Anatomy of a flop: Why Trump's US-China phase one trade deal fell short
Hexuan Li provided outstanding data assistance, and Melina Kolb, William Melancon, and Oliver Ward assisted with graphics.
The Biden administration plans to review the phase one trade agreement President Donald Trump forged with China in late 2019. Good. Much of the deal was a failure. Its centerpiece was China’s pledge to buy $200 billion more of US goods and services split over 2020 and 2021.
According to evidence from the deal’s first year, China was never on pace to meet that commitment, with the economic devastation of the COVID-19 pandemic only partly to blame. Attempting to manage trade—to meet Trump’s objective of reducing the bilateral trade deficit—was self-defeating from the start. It did not help that neither China nor the United States was willing to deescalate their debilitating tariff war.
The phase one deal should not be ripped up, however. Several elements are worth keeping and building upon—such as China’s commitment to reduce nontariff barriers related to food safety and open up to foreign investment. China’s agreeing to crack down on intellectual property violations and the forced, insufficiently compensated, transfer of American technology will also prove beneficial if enforced.
But the dubious policy objective of reducing the bilateral trade deficit—the heart of Trump’s phase one deal—should be scrapped. The purchase commitments only sowed distrust in the very same like-minded countries with which the new US administration must work to tackle their mutual concerns involving China.
Comparing China’s purchases of phase one goods in 2020 with 2019 is irrelevant for the legal agreement
China actually did import more phase one goods from the United States in 2020 compared with the previous year—imports were 13 percent higher (figure 1). That was also much better than the flat growth of China’s imports of the same goods from the rest of the world.
While economically meaningful, both comparisons, however, are irrelevant for the phase one legal agreement. US exports to China in 2019 were extraordinarily low, in part because China imposed retaliatory tariffs in response to Trump’s trade war. US exports had declined by 10 percent in 2018 compared with 2017, shrinking by another 8.5 percent in 2019. Thus, under the threat of continued tariff escalation, the Trump administration convinced Beijing to commit to purchasing an additional $200 billion of US goods and services, in prescribed amounts split over 2020 and 2021, on top of the baseline of 2017 trade flows—not 2019.
Relative to that precise legal commitment, US exports of phase one products to China in 2020 failed spectacularly—falling more than 40 percent short of the target (figure 2).
US manufacturing suffered throughout the trade war and has not recovered
President Trump cast his trade policies as designed to help American manufacturing, citing China’s large bilateral trade surplus in goods as a problem. He also complained that Beijing was forcing US companies to turn over their technology and manufacture in China instead of exporting from America. Other targets of his ire were China’s higher tariffs on cars and its subsidies to steel and aluminum, which squeezed American companies out of export markets. Though some of the policy complaints were valid, Trump’s approach of starting a trade war and then settling it with a demand for purchase commitments backfired.
US manufacturing exports to China, which had nearly doubled between 2009 and 2017, flattened in the second half of 2018 and fell by 11 percent in 2019 (figure 3). This was partially a result of Chinese retaliation against Trump’s tariffs. But in the first year of the phase one agreement, US manufacturing exports continued to suffer, declining another 5 percent. Overall, they fell 43 percent short of the legal commitment for 2020, remaining more than 14 percent below pre–trade war levels. Because manufacturing constituted 70 percent of the value of all goods covered by the purchase commitments, this shortfall essentially guaranteed that the total targets would disappoint.
The US auto sector provides an excellent illustration of how even temporary trade war tariffs can inflict long-term damage. By 2017, China had become the second largest export market for American vehicles. Then, in July 2018, China retaliated with a 25 percent tariff on US autos. (In a savvy economic maneuver, it simultaneously lowered its auto tariff on imports from the rest of the world.) US exports to China fell by more than a third (see figure 3). Tesla accelerated construction of a new plant in Shanghai, arguing that Trump's tariffs on auto parts, China's retaliation on cars, and the resulting uncertain trade picture, made it no longer competitive to manufacture electric vehicles destined for China in the United States. For similar reasons, BMW shifted some production destined for China out of South Carolina. By the end of 2020, US auto exports had still not recovered.
Aircraft provides another cautionary tale about the limits of the purchase commitment approach to policymaking. Boeing had been a large historical exporter to China, and aircraft did not face Chinese retaliation during the trade war. Yet, following two crashes of Boeing's 737 MAX airplane, sales declined from over $18 billion in 2018 to less than $11 billion in 2019. (The model was grounded globally between March 2019 and December 2020, and Boeing shut down production between January and May 2020.) In April 2020, China canceled purchase orders for undelivered planes. US aircraft exports in 2020 were only $4.6 billion, less than one-fifth of the estimated target.
Semiconductors and semiconductor manufacturing equipment provide a different lesson. The high-tech sector’s exports outperformed their estimated targets in 2020, also for reasons likely unrelated to the phase one deal’s legal commitments. In 2019 and 2020, the United States announced it would soon limit exports of semiconductors and equipment, citing national security concerns. Exports accelerated in part because Chinese buyers such as Huawei and SMIC reportedly stockpiled in 2020, anticipating that US export control policy would soon cut them off.
Also on the (export) bright side were US sales of medical products to China in 2020. Pandemic-induced demand likely helped growth continue despite the otherwise challenging economic environment.
US agriculture also suffered, but received subsidies, then recovered
China has long been an important market for US agricultural exports such as soybeans. US exports to China through 2017 remained strong, though slightly lower than the peak years of 2012–14, driven in part by a global increase in certain commodity prices, including corn and wheat.
But as with manufacturing, the trade war devastated US agricultural sales to China. Exports were cut in half in 2018, with 2019 levels remaining nearly 30 percent lower than in 2017 (figure 4). The Trump administration responded by dishing out tens of billions of dollars of taxpayer-funded federal subsidies to farmers in 2018 and 2019, a step it never took for the manufacturing sector. As a result, the US Department of Agriculture (USDA) estimated that American farm income—subsidized by the government—was 11 percent higher in 2019 than 2017, achieving its highest level since 2014.
Overall, China did ramp up farm purchases in 2020 and by September was back on pace to reattain 2017 levels. Nevertheless, US agricultural exports ended up both 18 percent short of the 2020 legal commitment and considerably lower than the Trump administration’s political aspirations. The Trump administration touted the deal as just the start of what China would do to help US farmers, boasting China had agreed to “strive” to buy $5 billion more per year on top of the already hefty purchase commitments. In 2020, China did not.
Soybeans had been nearly 60 percent of US farm exports to China prior to the trade war and one of the first products China hit with 25 percent retaliatory tariffs. US exports fell from $12 billion in 2017 to $3 billion a year later, as China shifted purchases toward Brazil and Argentina. Despite President Trump’s repeated assurances beginning late in 2018 that China would soon be “back in the market” for soybeans that American farmers were being forced to stockpile in record amounts, 2019 sales also remained more than a third lower than in 2017. Part of China’s reduced demand for the animal feed in 2019 derived from a devastating outbreak of African Swine Fever that cut the world's largest pig herd by 40 percent. US soybean exports picked up again only in 2020, reaching pre–trade war levels (though falling short of the estimated target), as the Chinese pig herd recovered.
Consider pork. China had begun to import more pigmeat from the United States in 2019 to deal with its local pork shortages, even before the phase one agreement was signed. The shortage was so bad, China's pork imports from the rest of the world in 2020 were also more than 500 percent higher than 2017 levels.
A few other US farm exports also beat their estimated targets in 2020, including corn and wheat. Here, Beijing began complying with a 2019 World Trade Organization (WTO) dispute settlement ruling against its unfilled tariff rate quotas; compared with 2017, China’s imports from the rest of the world in 2020 increased by more than 340 percent for corn and 280 percent for wheat. US cotton sales to China also improved in 2020, and sorghum recovered from the effects of trade war tariffs.
But many other food products did not recover. US lobster exports to China remained 18 percent lower in 2020 than 2017. Beijing both imposed tariffs on US lobster during the trade war and encouraged Chinese consumers to shift to other suppliers by lowering tariffs on lobster from Canada and other countries. (China’s lobster imports from the rest of the world increased by nearly 250 percent in 2020 compared with 2017 levels.) Maine’s lobster industry suffered but was ineligible for the tens of billions of dollars of USDA trade war payments of 2018 and 2019. The Trump administration granted it subsidies only in the immediate runup to the 2020 election.
The energy commitments were a riddle
Energy made up only 8 percent of the total goods covered by the phase one agreement yet is also characterized by contradictions. US energy exports to China performed the worst of the three goods sectors (figure 5), reaching less than 40 percent of the 2020 legal commitment. Low oil prices hampered export commitments measured in dollars, not volume (e.g., barrels of oil). On the other hand, the export performance of both crude oil and liquefied natural gas was considerably higher in 2020 than even 2017 levels, as each started from a very low pre–trade war baseline.
Yet, the extremely large energy purchase commitments in the phase one agreement raise at least two additional questions for policymakers. First is legal good faith. As Bloomberg reported, only after the agreement was signed did the administration learn from the US industry that it lacked production capacity to fulfill the targets. Can the consumer be found at fault for insufficient purchases if the American industry lacked short-run ability to provide the supplies?
Second is climate change concerns. The unrealistic targets were only for crude oil, liquefied natural gas, coal, and refined products. The Trump administration did not share climate concerns. But President Joseph R. Biden Jr. rejoined the Paris Agreement on climate change on day one of his administration and has prioritized climate change mitigation, which could have implications for fossil fuel exports.
The trade war and phase one agreement in the macroeconomic context
Did the COVID-19 pandemic doom the phase one deal? The United States fell into recession, with GDP contracting by 3.5 percent in 2020. China was the first country hit by the pandemic, yet its economy recovered both more quickly and more robustly than most, with GDP expanding 2.3 percent in 2020. And according to World Trade Monitor, China’s imports reached prepandemic levels by June; the rest of the world’s trade recovered only near the end of 2020.
We will also never know, of course, what the world economy would have looked like without the trade war and phase one agreement. But what if US exports in 2018, 2019, and 2020 to China of phase one products had grown at the same pace as China’s imports of those same goods from the world? (China’s imports from the United States and the rest of the world grew roughly in tandem over 2009–17; see again figure 1.)
Without the US-China trade war, US exports to China would have ended up roughly 19 percent higher than actual 2020 levels (figure 6). The United States would also not have suffered those export losses—US sales to China would have been $29 billion (33 percent) higher in 2018 and $34 billion (43 percent) higher in 2019 than actual levels. Without the export losses, American taxpayers would also not have needed to fund tens of billions of dollars of farm subsidies.
Looking more broadly, the trade war was costly. As economists have documented extensively, the tariffs raised prices and hurt American consumers. American companies (not just exporters) faced higher input costs because of the tariffs, hurting competitiveness and reducing their employment and sales. A handful of sectors and workers may have benefited, but the overall damage to the US economy was inarguable.
The lingering puzzle of the uncovered products
Oddly, the purchase commitments in the phase one agreement did not cover 27 percent of US goods exports to China in 2017. China had little incentive to buy such goods from the United States in 2020, as the purchase targets would not be credited.
Unsurprisingly, US exports of uncovered products to China performed even worse than covered products (see again figure 6). The Trump administration may have thought them unimportant—indeed, nine of the top 20 uncovered products by value included the words “waste” or “scrap” or “not elsewhere specified or indicated” in their descriptions. But declines in these exports simply offset one-for-one any export increase in covered products. Their omission from the deal remained a mystery.
The United States needs a new China policy
The lesson from year one of the US-China agreement is that purchase commitments did not work. Looking ahead to year two, some products that performed well last year could fall short in 2021. Will China continue to accelerate soybean purchases, once its stockpiles have been replenished? Will it import as much pork, once its domestic herd has recovered? Will it be cut off from American semiconductors and equipment as US export controls start to bind? Will it demand less American cotton, since the January 13, 2021 ban on US imports from Xinjiang means more domestic cotton is now available in China? As part of its decarbonization policy, will the Biden administration seek cuts in carbon-intensive energy production and exports?
The phase one agreement was never the long-term fix to what ails the US-China trade relationship. But attempting to manage trade with purchase targets and an intention to reduce the bilateral deficit is the wrong approach. It distracts from the engagement necessary to address the costly incompatibilities of the Chinese economic system with the more market-oriented economies of the United States, European Union, Japan, and other like-minded countries.
The other positive parts of phase one—such as China’s removal of technical barriers to trade, baby steps of market-oriented reform, and additional market access—can be a foundation for future progress. For example, the United should press for more Chinese commitments to permanently eliminate the retaliatory tariffs on US exports and bind its significant most favored nation (MFN) tariff reductions of 2018 and 2019. The Biden trade team should look at the European Union’s recently signed EU-China Comprehensive Agreement on Investment (CAI). It achieved many of the same things the United States attained with China in the phase one deal, including new financial services market access and some Chinese commitments not to forcibly transfer technology—an achievement won without the costs of a trade war.
Indeed, the European Union’s export performance with China is humbling. Even limiting focus to America’s priority products covered by the phase one agreement, China’s imports from the EU ended up 21 percent higher in 2020 than 2017 (figure 7). China’s imports from the United States ended up 8 percent lower in 2020, in addition to all the losses US exporters suffered in 2018 and 2019.
Yet, even with CAI, the Europeans learned that core issues with China cannot be tackled bilaterally. Sabine Weyand, director-general for trade, stated recently that “what we are looking at is to work with the US, Japan, but also other like-minded countries to agree on an update of the WTO rulebook.” That includes rules for industrial subsidies, a thorny issue neither the United States or the European Union was able to address with China through their respective deals.
In other words, only a group of countries working together and sharing the burden will be able to make progress with China. For the United States, the purchase commitments both worked against collaboration and reflected an approach divorced from economic reality.
1. See Chapter 6 and Annex 6.1 of the Economic and Trade Agreement between the Government of the United States of America and the Government of the People’s Republic of China.
2. Annex 6.1 also sets targets for $37.9 billion (18.9 percent of $200 billion) of additional Chinese purchases of certain traded services from the United States. Those data have not been released and so are not covered here.
3. For specific products, phase one targets are only estimates, apportioned as the share of that product in total US exports to China in 2017 of goods covered by the purchase commitments. Neither the United States nor China made targets publicly available at a more disaggregated level than manufacturing, agriculture, energy, or services.
4. Annex 6.1, footnote b, states, “At the request of the United States, China will strive to purchase and import $5 billion per year of the US agricultural products covered by this Chapter, in addition to the minimum amounts set forth herein.”
5. In a more traditional chapter 3 of the phase one deal, China also agreed to address technical barriers that had previously slowed pork imports.
6. See Joseph Glauber and Simon Lester, China–Tariff Rate Quotas for Certain Agricultural Products: Against the Grain: Can the WTO Open Chinese Markets? Forthcoming in World Trade Review.
7. Jennifer A. Dlouhy, Shawn Donnan, and Nick Wadhams, "Big Oil Warned Trump Team China Trade Deal Was Unrealistic," Bloomberg, February 12, 2020.
8. See, for example, Article 6.2.7, “If China believes that its ability to fulfill its obligations under this Chapter is being affected by an action or inaction by the United States or by other circumstances arising in the United States, China is entitled to request consultations with the United States.”
9. See Mary Amiti, Stephen Redding, and David Weinstein, 2019, The Impact of the 2018 Trade War on U.S. Prices and Welfare, Journal of Economic Perspectives 33, no. 4: 187-210; Pablo D. Fajgelbaum, Pinelopi K. Goldberg, Patrick J. Kennedy, and Amit K. Khandelwal, 2019, The Return to Protectionism, Quarterly Journal of Economics 135, no. 1: 1-55; Kyle Handley, Fariha Kamal, and Ryan Monarch, 2020, Rising Import Tariffs, Falling Export Growth: When Modern Supply Chains Meet Old-Style Protectionism, NBER Working Paper No. 26611, January; and Aaron Flaaen and Justin R. Pierce, 2019, Disentangling the Effects of the 2018-2019 Tariffs on a Globally Connected U.S. Manufacturing Sector, FEDS Working Paper No. 2019-086, December.
10. See Chad P. Bown and Soumaya Keynes, 2021, The EU’s new trade policy, with Sabine Weyand of DG Trade. Trade Talks, episode 148, January 17.