Remarks by Vice Chairman Roger W. Ferguson, Jr.
At the Conference on Trade and the Future of American Workers, Washington, D.C.
October 7, 2004
Free Trade: What Do Economists Really Know?
The role of trade in the U.S. economy has moved well into the spotlight in recent years, and I am pleased to be here today to share my thoughts on this important topic with such distinguished and knowledgeable colleagues. Over the course of this day, you will be hearing from leading analysts, policymakers, and commentators about recent developments in the U.S. economy, past and prospective trends in job creation, the role of sourcing (both out- and in-), and the implications of trade for the coming elections. In my remarks this morning, I would like to put these issues into the broader context of the debate over free trade and its implications for the American economy.
Though my focus will be on free trade, we must remember that prospects for the average American depend on many other factors as well, including technological progress, the education required to exploit this progress, a dynamic market-oriented economy, a framework of limited but effective regulation, healthy and well-governed financial institutions, and a stable macroeconomic environment. And free trade is not necessarily the most important item on this list. Even so, it has been a focus of interest and aspiration for economists dating back to Adam Smith and David Ricardo. As you know, finding overwhelming agreement on issues is difficult among economists, but free trade is an exception.1
The supporters of free trade have not been ignored. In the past half-century, global trade has become freer and has expanded rapidly. The ratio of trade (exports plus imports) to worldwide gross domestic product rose from only 16 percent in 1960 to 40 percent by 2001. In 1960, the United States, Germany, and Japan had average tariff rates of around 7 percent; these rates were more than halved by 1993. The number of members of the World Trade Organization (WTO), or its predecessor, the General Agreement on Tariffs and Trade, rose from 18 in 1948 to 146 in 2003, and the number of regional trade agreements in the world ballooned from only 1 in 1958 to 161 in 2003.
Although most economists welcome these trends, the public at large has been much more ambivalent about international trade. Attitudes toward free trade in principle remain generally positive, but a substantial--and, perhaps, growing--minority of Americans hold more negative views. According to a poll completed around the beginning of this year, 41 percent of respondents viewed the process of increasing international trade through reduction of barriers as proceeding too quickly; this number was up from 30 percent in 1999. And 43 percent of respondents believed that the government should try to slow or reverse the expansion of international trade, up from 39 percent in 1999.2
What accounts for the apparent deterioration in public support for free trade over the past five years? The widening of the U.S. trade deficit may have exacerbated concerns about the country's international competitiveness. More important, some have blamed overseas competition for the job losses associated with the economic slowdown earlier in this decade.
Without solid public support for free trade, achieving continued progress in reducing protectionist barriers, both at home and abroad, may become more difficult. In the remainder of my remarks, I'd like to review the arguments for and against free trade, explore why it has been difficult to muster more widespread public support for this goal, and address some of the consequences of trade protection as it has been implemented in practice.
for Free Trade
First, and most obviously, trade increases the variety of goods available to consumers. Trade provides some products that otherwise would be beyond the reach of most American households, such as roses for Valentine's Day, or peaches and nectarines during the winter. More generally, international trade allows us to choose from a wider array of goods than would otherwise be available: Japanese and German cars in addition to American, Chilean apples as well as Washington state, French and Australian wine as well as Californian. It is difficult to put a dollar figure on the value of this increased variety to the consumer, but estimates range as high as nearly 3 percent of GDP.3
A second benefit of international trade is its role in reducing the cost of goods and hence in raising our standard of living. To anyone who has walked into a large discount store and surveyed the range of low-priced items produced in any number of distant economies, this benefit is abundantly clear. However, actually measuring the extent to which trade holds down consumer costs is tricky. Between 1990 and 2003, for example, the overall consumer price index rose 41 percent, whereas prices declined for many highly traded goods, including toys (whose prices fell 26 percent), televisions (53 percent), and clocks and lamps (15 percent); in just the past five years, the price of telephones, calculators, and other such items has fallen 42 percent. Yet, we do not know how much of the decline in these prices can be attributed to trade, as most traded products are manufactures and are subject to greater productivity growth (and hence steeper declines in costs) than nontraded products such as services.
A more fruitful approach may be to compare the prices of goods that are protected from international competition with what they would be in the absence of such barriers. A recent study by the U.S. International Trade Commission indicates that sectoral trade liberalization would lower the price of sugar for U.S. consumers by 8 percent, of apparel by 5 percent, and of footwear and leather products by 4 percent.4 Clearly, if international trade were curtailed for a much broader range of goods, the cost of living for American workers would be higher and the standard of living correspondingly lower.
A third key benefit of free trade is that it allows economies to specialize in the activities they do best. This notion was at the core of the classical economists' defense of free trade. By allowing England to specialize in cloth production and Portugal in wine, for example, international commerce leads to a higher income for both countries than if each tried to produce both goods for themselves. By the same token, no American today would object to trade between Massachusetts and Montana, or between Alaska and Alabama--the various U.S. states obviously have their own comparative advantages in producing a variety of different products, and trade among them makes such specialization possible. Extending the example of trade among states to trade among countries is not much of a stretch.
Can we measure the extent to which the specialization associated with free trade may boost incomes and welfare? Such an estimate is obviously no simple thing to calculate. Economists frequently use so-called computable general equilibrium models, often consisting of hundreds of equations, to address this issue. A recent analysis of the effects of past trade liberalizations on the U.S. economy puts the gains to U.S. welfare at about 1/2 percent of GDP.5 A separate analysis of a hypothetical 33 percent reduction in trade barriers around the world suggests it would raise welfare by 1-1/2 percent of global GDP.6
In addition to promoting specialization, trade boosts productivity through a fourth channel of influence: opening the economy to heightened competition. This effect could occur either as firms are spurred by foreign competitors to become more efficient, or as the least productive firms are forced to close, thus raising the average level of productivity for the economy as a whole. Again, most Americans likely recognize the importance of competition in boosting performance--the ascendancy of Japanese automobiles, for example, has been cited as a factor that has spurred Detroit to greater innovation and better quality. By heightening competitive forces and thus incentives for productivity and innovation, international trade has likely accelerated the process of "creative destruction" by which outdated and less productive activities are replaced by new technologies and more dynamic enterprises.
Academic research supports the view that import competition has led U.S. manufacturing firms to become more capital intensive;7 trade liberalization apparently has enhanced productivity in some import-competing firms in foreign countries as well.8 Producing for export markets may also yield dividends: Research suggests that exporters are more productive than non-exporters in the same industry and that they grow more rapidly as well.9 Finally, many studies suggest that countries that are more open to international trade have enjoyed higher rates of economic growth.10 Our sad experience after adoption of the Smoot-Hawley tariff of 1930, as well as the record of Latin America, India, and other regions that experimented with "import-substituting industrialization," point to the deterioration in economic performance that occurs when countries erect barriers to trade.11
Arguments against Free Trade
One concern about free trade may be that it has given rise to large trade and current account deficits, thereby adding to the nation's debt and putting future prosperity at risk. Now at more than 5 percent of GDP, the current account deficit is in record territory, it is growing, and it cannot be sustained indefinitely. We cannot foresee when the deficit will stop growing and return to more-sustainable levels, through what mechanisms this adjustment will occur, or whether this adjustment will be smooth or disruptive for financial markets and the economy more generally. No matter how a correction of the external imbalance proceeds, however, it will involve a range of adjustments to investment, saving, and asset prices, both for the U.S. economy and for our trading partners. Research suggests that past corrections of large external imbalances in industrial countries generally have occurred without crisis.12 Whether or not this will remain the case, I am confident that protectionism is not the appropriate response to our growing current account deficit. The amount of current account adjustment that would be gained from a given tightening of import controls is questionable. Yet, it is certain that such actions would impose costs on the economy that would persist long after concerns about the deficit dissipated.
A second concern about free trade that is frequently voiced, and probably a more important one to many people, is that trade destroys American jobs and creates unemployment. The same survey I mentioned earlier, showing a deterioration in general attitudes toward trade, also indicated that 40 percent of respondents believed that trade barriers should be maintained because of the threat to U.S. jobs, up from 31 percent in 1999.13
It is worth distinguishing among several variants of the concern about trade and jobs. The first variant holds that the rise in imports lowers employment and raises the unemployment rate by shifting jobs overseas. This claim is strongly contradicted both by theory and by experience. Make no mistake: Import competition clearly has cost some American workers their jobs and has caused them considerable hardship as a result. However, economywide equilibrating forces, including monetary policy, ensure that over time such employment losses are offset by gains elsewhere in the economy, so that the nationwide unemployment rate averages around its equilibrium level. In fact, the inflow of foreign capital that finances our trade deficit provides the funding for investment projects that employ U.S. workers just as surely as does any other productive activity in the economy. Between 1960 and 2003, the trade balance moved from a slight surplus to a deficit of 4-1/2 percent of GDP, and nominal imports rose from about 4 percent of GDP to 14 percent--yet, the current unemployment rate of about 5-1/2 percent is little changed from its 1960 level, while nonfarm private employment has grown by more than 60 million jobs.14
It has also been suggested that import competition has caused a significant portion of the decline in employment since the recession of 2001. Yet, the ratio of the nominal trade deficit to GDP widened less than 1 percentage point between 2000 and 2003. Moreover, this deterioration came entirely from a decline in the ratio of exports to GDP, from 11.2 percent in 2000 to 9.5 percent in 2003; the ratio of imports to GDP actually declined about 1 percentage point over this period.
A second variant of the concern over trade and jobs is certainly valid: Import competition can be highly disruptive and cause considerable pain for those who lose their jobs. One study of worker displacement indicates that only about two-thirds of displaced workers found another job within three years, and even when they were successful in finding full-time work, the earnings of these workers on average declined 8 percent.15 Another study found that job losers in industries facing heavy import competition were slightly less likely to be reemployed, and suffered greater earnings losses, than workers who lost their jobs in industries facing less import competition.16
We cannot and should not minimize the hardships of workers displaced by imports. However, we must also keep in mind that their numbers are relatively small compared with either the total labor force or even the total number of jobs lost in the United States. Estimates of the gross number of jobs lost to imports vary, but one representative estimate puts them at a bit more than 300,000 per year during the 1980s and 1990s.17 This number, while hardly negligible, is dwarfed by the roughly 15 million job losses estimated to occur each year in the United States. As our dynamic market economy evolves, it generates substantial churning in labor markets as jobs are gained in some sectors and lost in others; jobs gained and lost because of trade are only a small part of that process.
It is understandable that concerns about job losses from import competition may extend far beyond their actual incidence in the labor market, given more general anxieties about employment security among American workers. However, to echo a point that has been made before, the proper response to the disruptions associated with trade is not to reduce trade, but rather to ameliorate the pain associated with those disruptions through enhanced assistance and retraining for displaced workers.
A final concern about trade that I would like to discuss is that import competition, whether or not it affects the number of jobs, shifts the employment mix from high-quality jobs to low-quality jobs. For example, critics have long held that international trade pushes workers out of manufacturing jobs and into less desirable service-sector jobs. However, no conclusive evidence has shown that, over the long haul, the service jobs being created pay less or are otherwise less desirable than manufactured jobs being displaced. Moreover, the declining share of manufacturing in U.S. employment most likely stems less from import competition than it does from the rapid pace of productivity growth in manufacturing; this growth outpaced the productivity growth of the overall economy by about 1-1/4 percentage points annually from 1973 to 1994 and by 1-1/2 percentage points from 1994 to 2000. The higher rate of productivity growth in manufacturing has restrained both price increases and employment in the sector, thus leading the services area of the economy to expand its share of spending and jobs. This phenomenon is hardly unique to the United States--the share of manufacturing has declined in most of our major foreign trading partners as well.
More recently, the outsourcing of service jobs to developing countries has come under the spotlight. The increasing use of computer programming talent in India and other low-wage countries has, understandably, struck a chord of anxiety among American workers. For years, the response of pro-trade advocates to the loss of low-wage jobs in manufacturing has been that they are being made up by the creation of higher-paid, higher-skilled jobs in the service sector. The loss of highly paid programming jobs to lower-paid workers abroad now appears to suggest that there is no place where American workers can hold their own.
Yet, as in the case of import competition more generally, we must not exaggerate the importance of outsourcing to the nation's overall employment picture. There are no conclusive data, but a prominent study puts the number of jobs displaced through services outsourcing over the next decade or so at fewer than 300,000 annually, or less than 2 percent of the 15 million in total gross job losses I noted earlier.18 Moreover, only a fraction of those jobs represent high-skilled, high-wage jobs; these numbers are quite difficult to pin down, but one study puts the number of software jobs lost to India since 2000 at fewer than 50,000 annually.19 Finally, we should remember that the United States gains jobs through what is often referred to as "insourcing," that is, performing service jobs for other countries. In fact, the United States has consistently run a surplus in those categories of the balance-of-payments associated with trade in business services.
Turning from the sectoral job mix to the impact of import competition on wages, the evidence is particularly unclear. Some studies have suggested that import competition from low-wage countries has depressed wages for low-skilled workers relative to those for higher-skilled workers in recent decades. However, other studies have argued that the rise in skill premiums is attributable to technological developments that have raised relative demands for educated workers. Focusing on the past few years, we see no consensus on how the mix of low- and high-wage jobs in the economy has evolved; estimates are extremely sensitive to the definition of job classes, the source of data, the time period, and method of calculation. In any event, it is doubtful that changes in the pattern of wages in the U.S. economy can be explained by any single factor--trends in trade, in population and immigration, in unionization and labor market competition, in minimum wage policy, in the skill mix of the labor force, and in technology all play a role.
Drawbacks of Protectionism
Rather than arguing the merits of international trade in the abstract, advocates of free trade might gain more traction by arguing against concrete examples of protectionism. Each year brings new actions by the U.S. government to protect individual sectors from imports. Antidumping duties are imposed when domestic industry is believed to be injured by the sale of imported goods at less than "fair value." Countervailing duties are intended to counteract subsidies to foreign producers. Safeguard actions are intended to protect a domestic industry that has been seriously injured by a surge in imports.20 As of August 2004, 359 antidumping and countervailing duty orders were in place in the United States against imports from 51 countries.21
By discouraging unfair commercial practices, such actions, in principle, promote a more stable and competitive environment for international trade. In practice, identifying anticompetitive practices is a murky process. For example, in antidumping cases, determining the "fair value" of a good may involve a degree of discretion, thereby complicating the assessment of whether foreign goods are being sold below their appropriate price.22 Domestic producers have a strong incentive to lobby for trade actions regardless of whether such actions are merited.
Because they inhibit free trade, protectionist actions have an array of adverse consequences that one would expect: They reduce variety and raise costs for consumers; they distort the allocation of resources in the economy by encouraging excessive resources to flow into protected sectors; and they foster inefficiency by reducing the extent of competition. Perhaps more important in the eyes of the public, however, may be several related and highly egregious consequences of protectionist actions.
First, by raising the cost of goods that are inputs for other producers, import barriers may destroy more jobs in so-called "downstream" sectors than they save in protected sectors. According to one study, the 2002 steel safeguard program contributed to higher steel prices that eliminated about 200,000 jobs in steel-using industries, whereas only 187,500 workers were employed by U.S. steel-producers in December 2002.23
Second, trade protection may lead to very large payouts to a small number of producers and hence is often inequitable. Any time a product receives import protection, of course, a relatively small number of domestic producers receive benefits--through higher prices--at the cost of all domestic consumers in the economy. On top of this, a disproportionately small number of sectors, and often a disproportionately small number of firms within a sector, tend to enjoy the gains from protection. For example, more than one-half of the antidumping and countervailing duty orders in place as of August were on iron and steel-related products alone; by contrast, less than one-half of 1 percent of total private nonfarm employment is accounted for by iron and steel producers.24 As another example, according to a 1993 General Accounting Office study, 42 percent of the benefits to growers from sugar protection went to just 1 percent of growers.25 Although Americans favor policies designed to help the small farmer, much larger enterprises are also benefiting from agricultural trade protection.
This disturbingly inequitable distribution of the benefits of protectionism is exacerbated under current law by provisions allowing antidumping and countervailing duties to be disbursed to the companies that petitioned for the duties. These provisions, which have been ruled illegal by the WTO, lead to protected producers being rewarded twice: Once through the higher prices stemming from the trade protection and again through the disbursal of the higher duties paid by importers. The distribution of these payouts has been extremely skewed: For fiscal year 2003, a single firm received more than one-fourth of the $190 million in countervailing and antidumping duties that were distributed to U.S. firms.26
Import quotas (as opposed to tariffs) raise a third concern about trade protection. By restricting the supply of certain types of imports within the United States, quotas may benefit those foreign producers who retain the right to sell to U.S. markets by raising the prices of their goods. For example, one study found that, of the $8.6 billion in net welfare costs induced by the Multi-Fiber Agreement, which restricts textile and apparel imports, about $6 billion accrued to those foreign producers who were allotted shares of the import quotas.27 Surely, many Americans would cease to support certain types of import protections if they knew that such actions were serving to prop up the profits of foreign producers.
Finally, we must not forget that trade actions, while sometimes protecting some American workers in import-competing industries, often invite the threat of foreign retaliation that would hurt American workers in export industries. For example, after the imposition of steel safeguard duties in March 2002, eight of our trading partners initiated safeguard investigations of their own on steel imports. Given the importance of export markets to the most dynamic areas of U.S. manufacturing, we cannot afford to jeopardize them by inviting foreign barriers to our products.
Thus, it is crucial to maintain public pressure for free trade. First, it is important to continue to educate the public and create a political environment supportive of free trade. In this respect, targeted criticisms of protectionist actions may be more effective than general paeans to free trade. In a recent speech, my colleague, William Poole, urged journalists describing trade restrictions to ask who gains, who loses, and what is the net gain or loss for the economy as a whole?28 I very much support that sentiment. Second, it is crucial to implement policies that foster stability and economic growth. Reducing unemployment and diminishing economic insecurity will likely be more effective against protectionism than a thousand speeches like this one. Toward that end, the Federal Reserve will do its part by working to promote stable financial conditions and sustainable, noninflationary growth.
1. Ninety-three percent of economists surveyed agreed, to a least a limited degree, with the statement that "tariffs and import quotas usually reduce general economic welfare." Richard M. Alston, J.R. Kearl, and Michael B. Vaughan (1992), "Is There a Consensus Among Economists in the 1990s?" American Economic Review, vol. 82 (May, Papers and Proceedings, 1991), pp. 203‑09. Return to text
2. Steven Kull and others (2004), "Americans on Globalization, Trade, and Farm Subsidies," The American Public on International Issues, the PIPA/Knowledge Networks Poll (Program on International Policy Attitudes, University of Maryland, January 22), www.pipa.org. Return to text
3. Christian Broda and David E. Weinstein (2004), "Globalization and the Gains from Variety," NBER Working Paper Series 10314 (Cambridge, Mass.: National Bureau of Economic Research, February). Return to text
4. U.S. International Trade Commission (2004), The Economic Effects of Significant U.S. Import Restraints: Fourth Update, Investigation 332‑325, Publication 3701 (Washington: ITC, June). Return to text
5. U.S. International Trade Commission (2003), The Impact of Trade Agreements: Effect of the Tokyo Round, U.S.-Israel FTA, U.S.-Canada FTA, NAFTA, and the Uruguay Round on the U.S. Economy, Investigation TA-2111-1, Publication 3621 (Washington: ITC, August). Return to text
6. Drusilla K. Brown, Alan V. Deardorff, and Robert M. Stern (2002), "Computational Analysis of Multilateral Trade Liberalization in the Uruguay Round and Doha Development Round," Discussion Paper 489, Research Seminar in International Economics, Gerald R. Ford School of Public Policy (Ann Arbor: University of Michigan). Return to text
7. Andrew B. Bernard, J. Bradford Jensen, and Peter K. Schott (2002), "Survival of the Best Fit: Competition from Low Wage Countries and the (Uneven) Growth of U.S. Manufacturing Plants," NBER Working Paper Series 9170 (Cambridge, Mass.: National Bureau of Economic Research, September). Return to text
8. Petia Topalova (2004), "Trade Liberalization and Firm Productivity: The Case of India," IMF Working Paper WP/04/28 (Washington: International Monetary Fund, February); Nina Pavcnik (2002), "Trade Liberalization, Exit, and Productivity Improvements: Evidence from Chilean Plants," Review of Economic Studies, vol. 69 (January), pp. 245-76. Return to text
9. Andrew B. Bernard and J. Bradford Jensen (1995), "Exporters, Jobs, and Wages in U.S. Manufacturing, 1976-1987," Brookings Papers on Economic Activity: Microeconomics, 1995, pp. 69-119. Return to text
10. David Dollar and Aart Kraay (2001), "Trade, Growth, and Poverty," unpublished paper, World Bank, June; Sebastian Edwards (1998), "Openness, Productivity and Growth: What Do We Really Know?" Economic Journal, vol. 108 (March), pp. 383-98. Return to text
11. Alan M. Taylor (1994), "Three Phases of Argentine Economic Growth," NBER Historical Paper 60 (Cambridge, Mass.: National Bureau of Economic Research, September); Jagdish Bhagwati (1993), India in Transition: Freeing the Economy (New York: Oxford University Press); Douglas A. Irwin (2002), Free Trade Under Fire (Princeton: Princeton University Press); Mario J. Crucini and James Kahn (1996), "Tariffs and Aggregate Economic Activity: Lessons from the Great Depression," Journal of Monetary Economics, vol. 38 (December), pp. 427-67. Return to text
12. Caroline Freund (2000), "Current Account Adjustment in Industrialized Countries," International Finance Discussion Paper 2000-692 (Washington: Board of Governors of the Federal Reserve System, December). Return to text
14. My colleague, Ben Bernanke, has discussed many of the issues linking trade and jobs in a speech earlier this year ("Trade and Jobs," at the Fuqua School of Business, Duke University, March 30, 2004). Return to text
15. Henry Farber (2003), "Job Loss in the United States, 1981-2001," Industrial Relations Section Working Paper 471, (Princeton: Princeton University, January). This study covers workers displaced for any reason, including import competition. Return to text
19. Martin Neil Baily and Robert Z, Lawrence (2004), "What Happened to the Great U.S. Job Machine? The Role of Trade and Offshoring," paper prepared for the Brookings Panel on Economic Activity, September 9-10. Their rough estimate of nearly 45,000 software jobs relocated annually to India is consistent with an estimate by Charles L. Shultze that "the number of workers employed in producing computer and related services relocated from the United States to India could have increased by roughly 185,000 over the past four years" (Charles L. Schultze, "Offshoring, Import Competition, and the Jobless Recovery," Policy Brief 136, Washington: The Brookings Institution, August). Return to text
20. See World Trade Organization, "Anti-dumping, Subsidies, Safeguards: Contingencies, etc.,"; U.S. International Trade Commission (1998), Summary of Statutory Provisions Related to Import Relief, Publication 3125 (Washington: ITC, August). Return to text
21. U.S. International Trade Commission (2004), "Antidumping and Countervailing Duty Orders in Place as of August 9, 2004, by Product Group (111 KB PDF)," Five-Year (Sunset) Reviews, General Information, item 7: AD and CVD orders in place. Return to text
22. Bruce A. Blonigen (2003), "Evolving Discretionary Practices of U.S. Antidumping Activity," NBER Working Paper Series 9625 (Cambridge, Mass.: National Bureau of Economic Research, April). Return to text
23. Joseph Francois and Laura M. Baughman (2003), "The Unintended Consequences of U.S. Steel Import Tariffs: A Quantification of the Impact During 2002," CITAC Job Studies (Washington: Consuming Industries Trade Action Coalition). Return to text
28. William Poole (2004), Free Trade: Why Are Economists and Noneconomists So Far Apart? Speech prepared for the Trade, Globalization, and Outsourcing Conference, Reuters America, Inc., New York, June 15. Return to text