International trade, gravity equations and tourism: an application for intra-latin American trade flows

Abstract

The main objective of the present chapter is to revisit the traditional estimates from the trade gravity equation in two ways. First, by recognizing the presence of "zero" trade flows between countries by following the methodology proposed by Helpman, Melitz and Rubinstein (2008), thereafter HMR. Disregarding countries that do not trade with each other implies that important information is not being considered and hence estimates could be biased. Second, the potential omission of a relevant variable that may lead to biased estimates is addressed. In particular, tourist arrivals is introduced as a potential omitted factor in traditional trade gravity specifications since international visits could reduce both, fixed and variable costs of exporting. In that sense, the framework proposed by HMR is modified to theoretically justify the inclusion of tourism as an explanatory variable for explaining international trade. Finally, an empirical exercise is carried out to describe intra-Latin American trade flows for the period 1995-2010. Results suggest that not considering "zero" trade flows and the omission of tourism may contribute to explain the presence of an upward bias in the estimation of the effect of key variables of gravity equations.