This paper looks at the interrelationship between trade in goods and asset holdings, as brought forward by some contributions to the empirical literature in international economics. These contributions argue that single-equation gravity models suffer from an endogeneity bias, culminating in a request for the estimation of systems of gravity equations. Yet, the theoretical basis for such an interrelationship is weak. In this paper we present baseline models of international trade in goods and bank asset holdings, which yield gravity equations that can be tested empirically. We then use these models to test three different explanations for the interrelationship between trade in goods and asset holdings that have been brought forward by the literature: (i) consumption hedging, (ii) sovereign risk, (iii) information spillovers. Our results indicate that none of these channels can explain the interrelationship. We therefore conclude that single-equation gravity models, as opposed to systems of gravity equations, are justified: All you need is trade!
International trade international banking gravity equations