Trump's phase one trade deal with China and the US election
Hexuan Li provided outstanding data assistance, and Melina Kolb, William Melancon, and Oliver Ward assisted with graphics.
Minor revisions and updates to the figures in the original post of October 26, 2020, have been made in light of the November 4, 2020 release of revised US export data for September.
President Donald Trump has staked a claim to success in his trade war with China on his phase one trade agreement of January 2020. In his self-proclaimed "historic" deal, China committed to purchasing an additional $200 billion of American-made goods and services over 2020 and 2021. Trump has even boasted that the deal "could be closer to $300 billion when it finishes."
As the US election approaches, however, the evidence indicates that the deal is failing to live up to Trump's pledge.1 The latest official data show that China is falling short, reaching only 53 percent of the expected purchase target through September 2020. In fact, Chinese imports of US goods are now lower than they were before Trump started his trade war with a blitz of US tariffs on Chinese products in 2018.
No single reason explains why China has not met the phase one goals. The COVID-19 pandemic did initially knock China's economy on its heels, but its trade has recovered faster than most. And some US exports to China—including medical supplies, pork, and semiconductors—actually accelerated in 2020.
From the start, $200 billion of additional sales to China was a worrisome target. Nearly 30 percent of US goods exports to China are not even covered by the phase one deal. And for those that the agreement covered, a look at 15 sets of products shows that their sales to China have been affected by disparate factors, including plane crashes, animal disease outbreaks, export controls, legal rulings by the World Trade Organization (WTO), the lingering effects of the trade war's tariffs, as well as the pandemic.
One conclusion from the data is obvious. Americans suffered as China's retaliation devastated US exports. Trump's higher tariffs have raised prices for American consumers and costs for American businesses. His politically motivated purchase commitments may have created more problems than they solved. After the election, the United States needs a new approach to address its trade concerns with China.
Chinese imports of US goods remain well below 2017 levels
The tariffs that the United States and China imposed on imports from each other decimated bilateral trade in 2018 and 2019. The Trump administration thus negotiated the legal text of the phase one agreement to commit China to buy an additional $200 billion of US goods and services over 2020 and 2021 on top of 2017 (and not 2019) levels, when bilateral trade was more robust.2 Evaluating the deal on legal grounds thus requires comparison of 2020 with 2017.
Through September 2020, China had bought only 53 percent of what would have been expected by this point in the year (figure 1, panel a).3 Year-to-date imports of all covered goods were only $65.9 billion relative to a target of $124.9 billion. Through three-quarters of 2020, China had purchased just more than one-third of what it pledged in Trump's deal it would buy this year. (The target for full year purchases is $173.1 billion.) China's imports from the United States have failed to catch up to pre-trade war levels, running 16 percent lower than at the same point in 2017.
But China's imports of covered goods from the rest of the world are 20 percent higher in 2020 than 2017 (figure 1, panel b). While the pandemic has hurt demand, China's overall imports in 2020 have done better than imports entering the rest of the world, according to data from the World Trade Monitor.
The phase one deal does not cover nearly 30 percent of products that the United States exports to China. Not surprisingly, China's imports of uncovered goods have done even worse in 2020—26 percent lower than 2017 levels (figure 1, panel c). On the other hand, China's imports of uncovered products from the rest of the world are up 3.4 percent (panel d).
Manufacturing is the largest part of the phase one deal and yet has not done well
Trump's complaints about China's imports historically focused on manufacturing, which makes up 70 percent of goods covered by the purchase commitments.4 Grievances included (among others) China's large trade surpluses in goods, higher tariffs on cars, forcing US companies to turn over their technology and manufacture in China instead of exporting from America, and subsidizing sectors like steel and aluminum, which would thereby squeeze American companies out of export markets.
Overall, US exports of manufactured products to China have done poorly in 2020 (figure 2). Through September, exports remained 17 percent lower than pre-trade war levels. And relative to the purchase commitments, manufacturing was only at 56 percent of its seasonally adjusted target for 2020.
For reasons unrelated to the purchase commitments, US sales of some manufactured products to China have done well. Semiconductors and semiconductor manufacturing equipment were actually above year-to-date targets. But here the 2020 export surge may be explained by China stockpiling in anticipation of new Trump administration export controls that threaten to cut off American sales of high-tech products to Chinese companies like Huawei and SMIC. Trump's own policies could thus just as quickly end this sector's booming sales.
US sales of medical supplies to China also excelled in 2020. Exports of products needed to diagnose or treat illness related to COVID-19—such as testing kits and pharmaceuticals—have also kept pace with the estimated targets (see again figure 2). But here too the likely explanation was increased Chinese demand caused by an unanticipated pandemic. These sales could also just as quickly disappear once the health situation stabilizes.
But other traditionally strong US exports to China have suffered. Aircraft has historically been the largest US export to China—yet 2020 sales were only at 18 percent of year-to-date targets (see again figure 2). Following crashes of Boeing's 737 MAX airplane in 2018 and 2019, China announced it was cancelling a number of purchase orders for undelivered planes in April 2020. (More broadly, the 737 MAX was grounded, and Boeing shut down production of the model between January and May 2020.)
US sales of autos, trucks, and parts similarly dropped to only 33 percent of the year-to-date target. Before the trade war, China was the second largest export market for American vehicles. In July 2018, China retaliated against Trump's tariffs with a 25 percent tariff on US autos. American exports then fell off by more than a third, as production destined for Chinese consumers shifted to other locations, and US exports have not recovered since. Tesla, for example, announced in late 2018 it was accelerating construction of a new plant in Shanghai and would shift production for Chinese consumers out of the United States. The company indicated that Trump's tariffs on auto parts and China's retaliation on finished cars had made it uncompetitive to export to China from the United States. In response to these trade war policies, BMW similarly shifted some of its production from South Carolina to China.5
US agriculture plays a larger political than economic role
The US farm belt, with its outsized clout in American politics, was the most well-publicized target of China's retaliation to Trump's tariffs. Agriculture exports had been performing relatively well before the trade war, but they suffered as a result.6 Unlike hard-hit US manufacturers, American farmers received tens of billions of dollars in federal subsidies in 2018, 2019, and 2020. Agriculture makes up only 22 percent of the products covered by the purchase commitments, less than a third the size of manufacturing.
Despite a recent report from the Trump administration suggesting otherwise, US farm exports to China have not yet kept up with the phase one commitments.7 Though better than manufacturing, it took until September for farm exports to reach pre-trade war levels again (figure 3). Through September, they stood at only 66 percent of their seasonally adjusted targets. Put differently, China will need to import 62 percent of the total farm commitment in October, November, and December if it is to meet the 2020 target.
Soybeans are a good example of how the backlash erupted against Trump's trade policies. Before the trade war, they made up nearly 60 percent of US farm exports to China. China did not choose soybeans for retaliation at random. Eight of the top 10 soybean-producing states voted for Trump in 2016, including many crucial swing states.8
But ever since China imposed retaliatory tariffs, US soybean exports to China have lagged and remain at only 48 percent of the seasonally adjusted purchase target through September. One reason is that, despite US soybeans being a popular animal feed, Chinese customers turned to soybeans from Brazil and Argentina after China imposed its tariffs on US soybeans. They may now be loath to pay the costs of switching back to US suppliers, having already gone through the effort of switching once. Furthermore, Chinese demand for soybeans fell in 2019, after the devastating outbreak of African Swine Fever cut the world's largest pig herd by 40 percent, reducing demand for feed protein.
At the same time, China imported more pork to deal with local shortages caused by the outbreak, resulting in US pork exports exceeding their 2020 target (see again figure 3). China's pork imports from the rest of the world were also higher through September, by more than 400 percent than in 2017. And in one of the few parts of the phase one agreement involving policy commitments (Chapter 3), China agreed to address technical barriers to trade that had been slowing pork imports.
A few other US farm exports have also done better than their year-to-date targets, but on a smaller scale. Through September, US corn exports were above the commitment, partly because China had complied with a 2019 WTO ruling against its unfilled tariff rate quotas. (To comply, China has also increased corn imports from other countries by more than 200 percent in 2020 above 2017 levels.9) US cotton sales to China are also doing relatively better in 2020. But for Trump's aggregate metrics, even hundreds of millions of dollars of additional sales of pork, corn, and cotton pale in comparison to billions of dollars of lost soybean sales.
Sales of the lobster industry remain at only 39 percent of its year-to-date target, amid vociferous complaints by Maine residents, who lost a major customer (China) to Canada. (Beijing encouraged the shift by lowering its tariffs on lobster from Canada and other countries while retaliating with higher tariffs on US lobsters beginning in 2018.) American wheat is also struggling, and sorghum is showing signs of recovery, but it is still only at 73 percent of its year-to-date target.
Energy struggles to meet dubious targets
Energy made up only 8 percent of the total goods covered by the phase one purchase commitments, yet its targets were especially questionable. Bloomberg reported that only after the agreement was signed did the administration learn from the US industry that it lacked production capacity to fulfill the targets.10 Furthermore, evaluating the US-China trade relationship based on significantly scaled up fossil fuel exports—the targets include only crude oil, liquified natural gas, coal, and refined products—ignores global concerns over climate change.
Overall, US energy exports performed the worst of the three categories (figure 4), at less than 40 percent of year-to-date targets. (Exports of all covered energy products were higher than 2017 levels because historically the United States has not been a large exporter of energy products to China.) Hampering exports from reaching the target is the fact that the commitments are measured in dollars, not volume (e.g., barrels of oil). As a result, lower oil prices in 2020 have made it even more unlikely that China could reach the goal.
A new US approach is needed to fix ailing relationship with China
All these impediments and complications add up to a policy failure. No matter who is president, the United States needs to get China to liberalize its tariffs, remove nontariff barriers, and rationalize its subsidies and other practices that distort economic incentives. In response to Trump's trade war, China imposed additional tariffs on more than 50 percent of US exports in 2018 and 2019. It is puzzling that the legal text of the phase one agreement did not remove, reduce, or even mention the word "tariffs," and it did very little to tackle the major trade issues the United States has with China. Instead, the Trump administration has provided an excellent case study of why simple purchase commitments cannot do the trick.
Relying simply on purchase targets not only fails to address China's problematic policies that hurt Americans but also helps solidify state planning as opposed to market-determined outcomes. Specifically, because China continues to impose discriminatory retaliatory tariffs on American exporters, only its state-owned enterprises and not the Chinese private sector will increase many of the purchases to meet the commitments—the opposite of what US policymakers say they want. The agreement's targets also send signals to America's economic allies that the United States is mainly interested in China diverting imports away from their suppliers, as opposed to addressing China's problematic policies, undermining their confidence in US policy.
The United States and China need to get back to negotiating important policies that remain untouched by the phase one agreement. Trump's trade war has failed to address what really ails the US-China trade relationship. It is time for a new approach.
Methodology used in this analysis
The details of the basic approach to mapping the annual targets for 2020 to the trade data are available in Bown (2020). Additional assumptions made here involve constructing estimates for 15 separate product categories, since the agreement only provides aggregate targets for the four industries of manufacturing, agriculture, energy, and services. The approach here is to apportion the product-level targets based on the share of that product in total US exports to China in 2017 of goods covered by the purchase commitments. See also table below.
|Product||US exports as a share of total covered goods exports to China in 2017 (percent)||China's target as a share of total covered goods target in 2020 (percent)*||Schedule B codes|
|Total covered manufacturing||70.0||62.5|
|1. Aircraft, engines, parts||17.4||15.5||8800; 8411|
|2. Autos, trucks, parts||10.9||9.7||8703; 8704|
|3. Semiconductors and semiconductor manufacturing equipment||8.9||8.0||8486; 8541; 8542|
|4. Covid-19 related medical products**||2.9||2.6||2804400000; 2847000000; 3002130000; 3002140000; 3002150000; 3002190000; 3002200000; 3003100000; 3003200000; 3003900100; 3004101020; 3004101045; 3004105045; 3004105060; 3004200020; 3004200060; 3004490060; 3004600000; 3004901000; 3004909210; 3004909285; 3004909290; 3005100000; 3005900000; 3006700000; 8419200000; 8419390180; 9018113000; 9018118000; 9018120000; 9018194000; 9018195500; 9018197500; 9018310040; 9018310080; 9018310090; 9018320000; 9018390030; 9018903000; 9018907080; 9018908000; 9019200000; 9020008000; 9022120000|
|All other covered manufacturing||29.9||26.7|
|Total covered agriculture||22.0||21.0|
|9. Lobster||0.2||0.2||0306; 1605|
|All other covered agriculture||6.2||5.9|
|Total covered energy||8.0||16.5|
|12. Crude oil||4.6||9.4||2709|
|13. Liquefied natural gas||0.4||0.9||271111|
|15. Refined energy products||2.6||5.3||2710122500; 271112; 271113; 2711190020; 271311; 271312; 290511|
*Estimated based on methodology described in this appendix.
** These products are defined by USITC 2020 Covid-19 Related Goods: U.S. Imports and Tariffs, with Harmonized Tariff Schedule codes converted to Schedule B codes.
Source: Author's calculations.
1. For Americans voting early, the deal looked even worse. For example, as of the August data (released on September 25, 2020), China had purchased 51 percent of the year-to-date imports needed to meet the purchase target (see figure 1, panel a).
2. The Trump administration imposed its first tariffs on $34 billion of Chinese goods on July 6, 2018. China simultaneously retaliated. The two countries imposed tariffs collectively covering over $450 billion of bilateral trade by September 2019. The January 2020 agreement covers US exports of goods and services. Because detailed high frequency trade data for services are not available, those commitments will not be evaluated here.
3. See the appendix for the basic methodological approach. For more details, see Chad P. Bown, US-China phase one tracker: China's purchases of US goods (as of September 2020), PIIE Chart, October 26, 2020.
4. This is taken from the 2017 baseline. Taking into consideration the 2020 sectoral targets would decrease manufacturing's share of total US covered goods exports to China (to 62.5 percent) in favor of energy products. See appendix table.
5. BMW reportedly shifted production of its X3 sport utility vehicle to China. The X3 is one of the top models of US manufactured autos exported to China prior to 2018. It had been exported from the BMW's plant in Spartanburg, South Carolina.
6. There was some variation across products, of course. The Obama administration had filed two WTO disputes in late 2016 over China's policies toward corn, wheat, and rice. Unlike a number of other ongoing disputes it took over from the Obama administration, the Trump administration actually litigated these disputes at the WTO.
7. See US Trade Representative and US Department of Agriculture, Interim Report on the Economic and Trade Agreement between the United States of America and the People's Republic of China: Agriculture Trade, October 23, 2020. That report presents data on projected US exports—not actual US exports as specified in the legal text of the trade agreement—based on weekly sales data for some farm products reported to the US Department of Agriculture.
8. Illinois and Minnesota did not vote for Trump in 2016. Iowa and Ohio were the other top soybean-producing states in 2019 that had voted for the Democratic candidate (Barack Obama) in 2012 but switched to Trump in 2016. The other states in the top ten that voted for Trump in 2016 are Nebraska, Indiana, Missouri, Kansas, North Dakota, and South Dakota.
9. 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.
10. Jennifer A. Dlouhy, Shawn Donnan, and Nick Wadhams, "Big Oil Warned Trump Team China Trade Deal Was Unrealistic," Bloomberg, February 12, 2020.
The data underlying this analysis are available here.