents further robustness checks. Breaking up the sample into
developed and developing countries in columns (1) and (2), we see that the phenomenon
is especially prevalent in the developing countries. Column (3) checks whether our results
are driven by outliers. Dropping outliers improves the fit of the regression, and the results
remain significant. Columns (4)-(6) repeats the three previous exercises while including
the Herfindahl index of export shares. The trade openness coefficient remains positive and
significant. Columns (7) and (8) check if the results are robust to an alternative measure of
trade openness. We use total trade openness as a share of GDP from the Penn World Tables
instead of total manufacturing trade as a share of manufacturing output from our data. It
is clear that the main result is not driven by our particular measure of trade openness.
Finally, the specification in column (9) is based on the work of Kalemli-Ozcan et al. (2003),
and includes a wide variety of other controls, such as population density, population, share
13