Testing the “China Shock”: Was Normalizing Trade with China a Mistake?

  • Introduction
  • U.S.–Chinese Trade: Amplified Costs
  • The Reality of China’s WTO Accession
  • China’s Backsliding since Accession
  • The Curious Failure to Note the Problems
  • Trade and Globalization Bring Unambiguous Benefits
  • Protectionism Keeps Failing
  • Why Did Some Places “Move On”?
  • Conclusion
  • Citation
  • Downloads
  • Endnotes
  • Related Content

There is an emerging consensus among American politicians and many citizens that trade and globalization have undermined America’s working class, resulting in a rise in U.S. populism. This view frequently targets the 2000 U.S. law that granted China “permanent normal trade relations” (PNTR) and China’s 2001 entry into the World Trade Organization (WTO) as key drivers of the country’s rise and the now‐​famous “China Shock”—the period between 1999 and 2011 during which a sizeable increase in Chinese imports supposedly produced the loss of approximately 2.4 million U.S. jobs.

However, the view that PNTR was an erroneous policy choice that disproportionately benefited political elites and corporations, directly drove the China Shock, and, combined with other allegedly “laissez‐​faire” policies, permanently scarred America’s working class suffers from several flaws that collectively prove fatal for the anti‐​PNTR thesis.

As we approach the 20th anniversary of PNTR, criticism of the law and of the WTO more broadly will surely intensify, but a proper accounting of the relevant economics and history reveals most critics to be misguided. Labor market and cultural disruptions in the United States are real and important, as is China’s current and unfortunate turn toward illiberalism and imperialism. But it is a mistake to pretend that there was a better trade policy choice in 2000 than PNTR and engagement with China more broadly. It assumes too much, ignores too much, and demands too much. Worse, it could lead to truly bad governance: increasing U.S. protectionism; forgiving the real and important failures of our policymakers, CEOs, and unions over the past two decades; and preventing a political consensus for real policy solutions. Indeed, these are happening now.


Since Donald Trump’s surprising presidential victory in 2016, both conservatives and progressives have debated whether and to what extent “Washington elite” policy choices, in particular international trade liberalization, have systematically (and perhaps nefariously) harmed members of America’s working class, dooming them to lives of drug abuse, isolation, and despair and creating fertile ground for populists like Trump. In this increasingly popular view lies a nugget of truth: Americans today face serious and relatively new problems when forced to adjust to severe economic disruptions, whether those disruptions come from trade, technology, culture, or anything else. These problems are often caused or exacerbated by outmoded government policies in need of reform.1

However, champions of the emerging consensus that trade liberalization was a mistake err when targeting U.S. trade with China for particular scorn.2 The policy choice most commonly criticized in this regard is the 2000 U.S. law to grant China “permanent normal trade relations” (PNTR) and the country’s subsequent entry into the World Trade Organization (WTO) in 2001.3 These two events are considered key drivers of the now‐​famous “China Shock,” the period between 1999 and 2011 during which a sizeable increase in Chinese imports caused, according to economists David Autor, David Dorn, and Gordon Hanson, the loss of approximately 2.4 million U.S. jobs.4 A related analysis by Justin Pierce and Peter Schott specifically targets PNTR as the China Shock’s root cause, alleging the policy caused concentrated job losses between 2001 and 2007 in U.S. industries most exposed to Chinese import competition.5

Armed with these studies, it has become fashionable, especially on the political right, to blame PNTR and China’s WTO accession for the country’s economic rise and unfortunate recent turn toward illiberalism.6

However, the view that PNTR was an erro­neous policy choice that disproportionately benefited political elites and corporations, directly drove the China Shock, and, combined with other “laissez‐​faire” policies, permanently scarred America’s working class suffers from several flaws that collectively prove fatal for the anti‐​PNTR thesis. As PNTR approaches its 20th anniversary and as U.S.–Chinese rela­tions have deteriorated during the Trump era, a proper accounting of the economic and historical record is essential. This paper summarizes the flaws in the conventional wisdom on the China Shock. It finds that PNTR and trade with China are generally more benign—and far more complicated—than the story that PNTR critics now repeat.

U.S.–Chinese Trade: Amplified Costs and Ignored Benefits

Perhaps the simplest, yet most substantial, flaw in the PNTR thesis is that it ignores the documented benefits of increased U.S. trade with China over the past two decades—benefits that often accrued to the U.S. working class and manufacturing sector. For starters, even if one were to treat the China Shock literature as gospel, studies have found that trade with China in the 2000s also provided ample benefits for American consumers—a group that includes those directly harmed by the shock.

Economists Xavier Jaravel and Erick Sager, for example, found that Chinese import competition between 2000 and 2007—the peak of the “China Shock”—had substantial “pro‐​competitive effects” on U.S. firms and generated over $202 billion in consumer benefits via lower prices. That equals $101,250 in benefits to U.S. consumers per manufacturing job lost, as calculated by the China Shock papers.7 The following year, Liang Bai and Sebastian Stumpner concluded in the American Economic Review that Chinese imports “significantly reduced inflation,” cutting the price index for consumer goods by 0.19 percentage points per year between 2004 and 2015 as a result of both changes in the prices of existing goods and the entry of new goods—signaling strong pro‐​competitive effects and improved variety.8 A study by Mary Amiti and others found similarly impressive consumer gains,9 while Christian Broda and John Romalis found that the consumer benefits of trade, already tilted toward America’s poor and middle class, were even more so for Chinese imports because those consumers frequently shop at places that carry such goods, such as Target and Walmart.10 One can argue that those consumer benefits are cold comfort to someone who lost a job because of Chinese import competition, but they are nevertheless real, widespread, and important.11

Chinese imports have also been found to generate substantial benefits for American companies, including manufacturers and their workers. Germán Gutiérrez and Thomas Philippon, for example, found that Chinese import competition encouraged many American manufacturing firms to invest and innovate more—another “pro‐​competitive” effect.12 Using a general equilibrium model, Lorenzo Caliendo, Maximiliano Dvorkin, and Fernando Parro found net welfare benefits from the China Shock for U.S. manufacturing and nonmanufacturing firms across regions.13 Zhi Wang and others, after accounting for manufacturing supply chains and intermediate inputs, found that the overall effect of the China Shock on American jobs and wages has been quite positive.14 Simon Galle, Andrés Rodríguez‐​Clare, and Moises Yi found that while the China Shock produced losses for certain groups of Americans, it generated overall gains in social welfare.15

Meanwhile, researchers with the Federal Reserve Bank of San Francisco have estimated that about 56 cents of every dollar that Americans spent on “Made in China” imports in 2018 actually went to American firms and workers—the highest share of any country.16 Such benefits make sense: 2019 U.S. labor market data show millions more “blue collar” American jobs that might benefit from Chinese imports—in transportation, logistics, construction, and maintenance and repair, for example—than in manufacturing.17

Furthermore, this already benign assessment assumes that Chinese import competition potentially hurts all U.S. manufacturing jobs. That assumption is proven incorrect by the San Francisco Fed study, which found that one‐​third of all Chinese imports were intermediate goods that American companies used to produce globally competitive products. (Hundreds of manufacturing jobs at a Missouri custom hat company, for example, are threatened by President Trump’s tariffs on imported Chinese baseball caps.18) These imports have helped, not hurt, U.S. manufacturing workers. In fact, Pol Antràs, Teresa C. Fort, and Felix Tintelnot found that U.S. manufacturing firms that increased direct imports from China between 1997 and 2007 experienced growing or steady employment, likely because of the importers’ ability to lower prices and raise output (even as nonimporting competitors suffered).19 With respect to these types of complex value chains, the WTO estimates that China in 2015 was the third‐​largest user—behind only Mexico and Canada—of “Made In America” manufacturing inputs and the largest source of inputs for American manufacturers.20

Then there are the benefits that American farmers and workers have derived from exporting to China, still the United States’ third‐​largest export destination.21 According to the US‐​China Business Council, exports to China in 2019 supported over 1.1 million American jobs in a wide range of manufacturing, logistics, and services industries.22

Beyond the benefits of trade with China, a proper accounting of the China Shock also requires proper context. There is evidence, for example, that many U.S. manufacturers adapted during the shock and ended up hiring many Americans and increasing output. Summarizing a 2018 paper from Teresa C. Fort, Justin R. Pierce, and Peter K. Schott,23 the Financial Times’ Gillian Tett notes:

Between 1977 and 2012, the number of “manufacturing firm workers” employed in “manufacturing plants” halved from just under 20m to nearer 10m. However, the employees in “non‐​manufacturing plants” that were owned by “manufacturing firms” rose from 13m to 23m, primarily due to an explosion in service sector jobs such as design and IT. As a result, by 2012 the US’s “manufacturing” companies employed slightly more workers than in 1977. Moreover, that was not because of business churn: 75 per cent of the “manufacturing” job losses in this period occurred at companies which remained in business, and it was the incumbents which opened most of the non‐​manufacturing plants. In plain English, this means that as Chinese competition hit, America’s “manufacturing” groups quietly re‐​engineered themselves. Yes, they might call themselves “manufacturers”, and be defined that way in the data. But they increasingly hire service‐​sector workers, as their output soars.24

Nicholas Bloom and others found a similar trend among U.S. workers in “high human‐​capital areas,” such as the West Coast or New England, where manufacturers “remained open but changed to research, design, management or wholesale.”25 Low human‐​capital areas, by contrast, lost jobs on net—a regional discrepancy that might indict policies that help Americans gain skills or cope with disruption but not the disruption itself.

The evolution of American manufacturing—driven by trade, automation, or other factors—raises further concerns about attempting to isolate the effects of Chinese import competition on low‐​skill American manufacturing employment. Kerwin Charles, Erik Hurst, and Mariel Schwartz, for example, found that the decline in manufacturing employ­ment during the 2000s was a substantial cause of rising American unemployment, especially for less‐​educated prime‐​age workers.26 However, they also found that a mix of both import competition and nontrade factors caused these declines. They show that “manufacturing employment declined substantially over the 2000s, even in markets where there was essentially no manufacturing loss because of Chinese imports” and that “shocks to manufacturing that were unrelated to China or trade (including, presumably, things like rising automation) had very similar effects on local labor markets to the Chinese import shock.” As a result, they conclude that “policy efforts to address the adverse labor market effects of trade will not reverse the broader trend in manufacturing employment that has significantly weakened labor market options, particularly for less educated workers.” They further speculate that persistently depressed low‐​skill manufacturing employment in the United States was likely caused by nontrade issues such as a skills mismatch in the U.S. manufacturing sector (which is becoming more skilled compared to other low‐​skill professions such as retail and construction) and declining cross‐​region mobility among U.S. workers during the 2000s compared to earlier periods. As a result, “imposing trade barriers against the rest of the world is unlikely to substantially increase the employment prospects of workers with lower levels of accumulated schooling.”

Studies have similarly found it difficult to distinguish the employment effects of trade from those of technology. After documenting the evolution of American manufacturers in their aforementioned paper, for example, Fort, Pierce, and Schott acknowledge that the “data provide support for both trade‐ and technology‐​based explanations of the overall decline of [manufacturing] employment over this period, while also highlighting the difficulties of estimating an overall contribution for each mechanism.”27

Katherine Eriksson and others provide addi­tional China Shock context. They show that the China Shock was so “shocking” not because of China or PNTR but because of when it hit the United States: during regional shifts in the U.S. production of certain goods.28 In particular, “late stage” industries—with now‐​standardized processes and technologies that are susceptible to global competition (particularly in developing countries)—had moved out of higher education/​innovation U.S. regions to places with less education and innovative capacity, thus explaining “why the shock hurt in these areas to the extent that it did.” This timing adds to the uniqueness of the China Shock, as the authors find that previous U.S. trade shocks—involving Japan and the Asian Tigers, for example—had no such dynamic (and thus labor market effects that were far more limited). The analysis also shows that these “late stage” industries were well on their way out of the United States regardless of the China Shock.

Many other experts have questioned whether the China Shock literature tells the whole story about Chinese imports, U.S. manufacturing jobs, and related issues. As noted, numerous economists have found substantial net benefits for the United States when more fully accounting (e.g., through a general equilibrium model) for Chinese import competition. The Caliendo, Dvorkin, and Parro model further shows far fewer manufacturing job losses caused by the China Shock (only 15 percent of the observed decline between 2000 and 2007).29 Similarly, a pair of papers by lead author Robert Feenstra found offsetting job gains in U.S. manufacturing exports and services,30 while Brad DeLong estimated that China’s WTO entry resulted in a net loss of only 300,000 U.S. jobs—just 0.22 percent of nonfarm employment.31 Adam Jakubik and Victor Stolzenburg found one‐​third fewer manufacturing job losses and much different regional effects when using value‐​added, instead of gross, trade flows to measure the China Shock (and that the job losses basically ended in 2008),32 while Yuan Xu, Hong Ma, and Feenstra found 20–30 percent fewer job losses when accounting for booms and busts in the U.S. housing market.33

Other experts have voiced skepticism regarding the China Shock findings themselves34 (including on cultural effects35). The Cato Institute’s Alan Reynolds notes, for example, that the China Shock’s “microeconomic model designed for local ‘commuting zones’ cannot properly be extended to the entire national economy” and therefore misses important macroeconomic effects of U.S.-Chinese trade liberalization such as increased U.S. exports (to China and other countries).36 Reynolds adds that extending the period beyond 2011, during which the U.S. economy was still affected by the Great Recession, causes half of the job loss attributed to the China Shock to “disappear.” Charles Freeman, who ran the Office of the U.S. Trade Representative’s Office of China Affairs during the George W. Bush admin­istration, recalls:

Among the things that has troubled me about the Autor study is the lack of correlation between [Harmonized Tariff Schedule] level imports from China and US job losses in those sectors. We were deeply attuned to those losses at [the Office of the U.S. Trade Representative] at that time because we had such a powerful tool in the special safeguard in section 421. We just didn’t see any profound direct US job losses in sectors exposed to new direct competition from China. Most of the post PNTR surge in China imports was in sectors that had already shifted overseas. The small blip in accel­eration of manufacturing job losses was actually far below anything any of us could have anticipated. We were ready to be protectionist, but the numbers never justified it. We actually had the [International Trade Commission] prepped to do a study showing the lack of linkage between what were primarily productivity‐​related manufacturing job losses and China trade policy but a political decision was made to blame China rather than domestic [economic] realities.37

Phil Levy, a member of the George W. Bush administration’s Council of Economic Advisers, adds that the fungibility of Chinese and other developing country imports undermines the argument that Chinese imports—as opposed to imports more generally—were to blame for some of the manufacturing job losses that occurred during the China Shock period. The proof came in Levy’s personal examination of domestic industry petitions for relief from Chinese imports under the Section 421 special safeguard mechanism:

In each of the two Section 421 cases I heard, the importers made credible presentations that, were tariffs to be imposed, they would switch their sourcing from China to Vietnam, or to India, or Brazil. In one case, the factory move was estimated to take three weeks. In another, contingent contracts were already in place. Producing in those places cost a bit more than in China, which is why they weren’t the original sourcing countries, but they were cheaper than the United States. So what benefit would U.S. workers have seen in blocking China trade? None. That’s why we recommended against imposing tariffs.38

Levy concludes from this experience that it “calls into question the premise of [the China Shock] analysis. If the alternative to imports from China was imports from other developing nations, then the impact of China on U.S. workers was negligible.”

The data tend to corroborate Freeman’s and Levy’s claims. First, Figure 1 shows only a modest change in trend for manufacturing jobs as a share of the U.S. workforce before and after PNTR passed and China entered the WTO:39

Second, data indicate, pace Levy, that Chinese imports simply replaced other imports (as opposed to domestic production) during the China Shock. According to the Congressional Research Service, the total share of imports into the United States from Pacific Rim countries between 1990 and 2017 remained constant at 47.1 percent, but “the role of China as a supplier of U.S. manufactured products among Pacific Rim countries increased sharply, while the relative importance of the rest of the Pacific Rim (excluding China) for these products sharply decreased,” (see Figure 2) a result “partly due to many multinational firms shifting their export‐​oriented manufacturing facilities from other countries to China.”40

The San Francisco Federal Reserve also found that Americans’ total import consumption, as measured by 2017 personal consumption expenditures, remained relatively steady during the China Shock period. This further signifies that Chinese imports displaced other imports far more than American production. According to the report, “the fact that the overall import content of U.S. consumer goods has remained relatively constant while the Chinese share has increased demonstrates that Chinese gains have come, in large part, at the cost of other exporters, namely Japan.41

That economists repeatedly and openly express reservations—supported by various trade and employment data—about blaming China trade for massive declines in U.S. manufacturing employment should foment similar levels of caution among U.S. politicians and pundits.

Finally, there is the matter of putting the China Shock’s effects into perspective. For example, Douglas Irwin (citing a 2014 Robert Lawrence paper42) notes that “imports from China may have resulted in involuntary displacement of 97,000 manufacturing workers per year (on average, adjusted to account for voluntary separations), but that is less than one‐​fifth of total involuntary job loss in manufacturing and less than 5 percent of all involuntary job losses over the same period.”43 As previously noted, DeLong estimates that the China Shock resulted in the loss of less than 0.25 percent of all U.S. nonfarm jobs. Autor himself has called his estimate of 2 million jobs lost an “upper bound” (the more likely central estimate was about half that number), and it includes around 1 million nonmanufacturing jobs. Autor’s more recent paper on China trade and U.S. marriage trends, moreover, acknowledges that the “analysis does not imply that surging import competition from China over the last two decades has been the sole or primary driver of these [marriage and childbirth] trends” but only a “plausible contributor.”44 These analyses should make us skeptical of the claimed benefits of recent proposals for government to remake the U.S. economy because of the China Shock.

That said, the numerous academic studies discussed above are not intended to argue that Chinese import competition in the decade following China’s WTO accession was purely beneficial to the United States or that the U.S. labor market and certain communities are problem free. Instead, they reveal that the claims of harm from Chinese trade are likely wildly overstated while the substantial economic benefits are usually ignored. These studies also reveal that the China Shock issues are more uncertain and complex than the caricature painted by PNTR/​China critics.45

The Reality of China’s WTO Accession and Export Competitiveness

Critics also often distort the circumstances of China’s WTO accession and the effects of PNTR. First, PNTR did not actually open the United States to Chinese imports: China had previously held “most favored nation” (MFN) trade status, renewed on an annual basis, since 1980, meaning the country faced no greater trade barriers than most other (“favored”) U.S. trading partners. MFN status was even renewed right after the Tiananmen Square protests and the presidential election of Bill Clinton, who ran against MFN, which was subsequently renamed “normal trade relations” (NTR).46 Only once between 1990 and 2001 was China’s MFN/NTR status truly in doubt: in 1992, when a presidential veto was needed to maintain it. As a result, Chinese imports to the United States increased more than six‐​fold in the decade preceding PNTR, and by the late 1990s the rational expectation of most U.S. importers was more of the same. Indeed, a 1998 Congressional Research Service analysis of congressional votes and the broader annual MFN/NTR renewal debate concluded that, by the late 1990s, MFN/NTR was “a largely settled issue” in Congress:

In 1993, newly elected President Clinton announced he would link China’s MFN status to human rights progress beginning in 1994. Although ultimately the President reversed himself, the 1993 decision appears to have been a pivotal catalyst in the declining importance of MFN status as a tool with which to influence China policy. Neither the House nor the Senate has passed MFN‐​related legislation during the Clinton Administration. Instead, Members have turned to legislative alternatives, most of which have included more specific, more targeted sanctions on China’s activities.47

The run‐​up to the PNTR vote in 2000 permits the same conclusion. In 1999, the House vote to deny MFN/NTR for China was defeated by a 170–260 margin; the Senate vote was an even more lopsided 12–87. As former Office of the U.S. Trade Representative staffer Erin Ennis recalls: “I was part of the Clinton administration’s annual efforts to ensure that MFN was continued each year. We never took it for granted and contacted every House office each time a vote was in order, but the outcome was rarely in doubt—particularly since there was only one vote in the Senate during that time and it failed by a wide margin.”48 Table 1 lists these votes.

Nevertheless, there is evidence that the certainty of “permanent” trade relations accelerated the growth of Chinese imports into the United States. The most well‐​known paper on the effect of PNTR’s certainty on Chinese imports and U.S. manufacturing jobs, from Pierce and Schott, found a substantial connection among PNTR, Chinese imports in sectors that would have faced high tariffs in the absence of MFN/NTR, and U.S. jobs.49 Kyle Handley and Nuno Limão found similar results (along with substantial consumer gains) in their 2017 paper.50 Other experts, however, question the magnitude of the PNTR “uncertainty driver.” For example, George Allesandria, Shafaat Khan, and Armen Khederlarian in 2019 found that the annual MFN/NTR votes actually increased Chinese imports into the United States as a result of importers’ increasing shipments in advance of any potential tariff increases.51 They also found, consistent with the aforementioned congressional and anecdotal evidence, that the probability of NTR denial averaged only about 5.5 percent between 1990 and 2001, reaching a mere 1.4 percent in 2001 right before China joined the WTO. Based on these data, they found the trade‐​dampening effects of MFN/NTR uncertainty to have evaporated by the late 1990s.

Regardless of which expert is correct, the congressional record and Chinese trade flow data contradict the popular assertion that an isolated U.S. policy choice in 2000 first exposed the U.S. market and U.S. workers to Chinese import competition. At most, PNTR merely accelerated a bilateral economic integration that was already well underway.

More importantly, there is ample evidence that PNTR was not the only, and perhaps not even the main, driver of the China Shock that occurred in United States. Handley and Limão, for example, found that a reduction in trade policy uncertainty accounted for only about one‐​third of the growth in Chinese exports to the United States between 2000 and 2005.52 Amiti and others found similar results, attributing approximately two‐​thirds of the effect of China’s WTO entry on U.S. manufacturing not to PNTR but to China’s own tariff reductions resulting from WTO entry.53 As shown in Figures 3 and 4, average Chinese import tariffs went from about 15 percent in 2000 to less than 9 percent in 2006, and even lower on a trade‐​weighted scale.

The Autor, Dorn, and Hanson China Shock papers even emphasize that China’s internal reforms—on privatization, trading rights, and (again) import liberalization, often in response to new WTO commitments—were major contributors to China’s export competitiveness in the late 1990s and 2000s.54 Jakubik and Stolzenburg subsequently confirmed this view.55 Several papers have shown significant, though often different, effects of Chinese import competition on firms and workers across Europe, which was obviously not affected by PNTR (and had granted China MFN status years earlier).56 In other words, PNTR probably accelerated Chinese exports to the United States, but China’s own reforms—far beyond the control of Washington policymakers—also fueled the China Shock.

Furthermore, China’s WTO accession was not “shocking” for anyone paying attention to U.S. trade policy in the 1990s (a group that presumably included U.S. manufacturers, unions, and politicians). China first applied to join the WTO (under its predecessor, the General Agreement on Tariffs and Trade) in 1985, then reapplied in 1995 when the WTO came into being, and finally acceded to the body in 2001.57 As shown in Table 2 and Table 3, China’s accession over this time involved dozens of bilateral and multilateral (“working party”) meetings, negotiating texts, disclosures, and—as previously noted—internal reforms. China’s final accession package—a “Working Party Report” and “Protocol of Accession,” plus liberalization schedules for goods and services—contained hundreds of pages of commitments (by far the most of any acceding member to that point and considered still today to be some of the deepest ever). This included many “WTO‐​plus” commitments that the United States and other members dictated (via bilateral accession agreements) and have since been used, for example, to challenge Chinese laws through dispute settlement or to restrict Chinese imports.58

Notably, the United States was the final holdout among large industrialized nations to approve China’s WTO accession via bilateral negotiations, demanding ever more concessions from the Chinese government over a contentious 13‐​year negotiation.59 Contrary to allegations from President Trump and others, the United States did not simply “rubber‐​stamp” China’s WTO accession or base it on Pollyannaish dreams of Chinese democratization. In an interview about his book Schism: China, America, and the Fracturing of the Global Trading System, journalist Paul Blustein describes his research into the U.S.–Chinese bilateral accession negotiations:

I did a lot of research on the negotiations leading to China’s entry into the WTO, interviewing many of the key players on both sides. I found that both sides played extreme hardball—if anything, it was the Chinese who felt bruised and humiliated by the way the talks were conducted. The Americans … were usually the ones to walk away from the table.… China had to agree to open its economy and reform in ways that exceeded the requirements imposed on other countries. For example, China had to promise that it would reduce its tariffs on [manufactured] goods to an avg of about 9% in 2005. China had to lower its tariffs to less than 1/3 the comparable figures for Brazil & other comparable countries. China also had to agree that its trading partners could use several unusual mechanisms that could restrict the inflow of Chinese products. All in all, it’s hard to imagine how the US could have driven a harder bargain on [economic] issues and still gotten a deal. Chinese officials are resentful to this day; they feel China was forced to accept 2nd class citizenship on a [number] of issues.60

Beyond driving a hard bargain, U.S. trade representatives for multiple presidents from each major party also frequently consulted with Congress and the private sector, including labor unions, at every step (as required by U.S. law).61

With respect to the supposed U.S. dream of Chinese democratization, the Paulson Institute’s Neil Thomas has shown that creating a liberal democracy in China was not a primary reason for the U.S. government’s approval of China’s WTO accession. Instead, key Clinton administration speeches and policy documents demonstrate that U.S.–Chinese engagement “was a balancing act with multiple objectives”—most of them pragmatic—including “increasing bilateral dialogues, preventing [weapons of mass destruction] nonproliferation in East and South Asia, preventing the nuclearization of the Korean Peninsula, cooperating on disease and environmental issues, better market access [for U.S. companies] and intellectual property rules, fighting organized crime, ensuring stability in the Taiwan Strait, and WTO accession on ‘commercial terms,’ among others.” Democratization, on the other hand, was mentioned rarely. Thomas shows that the Clinton administration’s engagement policy with China was “neither a triumphant celebration of inevitable democratization nor a credulous declaration of China subsuming itself to American leadership,” and instead comported with “the dominant argument used by PNTR advocates to sway legislators” (i.e., that “engagement with China was not primarily aimed at changing China, but rather focused on benefitting America”).62

Thus, from a simple legal and historical perspective, there was nothing really “shocking” or “naïve” about PNTR and the China Shock.

It is also a stretch to assert that based on the facts at the time that Washington policymakers had much of a choice when deciding whether to grant PNTR to China (a move that every other WTO member had done years earlier). As Levy wrote in 2018, the two alternatives to PNTR—letting China in the WTO but continuing the annual NTR process (or even raising tariffs on Chinese goods) or keeping China from the WTO entirely—were inferior, in terms of both the economics and geopolitics, to granting PNTR:

A policy of denying MFN … would have forsaken the benefits of Chinese membership while having retained all the costs that accompanied low barriers toward Chinese goods. Further, this move would have divided the international community on China, given most [Organisation for Economic Co‐​operation and Development] countries supported its accession at the time. This split would have dramatically weakened the WTO in its early stages, thus undermining a major U.S. foreign policy goal to strengthen the global trading system. …

[Raising tariffs] would have not only hurt U.S. consumers and businesses that benefited from those imports, but would have also been interpreted as an act of enmity by Beijing. And on top of this, it would likely have been ineffective in stopping China’s rise. As China drove down the prices of toys and t‐​shirts in other global markets, it would have been very difficult for the United States to insulate itself from the effects. Further, China has ultimately emerged as a major global economic player by tapping into global value chains. Since China is the last stage in the chain, a finished product can appear to have come from China, even if Chinese value‐​added is relatively small. Since U.S. tariffs are applied based on where a good is finished, not based on value‐​added, China could have easily affected U.S. markets by performing earlier‐​stage tasks and then having the goods finished in Malaysia or some other neighboring country. This is the problem with conducting bilateral policy in a multilateral world. In sum, this second alternative is no better than the first, and decidedly worse than the current policy.

[Excluding China from the WTO entirely] appears dangerous, implausible, and infeasible: dangerous because trying to isolate China with the open intent of blocking Chinese growth would likely have elicited a hostile response; implausible because the United States was, in late 2001, trying to rally the world to respond to terrorism emerging from the Middle East; and infeasible because the United States has had a difficult time trying to isolate countries with much smaller economies, such as Iran and North Korea. Trying to isolate China would have been orders of magnitude more difficult.63

The Iran and North Korea examples are especially relevant to the current debate, given today’s foreign policy justifications for opposing PNTR. Neither country is a WTO member, but each has continued to militarize and progress toward nuclearization, even in the face of economic sanctions and isolation that would never have been applied to 1990s China (or to China today for that matter). Perhaps more importantly, China—unlike those rogue regimes—at the time of WTO accession had possessed nuclear weapons for decades. Though its recent military actions are concerning, it beggars belief that—given recent experience with Iran and North Korea (as well as other targets of U.S. sanctions such as Cuba)—unilateral attempts to isolate a massive nuclear power would have produced a better geopolitical outcome than did engagement.

Indeed, marshalling the necessary WTO‐​wide consensus to deny more than a billion people in a modernizing economy access to an open multilateral trade organization—one that already included communist Cuba and for decades had tolerated Eastern Bloc command‐​and‐​control economies and “socialist” countries with pervasive state‐​owned industries—was not realistic, especially given what U.S. policymakers could have known at the time about China’s relatively liberal leadership and impressive economic reforms. This last point bears emphasis: as previously noted, the reforms that China undertook during its WTO accession—along with additional reforms made shortly after accession—often in direct response to WTO require­ments (and member demands), were so substantial as to drive China’s incredible export competitiveness. To assert that U.S. policymakers in the 1990s should have somehow known that these reforms would cease or reverse a decade later under different Chinese political leadership, and thus either convince over 140 other WTO members to deny a nuclear China entry into the WTO or reject PNTR (becoming the only WTO member to do so), is applying an impossible standard.

China’s Backsliding since Accession Cannot Be Blamed on PNTR or the WTO

PNTR critics also ignore the missed oppor­tunities since China’s WTO accession, especially the unused checks on Chinese trade abuses that were among China’s allegedly weak WTO commitments. As previously noted, China undertook substantial trade and economic liberalization before and shortly after WTO accession and made dozens of “WTO‐​plus” commitments to become a member. Since then, problems have undoubtedly arisen, but as Cato’s James Bacchus, Simon Lester, and Huan Zhu have documented, most of those problems—for example, on industrial subsidies and intellectual property—are covered by WTO rules and can be litigated through dispute settlement.64 Moreover, such litigation has proven effective. According to the Peterson Institute for International Economics’ Jeffrey Schott and Euijin Jung, for example, the United States was undefeated at the WTO when challenging Chinese trade practices between 2002 and 2018.65 Four other cases were pending at the time of that paper’s publication, but the United States has since won two more—one on agricultural subsidies and one on barriers to U.S. imports of wheat, rice, and corn.66

Furthermore, Bacchus, Lester, and Zhu demonstrate that when China loses WTO disputes, it tends to comply with the decisions:

Of the 27 matters litigated against China, 5 are still pending, 12 were litigated all the way through, and 10 were resolved through some kind of settlement, or not pursued after the measure was modified. These cases addressed a wide range of issues: export restrictions, subsidies, intellectual property protection, discriminatory taxes, trading rights, services, and trade remedies. In all 22 completed cases, with one exception where a complaint was not pursued, China’s response was to take some action to move toward greater market access.67

Chinese compliance is not perfect (nor is any other WTO member’s), but it is arguably better than that of the United States, which has famously shirked WTO rulings on subsidies, antidumping rules, and internet gambling.

The refusal of the United States and other WTO members to pursue more disputes against China—or open “compliance proceedings” when China does not fully comply—is a policy choice worth criticizing, but this says nothing about the original decision to admit China to the WTO. Indeed, it is either mistaken or misleading to claim that China’s WTO accession terms were weak and that the WTO has utterly failed to discipline China’s unfair trade practices when the sole means of imposing such discipline—dispute settlement—and the “WTO‐​plus” rules that China accepted have never been fully utilized. This is declaring defeat before ever firing a shot.68

Other U.S. policy choices since the passage of PNTR also deserve scrutiny. Among these are the United States’ withdrawal from the Trans‐​Pacific Partnership, a treaty that was designed in part to counterbalance China’s economic and geopolitical ambitions; its failure to reform tax, trade, and immigration policies that inhibit American companies’ global competitiveness;69 its failure to modernize adjustment assistance and worker retraining programs intended to mitigate trade, technological, or cultural disruptions;70 or its continued imposition of tax, education, occupational licensing, criminal justice, zoning, and other policies that discourage labor adjustment and economic dynamism.71 Such policies are indeed worthy of criticism and debate, but they have nothing to do with the decisions to pass PNTR, allow China to join the WTO, or otherwise “normalize” trade with China. And blaming China for these policies’ inevitable failures relieves the policies—and the American politicians who implemented them—of the scrutiny that they deserve.

The Curious Failure to Note the Problems with U.S. Market Interventions

Those wishing to blame the problems of the American working class on PNTR and freer markets more broadly often ignore the United States’ own long history of market inter­ventions and their failures to help companies and workers. A core tenet of the current populist backlash against trade, and particularly trade with China, is that American “elites” opened the floodgates to Chinese imports with only a timid threat of the WTO dispute process to protect American workers from “unfair” competition. This ignores the mountain of government interventions that have been used—at the federal level alone—to restrict Chinese imports, otherwise protect or subsidize American manufacturers, and assist American workers.

On trade, the United States still maintains significant tariffs and tariff‐​rate quotas on imports of “sensitive” products such as trucks, apparel, footwear, and food.72 Moreover, according to Global Trade Alert, the United States has been one of the most frequent users of “harmful” nontariff government trade interventions—ones that far outnumber its “liberalizing” measures over the same period.73 This includes, as of late 2019, almost 190 special duties (“trade remedies” such as antidumping and antisubsidy measures) on a variety of Chinese imports, two‐​thirds of which (127 of 187) use a special “non‐​market economy” antidumping methodology that practically ensures prohibitive duty rates on those goods (a “WTO‐​plus” accession commitment special to China and a few others).74 These duties target “unfair” trade and subsidies that injure U.S. manufacturers and workers, and—as the numbers indicate—American companies and unions have been successful in petitioning for them. (There are around 500 duty orders in place overall.)

Dozens of other Chinese imports are barred from the U.S. market as a result of Section 337 actions that remedy intellectual property rights violations.75 Chinese investment in U.S. industries, meanwhile, can be (and has been) restricted by the Committee on Foreign Investment in the United States, and U.S. technology exports to China are often blocked on national security grounds.76

The U.S. government also has long provided financial and other support to favored industries and workers, for example through auto bailouts, steel industry bailouts, alternative energy subsidies, manufacturing tax credits, Export–Import Bank loans and other export assistance, procurement preferences such as the Buy American Act and the Davis–Bacon Act, shipping restrictions such as the Jones Act and the Passenger Vessel Services Act,77 and the billions of other taxpayer dollars that the United States has doled out to “blue collar” industries and workers over the past few decades at the federal level alone. As I documented in a 2012 paper on global subsidies and antisubsidy disciplines, “despite the obvious economic, legal, and political problems associated with domestic subsidies, the United States remains one of the world’s largest subsidizers.”78

The U.S. government has also repeatedly tried to fund and retrain workers, most notably through the Trade Adjustment Assistance (TAA) program, which offers generous subsidies to U.S. workers affected by import competition. Unfortunately, TAA has proven to be a “notorious failure”: as I noted in a 2016 article, “multiple studies commissioned by the Labor Department have found that TAA participants are worse off, as measured by future wages and benefits, than similarly situated jobless individuals outside the program.”79 The Wall Street Journal’s Eric Morath in December 2018 noted additional research into the failure of TAA to help workers allegedly displaced by trade in 2018 and a move by certain states to “de‐​emphasize programs such as TAA in favor of getting workers back into jobs more quickly.”80 Other federal job‐​training programs have been found to be similarly inefficacious, and related reform efforts have thus far been underwhelming.81

These and other government programs raise serious concerns when it comes to helping American workers adjust to trade and other shocks, and they need to be reformed. But that does not change the simple fact that these programs do exist and have for decades. These policies refute the claim that U.S. policymakers simply passed PNTR and walked away from the American working class out of some sort of “market fundamentalism” or rigid adherence to “laissez faire ideology.”

The real problem was that these interventions did not work very well. A classic example is the U.S. steel industry, whose companies and workers since the 1970s have arguably received more government assistance than any industry in the country. This includes hundreds of import restrictions; tens of billions of dollars in state, local, and federal subsidies and bailouts; exemptions from environmental regulations; special “Buy American” rules; federal pension benefit guarantees; and even its own caucus in Congress.82 The result: dramatic historical declines in employment and capitalization, numerous bankruptcies, and of course, continued demands for even more government protection (as the current Section 232 tariffs make clear).

The steel industry certainly is not alone. As a 2013 Congressional Research Service report concluded about the state of American manufacturing, “Although Congress has established a wide variety of tax preferences, direct subsidies, import restraints, and other federal programs with the goal of retaining or recapturing manufacturing jobs, only a small proportion of US workers is now employed in factories.”83

In short, there is scant evidence that Washington elites abandoned the American working class after liberalizing trade with China. The government’s interventions may have failed, but they were interventions nonetheless.

Trade and Globalization Bring Unambiguous Benefits

If the China Shock’s disruptions are unique to that country and time period, the debate over PNTR and China trade is academic. Most economists believe that the China Shock ended years ago and that if its effects are unlikely to happen again, then the historical analysis is not instructive regarding future trade policies. As Jakubik and Stolzenburg recently concluded:

The literature on the local labour market effects of Chinese import competition has been cited extensively as an argument for limiting trade with China despite the fact that the results do not support this conclusion.… Even if policy were narrowly focused on direct import competition effects ignoring price and indirect effects, there is no case for limiting trade with China [because] US local labour market adjustment to the China Shock has largely concluded.84

Put another way, if there will never be an “India Shock” or a “Vietnam Shock,” then there is no reason to reset American trade and labor policy or to reorganize the U.S. economy more broadly in preparation for such a moment.

On the other hand, if the supposed lessons from PNTR are, as some intend, a guide for future U.S. decisionmaking on trade and global­ization writ large—pundits such as Tucker Carlson seem to aim wider85—then the debate must consider the many factors supporting freer trade and opposing protectionism. Those factors include the following:

  • the wide body of research showing significant economic gains from import liberalization and multilateral trade—even trade “shocks”—and the overwhelming support for these policies from economists on the political left, right, and center;86
  • the unseen benefits of import competition on American economic dynamism, providing Americans with not merely cheaper goods and services but better (and once unimaginable) ones, as well as better jobs, better companies, and better lives;87
  • the morality of freer trade—both for Americans and the global poor—and the political problems that arise from government putting the desires of favored producers above those of all consumers;88
  • the fact that much of “globalization” is driven not by elite policy choices such as free trade agreements but by seismic changes in communications and technology, such as shipping containers, that are far beyond the control of any policymaker;89
  • the extensive research showing that trade complaints through the WTO are more effective than unilateralism in bringing about trade reforms;90
  • the longstanding geopolitical benefits of trade, including the WTO’s role in preventing world war and the strong connection between trade and peace;91
  • the unique position of American manufacturing and workers post–World War II, when much of the rest of the world was either recovering from war or descending into communism;92
  • the fact that, historically, productivity shocks such as automation have been just as disruptive as trade (if not more so) in terms of low‐ and middle‐​skill manufacturing job losses and that U.S. production occupations are today among the most vulnerable to future automation‐​led disruptions;93 and
  • that trade economists have for decades acknowledged that adjustment to trade shocks “may be neither quick nor easy.”94

A fuller accounting of these points is beyond the scope of this paper, but each has ample support. For example, a 2019 International Monetary Fund cross‐​country analysis of trade and technology shocks found that while both can have adverse regional employment effects (raising unemployment and lowering labor force participation), only automation has long‐​lasting harms and that regions hit by trade shocks actually ended up better off a couple years later. That same paper also found that national policies encouraging more flexible labor markets can improve both adjustment in regional labor markets and their resilience to shocks and that countries with less stringent product market regulation, lower administrative costs for starting a business, and greater trade openness had lower regional inequality.95

These points do not prove that free trade is seamless or that economists accurately predicted regional labor market frictions in response to large‐​scale trade disruptions, but—unless the China Shock is truly sui generis (and in that case, again, the historical and economic analyses are not instructive)—the broader economic and geopolitical benefits of trade and globalization are essential to any discussion of trade liberalization, elite policy choices, and the American working class. Unfortunately, these points always seem to be missing.

Protectionism Keeps Failing

When critics decry “normalized trade” as an elite policy “choice,” they necessarily imply—but rarely state—a trade policy alternative. The only apparent alternative is some form of protectionism (i.e., government restrictions on imports of goods and services), and it has repeatedly proven a failure. For example, International Monetary Fund economists in 2018 examined data for 151 countries over 51 years (1963–2014) and found that “tariff increases lead, in the medium term, to economically and statistically significant declines in domestic output and productivity” as well as more unemployment and higher inequality.96 The same is true for American protectionism, which has repeatedly been shown to impose immense economic harms that far outweigh any possible benefit to protected workers; to fail to protect American firms and workers over the longer term; and to breed elite corruption, cronyism, and political dysfunction.97 Two instances warrant mention in this regard:

  • Using the Section 421 safeguard mechanism tied to China’s WTO accession, President Obama in 2009 imposed 35 percent tariffs on Chinese tires. The result was, even under the best assumptions, a handful of jobs saved at an annual cost to U.S. consumers of over $900,000 per job, plus a substantial increase in non‐​Chinese imports instead of new U.S. production.98 Today, the industry’s prospects are no better.99 (Such inefficacy is precisely what Levy predicted because of the interchangeability of Chinese and other imports.100)
  • A 2017 review of all U.S. antidumping investigations against Chinese imports between 1998 and 2006 revealed that the duties reduced Chinese imports and increased prices of subject merchandise in the U.S. market.101 However, these effects “dissipate approximately 2 years after the antidumping decision,” and imports from other countries simply increased to replace the declining Chinese imports. Such results “cast doubt on the effectiveness of antidumping actions against China as mechanisms for protecting US producers.” Specific case studies, such as Daniel Ikenson’s review of antidumping measures on wooden bedroom furniture from China, show similar results: “Instead of preserving or returning domestic jobs … import restrictions will cause a shift in sourcing from China to places like the Philippines, Indonesia, Brazil, and Vietnam—places from which many of the petitioners have begun or are poised to begin importing themselves.”102

These studies reveal both the futility and cronyism surrounding past U.S. attempts to stop Chinese imports during the China Shock period. Thus, if one looks to restrict trade to solve the problems facing America’s working class, the “solution” will most likely be worse than the alleged problem.

Those who object to “normalizing” trade with China (i.e., removing U.S. restrictions on Americans’ consumption of Chinese imports) must also acknowledge that liberalization was not merely an economic and geopolitical decision but also a moral one that removed inequities in the previous, more protectionist system. The now‐​eliminated trade restrictions typically resulted from political “elites” seeking to support certain industries and workers at most Americans’ (especially poor ones) expense.103 American footwear workers, for example, benefited from a political decision (dating back to the Smoot–Hawley Tariff Act) to protect footwear jobs through hidden restrictions on the voluntary commercial decisions of other, unknowing Americans—restrictions that forced these citizens to subsidize U.S. footwear jobs by paying more for shoes.104 Today, former “big steel” lawyers and executives, now in the Trump administration, dole out tariff protection to their former colleagues who lobbied for it; those well‐​connected colleagues, in turn, get to decide the fate of their American customers’ requests for steel‐​tariff relief, even though the steel‐​consuming customers are a far larger share of the U.S. economy and workforce than is the steel industry. Trade‐​related lobbying expenditures over the past two years of “trade populism” have unsurprisingly skyrocketed.

Trade liberalization cures this malady, whether intended or not, while also improving the living standards of most Americans. And though it is legitimate to ask after government removes import protection whether it owes the affected workers more in terms of adjustment welfare or job training, removing the protection was clearly the right thing to do.

Why Did Some Places “Move On”?

Finally, those seeking to blame PNTR or Chinese imports for the current plight of the American working class ignore the many places in the United States that were affected by Chinese import competition but did adjust and have thrived economically—often with the help of trade and foreign investment. Indeed, the fact that the longer‐​term effects of Chinese import competition vary dramatically from place to place—even in states or regions that face intense competition105—undermines the notion that the China Shock was a national trade problem (necessitating national protectionism) as opposed to a local adjustment problem (necessitating local solutions).

Many cities and towns in America that were once known for low‐​skill manufacturing and faced intense import competition in the 1990s and 2000s have since adapted and thrived. As previously noted, several studies show that most U.S. regions ended up better off following the China Shock, though some areas—particularly those with low human capital—struggled. A 2018 Brookings Institution report, moreover, finds that 115 of the 185 U.S. counties identified as having a disproportionate share of manufacturing jobs in 1970 had “transitioned successfully” from manufacturing by 2016 and that of the remaining 70 “older industrial cities,” 40 exhibited “strong” or “emerging” economic performance between 2000 and 2016.106 The “strong” localities, achieving high marks for growth, prosperity, and inclusion, include not only well‐​known success stories such as Pittsburgh and cities close to Boston and Manhattan but also smaller places such as Beaumont, Texas; Waterloo, Iowa; and Bethlehem, Pennsylvania.

Anecdotal evidence reiterates these findings: towns that once depended on low‐​skill manufacturing, such as Greenville–Spartanburg, South Carolina; Hickory, North Carolina; Warsaw, Indiana; and Danville, Virginia, are now home to thriving companies that succeeded by adapting to the market, including through international trade and investment.107 Journalist James Fallows has documented many of these lesser‐​known success stories in his 2018 book, Our Towns: A 100,000-Mile Journey into the Heart of America, and in a regular column for The Atlantic.108

Anyone still doubting such successes need only drive down Interstate 85 from Charlotte, North Carolina, to Montgomery, Alabama, to see the multinational factories firsthand.109 The Federal Reserve Bank of New York emphasized this region in its examination of the recent surge in U.S. manufacturing jobs:

While job losses during the 2000s were fairly widespread across the country, manufacturing employment gains since then have been concentrated in particular parts of the country. Indeed, these gains were especially large in “auto alley”—a narrow motor vehicle production corridor stretching from Michigan south to Alabama—while much of the Northeast continued to shed manufacturing jobs.110

The contrast between now‐​thriving American towns and those still reeling from a trade shock that ended a decade ago again indicates that the problem the shock revealed was not import competition but—as Autor, Dorn, and Hanson themselves concede—many communities’ inability to adjust to seismic economic changes. The International Monetary Fund study on trade shocks, labor market policies, and regional adjustment reiterate these conclusions at a cross‐​country level. Thus, commentators and politicians who blame China trade for the difficulties of the American working class should stop asking, “Why did elites normalize trade with China in the 1990s?” and instead ask, “What did many American towns, companies, and workers do right in the face of intense import competition, and how can local, state, and federal policies encourage that important improvement?”


The historical record before and after PNTR and the numerous academic analyses of the “China Shock” provide a straightforward explanation for the past 20‐​plus years of U.S. trade policy toward China: engagement and liberalization, exemplified by PNTR and China’s WTO accession, were a pragmatic and bipartisan policy choice made in the face of nonexistent or inferior alternatives, especially given the information when the choice was made. Engage­ment, moreover, produced real economic benefits for most Americans while bolstering the multilateral trading system and removing historical inequities under the previous, more protectionist U.S. trade policy regime.

The resulting economic disruption and adjust­ment were difficult for some U.S. regions and workers—more difficult than many experts expected—and certainly post‐​liberalization policy mistakes were made (though often in the direction of less liberalization, not more). With the benefit of two decades of hindsight, one can legitimately claim that certain specific “WTO‐​plus” rules should have been drafted differently during China’s accession.

That said, the facts simply do not support popular assertions from American politicians and pundits that engagement with China in the 1990s and 2000s was an obvious mistake and that denying China admission to the WTO was realistic a policy choice that would have improved U.S. economic and geopolitical standing today or that the real labor and cultural issues in America today are the fault of “Washington elites” who blithely pursued normalized trade with China to benefit corporate donors and democratize communist China while dogmatically refusing to support—through trade, labor, or any other policies—the working class. Such narratives are unsupportable.

Labor market and cultural disruptions in the United States are real and important, as is China’s current and unfortunate turn toward illiberalism and imperialism. But pretending today that there was a better trade policy choice in 2000 than PNTR and engagement more broadly is misguided. It assumes too much, ignores too much, and demands too much. Worse, it could lead to truly bad governance: increasing U.S. protectionism; forgiving the real and important failures of our policymakers, CEOs, and unions over the past two decades; and preventing a political consensus for real policy solutions. Indeed, these things are happening now.


Lincicome, Scott. “Testing the ‘China Shock’: Was Normalizing Trade with China a Mistake?” Policy Analysis No. 895, Cato Institute, Washington, DC, July 8, 2020. https://​doi​.org/​1​0​.​3​6​0​09/PA.891.

Scott Lincicome

Scott Lincicome is a senior fellow in economic studies at the Cato Institute and a Senior Visiting Lecturer at Duke University Law School. Previously he spent two decades practicing international trade law.