We show that any well-behaved demand function can be represented by its “demand manifold,” a smooth curve that relates the elasticity and convexity of demand. This manifold is a sufficient statistic for many comparative statics questions; leads naturally to characterizations of new families of demand functions that nest most of those used in applied economics; and connects assumptions about demand structure with firm behavior and economic performance. In particular, the demand manifold leads to new insights about industry adjustment with heterogeneous firms, and can be empirically estimated to provide a quantitative framework for measuring the effects of globalization. (JEL F12, L11)

We provide a general characterization of which firms will select alternative ways of serving a market. If and only if firms' maximum profits are supermodular in production and marketaccess costs, more efficient firms will select into the activity with lower market-access costs. Our result applies in a range of models and under a variety of assumptions about market structure. We show that supermodularity holds in many cases but not in all. Exceptions include FDI (both horizontal and vertical) when demands are "sub-convex" (i.e., less convex than CES), fixed costs that vary with access mode, and R&D with threshold effects.

The GATT/WTO Article XXIV prevents a customs union (CU) from raising its common external tariff. In this paper, we compare CU formation with and without this Article XXIV constraint. We show, in a multi-country oligopoly model, that for a given CU structure, Article XXIV improves world welfare by lowering trade barriers.However, we also show that Article XXIV has a composition effect on CU formation: it affects the endogenous choice of CUs. By encouraging more symmetric CUs, Article XXIV causes more trade to be subject to trade barriers. As a result, Article XXIV may be 'bad' for world welfare.

Gravity as both fact and theory is one of the great success stories of recent research on international trade, and has featured prominently in the policy debate over Brexit. We first review the facts, noting the overwhelming evidence that trade tends to fall with distance. We then introduce some expository tools for understanding constant-elasticity-of-substitution theories of gravity as a simple general-equilibrium system. Next, we point out some anomalies with the theory: mounting evidence against constant trade elasticities, and implausible predictions for bilateral trade balances. Finally, we sketch an approach based on subconvex gravity as a promising direction to resolving them.