Finance & Development, December 2016, Vol. 53, No. 4

PDF version

Large gains, small losses

The June 2016 issue of Finance & Development contained some comments by Dani Rodrik about the gains and losses that arise from freer trade (“Rebel with a Cause”). Rodrik is quoted as saying that trade theory “shows that the larger the net gains, the larger the redistribution [that is needed]. It is nonsensical to argue that the gains are large while the amount of redistribution is small.” I wish to point out that these statements are not true in general for the reasons below. It is important to qualify these statements because they could be used to postpone liberalization.

Suppose that when a particular market is opened to trade, the domestic price falls and the good is imported. Since the price falls, producers are harmed, while consumers benefit. In technical terms, there is a loss in “producer surplus” and a gain in “consumer surplus.” The magnitude of the gain in consumer surplus and the loss in producer surplus depend, among other things, on the price elasticities of demand and supply, and there is no reason to believe that these are systematically related to each other. So, for a given price decline, the gain in consumer surplus will be very large and the loss in producer surplus will be very small when the price elasticities of demand and supply are both large. The gains and losses experienced by different groups from opening to trade depend on elasticity values, among other things, and it is an empirical matter just what these magnitudes are. It is quite possible that these elasticity values are such that the gains are large, while the losses, that is, “redistributions,” are small.

The level of domestic production relative to consumption, and hence imports, also matters for the size of the gains and losses. Suppose there is no local production of a product, so consumption is satisfied only by imports. In this case, reducing a trade barrier will increase consumer surplus, without any loss in producer surplus. Thus, in this case, there is no redistribution that would arise from opening a market to trade. How realistic is this? Quite.

To mention just one example, many developing economies do not produce automobiles, yet they impose various restrictions on their importation. So removing restrictions on imports of automobiles would generate a consumer benefit but no producer loss, and thus no redistribution. So when domestic production is very small relative to domestic consumption, and thus imports very large, the gains to consumers could be quite large, while the losses to producers quite small. Again, this is an empirical matter across countries. Computers and smartphones are other good examples for which the gains could be large while the losses small.

In short, the magnitudes of the gains and losses incurred by opening a market to trade depend on elasticities of demand and supply, as well as the size of domestic production relative to consumption. These are critical factors that qualify Rodrik’s remarks. It would be interesting to delve further into these matters using actual data across countries.

Stephen Tokarick
Senior Economist, IMF Research Department

Opinions expressed in articles and other materials are those of the authors; they do not necessarily reflect IMF policy.