13 Krugman/Obstfeld/Melitz • International Economics: Theory & Policy, Ninth Edition
economy producing two goods: cloth and food. Cloth is produced using labor and its specific factor, capital.
Food is produced using labor and its specific factor, land. Given that capital and labor are specific to their
respective industries, the mix of goods produced by a country is determined by share of labor employed in
each industry. The key difference between the Ricardian model and the Specific Factors model is that in the
latter, there are diminishing returns to labor. For example, production of food will increase as labor is added,
but given a fixed amount of land, each additional worker will add less and less to food production.
As we assume that labor is perfectly mobile between industries, the wage rate must be identical between
industries. With competitive labor markets, the wage must be equal to the price of each good times the
marginal product of labor in that sector. We can use the common wage rate to show that the economy will
produce a mix of goods such that the relative price of one good in terms of the other is equal to the relative
cost of that good in terms of the other. Thus, an increase in the relative price of one good will cause the
economy to shift its production toward that good.
With international trade, the country will export the good whose relative price is below the world relative
price. The world relative price may differ from the domestic price before trade for two reasons. First, as
in the Ricardian model, countries differ in their production technologies. Second, countries differ in terms
of their endowments of the factors specific to each industry. After trade, the domestic relative price will
equal the world relative price. As a result, the relative price in the exporting sector will rise and the relative
price in the import competing sector will fall. This will lead to an expansion in the export sector and a
contraction of the import competing sector.
Suppose that after trade, the relative price of cloth increases by 10 percent. As a result, the country will
increase production of cloth. This will lead to a less than 10 percent increase in the wage rate since some
workers will move from the food to the cloth industry. The real wage paid to workers in terms of cloth
(w/PC) will fall while the real wage paid in terms of food (w/PF) will rise. The net welfare effect for labor is
ambiguous and depends on relative preferences for cloth and food. Owners of capital will unambiguously
gain since they pay their workers a lower real wage while owners of land will unambiguously lose as they
now face higher costs. Thus, trade benefits the factor specific to the exporting sector, hurts the factor
specific to the import competing sector, and has ambiguous effects on the mobile factor. Despite these
asymmetric effects of trade, the overall effect of trade is a net gain. Stated differently, it is theoretically
possible to redistribute the gains from trade to those who were hurt by trade and make everyone better off
than they were before trade.
Given these positive net welfare effects, why is there such opposition to free trade? To answer this question,
the chapter looks at the political economy of protectionism. The basic intuition is that the though the total
gains exceed the losses from trade, the losses from trade tend to be concentrated, while the gains are diffused.
Import tariffs on sugar in the United States are used to illustrate this dynamic. It is estimated that sugar tariffs
cost the average person $7 per year. Added up across all people, this is a very large loss from protectionism,
but the individual losses are not large enough to induce people to lobby for an end to these tariffs. However,
the gains from protectionism are concentrated among a small number of sugar producers, who are able to
effectively coordinate and lobby for continued protection.
While the losers from trade are often able to successfully lobby for protectionism, the chapter highlights
three reasons why this is an inefficient method of limiting the losses from trade. First, the actual impact
of trade on unemployment is fairly low, with estimates of only 2.5 percent of unemployment directly
attributable to international trade. Second, the losses from trade are driven by one industry expanding
at the expense of another. This phenomenon is not specific to international trade and is also seen with
changing preferences or new technology. Why should policy be singled out to protect people hurt by trade,
and not for those hurt by these other trends? Finally, it is more efficient to help those hurt by trade by
redistributing the gains from trade in the form of safety nets for those temporarily unemployed and worker
retraining programs to ease the transition from import competing to export sectors.
Finally, the chapter uses the framework of the Specific Factors model to analyze the distributional effects
© 2012 Pearson Education, Inc. Publishing as Addison-Wesley