Report | Economic Growth

Trade, jobs, and wages

Issue Brief #244

Download PDF

A wide gulf exists today in American politics. On one shore are voters increasingly anxious about globalization and its effect on their jobs and communities. On the other are economists, policy makers, and pundits who maintain that trade is good for the economy, that the wider public is simply misguided about its benefits, and that politicians who sympathize with those concerned about globalization are pandering to special interests at the expense of the wider economy. This latter group relies heavily on the suggestion that “all economists believe” globalization is good for the vast majority of American workers.
This reliance is odd given that mainstream economics actually argues that there are plenty of reasons for concern about globalization’s effect on the majority of American workers. This primer highlights two issues in particular that should worry American workers about globalization: job losses stemming from growing trade deficits; and downward wage pressure for tens of millions of American workers. These problems are not unexpected consequences of expanded trade; quite the opposite, they are exactly what standard economic reasoning predicts.

Trade and jobs

Job loss is by far the most visible and easily understood way that international trade can affect American living standards. The effect of trade flows on American jobs is actually pretty complicated and so requires a bit of untangling. First, trade creates new jobs in exporting industries and destroys jobs when imports replace the output of domestic firms. Because trade deficits have risen over the past decade, more jobs have been displaced by imports than created by exports.

In a sense, a trade deficit is the difference between a country’s production (exports) and its consumption (imports). Each year that the United States runs a trade deficit is a year that it must borrow from abroad to finance this excess of consumption over production. This borrowing leads to growing foreign debt that must be paid, with interest. In 2007, U.S. borrowing was on the order of $2 billion every day.

Australia provides a cautionary tale on the consequences of such borrowing. In recent years, the Australian trade deficit has averaged around 2% of gross domestic product, yet Australia’s total deficit of international credits over debits reached 6% of GDP. The 4% gap between the trade and total deficit
was debt service (i.e., interest) paid on the borrowing to cover previous years’ accrued trade deficits. This large income flow leaving Australia to pay interest on accumulated foreign debts should be a red flag for the future of the U.S. economy.

There are, however, some possible offsets to this job loss resulting from trade flows. As the trade deficit grows, dollars piled up by our trading partners come back to the U.S. economy, and this increases the supply of funds available for U.S. business and households to borrow. This increase drives down the price of borrowing (interest rates), just as an increase in supply in any other market drives down prices. Lower interest rates spur job growth in interest-sensitive industries (like housing); and these can offset some of the job losses from trade.

Can these jobs created through capital inflows completely balance jobs lost to growing trade deficits? It is possible, but unlikely. Of course, other macroeconomic influences may push an economy to full-employment even in the face of trade deficits. In the late 1990s, for example, manufacturing jobs were lost to trade while construction jobs (at least partially spurred by foreign capital inflows) boomed. In the early 2000s, conversely, manufacturing hemorrhaged jobs due to trade faster than any other industry (even interest-sensitive industries) could replace them.

The Economic Policy Institute and other researchers have examined the job impacts of trade in recent years by netting the job opportunities lost to imports against those gained through exports.1 One criticism of these studies is that they do not try to estimate the jobs gained from capital inflows. However, this criticism misses the point of these studies: estimates of jobs displaced by growing trade deficits are not a declaration of exactly how many more jobs the economy would have today if these deficits had not grown. Rather, they are a conservative measure of the involuntary job displacement caused by these growing deficits and an indicator of imbalance in the U.S. labor market and wider economy. These studies also provide an indicator of how trade has affected the composition of jobs in the U.S. labor market.

Economists may cheerfully label it a wash when the loss of a hundred manufacturing jobs in Ohio or Pennsylvania is offset by the hiring of a hundred construction workers in Phoenix, but in the real world these displacements often result in large income losses and even permanent damage to workers’ earning power.2

Lastly, and importantly, even if trade deficits and capital inflows were to fight to a draw and there was no effect on the total number of jobs, job quality could still suffer. Manufacturing jobs (disproportionately lost to trade) tend to pay more and have better benefits, especially for workers without a four-year degree.

Trade and wages

While job-loss caused by rising trade deficits is the most visible effect of globalization, its impact on wages is a concern to an even much larger number of workers. Even if trade flows begin to balance and there is less job loss in the future, the integration of the U.S. economy with those of its low-wage trading partners will pull down wages for many American workers, and will contribute to the ever rising inequality of incomes in the U.S. economy.

The ongoing dispute over the effects of the North American Free Trade Agreement (NAFTA) on the U.S. economy raises a narrower issue than addressed above: do trade agreements (and not just trade flows) impact American jobs and wages?

As described in this overview, increased trade flows affect jobs and wages in the United States. Given that a key purpose of trade agreements (like NAFTA) is to increase these trade flows—and all evidence indicates that they have succeeded—it is safe to say that trade agreements have indeed increased pressure on American jobs and wages by increasing trade flows.

It is, however, hard to disentangle the precise influence of trade agreements apart from all other

economic influences. Given this difficulty, researchers (and editorialists) frequently compare trade levels and other economic outcomes in periods before and after the implementation of trade agreements to assess their impact. While these “before-and-after” comparisons are assessments of the impact of increased trade generally, not trade agreements alone, this general method of assessing the outcomes of trade agreements is essentially an i
ndustry standard employed by nearly all commentators in the debate over trade agreements.3

While global integration is usually “win-win” between countries, it can still translate into steep losses for tens of millions of workers in the U.S. economy. Crucially, this wage-loss is not restricted to just workers in sectors exposed to trade, but is experienced by all workers who resemble those displaced by imports in terms of education, skills, and experience. Many of these workers probably do not even know that they are being affected by globalization, but they are. Landscapers may not get displaced by imports, but their wages do indeed suffer from job competition with import-displaced apparel workers.

Take the case of China and the United States. Reducing trade barriers allows each to specialize in what they do more efficiently, and this specialization generally leads to national-level gains for both countries—that is, increased efficiency, worldwide production, and total consumption. This is essentially chapter one in trade textbooks.

However, a later chapter in the textbook points out that, when the United States exports financial services and aircraft while importing apparel and electronics, it is implicitly exchanging the services of capital (physical and human) for labor. This exchange bids up capital’s price (profits and high-end salaries) and bids down wages for the broad working and middle-class, leading to rising inequality and wage pressure for many Americans. In the textbook’s index, this is called the Stolper-Samuelson Theorem. (For those more convinced by appeals to authority, the text box Interpreting Wage Impacts provides some quotes from standard economics texts.)

How big is this impact on wages? A reasonably cautious estimate is that between 1973 and 2006, global integration lowered the wages of U.S. workers without a four-year college degree (the large majority of the U.S. workforce) by 4%. College-educated workers saw 3% gains from trade, so inequality increased in this time as well.4

Four percent might not sound like that big a deal, but to put it in some perspective, wages of workers without a college degree rose by only 2% over the entire 1973-2006 period. If not for the effects of trade, then this group’s wage increase could have been 100% larger.

An honest debate on globalization

American workers are perfectly rational to worry about what globalization means for their living standards, and actually have a much better grasp of the underlying economics than do the elite policy making class who routinely tells them otherwise. Furthermore, the globalization status quo is at least as stingy to the poor trading partners of the United States as it is to American workers. It is time we had a national debate that acknowledged these facts and treated views dissenting from the elite consensus on globalization with the respect they deserve. This debate needs to include responses to globalization that match the scale of the economic insecurity, the wage losses, and the re-distribution it leaves in its wake. Simply put, this scale is not appreciated or acknowledged in today’s globalization debate, and policy responses reflect this failure.

The first thing to note is that the losses described above are not the unemployment spells suffered by

workers displaced by imports. These unemployment costs are not even considered in most trade theory, although in the real world they obviously should be. Rather, the biggest losses are the permanent wage cuts resulting from America’s new pattern of specialization made possible by globalization. These wage losses, it should be reiterated, are suffered by all workers who resemble import-displaced workers in education, skills, and experience.

Second, the wage losses discussed in this overview factor in the ability of all workers to buy cheaper

imports or find new job opportunities in expanding export sectors. Too often even professional economists imply or even state outright that cheaper imports or expanding opportunity in export sectors make the net outcomes of globalization for American workers impossible to predict. This is wrong.

Third, the channels described above are, of course, not the only way trade affects U.S wages. Just

the threat of substituting foreign labor and imports for U.S. workers (made more credible as global

integration proceeds) reduces the bargaining power of U.S. workers—even of high-wage, high-education workers who are generally helped by the effects described above (e.g., college-educated

accountants buying cheap imported shirts at Wal-Mart). These threat effects are all but impossible to

measure, but are nevertheless important.

Finally, for those more convinced by appeals to authority on the issue of trade and wages, below are

two quotations, one from Kenneth Rogoff, economics professor at Harvard and former chief economist

for the International Monetary Fund (IMF), and another from a standard undergraduate international

trade textbook authored by Paul Krugman and Maurice Obtsfeld:

“From a policy perspective, the major result of [the SST] was to confirm the intuitive analysis of Ohlin about who wins and who loses when a country opens up to trade. The answer, as we now well understand, is that the relatively abundant factor gains, and the relatively scarce factor loses, not only in absolute terms but in real terms. Thus if capital is the relatively abundant factor (compared to the trading partner), then an opening of trade will lead the return on capital to rise more than proportionately compared to the price of either good, whereas the wage rate will fall relative to the price of either good.” 5

“….International trade has a powerful effect on income distribution….This means that international trade tends to make low-skilled workers in the United States worse off —not just temporarily but on a sustained basis.” 6


[1] The latest such report from EPI is: Scott, R. Bruce Campbell, Carlos Salas, and Jeff Faux (2007), Revisiting NAFTA: Still not working for North America’s workers. Economic Policy Institute Briefing Paper #173. Other reports using the all-but-identical methodology include: Groshen, Erica, Bart Hobijn, and Margaret M. McConnell (2005); U.S. Jobs Gained and Lost through Trade: A Net Measure, Current Issues in Economics and Finance, Federal Reserve Bank of New York; and, Bailey, Martin N. and Robert Z. Lawrence (2004), What Happened to the Great U.S. Jobs Machine: The Role of Trade and Electronic Offshoring, Brookings Papers on Economic Activity, Volume (2). Further, it should be noted that pundits use this implicit logic of counting jobs embodied in trade flows all the time. The April 10 editorial of the Washington Post argued for passage of the U.S./Colombia Free Trade Agreement partly on the basis of jobs created in the U.S. through exports to Colombia: “The trade agreement would…give U.S. fi rms free access to Colombia for the first time, thus creating U.S. jobs.”

[2] In fac
t, one study (Philip Oreopolous, Marianne Page, and Ann Huff Stevens (2005), The Intergenerational Effect of Worker Displacement. NBER Working Paper No. 11587) has actually shown that involuntary job displacement leads to lower lifetime income for the displaced worker’s children. Involuntary job-loss, in short, is costly to workers in the real-world.

[3] EPI, for example, is careful to identify just what is being measured. For example, the EPI report referenced above (Scott et al. (2007) notes that “Growing trade deficits with Mexico and Canada have displaced production that supported 1,015,291 U.S. jobs since NAFTA took effect in 1994” [emphasis added].

Some recent examples of this “before and after” assessment of NAFTA’s effect from pro-NAFTA sources follows: Trade Distortions, Washington Post Editorial, 12/3/2007, “…[The impact of NAFTA seems to have been both larger and more positive in Mexico…. Mexico’s gross domestic product…more than quadrupled since 1987.” (It should be noted that this particular “before and after” snapshot is wrong in almost every way: Mexican GDP has not quadrupled since 1987, and NAFTA took effect in 1994, not 1987.

NAFTA – Myth vs. Facts, Office of the United States Trade Representative, March 2008, “Myth #2: NAFTA has cost the U.S. jobs…Fact: U.S. employment rose from 110.8 million people in 1993 to 137.6 million in 2007.”

[4] For this number, see Bivens, L. Josh (2007), Globalization and American Wages: Today and Tomorrow. Briefing Paper, Economic Policy Institute, Washington, D.C.

[5] Rogoff , Kenneth (2005), “Paul Samuelson’s Contributions to International Economics,” chapter in volume edited by Szenber in honor of Paul Samuelson’s 90th birthday.

[6] Krugman and Obstfeld (1994), International Economics: Theory and Policy. 3rd Edition. Harper-Collins.

See related work on Economic Growth

See more work by Josh Bivens