Commentary | Trade and Globalization

The Aerospace Sector as a National Asset—Viewpoints | EPI

Opinion pieces and speeches by EPI staff and associates.

THIS TESTIMONY WAS GIVEN BEFORE THE COMMISSION ON THE FUTURE OF THE UNITED STATES AEROSPACE INDUSTRY ON MAY 14, 2002.

The Aerospace Sector as a National Asset

by Jeff Faux

The aerospace industry is a unique strategic asset for America. In addition to its obvious national security benefits, the industry makes, and must continue to make, a critical contribution to our economic growth and rising living standards.

U.S. aerospace is a major source of:

  • Technological innovation with substantial spillovers to other industrial and commercial sectors.
  • High wage employment, which spreads the benefits of rising productivity throughout the U.S. economy.
  • Exports, which America will need to substantially increase in order to resolve the growing problem of our current account deficit and rising foreign debt.

Thus, a healthy aerospace industry ought to be a primary goal of our nation’s economic policy. It represents the cumulative private and public investments of past decades. Allowing it to wither is, in effect, a national decision to abandon those investments.

Lose the People, Lose the Industry

We cannot separate the industry from its labor force. Indeed, it is the highly trained technical workers and their tradition and culture of technological innovation that comprise the core value of this asset. People are not commodities that can be discharged this year and immediately replaced “off the shelf” next year. If we lose the people, we lose the industry.

In recent years, we have been losing the people. Between 1989 — the peak of the last business expansion — and 2001, the aerospace industry lost 524,000 jobs. Total employment in the aerospace sector at the end of this period, 2001, was 790,000. This includes workers developing and manufacturing aircraft and engines, guided missiles, space vehicles, component parts, and other aerospace related employment (Table 1).

Table 1
Aerospace Industry Employment, 1982-2000 (thousands)
               
               
Aircraft, Engines, and Parts Guided Missiles, Space Vehicles, & Parts All Other Aerospace Related Employment
           
  Total Aerospace Total Aircraft Engines & Engine Parts Other Aircraft Parts
 
               
1982 1,038 584 320 149 115 131 323
1983 1,019 562 305 140 117 141 317
1984 1,058 575 306 140 129 154 329
1985 1,151 616 326 148 143 177 358
1986 1,241 656 339 154 163 200 386
1987 1,282 678 356 158 163 206 399
1988 1,294 684 369 156 159 208 402
1989 1,314 711 382 154 175 194 408
1990 1,302 712 381 152 180 185 405
1991 1,214 669 356 143 170 168 378
1992 1,100 612 332 127 153 146 342
1993 966 542 301 109 131 124 300
1994 855 482 271 95 115 108 266
1995 796 451 244 93 114 98 248
1996 796 458 243 95 120 90 248
1997 859 501 262 100 138 91 267
1998 896 525 272 103 150 92 279
1999 847 496 254 101 141 87 263
2000 800 465 233 101 131 86 249
2001 790 463 228 101 134 82 246
               
 
Change in Employment, 1995-2001
Number -6 12 -16 8 20 -16 -2
Percent -0.8% 2.7% -6.6% 8.6% 17.5% -16.3% -0.8%
               
Source: EPI Analysis of AIAA (1996, 1997, 2000 and 2002)

These high-wage, high skill jobs are part of the core foundation of the modern U.S. economy. Technologically-intensive aerospace production creates extremely important positive spillover effects for the rest of the economy. For every one aerospace worker, 3.9 other jobs are created. In comparison, every job in the retail trade sector generates only .9 other jobs.

While aerospace workers are losing their jobs, the rest of the U.S. economy is not producing jobs that demand these workers’ high skills. The Bureau of Labor Statistics reports that Aerospace production workers have hourly earnings roughly 1.4 times that of the average hourly earnings for all manufacturing workers and 2.2 times that of hourly earnings for retail trade workers. A 1996 survey of aerospace workers laid off from the Lockheed-Martin plant in Marietta, Georgia and from the Boeing plant in Seattle, Washington, found that less than 17 percent were re-employed in the aerospace industry. Of those workers who found any new employment, nearly half reported earning less than 75 percent of their previous hourly wage.

Inasmuch as the time period in Table 1 compares business cycle peaks, the dramatic reduction in employment cannot be attributed to general business conditions. The major sources of the decline are cuts in defense spending, cyclical fluctuation in the demand for civil aircraft, loss of market share to Airbus, technical change and productivity improvements, and offset arrangements that transfer high technology abroad.

Given the environment of weakening sales, the robust productivity which is a characteristic of the industry produced large cuts in the workforce. It is difficult to separate the various sources of Aerospace’s declining employment, but we can say for sure that increased foreign competition — both in the domestic and overseas markets — accounted for a substantial share of lost jobs. Between 1989 and 2001, the import share of domestic sales rose 8.5 percentage points. If the share of imports had remained constant, U.S. aerospace employment would be at least 15.3 percent higher today: an additional 80,000 jobs.

Falling domestic demand, particularly in the military sector, also reduced U.S. aerospace employment. Total industry revenues, in real terms, declined 12 percent during this business cycle as shown in (Table 2). If demand were unchanged in 2001, relative to 1991, aerospace employment would be 17.5 % higher – another 92,000 additional jobs.

Table 2
Aerospace Industry Revenues (millions of constant 1987 dollars)
  Total Aerospace Total Aircraft Civil Aircraft Military Aircraft (a) Missiles (a) Space (a) Related Products & Services
1979 71,528 41,546> 20,830 20,717 7,524 10,307 12,150
1987 110,008 59,188 15,465 43
,723
10,219 22,266 18,335
1988 112,869 59,986 18,738 41,248 10,118 23,953 18,812
1989 114,250 58,341 20,761 37,579 12,912 23,956 19,042
1990 123,734 65,703 28,786 36,916 13,057 24,352 20,622
1991 124,998 68,149 33,611 34,538 9,847 26,169 20,833
1992 118,555 63,221 34,129 29,092 10,057 25,518 19,760
1993 102,482 54,766 27,551 27,215 7,031 23,604 17,081
1994 90,104 46,983 20,861 26,122 6,164 21,941 15,017
1995 86,572 44,215 19,249 24,965 5,933 21,996 14,429
1996 92,196 47,590 21,207 26,383 6,320 22,920 15,366
1997 102,959 55,402 29,286 26,116 6,289 24,109 17,160
1998 115,168 65,332 38,658 26,673 6,016 24,627 19,195
1999 118,784 68,571 40,905 27,666 6,820 23,596 19,798
2000 109,828 61,662 36,359 25,303 7,109 22,753 18,305
2001p 110,480 61,551 36,887 24,663 7,875 22,642 18,413
2002e 104,572 55,111 28,375 26,736 9,355 22,678 17,429
               
Changes in Constant Dollar Revenues
1979-1990 70% 55% 36% 75% 71% 132% 67%
1990-1995 -30% -32% -33% -32% -54% -9% -30%
1995-2002e 21% 25% 47% 7% 58% 3% 21%
1989-2002e -8.5% -5.5% 36.7% -28.9% -27.5% -5.3% -8.5%
Source: EPI analysis of AIAA (1996, and AIA Statistics)

Productivity growth, broadly defined as output per worker, also probably eliminated significant amounts of aerospace employment between 1989 and 2001. However, the available data do not allow us to separate with confidence the effects of rising productivity from the effects of weakening demand. In the absence of data on capital inputs employed by aerospace producers in this period it is impossible to estimate the effects of total factor productivity growth, as it is usually measured by economists. Even if we make all of the technical assumptions that favor the “productivity” explanation for job decline, it leaves us with a minimum of more than 170,000 jobs lost for other reasons.

The Airbus Challenge

The major, but not the only, source of this challenge has been the expansion of the European aerospace industry reflected in the loss of market shares to Airbus. Direct employment in the European aerospace industry surpassed 429,000 in 2000. It is estimated that the European aerospace industry generates twice this number of jobs in related industries within the aerospace supply chain. Between 1995 and 2000, European direct aerospace employment grew steadily at 2 percent annually, adding 42,000 jobs (Table 3).

Table 3
European Aerospace Sales and Employment
 
 

EU Revenues
(Bil. 2000 Euros)

Employment
(‘000s)
1982 42.1 560.2
1983 43 559.5
1984 45.8 540.2
1985 48.8 558.4
1986 54.3 565.6
1987 56.5 569.4
1988 58 564.9
1989 62.7 563.1
1990 64.3 561.1
1991 60.9 524.7
1992 58.4 480.3
1993 52.2 437.3
1994 49.2 407.5
1995 47.2 386.7
1996 53.7 382
1997 62.9 395.5
1998 66 422.5
1999 67.4 426.7
2000 72.3 429.1
 
% 1995-2000 53.2% 11.0%
Annual growth    
1995-2000 7% 2%
 
Source: European Association of Aerospace Industries.

The pattern of competition from the European aerospace industry can be seen clearly in the employment numbers (Tables 1 and 3). Both EU and U.S. aerospace employment declined in the 1980s and early 90s because of a downswing in demand for civil and military aircraft that bottomed out in 1995 (Table 2). In addition, the European industry was at an early stage in its development, in which one would expect bigger than normal gains in worker productivity and firm efficiency as they progressed up the learning curve. Between the cyclical peak in 1989 and the trough in 1995, U.S. aerospace employment fell 39 percent and European employment fell 31 percent (Tables 1 and 3).

But by the mid-1990s airbus had made serious inroads on U.S. market shares both here and around the world. The result was that between 1995 and 2000 EU aerospace employment rose 11 percent while U.S. employment remained essentially flat, gaining .5 percent, and then in 2001 fell again.

Aerospace Employment Outlook

The future is not bright. It is true that the aging U.S. civilian air fleet will have to be refurbished at some point, but U.S. producers’ share of both world and domestic markets is eroding. Airbus continues to make inroads and the foreign content of domestically produced aircraft is expanding. Both aspects of growing international competition pose threats to future employment in the U.S. aerospace industry.

Table 4 shows the outlook for U.S. aerospace employment through 2010 under three different growth scenarios. The high growth scenario assumes that Boeing’s share of the world market for commercial airplanes will grow while the foreign content of U.S. aerospace products, through outsourcing and offsets, remains constant. Under this scenario, U.S. aerospace employment would add 110,600 jobs between 2002 and 2010. The medium growth scenario assumes that Boeing’s current market share remains unchanged while the foreign content of U.S. output is held constant. Under the medium growth scenario, U.S. aerospace employment will fall 76,300 jobs between 2002 and 2010.

Table 4
Aerospace Employment Outlook
 
  Historical Data Boeing ’00 Market Outlook Teal, constant share, no int’l outsourcing Teal, falling share,w/int’l outsourcing
 
1986 655.8 655.8 655.8 655.8
1987 678 678.0 678.0 678.0
1988 683.5 683.5 683.5 683.5
1989 711 711.0 711.0 711.0
1990 712.3 712.3 712.3 712.3
1991 669.2 669.2 669.2 669.2
1992 611.7 611.7 611.7 611.7
1993 542 542.0 542.0 542.0
1994 481.5 481.5 481.5 481.5
1995 450.5 450.5 450.5 450.5
1996 458.1 458.1 458.1 458.1
1997 500.6 500.6 500.6 500.6
1998 525.1 525.1 525.1 525.1
1999 494.9 494.9 494.9 494.9
2000 458.3 440.8 426.8 402.1
2001   452.7 415.9 373.1
2002   464.9 405.2 345.4
2003   477.5 394.8 319.9
2004   490.4 384.7 294.9
2005   503.7 374.8 271.0
2006   517.3 365.2 247.9
2007   531.3 355.8 226.2
2008   545.7 346.6 205.5
2009   560.4 337.6 185.5
2010   575.6 328.9 166.5
 
Change in Employment
2002-10   110.6 -76.3 -178.9
 
Source: AIAA, Boeing 2000, Teal Group.

For reasons discussed below, the high and medium growth trends seem overly optimistic and ignore the dynamic competitive environment in the aerospace industry. A more credible scenario would recognize the realities of increasing foreign competition and shifts in the composition of demand. The third, low-growth, scenario, therefore represents our current trajectory. It reflects the current trends in Airbus’s growing world market share and the increasing foreign content of U.S. aerospace products. Under this scenario, the U.S. aerospace sector will shed 178,900 jobs between 2002 and 2010

Of the factors affecting our low-growth of estimates of future U.S. aerospace employment, increasing competition in world markets is by far the most significant, accounting for 41.5 percent of the change in employment, or 74,244 lost jobs between 2002 and 2010. Job losses due to import penetration of foreign engines and parts account for an additional 11.5 percent of total job losses. Together, growing foreign competition for U.S. aerospace products explains 53 percent of the expected lost jobs through 2010.

Aerospace Trade, Outsourcing and Offsets

The U.S. aerospace trade surplus peaked at $41 billion in 1998 and has since fallen $15 billion, or 37 percent, (more than 11 percent annually) (Table 5). Between 1989 and 2001 U.S. aerospace imports grew on average twice as fast as aerospace exports. While U.S. aerospace exports have been declining since their high in 1998, imports of foreign made parts and equipment continue to accelerate steadily. These trends in aerospace trade can be explained by three factors.

Table 5
The Relationship of U.S. Aerospace Revenues to Exports, Imports, and the Balance of Trade (Billions of Current Dollars)
             
Year Total Aerospace Revenues Total Aerospace Export Revenues Exports as % of Total Aerospace Total Aerospace Import Revenues Imports as % of Total Aerospace Aerospace Balance of Trade
             
1979 $45.4 $11.7 25.9% $1.6 3.6% $10.1
1989 $120.5 $32.1 6.6% $10.0 8.3% $22.1
1990 $134.4 $39.1 29.1% $11.8 8.8% $27.3
1991 $139.2 $43.8 31.4% $13.0 9.3% $30.8
1992 $138.6 $45.0 32.5% $13.7 9.9% $31.4
1993 $123.2 $39.4 32.0% $12.2 9.9% $27.2
1994 $110.6 $37.4 33.8% $12.4 11.2% $25.0
1995 $107.8 $33.1 30.7% $11.5 10.7% $21.6
1996 $116.8 $40.3 34.6% $13.7 11.8% $26.6
1997 $131.6 $50.4 38.9% $18.1 14.0% $32.3
1998 $148.0 $64.1 43.3% $23.1 16.1% $41.0
1999 $151.1 $62.4 41.3% $25.1 16.6% $37.3
2000 $143.7 $59.0 41.0% $25.7 17.9% $33.3
2001 $169.4 $58.5 34.5% $32.5 19.2% $26.0
             
Source: EPI analysis of Aerospace Facts and Figures (1999, 2000 and 2002)

First, the U.S. aerospace industry is facing increasing competition in global markets from firms like Airbus that have been specifically developed through state supported industrial policies to be national industrial “champions.” From its inception in 1970 through 1990, Airbus received government subsidies in the form of grants and soft loans totaling $26 billion. More recently, the British, French, German and Spanish governments have spent $4 billion in development loans on the Airbus A380, which are to be repaid with future sales revenues. The total cost of developing the A380 is estimated at $10.7 billion.

Over the long run, we can expect that the commercial sector of the aerospace industry will represent a larger share of the total market. The market share of Airbus and other European aircraft producers began to rise sharply in all significant markets after 1992. Measured by the number of aircraft, the European aerospace industry is already reaching parity with the United States. Airbus expects to provide fully 50 percent of world aircraft deliveries in 2003. Looking into the near future, Airbus may be pulling ahead of Boeing. Airbus booked 53 percent of orders for new commercial aircraft in 2001 worth $44.7 billion, while Boeing booked $29.6 billion.

We should also note that currently Airbus is competing with Boeing in markets for small and mid-size aircraft, such as Boeing’s “bread and butter” 737 line. At present, Boeing’s line of 747 jumbo jets enjoy little, if any, competition at present. But Airbus is online to introduce its new A380 super-jumbo in 2006 and has already booked 97 orders. Airbus expects to sell some 1,500 A380s in the next twenty years.

The competition to sell new products into this market is now fierce, and failure to keep pace with the state subsidized Airbus system will be costly. Airlines have shown preferences for amassing fleets from a single supplier in order to realize economies of scale in supply chain management and to bargain for volume pricing. By sticking to one “brand,” airline managers need only deal with one set of replacement parts and its mechanics can specialize in maintaining fewer models of more similar aircraft. This
means today’s winner for market sha

Outsourcing has become more than just a way to get cheaper components and parts. In order to off-load some of the costs of research and development, U.S. firms are entering into joint partnerships with state-subsidized firms in other nations. Under such arrangements, for example, a U.S. engine producer may outsource the design of a sub-component of a new engine to a foreign aerospace company. In return for sharing some of the up-front risks associated with developing new products, the part producer gets the right to produce those engine components. While engines might represent a quarter of the cost of a new aircraft, it is just a one-time cost. But, in order to maintain safety, aircraft engines are continuously overhauled and rebuilt with replacement parts. Thus, in the life of a plane, the demand for parts far outpaces the demand for engines. Indeed, Boeing salespeople have boasted to the French airlines that their planes are more “French” than Airbus’s.

One measure of the impact of import penetration in the U.S. aerospace industry is the ratio of imported engines and parts to total aircraft sales (both commercial and military). As shown in Figure 1, the ratio has more than doubled from 8 percent of production in 1981 to more than 20 percent last year, and has accelerated in the past three years. This chart shows that the foreign content of U.S. aircraft is increasing dramatically.

Figure 1

Third, foreign aerospace sales are increasingly being linked to the practice of offsets. Offsets are arrangements to transfer high-skilled jobs and valuable technologies to other countries in exchange for market access for U.S. aerospace products, and are common in both the commercial and defense aerospace sectors. For example, in exchange for purchasing U.S. aircraft, many countries require U.S. aerospace firms to produce parts of that aircraft in the purchaser’s country. Offsets result in the loss of high-skill U.S. jobs and the loss of cutting-edge technologies. In some cases pose a threat to national security and aviation safety. In time, these transfers will help spawn new entries into the market for parts and components, and eventually into markets for fully integrated aircraft designs and manufactured systems.

China is one of the most aggressive countries in pursuing offset agreements and, with its market potential and minimal labor standards, it has substantial leverage in negotiating these agreements. Examples of how U.S. and European aerospace firms enter into voluntary and mandatory offset agreements with China include:

  • Xian Aircraft Co. builds fins, horizontal stabilizers and doors for the Boeing 737; ribs for the Boeing 747; doors for the Airbus A300, A310, A330 and A340; and carbon fiber fin ribs for the Airbus A320.
  • Chengdu Aircraft Co. makes vertical fins, horizontal stabilizers and tail sections for the Boeing 757, MD-80 and MD-90.
  • Shenyang Aircraft Co. makes cargo doors for the 757; wing ribs and emergency exit doors for the A320; and machine parts for the A300 and A310.
  • Shanghai Aircraft Co. makes components for the 737 and jointly assembles the MD-80 and MD-90.
  • Airbus has donated a $50 million flight simulator for training pilots in Beijing.

Ironically, while much technology and production has been transferred to China, U.S. aerospace exports to China remain quite small — 5.8 percent of total U.S. aerospace exports in 1998, the last year for which we have data. This indicates that the benefits from offsets have been limited, while the costs in terms of job losses and lost technologies are significant.

Offsets also transfer security-sensitive aerospace technology and production to other countries, fostering the proliferation of defense systems and undermining essential sub-tier defense production in the United States. Technology transferred to Brazil through an offset resulted in an improvement of targeting capability of the Iraqi Scud missile system. Under another offset arrangement, McDonnell Douglas sold machine tools to China for use in production of commercial aircraft, but some of these were diverted to the production of Chinese military equipment.

In this case, both U.S. and European aerospace workers are engaged in a “prisoner’s dilemma”: fear of losing market access to each other makes them willing to trade away critical technologies and value-added steps in the production process in return for immediate sales.

Aerospace and the Current Account Deficit

It is clear that the U.S. current account deficit — driving by the chronic excess of imports over exports — is unsustainable. In 2000 the U.S. current account deficit was $450 billion, 4.5% of GDP. The recession reduced the deficit somewhat last year, but it was still above $400 and 4% of GDP. As the economy recovers, the deficit will continue its relentless expansion. As last year’s report of the US Trade Deficit Commission showed, for an equal increase in national income in the US and foreign countries, the United States increases its imports proportionally more than its exports.

The Wall Street firm of Morgan Stanley recently warned of a current account deficit reaching 6 percent of GDP by the end of 2003. As the Economist magazine observed, studies — including one done at the U S Federal Reserve — indicate that when the current account deficit reaches 5 percent, international financiers begin to pull back.

In order to finance this deficit the United States has had to borrow from other countries and sell them more of its assets. Thus, each year its economy must devote more of its income to interest on the debt and the transfer of profits to investors in other countries. After 1988, these payments began to exceed foreigners’ remittances to the United States.

This net foreign “debt” is now 22 percent of GDP. Assuming a recovery, the U.S. economy is on trajectory of a debt of roughly 40 percent of GDP within five years. Nobel prize-winning economists Franco Modigliani and Robert Solow last year characterized the large and growing deficit in the U.S. international trade balance as “the greatest potential danger facing the economy in the years to come.”

The dollar is the world’s most important reserve currency and America has better credit and more assets to sell. But it is a matter of simple arithmetic that it cannot forever borrow in order to buy more from the rest of the world than it sells. The interest burden will eventually be so heavy that foreign investors will be unwilling or unable to keep financing the rising debt. When that happens, the dollar will drop and interest rates will spike upward. The United States will then be forced to run a trade surplus with a drastic devaluation of the dollar and/or a draconian deflation in real incomes in order to reduce demand for imports and make U.S. goods cheap enough to run a surplus in world markets. An overvalued dollar makes it hard for even the most productive U.S. companies to compete in global markets. As an aerospace industry executive recently remarked: “We still probably do less off-shore sourcing than most industries, but with the dollar where it has been the last couple years, there’s even more pressure to look offshore.”

Economists at Goldman-Sachs estimate that cutting the current account deficit in half by devaluation alone would require a more than 40 percent drop in the dollar’s value.

Clearly the preferred way to address this danger is by increasing our exports rather than decreasing our incomes to force Americans to import less.

But as the U.S. industrial base has shrunk, so has our capacity to expand exports in the tradable goods sectors. Moreover, we clearly would prefer to expa
nd the export of high value goods. This makes the aerospace sector crucial if the U.S. is to weather the inevitable adjustment in its current account deficit. Today, rather than watch the aerospace export surplus shrink, we ought to be devising ways to expand it, not for some mercantilist notions of trade, but because we will soon find ourselves rather desperate for industries that can help us export our way out of the problem rather than adjusting through a lowering of living standards.

The more that we hollow out the aerospace sector, the more that we allow its U.S. base to be eroded by other nations’ industrial policies, the more difficult it will be to export our way out of the problem. Both labor and capital need a sense of commitment in order to make the sustained investments of money and training required for a sustained expansion.

Recommendations

Recognizing that every other nation seriously involved in aerospace has a high-level commitment to promoting a healthy industry, a formal inter-agency group on aerospace should be established in the Federal Government. Members of the groups should include at least: the Secretaries of Commerce, Defense, Labor, Transportation, and Treasury and the US Trade Representative.

Among its tasks would be to develop:

  • Policies and procedures to discourage US firms from engaging in offset agreements. At the same time, reopen discussions with the EU on a common front towards developing country offset rules.
  • Programs that could help assure that US manufacturers are positioned to take advantage of the expected wave of new domestic orders for the replenishing of the US commercial aircraft fleet.
  • Strategies for increasing US aerospace exports, beginning with a study of the effect of the high dollar on the competitiveness of the US aerospace sector.
  • Expanded investments in training and workforce development programs to assure that the economic opportunities from an expanded aerospace sector accrue to US workers.
  • Support trade policies that give the U.S. an even playing field and require enforceable labor rights and standards in all international trade and investment agreements.

Conclusion

Anyone of the economic reasons — hi-wages, technological innovation and concerns with our deteriorating balance of payments — would justify public action to stabilize and expand the US aerospace sector. The existence of all three make such action imperative.


Notes

1. This employment figure includes total employment for aircraft, engines, parts, missile and space industries from the BLS plus aerospace-related employment in communications equipment, instruments and other industries calculated with an input-output matrix by the AIAA. Just looking at the BLS data (SIC codes 372 and 376) significantly underestimates aerospace employment (approx. 545,000 in 2001 for these two categories).

2. There is a time lag between aircraft production and final sales that reflects the time and complexity involved in manufacturing aircraft. The lag can be seen in Tables 1 and 2 where employment peaks a year or two before a peak in total revenues. Because of this time lag, it is appropriate to use 1989 as a base year for examining the employment effects of changing imports and 1991 as a base year for examining the employment effects of changing demand.

3. The European and U.S. job to job totals are not strictly comparable. Therefore, the direction of change in both is more reliable than the absolute levels.

[ POSTED TO VIEWPOINTS ON JUNE 13, 2002 ]

Jeff Faux is president of the Economic Policy Institute in Washington, D.C.


See related work on Trade and Globalization