Chapter Twelve

Trade Shocks and Response, 1979–1992

The 1980s were one of the most difficult periods in the history of US trade policy. The combination of two powerful macroeconomic forces—a severe recession from 1979 to 1982 and the significant appreciation of the dollar against other currencies from 1980 to 1985—squeezed domestic producers of traded goods, particularly in manufacturing. The United States also began running large trade deficits, which became a symbol of the country’s troubles with trade. The intensification of foreign competition meant that the political pressures for import restrictions increased dramatically.1 The Reagan administration responded by limiting imports in many sectors, but also resisted congressional pressure to do more, particularly with respect to Japan. The economic recovery starting in 1983 and the fall in the value of the dollar starting in 1985 eventually helped relieve the pressure on producers of traded goods and enabled the import restraints to be removed by the early 1990s. This period also saw the continued reversal of the historic partisan divisions over trade policy, as many Democratic constituencies were now hurt by imports, while Republicans constituencies stood to benefit from open trade.

Double Trouble: Deep Recession and Strong Dollar

The macroeconomic forces driving trade policy originated with a shift in monetary policy designed to stop inflation. In 1979, with consumer prices rising at about 12 percent a year, Federal Reserve Board chairman Paul Volcker started tightening monetary policy. This policy succeeded in reducing inflation, but also drove up real interest rates and produced the most severe recession since the Great Depression. The manufacturing sector was particularly hard hit: employment dropped 12 percent from 1979 to 1983, with massive layoffs in large, trade-sensitive industries such as automobiles and steel. The full force of the Federal Reserve’s policy was felt in 1982, when industrial production fell more than 7 percent, and the unemployment rate peaked at almost 11 percent by year’s end.2

The new administration of President Ronald Reagan also pursued an expansionary fiscal policy, cutting tax rates and ramping up defense spending. The combination of a tight monetary policy and a loose fiscal policy led to a growing fiscal deficit, high real interest rates, and a steady appreciation of the dollar on foreign-exchange markets. Between 1980 and 1985, the dollar rose about 40 percent against other currencies on a real, trade-weighted basis. The dollar’s appreciation dealt a crushing blow to the competitive position of domestic producers of traded goods. The strong dollar undermined exports by making American goods more expensive to foreign consumers and gave imports a significant edge in the domestic market by making foreign goods less expensive to consumers. Consequently, the merchandise trade deficit grew to reach nearly 3.5 percent of GDP in 1987. Only after the dollar began to depreciate in 1985 did the trade deficit eventually begin to subside.

Why did such large trade deficits, which were completely outside the range of previous historical experience, suddenly appear at this time? A fundamental change in the international financial system, discussed in chapter 11, now made large, sustained trade imbalances possible. In previous decades, trade imbalances had been small because the Bretton Woods system of fixed exchange rates involved government restrictions on the international movement of capital. When countries could only buy and sell goods with each other, exports and imports had to be roughly balanced. When the fixed exchange-rate system finally collapsed in 1973, and countries adopted floating exchange rates, these capital controls were no longer necessary. As governments began to permit greater international capital movements, investors in different countries were able to buy one another’s assets as well. Consequently, financial flows between countries increased enormously.3 The increase in capital movements between countries allowed large trade imbalances to emerge. In the US case, other countries wanted to use the dollars they earned exporting to the United States to buy US assets rather than American-made goods. As a result, the dollar appreciated in value and exports began to fall short of imports as foreign investment in the United States surged.

Changes in Japan’s policy were particularly important. Japan had long been a country with a high savings rate and low interest rates. In December 1980, Japan liberalized capital outflows and allowed Japanese investors to purchase assets in the United States, a country with a low savings rate and relatively high interest rates. As a result, Japanese financial institutions began selling yen to buy dollars, so that they could purchase higher-yield, dollar-denominated assets.4 This drove up the value of the dollar in terms of yen on foreign-exchange markets. The appreciation of the dollar (or, conversely, the depreciation of the yen) made Japanese goods more price-competitive and American goods less price-competitive in world markets.

Both the severe recession and the strong dollar put export-dependent and import-competing sectors of the economy under enormous pressure. In the early 1980s, as figure 12.1 shows, exports fell sharply as a share of GDP, while imports were roughly unchanged. Yet this figure is misleading in suggesting that imports were not of growing importance in the domestic market. The value of imports relative to GDP did not increase much in part because the price of imports was lower due to the strong dollar, even as the volume of imports rose significantly. Over the period 1982–85, the volume of imports of semi-finished and finished manufactured goods grew 50 percent and 72 percent, respectively. Meanwhile, the volume of exports of semi-finished and finished manufactures grew only 9 percent and 1 percent, respectively.5

Figure 12.1. Merchandise exports and imports as a percentage of GDP, 1950–1990. (US Department of Commerce, Bureau of Economic Analysis, www.bea.gov/.)

The strength of the dollar against other currencies also contributed to a significant change in the structure of the trade balance. During the 1970s, the overall trade deficit was driven by net imports of mineral fuels (petroleum) that slightly exceeded net exports of manufactured and agricultural goods. During the 1980s, as figure 12.2 shows, the trade surplus in agricultural goods continued, and the deficit in mineral fuels stabilized, but the trade balance in manufactured goods fell sharply into deficit. Starting in 1983, the United States became a large net importer of manufactured goods.

Figure 12.2. US balance of trade, by category, 1967–1990. (US Department of Commerce, Highlights of U.S. Trade, various issues.)

These developments led to an ongoing debate about the health of the manufacturing sector. Much of the concern focused on jobs. After rising by nearly 4 million during the 1960s, manufacturing employment oscillated between 18.5 and 21.0 million workers during the 1970s and 1980s. Large declines in manufacturing employment were seen in 1968–70, 1973–74, and 1979–82. In the first two periods, the declines were almost entirely cyclical, coinciding with recessions and largely unrelated to trade. But in 1981–82, when manufacturing employment fell 12 percent, a loss of nearly 3 million jobs, about a third of the employment decline was due to trade—the fall in manufactured exports and rise in imports—and the other two-thirds were due to the recession.6

Over the longer period from 1979–94, however, trade actually contributed to higher employment in manufacturing. Although manufacturing employment fell 13 percent during this period, Kletzer (2002) calculates that if exports and imports had been frozen at their 1979 level, manufacturing employment would have declined 16 percent. The reason is that both exports and imports of manufactured goods grew during this period, but exports are more tightly linked to job creation than imports are linked to job destruction. In particular, not all imports of manufactured goods are direct substitutes for domestic production: imports may be so different from domestically produced goods that they do not really compete with one another.7

Aside from these cyclical fluctuations, the economy was also undergoing long-term structural changes that resulted in significant employment shifts between industries. While some manufacturing industries were expanding employment, others were experiencing large, permanent declines in employment. Between 1977 and 1987, the number of production workers in the primary metals industry (blast furnaces and basic steel products) fell by 390,000, and employment in textiles and apparel fell by nearly 600,000. For these sectors, production cutbacks and plant closures led to mass layoffs of blue-collar workers. The term “deindustrialization” came into use, and images of shuttered factories across the Rust Belt, as the industrial Midwest came to be known, became etched in popular memory. On the other hand, employment in the transportation and electronics industries rose 350,000 over this period, and increased by 430,000 in printing and publishing.8

Despite the difficulties for workers, overall manufacturing output continued to grow through most of this period. Even during the severe recession of 1979–82 when manufacturing employment fell 12 percent, manufacturing production fell just 4 percent. Conversely, in the 1983–89 expansion, manufacturing production grew 36 percent, but employment rose only 4 percent. Production and employment were no longer coupled with one another: productivity improvements enabled output to grow without new workers being hired. This was due to changes in the composition of manufacturing output (the expansion of technology and capital-intensive industries, and the relative decline of labor-intensive industries), as well as the general improvement in labor productivity due to new technology and equipment.

Manufacturing also declined as a share of the economy during this period: between 1970 and 1990, manufacturing’s share of GDP fell from 24 percent to 18 percent. In view of the growing trade deficit, foreign competition was often blamed for the 6 percentage-point decline. But manufacturing’s share still would have fallen five percentage points over that period even if trade in manufactured goods had been balanced, as Krugman and Lawrence (1994) note. In other words, the overwhelming proportion of the declining share of manufacturing in the economy was due to non-trade factors, such as the shift in consumer demand from goods to services and the decline in the relative price of manufactured goods owing to rapid productivity growth. (Furthermore, manufacturing’s share of economic output was stable in real terms, suggesting that much of its declining share of nominal GDP was due to the falling relative price of manufactured goods due to productivity growth.)

However, the experiences of the 1970s and 1980s were different. In the 1970s, the United States had a growing trade surplus in manufactured goods. Trade expanded manufacturing’s share of the economy because exports of skill-intensive goods (aircraft and machinery) more than offset imports of labor-intensive goods (apparel and footwear). The story was different in the 1980s. During that decade, manufacturing’s contribution to GDP fell 3.1 percentage points, almost the same magnitude as in the 1970s, but manufacturing’s share would have fallen just 1.7 percentage points if trade had been balanced. Thus, more than half of the decline in manufacturing’s share of GDP in the 1980s can be attributed to the trade deficit.

The real issue confronting the manufacturing sector was an intensification of competition, driven as much by developments in technology as by foreign competition, which forced restructuring in almost every industry. Domestic firms responded to greater competition by trying to become more efficient, closing inefficient production facilities, and finding ways of maintaining production with fewer workers in order to reduce costs. Competition forced all domestic firms to reduce production costs, upgrade the quality of their products, or move into new lines of business in order to survive. Firms struggled to reduce costs and increase efficiency by adopting new technology, trimming the workforce, and reorganizing production. Of course, different industries adjusted in different ways. The automobile industry was driven to improve product quality and produce smaller, more fuel-efficient cars. The steel industry began to rationalize production by shutting down excess capacity. Labor-intensive industries modernized by substituting capital (machinery) for labor. In industries where capital or technology could not be substituted for labor, such as the assembly of consumer electronics or the manufacture of shoes, domestic production was likely to be sent abroad to take advantage of cheaper labor. Consequently, the labor-intensive assembly stage of production in many industries moved to other countries.

This restructuring across industries occurred regardless of its exposure to international competition: trade was only slightly related to cross-industry variation in worker displacement rates. Although industries with high displacement rates were often import-sensitive, not all import-sensitive industries had high displacement rates.9 Restructuring was usually achieved by reducing the number of workers employed rather than by cutting the wages of existing workers.10 Regardless of whether they lost their jobs because of changes in imports, improvements in technology, shifts in consumer demand, the displacement of workers from their jobs was a hard blow for those affected. The earnings of displaced workers often fell significantly when they lost their jobs. In particular, older, unionized workers received a substantial wage premium above the average worker in manufacturing and earned substantially less if employed elsewhere in the economy.11 On the other hand, workers in the labor-intensive sectors that were most vulnerable to competition from imports—such as footwear, leather products, and textiles and apparel—tended to be women and minorities with few skills. If displaced from their jobs, these workers often found employment at comparable wages elsewhere in the economy, because they were already among the lowest paid workers in the labor force.12

While unemployment rose sharply in the 1979–82 recession, the unemployment rate fell back down to 5 percent by the end of the decade. While trade did not affect total employment, it did affect the composition of employment across different sectors of the economy. Imports destroyed jobs in low-wage manufacturing industries (apparel, footwear, leather) and in some high-wage unionized sectors (autos and steel), while exports created jobs in high-wage industries (aerospace, machinery, pharmaceuticals). Unfortunately, the strong dollar prevented exports from keeping pace with imports in the early and mid-1980s, and both exporters and import-competing industries were squeezed. This pressure shifted employment out of the production of tradable goods and into the production of non-tradables, such as services.

Even in the absence of this pressure, the United States was increasingly becoming a service economy. As incomes rose, American consumers demanded more services, ranging from health care, education, and finance to recreation and leisure. Because labor-productivity growth in services was slower than in other sectors of the economy, the share of the labor force devoted to the production of services also had to increase to accommodate this demand. Just as workers in previous generations had transitioned from agriculture to manufacturing, the growing share of the labor force employed in services—which rose from 67 percent in 1970 to 77 percent in 1990—was part of a long-run trend. While the total number of workers in manufacturing was about the same in 1970 and 1990, their share in total employment fell from 27 percent to 17 percent. This development had little to do with trade: productivity improvements in manufacturing, due to the substitution of capital for labor in production and the advance of new technology, were far more important in explaining the declining share of employment in manufacturing than increased imports. Even if trade had been balanced, manufacturing’s employment share would have been only one percentage point higher than it was—18 percent instead of 17 percent—given the rapid growth in labor productivity in manufacturing.13

The confluence of these many different factors in the early 1980s led to concerns about the “competitiveness” of US manufacturing and fears about the “deindustrialization” of America. To be sure, some industries had fallen behind their foreign competitors in productive efficiency and product quality, and competition was forcing domestic firms to improve both or go out of business. However, the main problem facing manufacturers was not some deep-rooted structural issue, but an exchange rate that posed an enormous obstacle to its ability to compete in domestic and foreign markets. The 40 percent real appreciation of the dollar against other currencies over 1979–85 made it extremely difficult for both export-oriented and import-competing producers to remain price-competitive against foreign producers. The dollar’s appreciation reduced manufacturing employment in trade-impacted industries about 4–8 percent, on average.14 That domestic producers did not suffer from a structural “competitiveness” problem was demonstrated by the resurgence in manufactured exports and the pickup in factory employment once the dollar started depreciating in 1985.

In sum, increased imports were just one of many challenges facing the manufacturing sector in the early 1980s. Unlike a sharp decline in domestic demand, increases in productivity growth, intensified competition and technological change, and shifts in consumer demand, all of which significantly affected employment in manufacturing but were beyond the immediate reach of policymakers, restricting imports was an action that policymakers could take in order to help import-competing industries. Consequently, there was a sharp increase in protectionist pressures.

Emergence of Protectionism

The nation’s struggling economy was a key issue in the 1980 presidential election. While trade was not yet a major concern, the Republican nominee, Ronald Reagan, held out the prospect of import relief to drum up political support, particularly in the South. Reagan pledged to protect the textiles and apparel industry from further market disruption.15 Regarding automobiles, Reagan initially disavowed import quotas, saying that the industry’s problems stemmed from excessive regulation rather than Japanese imports, but campaign advisers floated the idea that Japan might “voluntarily” restrain its exports. While Reagan did not make an explicit pledge to reduce steel imports, he promised tax and regulatory relief for the industry and criticized the Carter administration’s decision to suspend the trigger-price mechanism, saying that trade had to be “fair.”

Reagan won the 1980 election on a platform of reducing government’s role in the economy. In their public pronouncements, the president and his administration appeared strongly committed to free trade.16 The administration’s July 1981 Statement on Trade Policy declared that free trade, a term that previous administrations had never explicitly endorsed, was critical to ensuring a strong economy. It vowed to “strongly resist protectionism,” yet warned that “the United States is increasingly challenged not only by the ability of other countries to produce highly competitive products, but also by the growing intervention in economic affairs on the part of governments in many such countries. We should be prepared to accept the competitive challenge, and strongly oppose trade-distorting interventions by government.”17

In fact, the Reagan administration was sharply divided over trade policy. Officials in some agencies (the Treasury and State Departments, the Office of Management and Budget, the Council of Economic Advisers) wanted to uphold free-market principles and reduce government intervention in the economy. Elsewhere, officials in the Commerce and Labor Departments representing the business community and labor wanted the government to help firms and workers struggling with foreign competition. Reagan himself was often conflicted between his strong belief in free enterprise and limited government and his desire to help out American industries and their workers.18

As a result, despite its free-trade rhetoric but in light of the tremendous shocks affecting traded-goods industries, the Reagan administration often accommodated domestic industries seeking relief from foreign competition.19 At critical junctures, the administration either made a political calculation about the electoral benefits of protecting large industries from imports, or restricted imports to forestall congressional legislation. Consequently, the share of imports covered by some form of trade restriction, after rising from 8 percent in 1975 to 12 percent in 1980, jumped to 21 percent in 1984.20 “For the first time since World War II, the United States added more trade restraints than it removed,” noted William Niskanen (1988, 137), a former economic adviser in the administration. He described policy in this period as “a strategic retreat,” in that the outcome, while not desirable in itself, was better than the most likely alternative, which was believed to be import quotas imposed by Congress. The administration’s strategy, he said, was “to build a five-foot trade wall in order to deter a ten-foot wall [that would have been] established by Congress.” This pattern can be seen by looking at trade policy with respect to automobiles, steel, textiles and apparel, and other goods.21

Automobiles

The automobile industry was the last major manufacturing industry to be affected by the intensification of foreign competition that began for most industries in the late 1960s. It was also the first to receive protection from the Reagan administration. The automobile industry was structurally similar to the steel industry: a few firms dominated the market (the Big Three: General Motors, Ford, and Chrysler), production was regionally concentrated (in the industrial Midwest), and a powerful union represented labor (the United Auto Workers). As in other industries, imports were not a major concern in the decades after World War II. In the 1960s, the foreign share of the domestic market was stable and less than 7 percent, mostly imports from Germany. As late as 1968, Japan’s market share was only about 1 percent. The Big Three ceded the low-margin, small-car segment of the market to foreign producers and concentrated their product line on the more profitable mid-size and large-car segment of the market.

This strategy was upended when the oil price shock of 1973 shifted consumer demand to smaller, less expensive, more fuel-efficient cars. Caught without a deep product line in this category of vehicles, the Big Three saw the foreign share of the domestic auto market nearly double between 1975 and 1980, as figure 12.3 shows, particularly from Japan.

Figure 12.3. Foreign automobiles as a share of US car registrations, 1960–1990. (Ward’s Automotive Yearbook [Detroit: Ward’s Reports], various issues.)

As Japan’s share of the market grew, the views of labor and management began to change. Unlike other unions, the United Auto Workers (UAW) had opposed the Burke-Hartke bill of 1971, but soon it was demanding that import quotas be imposed and that Japanese firms begin building cars in the United States. At this point, GM, Ford, and Chrysler did not want to restrict imports because they themselves had begun importing foreign-produced cars under their own nameplate. In 1975, the UAW charged twenty-eight foreign auto manufacturers in eight countries with dumping, but domestic producers did not support the petition because 40 percent of imported cars came from their subsidiaries, especially in Canada. The Treasury Department dropped the antidumping investigation after receiving assurances of corrective action from foreign manufacturers.22

A second oil price shock in 1979 combined with the severe recession of the early 1980s inflicted far more damage on domestic producers. The Big Three suffered enormous financial losses and cut back production, throwing about three hundred thousand auto workers out of work, while a greater number of workers in supplying industries lost their jobs as well. Chrysler was on the verge of bankruptcy until it received government-backed loan guarantees.

In the summer of 1980, Ford and the UAW filed a section 201 escape-clause petition for import relief. General Motors and Chrysler did not support the petition: GM imported small cars from Japan under its nameplate, while Chrysler did not want to alienate the Carter administration, which had given it financial assistance and was on record as opposing import restrictions. In November 1980, just days before the presidential election, the International Trade Commission (ITC) voted unanimously that the automobile industry had suffered serious injury as a result of imports, but in a 3–2 vote ruled that imports were not a “substantial cause” of serious injury and therefore the industry was not entitled to relief. This finding appeared to hinge on a technicality: under the Trade Act of 1974, a substantial cause is “a cause which is important, and not less important than any other cause,” meaning that imports had to be the most important cause of injury for relief to be granted. But the ITC determined that the growing demand for compact cars and the declining demand for large cars was a greater source of injury than foreign competition and, on this basis, the petition was rejected. Following that decision, both presidential candidates promised some form of aid for the automobile industry. The House also passed a resolution authorizing the president to negotiate limits on Japanese exports, and Senator John Danforth (R-MO) introduced a bill restricting the number of cars imported from Japan to 1.6 million per year.

In March 1981, shortly after President Reagan took office, Transportation Secretary Drew Lewis urged the president to “keep faith with our campaign pledge” and restrict auto imports from Japan. Budget director David Stockman (1986, 154) was appalled: “This preposterous idea was so philosophically inimical to what I thought we stood for that for a few moments I just sat back, concussed . . . here was a cabinet officer talking protectionism in the White House, not two months into the administration.” Reagan’s advisers were divided: Lewis, Commerce Secretary Malcolm Baldrige, and Trade Representative William Brock favored restricting automobile imports, while Stockman, Treasury Secretary Don Regan, and Council of Economic Advisers Chair Murray Weidenbaum were opposed. The split between the business advocates and the free-market proponents reflected a tension that was present throughout the Reagan administration.

The president was undecided but clearly sympathetic to the auto producers, noting that government regulation was part of the industry’s problem. He refused to threaten a veto of the Danforth bill, a clear signal to Japan that something was going to be done to limit imports if it did not restrain its exports. When presented with various options, Reagan favored the idea of asking Japan to “voluntarily” limit its exports of automobiles, and so this approach was taken.23 Stockman (1986, 158) remarked bitterly: “And so the essence of the Reagan administration’s trade policy became clear: Espouse free trade, but find an excuse on every occasion to embrace the opposite. As time passed, we would find occasions aplenty.” After being pressured by the administration, Japan soon announced that it would limit its auto exports to the United States to 1.68 million cars per year, a reduction of nearly 8 percent from the quantity exported in 1980. The voluntary export restraint would be in effect for three years, from April 1981 to March 1984, although the export limit could increase over this time.

The voluntary export restraint (VER) was not a new trade-policy instrument, particularly for Japan. More than any other country, Japan seemed to increase its exports in narrow product categories rapidly, causing problems for producers in importing countries. When cotton textile producers complained about excessive imports from Japan in the 1930s, Japanese producers decided to limit their own exports rather than face import restrictions imposed by the United States. In the 1950s, Japan adopted a number of VERs on products ranging from tuna to cotton textiles and stainless steel flatware.24 As noted in chapter 11, exporting countries generally favored VERs rather than having to face a tariff or quota imposed by the importing country. With a VER, exporters would profit from a quota-rent, the extra revenue that they received from charging a higher price in the protected market for their limited exports.

The auto VER was probably not binding on Japan’s exports in 1981 and 1982, when the severe recession depressed the demand for automobiles. As a result, the export restraint initially failed to provide much help for domestic producers. The UAW continued to demand that Japanese firms build production facilities in the United States to create more jobs at home. Congress also began considering domestic content legislation that would require all cars sold to contain a certain proportion (up to 90 percent) of US-made parts and labor or else face import quotas. In December 1982 and again in December 1983, the House passed a domestic-content bill with Democratic votes and Republican opposition, although in each case the Senate failed to take it up. While it was widely recognized that the president would veto the domestic-content bill, the House votes sent a signal to Japan about the domestic political problems caused by its exports.

The 1984 election played a role in Reagan’s decision to ask Japan to renew the VER, because he did not wish to alienate large numbers of voters in the industrial Midwest by lifting the restriction. As the economy recovered and automobiles sales rebounded, the VER became a binding constraint on Japanese auto sales in 1984 and 1985, though set at the higher level of 1.85 million vehicles. The economic effects of the VER are considered later in the chapter, but the export restraint and the opening of Japanese production facilities in the United States stabilized the import share of the market by the end of the decade.

Steel

The steel industry was also hit hard by the recession in the early 1980s. The large, integrated producers, including US Steel and Bethlehem, suffered enormous financial losses that forced them to reduce output and lay off hundreds of thousands of workers. Although steel imports declined with the collapse in demand, domestic production fell more rapidly. As figure 11.3 showed, the share of imports in domestic consumption rose from 15 percent in 1979 to nearly 22 percent in 1982.25

In January 1982, major steel firms filed 155 antidumping and countervailing duty petitions against forty-one different suppliers of nine different products from eleven countries, but aimed primarily at the EEC.26 The ITC ruled affirmatively in about half of these cases, but the prospect of long-lasting and severe tariff penalties on European producers, as well as highly varied antidumping duties being imposed across a range of countries and producers, was unattractive to all parties. To persuade the firms to withdraw their petitions, the Reagan administration brokered a new voluntary restraint agreement (VRA) that limited EEC exports to 5.5 percent of the US market in eleven product categories. European producers preferred the quantitative restrictions, because those would allow them to avoid steep antidumping or import tariffs and enable them to charge higher prices. The domestic steel industry also preferred this outcome, because it fixed the volume of imports (unlike the trigger-price mechanism or antidumping and countervailing duties, which affected the price) and applied to all EEC countries. Japan continued to restrict its steel exports by agreement, limiting them to 5–6.5 percent of the US market, depending upon the product.

The restraint agreements failed to provide as much help as the domestic steel industry had hoped, because imports grew from countries whose exports were not constrained by the VRA. The share of the market held by producers outside of Japan, the EEC, and Canada rose from 5 percent to 10 percent between 1982 and 1984. Thus, the overall foreign market share continued to climb, reaching 26 percent in 1984. Steel producers sought to plug the holes in this leaky system by filing more than two hundred antidumping petitions against imports from countries not party to the VRAs, such as South Korea, Spain, Brazil, Mexico, Poland, and South Africa.

In a further effort to block imports, Bethlehem Steel and the United Steel Workers filed a section 201 escape-clause petition in 1984. The ITC concluded that imports were a substantial cause of serious injury in five product categories, but found no injury in four others. The petition had been timed so that the president would have to make a decision about the ITC recommendations just eight weeks before the 1984 presidential election. The president was put in a difficult position: with about half of steel capacity located in Pennsylvania, Ohio, Indiana, and West Virginia, many potential votes were at stake. The Congressional Steel Caucus pressed for a mandatory five-year quota limiting imports to 15 percent of domestic consumption. Reagan’s Democratic opponent in the election, former Vice President Walter Mondale, proposed capping the foreign market share at 17 percent.

Reagan rejected the ITC’s proposed tariff on the grounds that it was “not in the national economic interest to take actions which put at risk thousands of jobs in steel fabricating and other consuming industries or in the other sectors of the US economy that might be affected by compensation or retaliation measures to which our trading partners would be entitled” in an escape-clause action. Instead, he directed USTR to negotiate “surge control” arrangements or understandings with countries “whose exports to the United States have increased significantly in recent years due to an unfair surge in imports—unfair because of dumping subsidization, or diversion from other importing countries who have restricted access to their markets”—with the goal of “a more normal level of steel imports, or approximately 18.5 percent, excluding semi-finished steel.”27 Thus, the president went well beyond the section 201 case in promising to secure export-restraint agreements covering all segments of industry considered in the petition, even those the ITC turned down, against all major foreign suppliers of steel.

Of course, the restraint agreements still had to be negotiated. Steel imports surged in late 1984 and early 1985 before the market-share quotas could be finalized, prompting additional dumping and subsidy complaints against various European and Latin American countries. By August 1985, the product- and country-specific quotas of the VRAs were in place and covered fifteen countries accounting for 80 percent of steel imports. The VRAs were scheduled to expire in December 1989.

Textiles and Apparel

As Reagan had promised in the 1980 election campaign, the Multifiber Arrangement (MFA) was renewed for a third time in 1981. In effect from 1982–86, MFA-III reduced the annual growth of textile and apparel exports from developing countries from 6 percent to 2 percent and included tighter country-of-origin requirements and anti-surge and market-disruption provisions to limit export growth in sensitive categories. Over the course of the 1980s, the restraints were expanded to include other countries. By 1985, the MFA covered imports of textile and apparel goods from thirty-one countries in 650 separate product categories.

Like its predecessors, however, MFA-III failed to stem the rapid growth in apparel imports. This growth occurred because the quota allocations had grown over time, and some product categories were vastly underutilized. There was ample room for foreign exporters to expand their shipments by shifting products between categories and years: apparel from countries with filled quotas could ship their products to countries with unfilled quotas for some minor processing and then be exported to the United States. (For example, if a country’s exports of shirts hit the limit, it could export sleeveless shirts to another country with an unfilled export quota for final stitching.) Foreign producers could also alter their production mix, upgrading their products to take advantage of different limits in different categories. As a result, the protection provided by the MFA was a “screen rather than a solid wall,” as Cline (1990, 169) put it. The porous nature of the MFA allowed imports to surge in 1983 and 1984, fueled by the economic recovery and the strong dollar, and import penetration increased sharply, as figure 12.4 shows.

Figure 12.4. Imports as a share of domestic consumption, textiles and apparel, 1960–1988. Note: Textile mill products (SIC 22), apparel and other mill products (SIC 23), and non-rubber footwear (SIC 314). (US Department of Commerce, Industrial Outlook, various issues.)

Figure 12.4 also shows that import penetration was quite different for the textile and apparel industries. While the two were often lumped together, they were actually quite distinct. The textile industry produced fabrics and found it relatively easy to substitute capital machinery for labor and thereby improve productivity. Textile mills manufactured yarn, thread, carpets, and upholstery in highly automated mills located mainly in Georgia and the Carolinas. By adopting advanced technology, the industry shed workers but remained competitive: the import share was neither high nor rising, and some segments of the industry were able to export.

By contrast, the apparel industry produced clothing and garments. This involved cutting textile fabric and sewing and assembly operations—including pressing, dyeing, washing, and packaging—to convert it into clothing and other finished goods. This was a labor-intensive process that employed mainly unskilled women and minorities in plants spread out across Pennsylvania, the South, and southern California. Because production was necessarily labor-intensive, domestic firms found it difficult to innovate, keep costs low, and remain competitive against foreign producers who had access to low-wage labor. As a result, the share of the market taken by imports was rising rapidly and a much larger share of the job losses in the industry was due to foreign competition. Meanwhile, import penetration in the non-rubber footwear market, which was not protected after the failure of the OMAs in the late 1970s, surged as domestic shoe production plummeted.28

The MFA’s failure to stop apparel imports explains why apparel producers and allied textile firms, and particularly labor unions, made enormous political efforts to secure legislation that would tighten the restrictions on imports and slow the decline in employment. The textile and apparel industries remained one of the country’s largest employers in manufacturing, with nearly two million workers in the mid-1980s. Industry representatives argued that tighter import restrictions were needed to save jobs and that domestic production was critical for national defense. It succeeded in getting Congress to approve import limits in 1985–86, 1987–88, and again in 1990, only to have each bill vetoed by the president.

The first battle was over the Textile and Apparel Trade Enforcement Act of 1985, introduced by Rep. Ed Jenkins (D-GA). In pleading the industry’s case, Jenkins stated, “I know of no other industry or group of workers that has suffered more hardships than has the textile industry, as a direct result of cheap foreign imports.”29 The legislation would have reduced textile and apparel imports from 10 billion yards to 7 billion yards from twelve countries, mainly in Asia. The bill had more than 260 House cosponsors, and support came not just from the South, but also Pennsylvania (the home to many small mills) and even New England (although mills were fast dying out there).

This campaign ran up against widespread opposition. Some apparel manufacturers, such as Levi Strauss, had become importers and wanted the freedom to source from abroad. Retailers, such as Gap, JC Penney, and Kmart, opposed import limits and stressed the consumer interest in inexpensive clothing. Agricultural producers, represented by the American Farm Bureau Federation and other groups, feared foreign retaliation against their exports if the bill passed. Administration officials also rejected new import restrictions as a protectionist move that would jeopardize US negotiating goals in the next GATT round. The industry also suffered from bad press. The media portrayed the industry as an uncompetitive one that employed low-wage, unskilled workers who could get jobs in other sectors of the economy. The implication was that the shrinkage of the industry was inevitable and that the United States could do without domestic apparel production if it wanted to compete in high-technology sectors in the twenty-first century. Many members of Congress saw it as a low-wage, sunset industry of the past, not a sunrise, high-technology industry of the future.

Still, the House passed the Jenkins bill in November 1985 by a vote of 262–159; Democrats voted 187–62 in favor, while Republicans split 79–75 in opposing it.30 A month later, the Senate passed a less stringent version by a vote of 60–39, which the House accepted to avoid a protracted reconciliation process. To no one’s surprise, Reagan vetoed the bill. While he was “well aware of the difficulties” facing the industry and “deeply sympathetic about the job layoffs and plant closings that have affected many workers in these industries,” the president concluded, “It is my firm conviction that the economic and human costs of such a bill run far too high—costs in foreign retaliation against US exports, loss of American jobs, losses to American businesses, and damage to the world trading system upon which our prosperity depends.”31

Supporters of the bill delayed an override vote until the upcoming midterm election, hoping to force the administration to strengthen the expiring MFA. In July 1986, the Reagan administration announced a new, five-year MFA that included export limits on new fibers such as ramie, linen, and silk blends, new bilateral agreements with Hong Kong, South Korea, and Taiwan, the countries targeted in the Jenkins bill, and new safeguards to stop import surges. This was not enough to satisfy the labor unions, however, and the prospect of a close congressional override vote led the president to spend time phoning members of the House and asking for their support. In the end, the House failed to override the veto.32 In 1987, the industry and its workers again tried to get Congress to enact legislation that would cap overall textile and apparel imports at 1986 levels. The House passed the bill in late 1987 and the Senate followed a year later, but once again it was vetoed by President Reagan.

What explains the failure of the apparel industry and its workers to receive protection through legislation beyond that given in the MFA negotiated by the executive? Despite the efforts of the congressional Textile Caucus, the industry and its workers did not have as much political strength as might appear from the number of workers in the industry. Textile and apparel firms were divided about the merits of trade protection; textile firms were embracing new technology that enabled them to remain competitive, while apparel firms were increasingly sourcing their production from other countries. Advocates of import relief also encountered unexpectedly strong opposition from retailers, who put up a strong fight on behalf of consumers. Finally, the industry was already the beneficiary of the MFA, and its decline was not due to “unfair trade practices” by foreign governments. Import restraints were seen, at best, as a costly way of slowing the inevitable contraction of the industry.

Agriculture

Agricultural producers were also affected by the economic hardship caused by the recession and strong dollar. While farmers prospered during the commodity-price boom of the 1970s, they suffered when commodity prices collapsed in the early 1980s. Net farm income dropped by a third between 1979 and 1982, pushing farm indebtedness to record levels. Lower prices meant that it was more attractive for farmers to sell their crops to the government at the fixed price support than to sell them at the prevailing market price. As a result, the cost of federal farm programs escalated rapidly.

As government outlays to purchase and hold surplus crops grew, the Agriculture Department attempted to boost farm prices by reducing domestic production. This was done through acreage set-asides, in which farmers were paid to keep land idle. In addition, export subsidies were sometimes used to dispose of the government-held commodity stocks. These export subsidies put the United States on a collision course with the EEC, which had long been doing the same thing. One commodity, wheat, took on particular importance. Formerly a net importer, the EEC became a net exporter of wheat in the 1980s after it set domestic target prices so high that large production surpluses appeared. These surpluses were dumped on the world market using export subsidies. That policy, as well as the strong dollar, led to a sharp decline in foreign demand for American wheat. In 1983, the Reagan administration debated using selective export subsidies to reduce wheat stocks and punish the Europeans for distorting the world wheat market. The 1985 farm bill created the Export Enhancement Program to combat the EEC’s subsidies by providing for targeted export assistance to help American farmers increase sales in foreign markets.33 In one case, the United States displaced French exports by selling wheat to Egypt at $100 a ton when the US market price was $225 ton. Taxpayers were left to make up the difference.34

The United States was not alone in intervening in agricultural markets. The increasing use of domestic price supports, import restrictions, export subsidies, and other policies led to massive distortions in world agricultural markets. The Organization for Economic Cooperation and Development (OECD) documented the extensive support that governments gave to agricultural producers: in 1986, about 23 percent of US farm income, 39 percent of EEC farm income, and 65 percent of Japanese farm income came from policy measures.35 Interventions in one country led to spillover problems in other countries. For example, the Common Agricultural Policy enabled the EEC to increase its share of world food exports from 8 percent in 1976 to 18 percent in 1981.36 This depressed world prices, which led to more import restrictions and more costly price supports in other countries. The agricultural subsidy wars of the early 1980s made farm reform a major US negotiating priority in the next GATT round.

With most American farmers trying to export their produce to world markets, the demand to cut agricultural imports was not strong, with the exception of sugar.37 In the 1970s, raw sugar was protected only by a modest import tariff. The 1981 farm bill established a new domestic price support program for sugar at a time when world prices were relatively high, but mandated that it involve no federal outlays. In 1981–82, the falling price of sugar exposed the contradiction of having a government price support program that did not allow for any budgetary expenditures. The only way to keep the domestic price high and avoid government payments was to restrict imports in a bid to keep the domestic price at the government’s target price.

To do so, the Department of Agriculture imposed emergency quotas on imported sugar in May 1982. This slashed sugar imports from 5 million tons in 1981 to 3 million tons in 1982. To comply with GATT provisions about non-discrimination, the quotas were allocated to countries based on their share of imports in 1975–81, when a non-discriminatory tariff was in place. As sugar prices kept falling, officials reduced the import quotas six times over the next two years, resulting in a 75 percent reduction in the sugar imports allowed from Caribbean and Central American countries.

The government’s attempt to restrict sugar imports to balance supply and demand at the target price sometimes pushed the domestic price of sugar to more than five times the world price. This enormous price gap made it profitable to import refined sugar products that were not covered by the quotas—including packets of iced tea and cocoa, boxes of cake mix, tins of maple sugar, and other high-sugar content products—and then extract the sugar for sale at the high domestic price. In 1983, the Reagan administration banned imports of certain blends and mixtures of sugar and other ingredients in bulk containers on the grounds that they interfered with the price support program. Foreign exporters of processed foods, particularly in Canada and the EEC, vehemently protested the move.

The high domestic price of sugar also accelerated the substitution of high-fructose corn syrup for sugar in food manufacturing. In 1984, Coca Cola and Pepsi announced that they would use corn syrup instead of sugar as the sweetener in their soft drinks. This sharply reduced domestic demand for sugar and required the Agriculture Department to slash the import quota by another 20 percent for all exporting countries. The high domestic price of sugar meant that sugar-using industries faced higher production costs in comparison to their foreign competitors, driving many candy and confectionary producers to other countries and reducing domestic employment in the food-manufacturing industry.

Finally, the sharp reduction in sugar imports led to foreign-policy problems. The quotas reduced the export earnings and damaged the economies of Caribbean and Central American countries at a time when many of them were struggling with the global debt crisis and even fighting Communist-backed insurgencies. The US move hurt relations with those countries, and so the United States tried to help them with other trade preferences, such as the Caribbean Basin Initiative of 1983. The sugar quotas also led farmers in the region to stop producing sugar and start cultivating illegal narcotics that were smuggled into the United States, starting a war with drug traffickers.

The United States compounded all of these problems in August 1986 when it decided to subsidize the sale of the entire accumulated stock of 136,000 tons of sugar to China. The government took an enormous loss; sugar had been purchased at 18 cents per pound but was then sold at just under 5 cents per pound. Within two days of the sale, the world price of sugar dropped more than 20 percent, from 6.3 cents to 5.0 cents per pound, to the outrage of sugar-exporting countries.38

The Rise of Administered Protection

As we have seen, three large and politically powerful industries—automobiles, steel, and apparel—all benefited from executive-negotiated agreements with foreign countries to limit their exports. For their part, farmers relied on government price supports to insulate them from fluctuations in world commodity prices. But what about the many smaller, less politically influential industries that also felt the pain of the recession and increased foreign competition? What could they do to obtain government relief from imports? Since these producers could not command the attention of Congress or the executive branch, they fell back upon the system of trade laws that allowed domestic firms to petition the government for temporary duties on imports.39

As we saw in chapter 11, the main legal avenues by which domestic firms could request trade protection were the escape clause, antidumping duties, and countervailing duties. The escape clause, based on section 201 of the Trade Act of 1974, was supposed to be the principal means by which industries harmed by imports could receive temporary relief from foreign competition. If imports of a particular good were found to be “a substantial cause of serious injury,” the ITC would recommend that the president impose a higher tariff, phased out over five years, on imports from all sources. The president had complete discretion about whether to grant import relief or not.

However, we also saw that, in most escape-clause cases, either the ITC failed to find that imports were a substantial cause of serious injury, or the president rejected the provision of any relief. After the auto petition was turned down in 1980, the ITC dismissed petitions on fishing rods in 1981, tubeless tire valves in 1982, stainless steel table flatware and non-rubber footwear in 1983, canned tuna and potassium permanganate in 1984, and electric shavers, metal castings, and apple juice in 1985. President Reagan did grant escape-clause protection to heavyweight motorcycles, some steel products, and wood shakes and shingles, but turned down relief for unwrought copper and another non-rubber footwear case that the ITC had approved. Given this record, domestic firms knew that they were unlikely to obtain much assistance using the escape clause and therefore few bothered to initiate cases. As Senator Fritz Hollings (D-SC) put it, “going the 201 route is for suckers.”40

Consequently, domestic producers turned to antidumping and countervailing duties. Figure 12.5 presents the number of antidumping and countervailing duty petitions filed from 1960 to 1990 and shows that the demand for trade remedies increased significantly in the 1980s. From 1980–93, 682 AD cases and 358 CVD cases were filed. Many of them (38 percent of AD and 55 percent of CVD cases) were filed by the steel industry as a way of forcing the president to negotiate VRAs with foreign exporters (after which the petitions were withdrawn) or as a way of closing the market to countries or products not covered by the VRAs.41 Regarding countervailing duties, petitioning firms had a significant burden of proof in having to demonstrate that the foreign exports were subsidized and a cause of material injury. The difficulty in proving the existence of subsidies meant that relatively few CVD cases were filed, except by the steel and chemical industries. Just 21 percent of the CVD cases resulted in duties being imposed, because they were often terminated or suspended when replaced by VRAs.42

Figure 12.5. Number of trade-remedy cases initiated, 1960–1990. (Baldwin 1998, tables 11.1–4.)

By default, the antidumping law became the principal means by which small industries received some relief from foreign competition. The low burden of proof needed to find dumping and the high likelihood that duties would be imposed were the main attraction to potential filings. Congress also changed many of the provisions in the trade laws to encourage the filing of petitions and increase the probability of import duties as being the final outcome. Perhaps the most important procedural change came in 1979, when authority over dumping cases was shifted from the Treasury Department (which had little interest in the enforcing the statute) to the Commerce Department (which championed producer interests).

The antidumping process started with a firm or industry association filing a petition with the Commerce Department and the ITC alleging that imports from a particular country were being sold at “less than fair value” and causing “material injury.” Commerce made the “less than fair value” determination, and the ITC made the “material injury” determination. Under normal circumstances, foreign sales were considered “dumping” (sold at less than fair value) if a foreign exporter charged a lower price on its sales in the United States than in its home market.43 Commerce almost always ruled that dumping occurred: from 1980–92, dumping was found in 93 percent of all cases. Commerce often found large dumping margins: the average antidumping duty was 26 percent in the period 1980–84 and 41 percent from 1985–89. The average antidumping duty in effect in 1992 was 46 percent in non-steel cases and 27 percent in steel cases.44

Meanwhile, the ITC would determine if the petitioning industry had suffered from or was threatened with “material injury”—defined as “harm which is not inconsequential, immaterial, or unimportant”—as a result of the dumped imports. In making the injury determination, the ITC looked at such factors as changes in the industry’s output, employment, and capacity utilization. (Under the law, only the harm to domestic producers was considered, not the harm or injury to consumers or other domestic industries that might result from the imposition of additional duties.) The ITC made an affirmative injury finding in two-thirds of cases from 1980 to 1992, and about 40 percent of antidumping cases filed resulted in duties being imposed.45 In the 1980s, antidumping duties covered a wide array of products such as staples from Sweden, color television sets from Korea and Taiwan, raspberries from Canada, pistachios from Iran, candles from China, cut flowers from Colombia, and frozen orange juice from Brazil. However, because the goods subject to antidumping duties were narrowly defined products, the total share of US imports covered by such duties was less than 1 percent.

From the standpoint of domestic petitioners, the antidumping process had several advantages over the escape clause. First, the “material injury” standard in a dumping case was much less stringent than the “serious injury” requirement in an escape-clause case, making it more likely that the ITC would make an affirmative injury finding. The antidumping process also did not involve any presidential discretion: if Commerce and the ITC ruled in favor of the petitioner, antidumping duties went into effect automatically without further review. And unlike escape-clause relief, which was usually phased out over five years, antidumping duties could remain in place for an indefinite period. In the early 1990s, the mean duration of an antidumping duty was seven years, and about a fifth of such duties had been in place for ten or more years.46

However, unlike escape-clause duties, which applied to imports from all sources, antidumping duties only applied to imports coming from countries named in the petition. In other words, antidumping duties were selective and did not prevent other suppliers from expanding their exports when the targeted countries were hit. Such supply diversion made antidumping duties a leaky form of protection. For example, after antidumping duties were imposed on semiconductors from Japan, imports from Japan plummeted and production shifted to Taiwan. As Taiwanese semiconductor exports surged, they too were hit with antidumping duties. Then production shifted to Korea, where the same pattern repeated itself. Eventually, firms learned to file multiple petitions to cover imports from many different potential sources of supply.

As antidumping actions were increasingly used against imports, they came under criticism from economists. The administrative system strongly favored domestic petitioners and did not consider the interests of consumers, either downstream user industries or households, when such duties were imposed. Finger (1993, 13) argued that “antidumping is ordinary protection, albeit with a good public relations program” because it was based on the allegation of “unfair” foreign competition. The welfare cost of US antidumping and countervailing duty actions amounted to $4 billion in 1993—a considerable sum.47 Although US exports were adversely affected by the spread of antidumping to other countries, Congress favored the existing system so much that it refused to allow changes in the process to be negotiated at the multilateral level.

Assessing the Protectionism of the 1980s

Three questions can be posed about the new import restrictions imposed in the 1980s: What explains the type of policy instruments used? What were the economic effects of the import barriers? Did protection helped revitalize the protected industries?

The main policy instruments used to protect domestic firms from foreign competition were tariffs (antidumping, countervailing duties, or escape-clause actions) or quotas (export restraints). Generally speaking, the largest and most politically influential industries were protected from foreign competition through negotiated export-restraint agreements—textiles and apparel with the MFA, automobiles with the VER, and steel with the VRAs—whereas smaller, less politically influential industries had to file petitions for other trade remedies. Domestic producers liked the certainty that came with a specific quantitative limit on imports; foreign exporters also preferred an export quota because they could charge a higher price in the US market and earn a valuable “quota rent.”

The main difference between an import tariff and an export restraint is that a tariff generates revenue for the government while an export restraint generates a quota rent for exporting firms. Thus, an import tariff redistributed income from domestic consumers to domestic producers and the government, whereas an export restraint redistributed income from domestic consumers to domestic and foreign producers. Not surprisingly, foreign firms strongly preferred an export restraint to an import tariff, but the welfare consequences for the importing country were very different in the two cases. With a tariff, the losses to consumers exceeded the gains to domestic producers and the government by a deadweight loss that arose from the distortion of production and consumption. With a foreign export restraint or import quota, the losses included the deadweight loss and the much larger quota rent captured by foreign exporters. Furthermore, export quotas gave foreign firms an incentive to upgrade the quality of their products. This meant that they could move into the production of new, higher-end products and compete more directly with American producers.

In the 1980s, economists began producing quantitative estimates of the impact of various trade restrictions. For the first time, policymakers were confronted with an explicit calculation of the costs and benefits of trade protection. In almost every case, the costs of such restrictions to consumers and downstream industries exceeded the gains reaped by the protected domestic industry.48 For example, De Melo and Tarr (1992, 199) concluded that all major US trade restrictions in 1984 resulted in a net welfare loss of $26 billion, about 0.7 percent of GDP—about the same welfare loss that would have been generated by a 49 percent across-the-board tariff. Almost all of this loss—$21 billion of the $26 billion—was due to foreign export restraints in textiles and apparel, automobiles, and steel. About 70 percent of the $26 billion was due to the transfer of quota rents from domestic consumers to foreign producers; the remainder was due to the deadweight losses. Such findings suggested that the economic loss could have been significantly lower if tariffs had been used to protect domestic industries instead of export restraints or import quotas (or if the government had auctioned off the rights to import under the quota).49 Of course, foreign countries would have strongly objected if tariffs had been imposed, possibly even retaliating against them, whereas the quota rents compensated their exporters to some extent for accepting restrictions on their exports.

The Multifiber Arrangement (MFA) protecting the textile and apparel industry was the single most costly trade intervention of this period. De Melo and Tarr (1992) calculated that the welfare loss amounted to $10.4 billion in 1984, of which $6 billion was due to the transfer of the quota rent to foreign exporters.50 The MFA losses were large because of the high implicit barriers, the large volume of restricted imports, and the sizeable quota rents generated by the policy.51 The main purpose of the MFA, as with other import restrictions, was to slow the loss of jobs in the industry. Most estimates suggested that the MFA kept domestic employment in the textile and apparel industry higher than it otherwise would have been by about two hundred and fifty thousand jobs—about 10–15 percent of industry employment.52 Import restrictions could not stop the loss of jobs due to technological change; indeed, most of the fall in employment during this period was due to productivity improvements and the shift to more capital-intensive production methods, not declining output due to rising imports.53 The problem was that import restrictions were a costly and inefficient way of saving some jobs in the industry. The import restrictions were being used to save very poor jobs: average hourly earnings in the apparel and non-rubber footwear industries were among the lowest in all of manufacturing. The consumer cost of protection per job saved, which measured the total loss to consumers divided by the number of jobs saved in the protected industry, was more than $100,000 for industries in which the average worker earned perhaps $12,000 annually.54

By quantifying the consumer cost per job saved as a result of restricting imports, these studies put advocates of protectionist policies on the defensive. While some members of Congress were willing to have consumers pay this price with the hope that it would ensure the continued employment of their constituents, most policy analysts were less sympathetic. They pointed out that trade protection preserved jobs in relatively low-wage industries at the expense of high-wage jobs in export industries. Some analysts explicitly stated that these jobs were simply not worth keeping at that price. After studying the matter in relation to the textile, apparel, and non-rubber footwear industries, the Congressional Budget Office (1991, xi–xii) bluntly concluded that because “the estimated consumer costs are all higher than the average annual earnings of the workers. . . . it would generally be more efficient for government to allow the jobs to disappear and compensate any displaced workers who cannot find equivalent work.”

The second most costly trade restriction in the 1980s was Japan’s auto VER. If the VER had been removed in 1984, De Melo and Tarr (1992) calculate that the welfare gain would have been $10 billion, of which $8 billion was the quota rent transferred to Japanese producers. The restraint allowed Japanese exporters to increase their price by about $1,000 per car.55 The jobs saved in this industry were high-wage union jobs, but some analysts questioned the fairness of forcing consumers with lower average incomes than unionized workers to pay more for their cars to save the jobs of highly-paid auto workers. Others noted that export restrictions might ultimately hurt domestic producers because they gave foreign producers an incentive to upgrade the quality of products so that they could charge the highest possible markup on their constrained exports. For example, Japanese automobile producers, which had specialized in producing small, inexpensive, fuel-efficient cars, began producing larger, higher-quality vehicles that competed more directly with American brands after the VER was imposed.

Studies such as these provided greater information about the economic effects of trade restrictions, something that had been absent in previous discussions of trade policy. The findings of various studies bred widespread skepticism in policy circles about the wisdom and rationale for those restraints, giving members of Congress reason to pause before endorsing them. For example, in 1984 the Congressional Budget Office (CBO) was asked to evaluate the economic consequences of proposed legislation that would impose a five-year quota on imported steel that would cap the foreign market share at 15 percent. The CBO (1984) estimated that the quotas would raise the price of imported steel by 24–34 percent, increase domestic production by 6 percent, and boost steel-industry employment by 6–8 percent. However, by increasing the average price of steel by 10 percent (both domestic and imported), the quotas would reduce domestic steel consumption by 4–5 percent and lead to employment losses in steel-consuming industries that would roughly offset the employment gains in the steel industry. The fact that trade protection would not result in a net gain in employment (since it was an intermediate good used by other industries) was a strike against it. Furthermore, the CBO (1984, xv) argued that “there is little prospect that the quota would reverse the secular decline in the industry, since it does not address the underlying factors that have conditioned this decline.”

There was also little evidence that temporary trade protection helped protected industries adapt to the new world of global competition. In a report entitled “Has Trade Protection Revitalized Domestic Industries?” the CBO (1986, 101) concluded that “trade restraints have failed to achieve their primary objective of increasing the international competitiveness of the relevant industries.” Similarly, another study of the escape clause found that most industries receiving such protection were undergoing long, secular declines that limits on imports could not reverse. Looking back, the International Trade Commission (1982b, 86) concluded, “One observes how relatively little effect escape-clause relief had on firm adjustment either because so much of the firm’s injury was caused by non-import-related factors, or because the decline of imports following relief was small.”

These reports, among many others, identified two reasons for the failure of import restrictions to help struggling domestic industries. First, the restrictions were not very effective in reducing imports. The MFA was a porous sieve, Japan’s auto VER was not binding in its first two years and was then circumvented by foreign investment in the United States, and the steel VRAs and antidumping/countervailing duties could not prevent supply diversion. Indeed, most country-specific or product-specific trade restrictions were ineffective because of growing imports from new sources of supply or from new types of products. Despite an orderly marketing arrangement, machine tool imports were 10 percent higher in 1988 than in 1986 due to increased shipments from unconstrained suppliers.

Second, import barriers could slow but not stop the competitive pressures that were forcing producers to improve their efficiency. Like all labor-intensive industries, the textiles and apparel industry was modernizing and reducing employment only partly because of imports. In the textile industry, for example, the ITC report found that domestic producers of tufted carpets drove existing Wilton and velvet carpet producers out of business. The auto and steel industries also faced competitive pressure to increase productivity and improve product quality. By the late 1980s, an increasing number of Japanese auto producers had production facilities in the United States, meaning that import restrictions could no longer significantly diminish foreign competition. Similarly, the large, integrated steel producers had to fend off the rapidly rising market share of the small but efficient domestic mini-mills. The mini-mills had much lower costs than the integrated producers because they could process scrap metal instead of forging it from raw materials. By the early 1980s, mini-mills had captured about 20 percent of the steel market. The mini-mills were also responsible for the dramatic increase in the steel industry’s productivity in the 1980s and 1990s. While the industry lost about 75 percent of its workforce between 1962 and 2005, about four hundred thousand workers, shipments of steel were roughly the same in the two years, meaning that output per worker rose by a factor of five.56 Thus, steel producers would have faced massive restructuring even in the absence of foreign competition.

All of these cases illustrated what Robert Baldwin (1982) called the “inefficacy of trade policy” in helping struggling domestic industries.57 A classic example of the limitations of import restrictions was the celebrated Harley-Davidson motorcycle case, often heralded as the import-relief success story of the decade. As conventionally told, Harley-Davidson was pushed to the brink of bankruptcy by Japanese competition, but the company recovered quickly after it received temporary import relief in 1983 in an escape-clause case. In fact, import relief had little to do with Harley-Davidson’s turnaround. The early 1980s recession, rather than imports, had been the primary cause of the steep decline in demand for Harley’s products. The company’s resurgence came with the general economic recovery that began in 1983. Furthermore, Harley-Davidson mainly produced “heavyweight” motorcycles with piston displacements of more than 1000 cc, which were not imported because Honda and Kawasaki already produced them in the United States. Japanese producers mainly exported medium-weight bikes of 700–850 cc piston displacement, but Suzuki and Yamaha simply evaded the tariff by producing a 699 cc version that was not subject to the duty (initially set at 45 percent). Thus, protection had almost no impact on Harley-Davidson because Honda and Kawasaki were already manufacturing heavyweight motorcycles in the United States and other Japanese producers easily evaded the escape-clause tariffs on medium-weight bikes.58

If import barriers were so ineffective, why were they so often used? They arose as a second-best response to the inability or unwillingness of the president or Congress to adjust the underlying macroeconomic policies that were responsible for the strong dollar and the large trade deficit. The reality was that trade policies alone could do little to make the adjustment process less painful as long as the dollar continued to strengthen. Trade protection may have bought some firms more time to adjust, but ultimately it did not prevent large employment losses in labor-intensive industries. The main reason that most trade-sensitive industries began to perform better was that the economy began to recover in 1983, and the dollar began to depreciate in 1985. Trade protection played a very limited role in helping these industries, but it often imposed large costs on consumers.

Congress Threatens Action

While the Reagan administration protected several large American industries—automobiles, steel, and textiles and apparel—from foreign competition, it was still doing far too little to address the trade situation in the eyes of many in Congress. Legislators were frustrated by the administration’s apparent indifference to the growing trade deficit and the problems of struggling industries. In response, some administration officials argued that the trade deficit reflected the strength of the American economy and that the strong dollar reflected international confidence in the nation’s economy. To members of Congress, particularly those from regions struggling to cope with the recession and strong dollar, this benign interpretation showed a callous disregard for the difficult situation facing industries in their states and the lost jobs of their constituents. The pressure in Congress to “do something” about the trade situation came most strongly from the Midwest and the South Atlantic, where half of all import-sensitive employment was located, as table 12.1 shows. Both areas of the country saw large numbers of plant closings and the layoff of thousands of workers in the early and mid-1980s.

Table 12.1. Distribution of employment in trade-sensitive manufacturing industries, by region, 1990

Region	All manufacturing	Industries sensitive to			Factory workers receiving trade adjustment assistance (1987–92)
		Imports only	Exports only	Both imports and exports
Employment (thousands)	19,143.3	1,391.9	2,117.6	412.9	314.9
Percent	100.0	100.0	100.0	100.0	100.0
New England	6.4	8.2	8.7	11.7	8.0
Mid-Atlantic	14.3	19.7	10.9	18.4	21.3
South	23.9	26.0	14.2	15.8	28.0
Mid-West	29.5	24.4	22.3	41.5	25.5
Oil States	8.2	6.5	12.1	4.6	9.4
West	17.8	15.3	32.0	8.1	4.8

Source: Shelburne and Bednarzik 1993, 6–8.

Note: As a percentage of GDP, figures may not sum to totals due to rounding.

The nation’s old manufacturing belt, which stretched from upstate New York through Pennsylvania and Ohio and into Illinois, was particularly hard hit. This region, shown in figure 12.6, was the location of heavy industry production, particularly automobiles and steel. It became known as the “Rust Belt” because it was where most of the nation’s “deindustrialization” was occurring. While national manufacturing employment rose 1.4 percent between 1969 and 1996, manufacturing employment in the Rust Belt fell by a third.59 The manufacturing jobs lost in this region did not come back even after the recession had ended and the dollar had declined in value, leaving the Rust Belt region depressed. Although unemployment rates in the Rust Belt returned to the national average within five years of the massive job losses experienced in the early 1980s, this adjustment took place almost entirely through the out-migration of people rather than in-migration of jobs or a change in labor force participation.60 Of course, the United States remained a large exporter of manufactured goods, so the Midwest had significant export-sensitive employment as well, but these industries also suffered under the strength of the dollar.

Figure 12.6. The Rust Belt. (Map courtesy Citrin GIS/Applied Spatial Analysis Lab, Dartmouth College; based on work by Brendan Jennings 2010, Benjamin F. Lemert 1933, and John Tully 1996.)

The South Atlantic, the location of more than two-thirds of the nation’s production of textile mill products, also suffered from shuttered mills and mass layoffs. The apparel industry alone accounted for more than half of the trade adjustment assistance certifications granted during 1987–92. However, the South was able to absorb this blow more easily than the industrial Midwest. The South was attracting manufacturing investment from the Midwest and from foreign countries because of its favorable business climate.

Not surprisingly, representatives from the industrial Midwest and the textile South demanded that Congress and the administration take action to stop further employment losses. These regions were largely Democratic constituencies, and leaders such as Richard Gephardt from Missouri, John Dingell from Michigan, and Ernest “Fritz” Hollings from South Carolina were among the most vocal members of Congress advocating that policies be enacted to reduce foreign competition. The industrial Midwest was deeply scarred by the economic trauma of the early 1980s, and the region’s congressional representatives remained hostile to agreements that would reduce trade barriers well into the twenty-first century. Labor unions in the Midwest strongly pushed Democratic lawmakers into supporting import restrictions. The party became much more skeptical of trade, marking a major change from its traditional position.

Other regions of the country also suffered in the early 1980s but did not support import restrictions. The nation’s grain belt—the agricultural states of Kansas, Nebraska, Iowa, South Dakota, and Minnesota—still depended on exports. Farmers were slammed by high interest rates and falling commodity prices. On top of that, the strong dollar priced American agricultural goods out of world markets. But farm groups opposed new import restraints out of fear that their agricultural exports would be the first target of foreign retaliation: if the United States began limiting imports of clothing or steel, for example, other countries could easily shift their wheat or soybean purchases from the United States to Canada or other foreign suppliers. Consequently, Republicans from the agricultural Midwest, such as Robert Dole of Kansas, not only wanted to open foreign markets for agricultural exports through GATT negotiations, but fought efforts to close the domestic market to imports. If anything, they wanted export subsidies, lower interest rates, and a weaker dollar, much like the Midwest Populists in earlier decades.

The West was less affected by the trade difficulties of the early 1980s. As table 12.1 indicated, the West accounted for 18 percent of the nation’s manufacturing employment but 32 percent of the nation’s export-sensitive employment, with fewer industries facing direct competition from imports. California and Washington, in particular, were largely export-oriented, with a high concentration of production of high-technology goods, such as electrical equipment and aircraft, electronics and computers, as well as traditional exports of lumber and wood products. Although concerned about the strong dollar, representatives from the West, such as Robert Matsui of California, also tended to oppose protectionist measures sponsored by the industrial Midwest and textile South.

Because the industrial Midwest and the textile South were hardest hit by the trade shocks of the period, a new voting pattern began to be seen in Congress, one that first became evident in the 1970 vote over the Mills bill (shown in figure 11.2). In essence, the North-South division over trade policy seen throughout American history was reconfigured into a rough East-West pattern.61 In addition, the traditional partisan division over trade, in which Democrats favored open trade, and Republicans favored protectionist policies, began to flip. Now, many Democratic constituencies were harmed by imports while Republican constituencies tended to benefit from open trade.

Because Congress was divided, it was unlikely to pass protectionist legislation. (The auto VER, the steel VRA, and the apparel MFA were all negotiated by the executive branch with some pressure but little direct involvement from Congress.) In fact, somewhat surprisingly, given the growing pressure to address the trade situation, Congress enacted a moderate trade bill in 1984. The Trade and Tariff Act of 1984 contained no major innovations: it withheld from the president trade-negotiating authority and relaxed the statutory criteria for providing import relief, and it combined a popular idea (authority to reach a free-trade agreement with Israel) with a less popular program (the renewal of trade preferences for developing countries).62 The Reagan administration worked hard to remove any provision that might lead to special import restrictions for specific industries, such as for wine, copper, footwear, and dairy producers.63 The most controversial proposal was a reciprocity amendment sponsored by Senator John Danforth (R-MO) that would have set up an administrative mechanism to restrict imports from countries that did not provide the United States with the equivalent degree of market access, a provision clearly aimed at Japan. After Reagan threatened to veto the measure, the Danforth amendment was dropped, and the bill was passed with an overwhelming bipartisan majority in the House and unanimously in the Senate.

But this legislation was a misleading indicator of congressional sentiments on trade. Although a strong economic recovery began in 1983, Congress watched with growing alarm as the dollar continued to appreciate, and the trade deficit continued to grow. Members of Congress, particularly Democrats, complained that the Reagan administration was doing nothing to ease the pain of traded-goods industries. “President Reagan seems willing to preside over the de-industrialization of America,” House Speaker Thomas P. O’Neill (D-MA) complained. “We in Congress are not.”64

Of course, the basic problem was the strength of the dollar on foreign exchange markets. As Danforth put it, “No trade agreements, however sound, no trade laws, however enforced, will give Americans a fair chance to compete in the international marketplace if an overvalued dollar has the same effect as a 25–50 percent [foreign] tariff. To say this is not to belittle trade agreements. Rather it is to state the absolute necessity of dealing effectively with the exchange rate issue.”65 And yet the exchange rate was outside of Congress’s direct control. The appreciation of the dollar was the consequence of large capital inflows, due in part to the country’s monetary and fiscal policy. Congress could not directly influence monetary policy, which was the preserve of the Federal Reserve, and it was reluctant to adjust fiscal policy (i.e., cut the budget deficit) to reduce foreign capital inflows.

In 1984, as the trade deficit surged past $100 billion, an enormous figure at the time, trade pressures peaked with more than six hundred trade bills introduced in Congress. These bills proposed everything from the creation of a Department of International Trade and Industry (to mimic Japan) to sectoral reciprocity requirements, industry protection, and easier requirements in antidumping and countervailing duty cases.66 The proposal that received the most attention came from Rep. Richard Gephardt (D-MO) and would require countries running “excessive” trade surpluses with the United States—namely Japan, Taiwan, South Korea, and Brazil—to reduce their surpluses or faced a 25 percent surcharge on their imports.

Although the Democratic leadership endorsed the Gephardt amendment, this support may have been a strategic way of pushing the Reagan administration out of its benign neglect of the trade problems. Several decisions by the administration in 1984–85 fueled the perception that it was indifferent to the trade pressures being felt by members of Congress. In March 1985, with the presidential election of 1984 safely behind him, the president stated that Japan would not be asked to renew the auto export restraint. Although Japan announced that it would continue to enforce the limit for a fifth year, this time at the higher level of 2.3 million vehicles, members of Congress were upset that the administration let Japan off the hook. Reagan’s decision prompted the Senate to pass, by a vote of 92–0, a nonbinding resolution denouncing “unfair Japanese trade practices” and urging the president to retaliate against Japan unless it opened its market and started importing more goods to offset the additional auto exports.67 “We are in a trade war, and we are losing it,” Senator Lloyd Bensten (D-TX) complained.68

President Reagan also earned the ire of Congress for rejecting section 201 escape-clause petitions in which the ITC had found injury to domestic producers of copper and non-rubber footwear. The president justified these decisions on the grounds that import restrictions would impose a costly and unjustifiable burden on consumers, risk foreign retaliation against US exports, slow the industry from making necessary competitive adjustments, and only temporarily save jobs. Even so, his decisions frustrated many in Congress. As Destler (1995, 124) pointed out, Reagan’s actions were undertaken “without any apparent recognition that denying relief through established channels to an industry that was clearly damaged by imports was bound to increase pressure for statutory solutions”—that is, that constituent pressure would be diverted to members of Congress.

The only overtly protectionist legislation that passed Congress in 1985 was a bill to restrict imports of textiles and apparel. The Textile and Apparel Trade Enforcement Act of 1985 sought to address the failure of the MFA to stem the growth of imports, which had risen from 4.9 million square yards in 1980 to 10.2 million in 1984. As noted earlier, the MFA was a particularly porous form of protection, because restricted imports could shift between countries and between product types. As discussed earlier in the chapter, the Jenkins bill (named for its sponsor, Representative Ed Jenkins, a Georgia Democrat) mandated a 36 percent reduction in textile imports and a 20 percent reduction in apparel imports from existing levels. The bill would roll back imports from principal suppliers (Taiwan, South Korea, and Hong Kong) by establishing country- and product-specific export quotas and new limits on silk, linen, and ramie fiber goods. The bill would replace the bilateral arrangements in the MFA with comprehensive import quotas imposed unilaterally by the United States. It also included a provision to help the footwear and copper industries that had been denied import relief by the president. Imports of footwear would be capped at 60 percent of the domestic market for eight years, and the president was instructed to negotiate a five-year VRA with foreign copper exporters.

Although the House and Senate passed the Jenkins bill in late 1985, members of Congress knew this action was largely symbolic, because Reagan announced that he would veto it, which he did. In fact, most Democrats were not enthusiastic about enacting import restraints. Despite pressures from the industrial Midwest and textile South, the party’s leadership believed that such blatant protectionism was not good policy: it would violate GATT rules and lead to retaliation against US exports. As Ways and Means Committee Chair Dan Rostenkowski (D-IL) argued, “Don’t be dissuaded [from opposing the bill] by those who tell you that is a great political issue. Bad economics doesn’t make for good politics.”69

Recognizing that embracing protectionism was not necessarily a winning political strategy or an attractive policy option, Congress flirted with other fads. Some Democrats believed that an “industrial policy,” a national strategy of government support for investment in manufacturing, should be adopted, although the details of such a policy remained vague.70 Others pushed for an across-the-board surcharge on imports or something like the Gephardt proposal for higher tariffs on countries running large trade surpluses with the United States. Of course, none of these ideas was likely to become policy: it was extremely difficult to move legislation through Congress that would satisfy all regions of the country, let alone get the approval of the president.

These proposals reflected Congress’s searching for some means of relieving the trade pressures, given its inability to do anything directly about the strong dollar. Most members recognized that restricting imports would be a costly and ineffective way of addressing the trade deficit or reducing foreign competition. Even Gephardt conceded that only a small fraction of the trade deficit was due to foreign trade policies. Though sympathetic to those industries harmed by imports, Congress generally shied away from embracing outright protectionism, which was still far from being viewed as a desirable policy. As Rep. Don Bonker (D-WA) noted, “We [Democrats] have the stigma of protectionism, which comes by way of our closeness to labor and sponsorship of the domestic content bill.”71

Thus, Congress did not “go protectionist,” as Destler (1995, 66) points out. As Destler (1991, 277) put it, “Legislators became publicly engaged. They went on record in all sorts of ways; they forced the administration into a more aggressive international bargaining posture. But they did not—in the end—change American trade policy more than very slightly along its most important dimension, the openness of the US market to imports. And they did not themselves impose specific trade restraints on behalf of specific clients.” Rep. Don Pease (D-OH) explained, “We are trying to propose a respectable course between the Reagan policy of free trade at any cost and outright protectionism.”72 Or, as Rep. Thomas Downey (D-NY) stated, “Just because Reagan has no trade policy doesn’t mean we should have a bad one.”73

Reciprocity and Japan

As it searched for a solution to the trade pressures, Congress tried to put some of the burden on other countries by focusing on foreign “unfair” trade policies. The “trade hawks,” led by Gephardt, attacked other countries for engaging in unfair trade practices and keeping their markets closed. In this effort, Japan was singled out for special attention.74 Japan was an economic success story, a country that lay devastated after World War II but grew rapidly in the postwar period and eventually became the world’s second largest economy. Japan’s exports to the United States soared in the 1970s and 1980s and were highly concentrated in particular industries, leading to growing trade friction between the two countries. As Japan became an export powerhouse, the composition of its exports shifted from labor-intensive goods to more sophisticated and technologically advanced products. Industry after industry seemed to fall prey to Japanese competition, from cotton textiles and apparel in the 1950s to transistor radios, record players, and sporting goods in the 1960s; television sets, consumer electronics, and steel in the 1970s; and finally automobiles, semiconductors, and office equipment in the 1980s.

Americans were both awed by Japan’s technological prowess and fearful of its industrial dominance. Japan’s success was sometimes attributed to its Ministry of International Trade and Industry (MITI), which crafted targeted industrial policies—ranging from research and development subsidies to administrative guidance—to promote certain sectors of the economy. Some in this debate portrayed American producers as competing on an uneven playing field in facing government-supported Japanese firms, while others countered that American producers were careless about quality and slow to modernize. They also contended that MITI’s influence was exaggerated and that Japanese firms were successful because they were fiercely competitive at home and provided quality products at reasonable prices. For example, administrative guidance and industrial policy did not seem to account for the success of Japan’s automobile and consumer electronics industries, since neither was singled out for government assistance.75

The large bilateral trade imbalance symbolized the trade “problem” with Japan. From the US perspective, the issue was simple: Japan exported too much and imported too little. Under US pressure, Japan was sometimes willing to restrict its exports, so the question turned to getting it to import more. Unlike almost every other advanced economy, Japan’s imports of manufactured goods were strikingly small and failed to increase much in the 1960s and 1970s. In 1985, imports of manufactured goods as a share of manufacturing GDP was 32 percent in the United States but just 9 percent in Japan.76 Of course, this did not prove that Japan’s market was closed to foreign goods. Japan’s distinctive factor endowments may have been largely responsible: Japan lacked domestic supplies of food, fuel, and raw materials, so its imports of primary goods squeezed out imports of manufactured goods.77 Furthermore, Japan did not have many formal trade barriers: its tariff levels were comparable to those of the United States, and most import quotas had been abolished by the early 1980s.

And yet, as US Trade Representative William Brock acknowledged in 1983, “The perception, and indeed the reality, has been that the American market place has been much more open to Japanese goods than has the Japanese market to American ones.”78 Foreign businesses confronted a host of non-tariff barriers in selling to Japan, including arcane product standards and testing-certification requirements that affected imports of pharmaceuticals and agricultural products, as well as product-safety requirements that deterred imports of electrical goods, motor vehicles, and sporting equipment. There were frequent complaints about burdensome customs procedures, insufficient intellectual property protection, discriminatory government procurement practices, targeted administrative guidance, and collusive business practices, among other informal means of keeping foreign goods out of the market. Japan’s vertically integrated market structure, dominated by large, bank-centered industrial groups (keiretsu), as well as the country’s complex retail distribution system, created additional hurdles for foreign firms. As a result, foreign participation in Japan’s economy remained very low.

To address the complaints about market access in Japan, the Reagan administration initiated the Market-Oriented Sector-Selective (MOSS) negotiations in 1985. The MOSS talks focused on four sectors—telecommunications, electronics, forestry products, and medical equipment and pharmaceuticals—and concentrated on testing and certification requirements, government procurement rules, and other regulations and private practices that created obstacles to foreign producers in Japan.79 While Japan resisted any changes, arguing that foreign firms needed to improve the quality of their goods and redouble their efforts to sell in the market, it also made concessions in opening its market. And yet questions remained about whether these concessions had any real effect in promoting trade. Exporters were frustrated by the fact that their sales did not always increase when Japan’s trade barriers were reformed or removed. Indeed, subsequent analysis suggested the negotiations were a limited success. Only a small minority of the sectoral agreements—six out of twenty-seven—had a positive impact on trade, and even this was sometimes the result of diverting trade away from other partners rather than trade creation.80

The limited impact that such bilateral trade negotiations had on US exports led some to argue that the United States needed to adopt a “results-oriented” trade policy. The focus should not be on reducing trade barriers in principle, it was said, but on achieving a measurable increase in foreign sales. The Reagan administration had sharp internal debates about whether this was the right approach to take with Japan. On one side were the hard-liners (the black hats) of the Commerce Department and USTR, who wanted tough negotiations under the threat of retaliation to produce demonstrable results in terms of higher exports. On the other side were the soft-liners (white hats) of the Justice Department, the Office of Management and Budget, and the Council of Economic Advisers, who feared that results-oriented trade deals with fixed market shares would result in the cartelization of markets. The former group was accused of “Japan-bashing,” while the latter group was accused of being “soft” on foreign competitors.81

The failure of agreements with Japan to put a significant dent in its large trade surplus should not have been unexpected: opening up particular markets on a piecemeal basis might help some exporters sell more in Japan, but the bilateral trade deficit did not exist because Japan imported “too little” in protected sectors. Rather, it existed because Japan was experiencing large capital outflows, a matter entirely outside the scope of trade negotiations.

The 1985 Trade-Policy Shift

After President Reagan won reelection in 1984, key personnel changes dramatically transformed the administration’s approach to trade policy. In particular, White House Chief of Staff James Baker and Treasury Secretary Donald Regan switched jobs. At Treasury, Regan had maintained a policy of benign neglect with respect to the dollar, regarding its strength as a vote of confidence for the administration’s policies. But Baker, much like fellow Texan John Connally, Treasury secretary in the early 1970s, was concerned that the strong dollar was causing protectionist pressures to build in Congress. When he took over as Treasury secretary, Baker (2006, 427) recalled, “We confronted an overvalued dollar, measured against other currencies, and a trade imbalance that favored the Japanese, Germans, and other trading partners at the expense of US manufacturers and exporters. These two economic problems, in turn, had created a big political problem—a protectionist fever in Congress that grew hotter each time Honda or Mercedes won another customer from the Big Three or another pop economist wrote about the inevitable triumph of Japan, Inc.” Like Connally before him, Baker sought to relieve the pressures by helping to engineer a decline in the value of the dollar against other currencies.82

After convincing the president and Federal Reserve Board Chairman Paul Volcker of the need to adopt a new exchange-rate policy, Baker sought international agreement to bring about an orderly decline in the value of the dollar.83 Other countries resisted the call to strengthen their currencies, just as they had in 1971, but as Baker (2006, 429–30) noted, “Our leverage with them was that if we didn’t act first, the protectionists in Congress would throw up trade barriers. Auto makers and other industries were pounding the desks at the White House, Treasury, and Congress, demanding that something be done to save them from foreign competition, and Congress was listening. By late summer [1985], top foreign economic officials had begun to see that we were serious.”

The result was the Plaza Accord, named for the famous New York hotel where the September 1985 meeting was held. At the meeting, Japanese and European officials agreed to undertake measures that would lift the value of their currencies against the dollar, including coordinated central bank intervention in foreign exchange markets.84 Although the dollar had actually begun to depreciate in February 1985, it continued to fall against other major currencies over the next four years. Because exchange-rate changes have a lagged effect on trade flows, however, the trade deficit continued to grow for another two years before finally receding.

In addition to taking action on the dollar, Baker and US Trade Representative Clayton Yeutter came up with a new trade agenda to defuse protectionist pressure in Congress. The administration began taking a more aggressive stance against foreign unfair trade practices and attempted to revive bilateral and multilateral negotiations to reduce trade barriers. The package was designed to impress Capitol Hill, serve as a warning to key trading partners, and shift the debate from closing the US market to opening foreign markets.

On September 23, 1985, the day after the Plaza Accord was announced, President Reagan unveiled this new strategy. The president portrayed the United States as the victim of other countries’ trade policies and spoke of the responsibilities that they had in maintaining an open trading system. As he put it,

to make the international trading system work, all must abide by the rules. All must work to guarantee open markets. Above all else, free trade is, by definition, fair trade. When domestic markets are closed to the exports of others, it is no longer free trade. When governments subsidize their manufacturers and farmers so that they can dump goods in other markets, it is no longer free trade. When governments permit counterfeiting or copying of American products, it is stealing our future, and it is no longer free trade. When governments assist their exporters in ways that violate international laws, then the playing field is no longer level, and there is no longer free trade. When governments subsidize industries for commercial advantage and underwrite costs, placing an unfair burden on competitors, that is not free trade.

Therefore, the president continued,

we will take all the action that is necessary to pursue our rights and interests in international commerce under our laws and the GATT to see that other nations live up to their obligations and their trade agreements with us. I believe that if trade is not fair for all, then trade is free in name only. I will not stand by and watch American businesses fail because of unfair trading practices abroad. I will not stand by and watch American workers lose their jobs because other nations do not play by the rules.85

The new strategy involved a more aggressive use of section 301 of the Trade Act of 1974, until now a relatively neglected part of US trade law. Under section 301, USTR had the authority to investigate and respond to any foreign government acts, policies, or practices that were “unreasonable, unjustifiable, or discriminatory” and “burden or restrict US commerce.”86 Such an investigation usually started with a firm or industry association filing a petition with USTR, which had complete discretion to accept or reject the case. In any such investigation, USTR sought to resolve the problem by consulting with the foreign government whose policies were under scrutiny. If the consultations did not resolve the matter, a range of remedial enforcement actions could be taken, most importantly retaliation in the form of higher tariffs against the country’s exports.

Reagan announced that USTR would initiate several section 301 cases involving South Korea’s restrictions on foreign insurance firms, Brazil’s obstruction of trade in informatics (computer hardware and software), and Japan’s barriers on the sale of foreign tobacco products. The president set a short deadline for resolving existing cases on Japan’s market for leather and non-leather footwear and the EEC’s subsidies for canned fruit before moving to retaliation. USTR also began to look into whether Japan’s telecom procurement policies excluded foreign producers from the market.

A high-profile case involving Japanese semiconductors exemplified this new strategy of more aggressively defending US producer interests and attacking foreign trade barriers.87 US domination of semiconductor production was threatened, as were other industries, by Japan’s rapid penetration of global markets in the 1980s. In 1978, American firms accounted for 55 percent of worldwide industry revenues, and Japanese firms accounted for 28 percent; by 1986, Japan’s share had risen to 46 percent, while the US share had fallen to 40 percent. The relative decline of the semiconductor industry seemed to symbolize the country’s loss of technological leadership and was partly due, it was commonly thought, to industrial targeting by the Japanese government.

In June 1985, the Semiconductor Industry Association (SIA) filed a section 301 petition alleging that Japan’s semiconductor market was closed to foreign producers due to structural barriers in Japan. These structural barriers—including anticompetitive business practices among its firms condoned by the government and supported by official regulations—denied “fair and equitable market opportunities” to American firms in Japan’s market and therefore were deemed “unreasonable” under the meaning of section 301.

The semiconductor case was complex, because it also involved a dumping case. Although the US economy was in a strong recovery, the semiconductor industry suffered a severe cyclical slump in 1984–85 due to a slowdown in computer production. The plummeting price of semiconductors led to large losses for all producers in the industry. However, while American firms were pushed to the brink of bankruptcy and reduced output sharply, Japanese firms continued to produce because they had ongoing access to credit due to their close ties to large banks. Japanese competition forced almost every American producer of one particular device, dynamic random-access memories (DRAMs), out of the market. In general, the SIA did not want the Commerce Department to impose antidumping duties: they did not want to create obstacles to the shipment of various devices across borders, because the production of semiconductors was worldwide; even the electronic workers’ union feared the off-shoring of semiconductor production if import barriers were increased. Furthermore, the industry’s problem was not dumping in the United States in particular but the worldwide depression of prices in general. Antidumping duties would not solve that problem, but instead make the country a “high price island” to the detriment of semiconductor-using industries such as computer manufacturers and telecommunications equipment producers. At the same time, given the size of the US market, denying Japanese firms access to American consumers would be a way of forcing Japan to resolve the issue.

Despite the SIA’s opposition to tariffs, one small DRAM producer (Micron Technology) filed an antidumping complaint against four Japanese exporters of 64K DRAMs. Micron contended that the home market sales of the Japanese producers were below their costs of production. This opened a floodgate. Micron’s June 1985 petition was followed in September by an antidumping petition from three major semiconductor producers (Intel, Advanced Micro Devices, and National Semiconductor) concerning another device, Erasable Programmable Read-Only Memories (EPROMs). The Commerce Department then, for the first time, initiated its own dumping case against Japan on 256K and future generations of DRAMs.88

The fact that the US government had initiated a dumping case and gave strong backing to the section 301 case put Japan under enormous pressure to resolve the conflict. If it did not find a way to end the dispute before the statutory deadlines facing USTR and the ITC, Japan would face antidumping duties on its semiconductor exports and retaliatory tariffs on its other exports. As a result, in July 1986, the United States and Japan reached a five-year agreement to end dumping and open up Japan’s market, allowing the dumping and section 301 cases to be suspended. Japan’s government agreed to enforce export price floors (adjusted quarterly) on semiconductors sold in world markets and to encourage its businesses to purchase foreign semiconductors. In a secret side letter to the agreement, Japan accepted that the foreign share of its semiconductor market should increase to 20 percent from the existing level of about 8 percent.89

The market share target, which was generally acknowledged to exist, was controversial in both countries. In the United States, proponents argued that it provided “affirmative action” for foreign producers in a market that was difficult to penetrate. The 20 percent target was held up as an example of a “results-oriented” approach that some thought should be employed more generally, since previous efforts to open up Japan’s market had proven disappointing. But critics argued that the target constituted “managed trade,” in which the commercial outcome was determined by government deals without regard to market competition and productive efficiency. In Japan, businesses resented their government’s meddling with the prices they could charge and the sources from which they could buy. They feared that if this was done in the case of semiconductors, other industries would be next. The agreement sparked a nationalist backlash among some in Japan who urged the government to say “no” to foreign demands.

In fact, Japan’s government had difficulty in getting local firms to comply with the terms of the agreement. Government officials could not easily prevent “dumping” in third markets or persuade Japan’s industrial buyers to increase their purchases of foreign semiconductors. The SIA soon complained that Japan was failing to live up to both the dumping and the market-access provisions of the agreement. With trade pressures at a peak in Congress, the Reagan administration could not back down and risk another “failed” trade agreement with Japan.

In January 1987, just six months after the semiconductor agreement was signed, USTR gave Japan sixty days to demonstrate that it was enforcing the agreement. The administration wanted concrete signs that third-market dumping had ceased and foreign sales of semiconductors in Japan were increasing. In April, the United States stunned Japan by declaring it in non-compliance with the agreement and retaliated by imposing 100 percent tariffs on $300 million of computers, televisions, and power tools imported from the country. This unilateral action, one of the largest retaliations of the postwar period and the first such action against Japan, dramatized the seriousness with which the administration viewed the semiconductor agreement and the trade problem with Japan. The retaliatory tariffs were partially lifted later in the year after Commerce determined that dumping had ceased. (Semiconductor production migrated to Taiwan and South Korea to avoid the antidumping measures, but those countries were also later charged with dumping.) The remainder were eliminated in 1991 when the foreign share of Japan’s semiconductor market reached 20 percent.

The United States began taking a hard-line stance against the EEC as well. In 1985, higher duties were imposed on imports of European pasta in retaliation for the EEC’s restrictions on imports of citrus products. In 1986, Spain and Portugal joined the EEC and had to adopt its common external tariff. Although their tariffs on manufactured goods would fall, Spain and Portugal now had to restrict imports of agricultural goods as part of the Common Agricultural Policy. With American exports of feed grains to those countries expected to drop by $1 billion, USTR threatened the EEC with retaliation unless compensation was given for the lost sales. In December 1986, after Europe failed to act, the Reagan administration announced that it would impose 200 percent duties on $400 million worth of European exports to the United States. A month later, the two sides reached an agreement whereby the EEC would guarantee that Spain would import a certain amount of American corn and sorghum for the next four years and reduce its duties on other goods.90

The Reagan administration took other trade actions in 1986 to head off protectionist trade legislation on Capitol Hill. Section 232 of the Trade Expansion Act of 1962, which gave the president the authority to restrict imports on grounds of national security, was invoked to persuade Japan, West Germany, Taiwan, and Switzerland to reduce their exports of machine tools to the United States. The administration imposed a 35 percent countervailing duty on red cedar shakes and shingles from Canada. USTR also continued to press the EEC over its ban on beef that had been treated with growth hormones and Brazil over its restrictions on information technology products and pharmaceuticals, and pursued other cases as well.

Between the Plaza Accord and the new section 301 policy, September 1985 marked a dramatic shift in US trade policy. Did these policies work? With time, the depreciation of the dollar helped reduce the trade deficit and relieve protectionist pressures. And section 301, Bayard and Elliott (1994, 64) conclude, “appears to have been a reasonably effective tool” in opening up foreign markets. At least initially, however, these actions did little to reduce the trade pressures coming from Congress. The trade deficit continued to grow in 1986 and 1987, and Congress continued to demand that the administration address the trade situation, including providing more relief for domestic industries competing against imports. As Rep. John Dingell (D-MI) complained, “While our trading partners have relentlessly pursued their economic self-interest in determining import and export policies, this country has been hamstrung by a free-trade ideology that ignores the reality of the world trading system.”91 Treasury Secretary Baker conceded in 1987 that the administration had been “a little late” in addressing the trade deficit, but insisted that “for the last several years, no administration has worked harder than we have against subsidized imports and trade barriers abroad.” Responding to critics who complained that the administration was too wedded to a free-trade philosophy, Baker replied that the president had “granted more import relief to United States industry than any of his predecessors in more than half a century.”92

Meanwhile, Democrats in Congress continued to work on trade legislation. In May 1986, the House passed a bill that included the controversial Gephardt provision that would require countries with “excessive and unwarranted” trade surpluses with the United States (Japan, Germany, and Taiwan) be given six months to change their policies and start reducing their surpluses by 10 percent annually. If they did not comply, the president would have no option but to retaliate against them. The bill also expanded the avenues for firms and workers to obtain relief from foreign competition; for example, import relief under the escape clause would be allowed if imports were an “important” cause of injury rather than a “significant” cause. The House passed the bill by a 295–115 vote, with Democrats voting 236–4 in favor and Republicans voting against 111–59. Although Democrats took the harder line, many Republicans also voted for the bill. “Republicans are displaying the same kind of protectionist urges that Democrats are showing because their constituents are showing it,” William Frenzel (R-MN), a leading free trade advocate in Congress, admitted. “They are reflecting what their constituents feel.”93

But the president was unpersuaded. “This anti-trade bill isn’t protectionism, it’s destructionism,” Reagan responded. The president said that the legislation was not an “omnibus trade bill” but an “ominous anti-trade bill,” adding that “this bill is so potentially destructive that even many of those who voted for it did so in the expectation that it would be vetoed. . . . Well, if it comes to that, I assure them they’ll get their wish.”94 Speaker O’Neill countered, “We don’t believe we are protectionists. We believe we are patsies for the rest of the world, and we want to be fair-traders.”95

The Republican Senate adjourned before acting on the bill, but the Democrats captured the Senate in the November 1986 midterm election. This gave the Democrats unified control of Congress and paved the way for the passage of a trade bill that would reach the president’s desk. In January 1987, the new House Speaker Jim Wright (D-TX) declared that “the first imperative in the 100th Congress will be to come to grips with the steady decline in American competitiveness and the corollary increase in the trade deficit.”96 Recognizing that a new trade bill could have some desirable features, such as trade-negotiating authority, administration officials began working more closely with Congress in order to strip out objectionable provisions, such as the “procedural protectionism” that limited executive discretion in trade cases, and make the bill acceptable to the White House. In early 1987, the new House passed a bill similar to the previous one. On the House floor, Gephardt introduced a watered-down version of his provision requiring that countries running large trade surpluses with the United States be threatened with retaliation. The Gephardt amendment narrowly passed by a vote of 218–214, which, in view of the administration’s opposition to it, ensured that it would not survive in a conference committee. The Senate soon passed its own trade bill by 71–27, but the October 1987 stock market crash prevented the completion of the reconciliation process before Congress adjourned.

The bill’s prolonged path through Congress was finally completed in 1988, but not after more last-minute wrangling. In April 1988, the conference committee dropped the Gephardt amendment, and both chambers passed the bill, but the president vetoed it because of a plant-closing provision. Congress passed the bill again without the provision, and finally, after three years of work, the Omnibus Trade and Competitiveness Act of 1988 was completed in a form that the president could approve.97

At more than a thousand pages, the wide-ranging legislation made many incremental changes to the nation’s trade laws. It reduced presidential discretion and expanded avenues for firms and workers to receive relief from imports in administrative trade cases. More importantly, the president was granted new negotiating authority, which allowed the executive to reduce tariffs by up to 50 percent in trade agreements reached within five years. The authority also ensured that Congress would give “fast-track” consideration to any agreements that required changes in domestic legislation, meaning that it would have sixty days to approve or disapprove of an agreement without amendment, after the president submitted it to Congress. These provisions were needed to conclude the Uruguay Round of trade negotiations that had been launched in 1986 and will be discussed in chapter 13.

The bill also included a controversial “Super 301” provision. Under Super 301, USTR was required to identify priority countries in 1989 and 1990 and negotiate the elimination of barriers to key US exports within a fixed time period. Another “Special 301” provision called on USTR to examine the failure of other countries to respect intellectual property rights, particularly as it affected the high-technology (software and semiconductors), entertainment (movies and music recordings), and pharmaceutical industries.

The Omnibus Trade and Competitiveness Act of 1988 capped a long struggle to formulate a trade bill that would be acceptable to the Reagan administration. To avoid a presidential veto, the legislation did not come with special protection for any specific industries. However, the textile and apparel industry was still unsatisfied with the MFA. In September 1987, the House passed a bill to impose import quotas on 185 categories of textiles and apparel and 15 categories of non-rubber footwear, but the vote fell short of the two-thirds majority needed to override a veto. The Senate did not pass the bill until September 1988. As expected, Reagan vetoed it—the second textile quota bill that had been sent to him.

By this time, however, the enormous political pressures over trade that had marked the early- and mid-1980s had eased considerably. By 1987, the dollar had fallen back to its 1980 level, and exports were finally accelerating, as figure 12.1 showed. The growth in exports and moderation of imports, combined with the long economic expansion that brought down the unemployment rate, relieved much of the pressure on traded-goods industries. As a result, pressures in Congress to address trade problems also began to dissipate.

The Multilateral System in Disarray

So far, this chapter has said little about the multilateral trading system. In fact, the value of the GATT was severely questioned during the trade-policy turmoil of the early 1980s. Previous multilateral trade agreements, such as the Kennedy Round in the 1960s and the Tokyo Round in the 1970s, had cut and bound tariffs on manufactured goods at low levels for developed countries. While tariffs remained in check, the GATT failed to stem the spread of non-tariff barriers to trade around the world. This left trade-policy observers concerned that the postwar system of open and nondiscriminatory trade was being eroded and could even fall apart.

The trade barriers that had proliferated since the 1970s included variable import levies (in the case of European agricultural imports), antidumping and countervailing duties, voluntary export restraints, orderly market arrangements, and other forms of managed trade. Between 1966 and 1986, the share of imports covered by non-tariff barriers rose from 36 percent to 45 percent in the United States, from 25 percent to 54 percent in the EEC, and from 31 percent to 43 percent in Japan.98 The increase was particularly large in the case of the EEC due to the expansion of the Common Agricultural Policy, but non-tariff barriers were applied to an increasing share of manufactured goods as well. The spread of voluntary export restraints was particularly striking. In 1989, the GATT reported that 236 export-restraint agreements were in effect, 67 involving exports to the United States and 127 involving exports to the EEC. Most of these originated in the 1980s and operated outside GATT rules or disciplines.99

Considered either as an institution or as a set of rules, the GATT failed to contain the ever-changing forms of protectionism that spread during this period. This raised the question of whether the GATT should be taken seriously if its rules were so widely ignored and its provisions so easily circumvented. In a 1978 article “The Crumbling Institutions of the Liberal Trade System,” University of Michigan law professor John Jackson argued that “almost every rule of GATT is inadequate to the present problems of world trade,” sometimes because of evasion but often due to lack of enforcement. Trade rules were violated because there was no effective dispute-settlement mechanism to hold countries accountable if they restricted trade. “The tragedy is that a defective rule system tends to punish those who abide and reward the transgressors,” which only encourages “further erosion of rules.” All of this contributed to the deterioration in the world trading system.100 “The real danger to the GATT is not that a trade war will break out, but that the major signatories to the GATT will simply pretend that the General Agreement is not there,” Arthur Dunkel, the director-general of the GATT, remarked in 1982. He believed that “this would effectively end the GATT” and the rules-based system of trade. The only thing holding the system together, he argued, was a “sort of a balance of terror” in which countries feared imposing too many protectionist measures only because they might bring retaliation.101

After the conclusion of the Tokyo Round in 1979, officials at USTR began planning for the next round of multilateral trade negotiations. The agenda consisted of old issues (agriculture and safeguards) and new ones (services and high-technology products). The United States started pressuring other countries to endorse the agenda and start negotiations, but the moment was inauspicious because of the world economic slump. A GATT ministerial meeting in November 1982, in which the United States pushed to launch a new trade round, collapsed amid acrimony and dissension. As Richardson (1994, 641) put it, “The European Community was especially resistant to agricultural liberalization, India and other developing countries were dead set against services liberalization, and all accused the United States of ramming its agenda onto the table without adequate documentation, interpretation, persuasion, or quid pro quo.” At the same time, participants worried about mounting protectionism. The ministerial declaration warned that the multilateral trading system “is seriously endangered” because protectionist pressures had multiplied, and governments were disregarding GATT rules. The ministers called for a “standstill and rollback” on trade barriers—an immediate halt to the introduction of new trade barriers and the start of their removal—but these words went largely unheeded.102 American officials left Geneva disheartened that other countries had rejected opening new trade negotiations. Frustrated with the GATT process, they began to consider seriously, for the first time in the postwar period, alternatives to the multilateral approach to reducing trade barriers.103

Although it only became apparent with hindsight, the outcome of the 1982 GATT ministerial meeting played an important role in persuading US trade officials to consider alternative methods of reducing trade barriers. If GATT participants were not willing to strengthen the enforcement of existing rules and patch holes in the legal framework, the United States was prepared to bypass the multilateral process by undertaking unilateral actions to enforce trade rules and address foreign trade barriers, as well as starting bilateral and regional discussions to open trade further. In 1985, President Reagan again called for the start of a new multilateral trade round, but warned that “if these negotiations are not initiated or if insignificant progress is made, I’m instructing our trade negotiators to explore regional and bilateral agreements with other nations.”104 In fact, disenchantment with the GATT was one reason that Trade Representative William Brock argued that the Trade and Tariff Act of 1984 should include authority to reach a bilateral trade agreement with Israel. Concluded the next year, the trade agreement phased out all tariffs on bilateral trade over ten years. Because Israel was a popular ally, the agreement encountered no opposition in Congress.105

While the agreement with Israel was more of a foreign-policy gesture and hardly signaled a major change in trade policy, the possibility of a free-trade agreement with Canada soon appeared. Canada had long been concerned about maintaining its access to the US market, particularly in light of trade actions that had damaged its economy: Canadian exports had not been exempt from the 10 percent import surcharge in 1971, and its other exports were routinely harassed with antidumping and countervailing duties, which caused protracted problems for the bilateral relationship in the 1980s. Canadian producers often found themselves at the mercy of the Commerce Department and the ITC where an unfavorable ruling could disrupt their exports for an extended period. Canadian officials also feared that business investment suffered because its access to the US market could not be guaranteed.

At the same time, Canada had long worried about being dominated by its giant neighbor and losing its sovereignty and unique culture. These domestic sensitivities left Canadian officials quite ambivalent about whether to seek a closer economic relationship with the United States. Of course, while the United States could announce its willingness to explore a bilateral trade agreement, the initiative had to come from Canada; any such proposal from the United States would have aroused fears among Canadian nationalists that it was plotting to take the country’s resources and control its economy.106

A 1985 report by a Royal Commission on Canada’s economic future served as the catalyst for moving forward. Like the United States, Canada had growing concerns about its productivity performance and the competitive position of its exports in world markets. The report urged Canadian officials to embrace policies that would lead to a more flexible and dynamic economy, less sheltered from foreign competition and fragmented across provinces. To this end, the report urged a “leap of faith” with the negotiation of a free-trade agreement with the United States. Such an agreement would not only expose the Canadian economy to greater competition, thereby forcing producers to improve their efficiency, but also lock in Canada’s access to the US market and leave the country much less vulnerable to unpredictable changes in American policy. In October 1985, after much debate within the government, Conservative Prime Minister Brian Mulroney, who earlier in his career had opposed the idea of free trade with the United States, wrote to President Reagan and requested the negotiation of a free-trade agreement.

The economic stakes in such an agreement were much higher for Canada than for the United States. In 1985, nearly three-quarters of Canadian exports (about 20 percent of its GDP) were destined for the United States, whereas 20 percent of US exports (1 percent of its GDP) was sent to Canada. On average, Canadian tariffs were higher on American products (about 9 percent) than US tariffs were on Canadian products (about 4 percent). About 80 percent of Canada’s exports to the United States and 65 percent of US exports to Canada were already duty-free due to a special 1965 agreement that eliminated tariffs on bilateral trade in automobiles and auto parts. Yet the United States had something to gain beyond lower tariffs: Canada maintained many restrictions on foreign investment (particularly in financial services, energy, and natural resources) and subsidized many producers (the source of complaints from American firms).

Not surprisingly, given the imbalance in economic size between the two countries, Canada took the negotiations far more seriously than the United States.107 Neither the Congress nor the Reagan administration seemed to realize that the Canadian government had taken a huge political step in proposing a deal. Negotiations began in early 1986 and did not start well.108 Canadian negotiators were constantly frustrated at the lack of high-level attention given to the negotiations by the Americans. The lead USTR negotiator, Peter Murphy, was not in a position to provide leadership; he was a staff-level official who lacked the political authority to conclude a deal.109

The two countries also had different conceptions of the agreement. The United States wanted a simple accord to eliminate tariffs, reduce Canadian subsidies, and open up the country to foreign investment. Canada sought a broader agreement that would rein in the use of trade remedies, which routinely harassed Canadian exporters. It proposed harmonizing antidumping and countervailing duty rules, which USTR ruled out as something Congress would never agree to do.

Unable to get any change in US trade laws, Canadian negotiators proposed creating a binding, impartial system to settle disputes over the implementation of those laws. By mid-September 1987, Canadian officials were frustrated by the failure of American negotiators to take the idea seriously. (Two influential Senators, Lloyd Bensten and John Danforth, said that it was out of the question.) Believing that no agreement was better than a bad or limited agreement, the Canadians were prepared to declare the negotiations a failure. The deadline for concluding the talks was coming up rapidly: Congress had to be notified by midnight October 3, 1987, for the agreement to be eligible for “fast-track” consideration. In late September, Canadian negotiators said the talks were on the brink of failure and that dispute settlement was a make-or-break issue, all of which came as a surprise to higher-level Reagan administration officials.

Alerted to an imminent breakdown, Treasury Secretary James Baker took over the negotiations and eventually saved the deal. Sam Gibbons (D-FL), the chair of the trade subcommittee of the Ways and Means Committee, called Canadian ambassador Allan Gotlieb and suggested that, since joint rules on antidumping and countervailing duties were unacceptable to Congress, a tribunal be set up to review the application of the existing trade laws of each country in the case of a dispute.110 Although Canadians were worried that the United States could later simply strengthen its antidumping and countervailing duty laws, the idea of an impartial tribunal was a concession. The Canadians encouraged Senator Bill Bradley (D-NJ) and a few other strong congressional supporters of an agreement to call Baker and show their support for this idea. Despite the opposition of USTR, and after his initial skepticism, Baker came around to the proposal. He began building support in Congress for an independent review system, while extracting further concessions from the Canadians on banking services and cross-border investment.111

After receiving assurances that a deal was looking good, the Canadians returned to the bargaining table but found that nothing had changed. On the night that Congress had to be notified that an agreement had been reached, the Canadians were prepared to declare the negotiation a failure. Around 9:00 p.m., Baker burst into the negotiating room, flung down a piece of paper, and exclaimed, “All right you can have your goddamn dispute settlement mechanism.”112 Ten minutes before midnight, the agreement was finalized and a messenger dispatched to Capitol Hill, where the president’s notice was delivered one minute before the deadline. As Ambassador Gotlieb (2006, 493) later stated: “If it were not for Baker, there would be no agreement.”

President Reagan and Prime Minister Mulroney signed the accord in January 1988, but it was not formally submitted to Congress until July 1988 after extensive hearings. (The administration had been negligent in not consulting very much with members of Congress about the negotiations, a mistake that would not be repeated with the North American Free Trade Agreement.) In a bipartisan vote that summer, Congress overwhelmingly approved the agreement, the House by 366–40 and the Senate by 83–9.113 By contrast, Canada had an intense public debate over the issue, and the 1988 election hinged on the issue. The Conservatives won the election, the House of Commons approved the agreement, and it took effect in January 1989.

The US-Canada free trade agreement was a historic achievement that cemented economic ties and reduced trade friction between the neighboring countries. The 250-page text had twenty-one chapters and annexes, and included a ten-year phase-out of all tariffs on bilateral trade, an agreement on trade in services, a cautious opening of investment and financial services, an impartial panel to settle disputes over trade remedies, special protection for Canadian culture that restricted foreign investment in media and film, and various other provisions. However, because bilateral trade was already fairly free, the agreement had a limited impact on the overall US economy.

Although this was a major step in the direction of bilateral trade agreements, the United States did not give up on the multilateral system. In 1986, it managed to lead other countries in the GATT to start the Uruguay Round of trade negotiations. These negotiations were concluded in 1993 and are discussed in chapter 13.

The Protectionist Tide Recedes

The continued depreciation of the dollar against other currencies after the Plaza Accord in September 1985 eventually succeeded in providing relief for industries competing against imports and exporters selling on the world market. From its February 1985 peak to the April 1988 trough, the real trade-weighted value of the dollar fell nearly 30 percent. As a result, export growth picked up, and import growth moderated. In fact, export-related industries accounted for half of the increase in manufacturing employment in 1987–88 before the economy slowed in 1989.114 The merchandise trade deficit peaked at $159 billion in 1987 and then fell over the next three years. With continued economic growth, these developments greatly eased protectionist pressures. As figure 12.5 showed, the number of antidumping petitions filed by domestic firms fell sharply after 1987. Of course, many of the recently imposed import restrictions remained in place: nearly a third of Japan’s exports to the United States were still restricted in some way.

This environment gave the new administration of President George H. W. Bush, and his trade representative Carla Hills, the opportunity to roll back some of the protectionist measures that had accumulated during the 1980s. The three big protected sectors were automobiles, steel, and textiles and apparel, and the special protection given to each soon came to an end. First, Japan’s voluntary export restraint in automobiles faded away. Japan had repeatedly renewed the VER after 1981, but eventually it was no longer binding because major Japanese auto producers established production facilities in the United States that replaced exports from Japan. By 1991, Japanese firms accounted for 15 percent of domestic production of cars and light trucks. As a result, the VER was no longer needed: the export limit was 2.3 million vehicles that year, but the country shipped only 1.73 million from Japan. In 1992, Japan reduced the export cap to 1.65 million vehicles, but over the next year Japan exported only 1.4 million vehicles, well under the limit.

In early 1994, Japan announced that it would no longer enforce the export restraint. Union officials and automobile executives in the United States grumbled about the decision, but they were not in a strong position to demand that it be continued. The automobile industry had recovered from the dark days of the early 1980s in terms of profitability and production, if not employment. Furthermore, Japanese firms had done what both the industry and union had asked: namely, to “make cars where they sell them.” And having taken investment stakes in some Japanese producers, the Big Three were now among the leading importers of cars from Japan.

The steel industry posed a more difficult challenge. The comprehensive VRAs won by the steel industry in 1984 were set to expire in 1989, shortly after the 1988 presidential election. Like the automobile industry, the steel industry was in much better shape than it had been a few years earlier: capacity utilization had increased considerably, financial losses had turned into profits, and imports had receded as the weaker dollar helped bolster the competitive position of the industry. In addition, the large steel firms had gradually rationalized production by shutting down inefficient plants and modernizing production facilities to increase productivity and reduce labor costs.

However, in running for president in the 1988 election, Vice President George H. W. Bush responded to weak poll numbers in Ohio and Pennsylvania by pledging to renew the import restraints, although he did not make any specific promises. The prospect of renewing the VRAs sparked a fierce debate that now included opposition from steel-consuming firms, such as the heavy earth-moving equipment producer Caterpillar and other major steel consumers. They complained that steel protection raised their production costs and harmed their competitive position against foreign rivals. Noting that they employed many more workers than the steel industry itself, the steel users argued that these jobs were jeopardized by steel shortages and higher prices caused by import restrictions.

This put the integrated steel producers in the unusual position of defending the VRAs, at a time when they were earning large profits, against the opposition of another major group of manufacturers. With the steel industry demanding a five-year extension of the VRAs, and steel users insisting that they be abolished, the Congressional steel caucus had to mediate between the two groups. Congress itself was divided. (Sam Gibbons quipped that the 1989 congressional reauthorization of the administration’s authority to conclude agreements limiting steel imports was “the only bill I ever introduced that nobody liked and everybody agreed to vote for.”)115 The Bush administration proposed a compromise that would extend the import restraints for a transition period of two and a half years, after which they would be eliminated. To help steel consumers during this period, the VRAs would be relaxed if steel was in short supply. The administration also argued that the ongoing GATT negotiations, rather than export restraints, were the proper way to deal with global excess capacity and trade-distorting practices in the steel market.

The integrated steel producers and the United Steel Workers were disappointed with this decision. Both had benefited from import restraints since 1967, and losing the VRAs reflected the industry’s diminished political clout. When the VRAs expired in early 1992, the steel industry immediately swamped the Commerce Department and ITC with more than eighty antidumping and countervailing duty petitions. Although Commerce found dumping and subsidy margins as high as 109 percent, the ITC made an affirmative injury determination in only thirty-two cases; most were dismissed because imports were often selling at higher prices than domestic steel. The rejection of these cases was another blow to the industry’s hopes for continued protection.

The textile and apparel industry also saw its special protection taken away. By 1991, the MFA’s export quotas involved 41 countries and covered 69 percent of textile and 88 percent of apparel imports.116 However, as we have seen, the MFA was a sieve rather than a wall in keeping out imports, and domestic producers constantly fought to get the restrictions tightened. Hoping that the Bush administration would be more sympathetic than its predecessor, the industry made yet another attempt to get comprehensive trade protection from Congress. In April 1990, the Congressional Textile Caucus introduced the third major textile and apparel bill in five years. Like the 1988 legislation, the bill would set an overall cap on imports by product (rather than by country) and limited their growth to one percent per year. To defuse opposition to the bill from Midwestern farm states, which had previously opposed such legislation out of fear that foreign countries would retaliate against its agricultural exports, the bill included a provision that would allow a country’s textile quota to increase if it increased its imports of US agricultural goods. The bill also included a provision to freeze imports of non-rubber footwear at their 1989 level in order to win extra votes from members of Congress with shoe producers in their districts. Despite these efforts to broaden the bill’s appeal, the textile and apparel industry was still in a weak political position: the American Apparel Manufacturers Association did not support the measure, because many of its members had already moved plants overseas or imported goods from foreign garment makers.

Congress passed the bill without enthusiasm, because President Bush had already announced that he would veto it. The president called it “highly protectionist” legislation that would damage the economy, increase already high costs of clothing to consumers, and violate the rules of the GATT.117 All of this would occur, the president noted, without eliminating any unfair trade practices or opening any closed markets abroad. This veto cleared the ground for the complete abolition of the MFA in the Uruguay Round, as discussed in chapter 13.

Other trade issues of the 1980s also faded away during the Bush administration. When the 1986 US-Japan semiconductor agreement came up for renewal in 1991, opposition from domestic computer producers (consumers of semiconductors) helped end the antidumping provisions of the agreement. The orderly marketing arrangement for machine tools expired, and other trade restraints lapsed. As a result, the share of imports covered by special protection measures receded from 21.5 percent in 1984 to 10.4 percent in 1990.118 The trade protection of the 1980s proved to be a temporary and not a permanent part of US policy.

After the intense trade pressures of the 1980s, what explains the relatively easy reversal of protectionism in the early 1990s? The unwinding of import restrictions was largely due to three factors: changed circumstances, economic adjustments, and increased domestic opposition.

The changed circumstances included the economic recovery that began in 1983 and continued through most of the 1980s: the growth in domestic demand revived the manufacturing sector and reduced unemployment. The depreciation of the dollar against other currencies, which helped both import-competing producers in the domestic market and exporters in the world market, was also critical in relieving protectionist pressures and allowing existing measures to disappear.

The economic adjustments included the response of domestic firms to deal with the intensification of competition. Although many of these adjustments were painful, including the closing of less efficient, higher-cost production facilities and the layoff of many workers, the results were lower costs and greater efficiency. Many firms had to change their product mix, usually by upgrading product quality, such as the semiconductor industry’s shift from producing commodity memory chips to focusing on specialized devices and sophisticated microprocessors. In addition, many industries became globally diversified with investments in overseas production, which reduced their interest in erecting trade barriers around the US market. Many foreign firms also established production facilities in the United States, which further diminished the value of restricting imports to domestic firms.

Finally, domestic opposition to protectionist measures had strengthened. Firms that purchased imported intermediate goods to produce final goods became much more active in objecting to import restrictions on their inputs. Unlike previous decades, steel purchasers strongly opposed the steel industry’s demands for import limits, and the same was true for semiconductors. In the case of apparel, retailers spoke for domestic households in trying to keep the price of clothing low. Consumers of imported goods thus became politically active, not household consumers but business consumers whose profitability was at stake. This opposition dramatically changed the politics of trade policy: instead of just being presented with the view of import-competing producers and their workers, members of Congress and administration officials were now forced to confront and reconcile the conflicting views of different constituencies.

Another trade issue left over from the 1980s was the use of section 301 to attack foreign unfair trade practices. The use of this tool to open up foreign markets also began to fade with time, although initially it did not appear that way. As mentioned earlier, the Omnibus Trade and Competitiveness Act of 1988 included a “Super 301” provision in which USTR had to identify trade barriers and policy distortions in “priority countries” and retaliate if the practices were not eliminated within three years. This was Congress’s way of pushing the executive to use high-pressure tactics to open foreign markets, particularly in Japan.119

Bush administration officials were divided about how much to make demands on allies, such as Japan. In her confirmation hearings, USTR Carla Hills promised to take a tough line in addressing unfair trade practices and gave assurances that new negotiations, unlike those in the past, would get results. (Using the analogy of taking a crowbar to pry open foreign markets to American goods, Hills became known as “crowbar Carla.”) At the same time, administration officials did not want a results-oriented trade policy to degenerate into “managed trade,” in which explicit market-share targets would be set, as in the 1986 semiconductor agreement.

In May 1989, USTR named Japan, Brazil, and India as “priority countries” under Super 301. Japan was cited for discriminatory procurement policies with respect to supercomputers and satellites and for technical barriers in forestry products; Brazil was cited for various quantitative import restrictions; and India was cited for barriers on trade-related foreign investment and in services trade, particularly insurance. (South Korea and Taiwan offered enough concessions to stay off the priority list, and the EEC also went unnamed.) A year later, Hills announced that the negotiations with Japan had successfully opened up the market for American supercomputers and forestry products, which were of interest to Senator Max Baucus, an influential Democrat from Montana.

But then Super 301 faded away. To the frustration of Japan critics, USTR did not name it as a priority country in 1990. Instead, the Bush administration launched the Structural Impediments Initiative (SII) to address the market-access problems with Japan without the pressure arising from Super 301.120 In 1990, only India was named as a priority country, but it refused to negotiate with the United States. Recognizing the futility of the situation, the United States did not press the matter by retaliating. After that, Super 301 expired and was not renewed. Bayard and Elliott (1994, 313) concluded that Super 301 was “no more likely to produce results than regular section 301,” but was “unnecessarily inflammatory” in provoking confrontations with other countries.

By the early 1990s, concern about foreign unfair trade practices had diminished. Japan became yesterday’s problem when it entered a prolonged economic slump; in the face of a stronger yen and a weaker economy, Japanese producers were no longer the competitive threat they had been a decade earlier. Yet the new administration of President Bill Clinton initially sought to reinvigorate the push for a “results-oriented” trade policy.121 In March 1993 Japan announced that the foreign share of its semiconductor market had reached 20 percent, although it insisted that figure was neither a target nor a promise. The Clinton administration viewed the semiconductor agreement as a potential model for other sectors. Meeting with Japan’s prime minister that year, President Clinton called for a “focus on specific sectors and specific structures with a view towards getting results” on market access.122 The experience with the semiconductor agreement—namely, the difficulty in getting domestic producers to agree to purchase foreign products—made Japanese officials strongly opposed to ever again accepting numerical benchmarks for imports.123 While the Clinton administration struggled to come up with a framework for the bilateral trade relationship, Japan successfully avoided making any more commitments regarding imports. By 1995, the matter was dropped.

Thus, by the early 1990s, the political and economic environment for trade policy was much calmer than it had been in the early 1970s or the 1980s. The system of open trade had been put under enormous stress during those years, resulting in the proliferation of export-restraint agreements and other forms of trade protectionism. While the pressure to impose new trade barriers had been strong, there was no reversion to the high tariffs seen prior to 1934. Furthermore, the groundwork was laid for further liberalization. As the trade measures of the 1980s began to expire, the United States stood on the brink of a new era of trade expansion.