Determinants of Bilateral Trade: Does Gravity Work in a Neoclassical World?

TLDR

The study derives equations for bilateral trade values from two extreme Heckscher‑Ohlin cases, illustrating how simple gravity forms arise. The authors derive bilateral trade expressions using Cobb‑Douglas and CES preferences, showing how transport costs affect gravity deviations. In a frictionless trade setting, producers and consumers are indifferent among partners, while in a complete‑specialization scenario each country produces distinct goods. The paper shows that frictionless trade yields the simple gravity equation under homogeneous preferences, while in specialized economies the gravity form persists with systematic deviations driven by transport costs, leading to the conclusion that gravity is broadly derivable yet its empirical use to test specific models is questionable.

Abstract

This paper derives equations for the value of bilateral trade from two extreme cases of the Heckscher-Ohlin Model, both of which could also represent a variety of other models as well. The first case is frictionless trade, in which the absence of all impediments to trade in homogeneous products causes producers and consumers to be indifferent among trading partners. Resolving this indifference randomly, expected trade flows correspond exactly to the simple frictionless gravity equation if preferences are identical and homothetic or if demands are uncorrelated with supplies, and they depart from that equation systematically when there are such correlations. The second case is of countries that each produce distinct goods, as in the H-O Model with complete specialization or a variety of other models. Expressions are derived for bilateral trade, first with Cobb-Douglas preferences and then with CES preferences. The standard gravity equation with trade declining in distance continues to be a central tendency for these trade flows, with departures from it that are easily understood in terms of relative transport costs. The main lessons from the paper are two. First, it is not all that difficult to justify even simple forms of the gravity equation from standard trade theories. Second, because the gravity equation appears to characterize a large class of models, its use for empirical tests of any of them is suspect.