Does International Trade Kill American Jobs?
By Douglas A. Irwin
In 1824 the great British historian Thomas Macaulay remarked that “free trade, one of the greatest blessings which a government can confer on a people, is in almost every country unpopular.” The popularity of free trade has not changed much since Macaulay’s day. The 1990s were a period of expanding world trade, strong economic growth, and the lowest U.S. unemployment in 30 years. Yet the decade began with fears of the “giant sucking sound” of jobs lost due to the North American Free Trade Agreement (NAFTA), and ended with opponents of free trade taking to the streets in the “Battle of Seattle.”
While free trade has always been the subject of complaint, the rhetorical charges against it have stepped up in recent years. Critics like Ralph Nader and Pat Buchanan rail against international commerce and the World Trade Organization for serving the interests of corporations rather than people, harming workers, decimating manufacturing industries, weakening environmental protections, and undermining American sovereignty. A wide range of groups, from religious organizations to human rights activists to Greens, have joined in the protests. The litany of complaints placed on the doorstep of free trade is hugely extensive, going well beyond the perennial objection that trade forces painful economic adjustments such as plant closings and layoffs. Nader charges that free trade “would make the air you breathe dirtier, and the water you drink more polluted. It would cost jobs, depress wage levels, and make workplaces less safe. It would destroy family farms and undermine consumer protections.”
Ironically, it was the inequities of protectionism that populists railed against a century ago. They warned that big businesses used tariffs to stifle competition and exploit consumers. They argued that free trade was the best way to ensure competition, discipline the power of domestic monopolies, and prevent politicians from using tariffs to give favors to special interests.
The clear conclusion of economists of all stripes is that the populists of a hundred years ago were right, and the populists of today are wrong. Free trade is a desirable economic policy for the “little man” at least as much as for the national economy. This conclusion is supported by extensive empirical evidence. Yet protectionism is far from vanquished in the political arena.
Industries that compete against imports will always promote their own interests by seeking trade restrictions. But today, the general public also has concerns about foreign competition. The argument that resonates most strongly is the claim that imports destroy jobs. Is this accurate? And if so, are import restrictions the remedy?
The claim that trade should be limited because imports destroy jobs has been trotted out since the sixteenth century. And imports do indeed destroy jobs in certain industries. Employment in the Maine shoe industry and the South Carolina apparel industry, for example, is lower because both industries were exposed to competition from imports. So we can understand why the plant owners and workers and politicians who represent them might prefer to avoid this foreign competition.
But just because imports destroy some jobs does not mean that trade reduces overall employment or harms the economy. After all, the dollars that U.S. consumers hand over to other countries in purchasing imports do not accumulate there. Those dollars, as an economic fact, must eventually all return to purchase either U.S. goods (exports) or U.S. assets (foreign investment). And both exports and foreign investment create new jobs here.
The claim that imports destroy jobs ignores the creation of jobs elsewhere in the economy as a result of trade. Since trade both destroys and creates jobs, the pertinent question is whether trade has a net effect on employment. In truth, the overall effect of trade on the number of jobs in an economy is best approximated as zero. That’s because total employment is not a function of international trade, but of the number of people in the labor force. Historical data show clearly that the number of jobs in the United States closely tracks the number of people who are available to work. That is to be expected in a free-market economy. Any imbalance in the numbers of workers and jobs will be offset by an adjustment of wages or some other change in economic equilibrium.
The same adjustments apply in trade. Some participants in the trade debate imagine that a country’s exports and imports are independent of one another, and that we ought to strive to reduce imports while increasing exports. But in truth, you can’t reduce one without having an adverse effect on the other—for exports and imports are flip sides of the same coin. You can’t have one without the other.
The mechanisms that link a country’s exports and imports to one another are complex and not always readily apparent, but they can be illustrated by focusing on the foreign exchange market. If the United States unilaterally reduces its tariffs on Japanese goods, for example, one would expect U.S. demand for Japanese goods to increase. To make these purchases, consumers in the United States will (indirectly) have to sell dollars on the foreign exchange market to purchase yen. In response to the increased demand for yen from those holding dollars, the value of the dollar will fall compared to yen. That raises the price of Japanese goods in the United States, dampening demand for those goods.
And there is a flip side: Even though it was the United States that lowered its tariff while Japan left its tariffs unchanged, Japan will now purchase more goods from the United States. This is because the cheaper dollar lowers the yen price of U.S. goods, stimulating Japanese demand for them. Economic statistics going back more than a century document this sort of direct link between outgoing and incoming trade.
The foreign exchange market is just one of several mechanisms that link exports and imports. And together, these mechanisms ensure that exports and imports, rather than being separate forces, one “good” and one “bad,” are actually related phenomena that rise and fall together. Indeed, the best way to think of exports may be as goods that a country must give up in order to acquire the imports it desires. Economically, both imports and exports are good for an economy and for consumers.
This is why, throughout U.S. history, large tariff increases that choke off imports have failed to stimulate greater employment. Any increase in employment in import-competing industries is offset by a decrease in employment in export-oriented industries. The Smoot-Hawley tariff of 1930, for example, significantly reduced imports but failed to create jobs overall because exports fell almost one-for-one with imports, resulting in employment losses in those industries.
The connection between imports and exports cannot be overlooked when evaluating trade policy. Governments that undertake policies to reduce imports will find themselves reducing exports also. That merely trades one expansion for another contraction, and both transactions will end up being done less efficiently.
Not only do import restrictions reduce the number of jobs producing exports. They also directly destroy jobs in downstream industries that use imports. Keep in mind that the majority of U.S. imports are not final consumer goods, but intermediate goods used by domestic firms in their production processes. Any trade restriction that raises the price of an intermediate good directly harms downstream user industries, and this adversely affects employment in those industries.
Restrictions on imported sugar, for example, have reduced U.S. employment in the sugar refining and candy making industries. Because our food manufacturers who produce sugar-intensive products must pay a higher price for sweetener than their foreign rivals, their competitive position has suffered. In 1990, Brachs Candy Company announced that due to the high domestic price of sugar it would close a factory in Chicago that employed 3,000 workers and expand production instead in Canada—which does not artificially inflate the price of sugar to protect its sugar producers. In 1988, the Department of Commerce estimated that the high price of domestic sugar due to U.S. protectionism cost almost 9,000 jobs in food manufacturing because of increased imports of cheaper sugar-containing products, and 3,000 jobs in the sugar-refining industry because of lower demand for sugar. At the time of this study, U.S. sugar-producing farms employed about 35,000 workers—but the sugar-processing and sugar-using sectors employed about 708,000 workers! A great many workers in the sugar-using industries were put at risk, in other words, to save the jobs of the few workers in the sugar-producing industry.
We have seen that framing trade policy in terms of employment is ultimately an empty exercise. Blocking imports may protect some jobs, but it harms others. Yet even those who agree that the overall effect of trade on employment is essentially zero may oppose free trade because they believe that it shifts jobs into less desirable sectors. The gravest of such concerns is that in the last three decades good jobs in manufacturing have been traded for bad jobs of other sorts. Has trade actually had this effect on the U.S.?
The economics reveal that the popular perception that imports destroy good, high-wage jobs in manufacturing is almost completely erroneous. It is closer to the truth to say that imports destroy bad, low-wage jobs in manufacturing.
This is because wages in industries that compete against imports are well below average, whereas wages in exporting industries are well above average. The United States tends to import labor-intensive products, such as apparel, footwear, leather, and goods assembled from components. U.S. companies in these labor-intensive sectors tend to employ workers with lower than average educational attainment and relatively low wages. In 1999, average hourly earnings of Americans working in the apparel industry were 36 percent less than in manufacturing as a whole. Average hourly earnings were 30 percent lower than average in the leather industry and 23 percent lower in the textile industry.
By contrast, the products the United States tends to export are more skill-intensive, such as aircraft, construction machinery, engines and turbines, and industrial chemicals. Workers in these industries earn relatively high wages. In 1999, average hourly earnings in the aircraft industry were 42 percent above the average in manufacturing. Wages were 8 percent higher in industrial machinery, and 24 percent higher in pharmaceuticals. One study reports that even “after being adjusted for skill differences, wages in export-intensive industries are 11 percent above average, whereas wages in import-intensive industries are 15 percent below average.”
As a result, any policy that limits overall trade by reducing exports and imports tends to increase U.S. employment in low-wage industries and reduce U.S. employment in high-wage industries. Restricting trade shifts American workers away from things that they produce well (and hence export and earn high wages in producing), and toward things that they do not produce so well (and hence import and earn low wages in producing). Under trade restrictions, employment gains for low-wage textile factory operators would be offset by employment losses for high-wage machinists and engineers in aircraft and pharmaceutical plants.
Because exports increase the number of workers in relatively more productive, high-wage industries, and imports reduce the number of workers in relatively less productive, low-wage industries, the overall impact of trade in the United States is actually to raise average wages. This is the opposite of what many people fear: that international competition, particularly with low-wage developing countries, will unleash a “race to
the bottom” in which wages are cut to match lower labor costs elsewhere.
Such fears are misguided. High American wages are based on the high productivity of U.S. workers. Foreign competition cannot take away the advantages that give rise to this high productivity, namely, the use of sophisticated technology, a substantial investment in education and human capital, and the many other advantages of operating in the U.S. market.
Data show that the growth of U.S. worker compensation tracks growth in worker productivity almost perfectly over the last half century. Foreign competition does not suppress this growth. In fact, trade can foster growth in productivity in several ways, thus actually increasing worker compensation.
The truth is, firms that compete against imports cannot cut wages—because they do not determine what those wages will be. Those firms must pay the prevailing market wage as determined by the productivity of workers. All that a company facing import competition can do is reduce employment; if it tries to cut wages, its best workers just leave to take opportunities elsewhere in the economy.
Although average wages are determined by the underlying attributes that make American workers productive, trade can affect the distribution of wages in an economy. This is implicit in what we have discussed: Trade creates jobs in high-wage industries in which the United States exports (aircraft, machinery), and reduces jobs in low-wage industries in which the United States imports (apparel, footwear). Some economic interests are therefore bound to be harmed by free trade and will seek to keep imports out of the domestic market. But there are overall benefits to many workers from free trade that exceed the losses to other workers.
In any rapidly changing economy, jobs are continuously created and eliminated. Changes in consumer tastes, domestic competition, growth in productivity, technological innovation, and international trade all contribute to the churning of the labor market. It is virtually impossible to disentangle all of the reasons for job displacement because they are interdependent; for example, technological change may be stimulated by domestic or foreign competition.
Yet to the extent that such attributions can be made, the available economic evidence suggests that trade is a small factor in the displacement of American workers. According to the Bureau of Labor Statistics, import competition was responsible for only 1.5 percent of total layoffs from 1996 to 1999. And counterbalancing this, as we have seen, are many positive effects on workers from unshackled trade.
Free trade, it is clear, is not only very good for economies and for consumers. It is also good for the average American worker.
