Rescuing observed fixed effects: Using the Hausman-Taylor method for out-of-sample trade projections

In this analysis, we propose the Hausman-Taylor (1981) method as an alternative estimation technique for estimating the gravity model of trade. We use an application to highlight the benefits of this technique for panel data estimation in general. Specifically, we compare the Hausman-Taylor method for estimating the unrealized US-Cuban trade potential to the OLS, fixed-effects, and random-effects methods using the out-of-sample approach. The Hausman-Taylor method is ideal because it allows for the inclusion of time-invariant variables in trade projections and circumvents the problem of an ad hoc estimation of the country-specific dummy variable needed for a projection based on the fixed-effects estimator. In addition, it removes the correlation between the error term and included variables which often plagues random-effects estimation.