Unappreciated hazards of the US-China phase one deal
Hexuan Li and Eva Zhang provided outstanding data assistance, and Melina Kolb and William Melancon assisted with graphics.
The centerpiece of President Donald Trump’s much anticipated “phase one” trade agreement with China, signed January 15, is a commitment by Beijing to import an additional $200 billion worth of American goods and services over the next two years. Trump is certain to cite that pledge time and again in his re-election campaign. Many experts were, of course, quick to note that China’s promised purchases are bound to fall short. In fact, a close look at the data—presented below—shows that the numbers are even more unrealistic than first believed.
That matters, because with unrealistic export targets, the deal may be doomed from the start. Other beneficial aspects of Chinese commitments in the agreement could be put in peril. Even worse, hostilities might renew, leading to a reescalation of trade tensions currently on hold.
Fortunately for the White House, the deal is structured so that the evidence of shortfalls in those Chinese purchase pledges won’t be clear until after the presidential election in November 2020. But anyone who cares about establishing a stable and productive relationship between the world’s two largest economies should be concerned.
The phase one deal has many novel aspects. At the signing ceremony, for example, Trump’s US trade representative (USTR), Robert Lighthizer, respectfully reminded everyone that the United States and China have “two very different economic systems” and that new “trade rules and practices” must “allow us both to prosper.”
But many of these new rules and practices amount to a step backward for the global trading system. Notably, the agreement’s Chapter 6 on Expanding Trade tackles only the one-sided problem of insufficient US exports to China. It does so without mentioning “tariffs.” This was surprising, of course, because Trump’s trade war resulted in Beijing imposing retaliatory tariffs that continued to cover 56.7 percent of US exports to China as of the deal’s signing.
The $200 billion purchase pledge reflects an alternative “managed trade” approach to expanding US exports. Rather than tackling specific Chinese tariff or nontariff barriers, the agreement mostly emphasizes outcomes. China must meet specific US export targets or face the consequences.
The real problem with managed trade is that it may divert, rather than expand, international commerce. For example, China could purchase more American soybeans by cutting back on imports of oilseeds from Brazil. At the same time, Beijing may choose to import significantly less of the tens of billions of dollars of American exports that are not covered by the legal agreement. In the end, mismanaging trade could hurt unwitting American companies as well as third countries.
Sound complicated? What follows is a closer analysis of the actual agreement and exploration of some of the unintended consequences of managing trade.
What the agreement commits China to buy is unrealistic
Chapter 6 of the agreement contains legal commitments for China to make additional purchases of US exports in both 2020 and 2021 that would total $200 billion, split across the two years. China would increase its imports of only certain US goods and services by $76.7 billion in 2020 and by $123.3 billion in 2021 from the 2017 pre-trade war baseline level of $134.2 billion (figure 1).
Chapter 6 does not cover all US exports of goods and services to China. That covered sectors add up to only $134.2 billion of the $185.8 billion of total US exports to China in 2017 implies that $51.6 billion of US export products to China are not covered by this agreement on expanding trade. (More on the unintended consequences of this omission below.)
Putting the magnitude of the US export targets in perspective
China’s commitment is to increase purchases from the United States valued at $134.2 billion in 2017 to a level of $210.9 billion by 2020 and $257.5 billion in 2021.
If fulfilled, the deal would result in a 92 percent increase—a near doubling—of US exports to China of the covered products between 2017 and 2021. This works out to US export growth of 18 percent on an annualized basis over 2017-21. For perspective, US export growth to China averaged only 21 percent per year when China’s economy was booming at more than 10 percent annually over 2000-07. With China’s economy currently growing much more slowly, for reasons unrelated to the trade war, sustaining 18 percent annual export growth over a four-year period would be a challenge.
But then there are the consequences of Trump’s trade war. The actual starting point to reignite US exports is January 2020. The American companies and farmers tasked with meeting the new targets have suffered through two damaging years of tariffs since those robust export numbers of 2017. US exports to China through 2019 are currently estimated to be $20 billion lower than in 2017.
Thus, China is not being asked to increase imports from the United States by $200 billion over four years (2017-21). It’s even more unrealistic. Beijing must now suddenly increase purchases by $240 billion over two short years (2019-21)—by $96.7 billion in 2020 and by $143.3 billion in 2021.
One potential political benefit of Trump’s deal is its timing. The American public will be unlikely to evaluate whether China has even met the 2020 target until well after the November presidential election. The official US trade statistics for 2020 won’t be available until March 2021.
At the same time, the clock is ticking. Failure to meet the target could result in American retaliation. The phase one agreement also contains a unique and unprecedented dispute settlement chapter with two key elements. The first is a process of bilateral consultations if, say, China has not lived up to the targets. There is no outsourcing of disputes to third party arbitrators, as exists in other agreements, such as the World Trade Organization (WTO) or the US-Mexico-Canada Agreement (USMCA).
Ultimately, if USTR believes China has not purchased the amounts inscribed in Chapter 6 of the agreement, then the second part kicks in—a proportionate US retaliation. Again, USTR, not a neutral third party, would unilaterally determine the level of that retaliation.
While China has agreed to this process, the agreement does contain a termination clause. China can pull out of the deal if it feels USTR has pushed disputes too far. Thus, conflict outside of the agreement—a potential reintroduction of tens of billions of dollars of additional tariffs—could resume at any time.
US export targets confront economic reality
Within the overall target of hundreds of billions of dollars of additional US exports to China in 2020 and 2021, there are four explicit subtargets for covered products in the manufacturing, agricultural, energy, and services sectors. A decomposition by sector reveals why achieving the growth target is so difficult (figure 2).
Consider each of the four subtargets for 2021 against two simple forecasting benchmarks designed to highlight important countervailing macroeconomic forces. The first is if the pre-2017 US export trend had continued unimpeded through 2021 (blue dotted line). The second is if Chinese demand for imports from the United States over 2017-21 grew at the same rate as overall Chinese demand, measured by Chinese GDP growth (orange dotted line). Implicit is the charitable assumption that there was no trade war, and thus no extra Chinese retaliatory tariffs to constrain US exports in 2018 and 2019. Even then, the targets look extraordinarily ambitious.
Take the agricultural products covered by the agreement (figure 2a). The deal commits China to a $19.5 billion increase in agricultural trade from 2017 to 2021, arriving at $40.4 billion in annual US farm exports to China in 2021. Projecting trade based on Chinese economic growth between 2017 and 2021 suggests $28.8 billion of US exports of covered agricultural products in 2021, a shortfall of $11.7 billion relative to the legal agreement’s target. The alternative, simple trend projection leaves an even bigger US export shortfall of $24.8 billion in 2021.
Similar export shortfalls arise in manufacturing and energy. Of the four sectors, services may get closest to its target (figure 2d).
Aggregating the sectoral shortfalls provides little reason for optimism that the overall US export target of $257.5 billion in 2021 can be met. Forecasting based on expected Chinese economic growth results in an expected US export shortfall of $73 billion in 2021 relative to the target. Alternatively, if US exports continue on their pre-2017 trends, there is a shortfall of $99.6 billion in 2021.
Finally, consider the implications of using Chinese import statistics, as opposed to US export statistics, to evaluate these projections. The agreement text allows for possible use of either country’s reported trade statistics.
The broad qualitative pattern of results in figure 2 is unchanged when relying on Chinese import statistics. Though not unique to the US-China relationship, Chinese import statistics tend to report more trade than the US export data. The magnitudes of the expected shortfalls are a bit smaller, partly reflecting higher baseline levels of 2017 Chinese import data. Yet, the total expected shortfalls remain sizable.
China could divert imports away from other foreign sources to try to meet some US targets
To overcome these strong economic forces, China could take action. Yet, specific Chinese policy changes that would liberalize trade and reduce barriers are mostly not mentioned in the text of the agreement, especially for manufacturing and energy.
The resulting managed trade approach thus creates problems for the trading system. It incentivizes Beijing to shift purchases away from other foreign suppliers and toward the United States to overcome the shortfalls anticipated in figure 2. Trading partners are right to be wary, and their concerns vary depending on the covered product category (figure 3). 
Take oilseeds, which includes soybeans, the largest US farm export (figure 3a). US exports to China in 2017 were $13.9 billion, more than one-third of China’s imports in the sector. Expanding US soybean exports could come at the expense of countries like Brazil and Argentina. Increased US exports of cereals—sorghum, wheat, and corn—could hurt Australia, Vietnam, or Thailand. Export of seafood products like fish and lobster could pull Chinese demand away from Russia or Canada.
Or consider the manufacturing products covered by the agreement (figure 3b). US exports of motor vehicles to China in 2017 were $13 billion, roughly one-fourth of China’s imports in the sector. The United States could expand car exports at the expense of the European Union (EU) or Japan. Aircraft orders from Boeing might increase to the detriment of the EU’s Airbus. Additional US industrial machinery exports could reduce Chinese trade with the EU, Japan, and Korea, and in pharmaceuticals the EU and Switzerland could be squeezed out.
In almost all categories of energy, except refined products, the United States exports relatively little to China (figure 3c). But additional US exports of coal could affect Australia and Indonesia. More US liquified natural gas to China could hurt exporters in Australia and Qatar.
This agreement may test the exports of many countries to China, several of which are traditional US allies (figure 4). The new EU trade commissioner, Phil Hogan, has already indicated that “if there’s a WTO compliance issue, of course we will take a case.” Such a WTO dispute, if brought, would most likely confront China—and not the United States—for failing to adhere to the most-favored nation principle of nondiscriminatory treatment of imports.
Incompleteness of Trump’s managed trade deal could hurt other American exporters
Surprisingly, China apparently won’t receive credit for US export growth in goods or services not covered by the legal agreement. The legal text also does not seem to contain any penalties for China if US exports of those uncovered products were to fall. Put differently, 28 percent of US exports to China, $51.6 billion of American exports in 2017 (see again figure 1), are not covered by the official managed trade part of the agreement (Chapter 6).
Thus, in legal terms, China has little incentive to import those $51.6 billion of uncovered products from the United States in 2020 or 2021. To compensate angry trading partners aggrieved because of the trade diversion (see again figures 3 and 4), China could purchase more uncovered products from them and reduce imports from the United States. That would be painful for American companies and workers whose products Trump has chosen not to put into this agreement.
Going to such lengths may seem ludicrous, but once the door to managed trade has been opened, this is how it can work.
Even more economic and policy forces are at work against Trump’s export targets
There are at least five other economic and policy-related reasons to be skeptical of a sudden massive US export boom to China.
Uncertainty remains a lingering problem. By nodding to a “phase two,” even Trump acknowledges that major areas of conflict remain untouched by his deal. Industrial subsidies and state-owned enterprises are unresolved sources of tension between “two very different economic systems.” Trump’s trade approach has created an unprecedented environment in which American companies and workers are unsure about what markets will be open to them in the future. Knowing that information affects their investment decisions. American farmers who lost their exports to China have been burned in two consecutive growing seasons. Will they really scale up purchases of seed, fertilizer, and capital equipment in early 2020 and plant for sales to China that may not be delivered until after the election?
Second, the US government is quickly further restricting its export control regime on national security grounds, limiting sales of many technology and telecommunications products for which there is considerable demand in China. China cannot increase imports of products that the United States refuses to export.
Third, the trade war may be contributing to China’s growth slowdown. In July 2019 especially, Trump was bragging about how his tariffs were hurting the Chinese economy, causing growth to slow and unemployment to rise. To meet his export targets, Trump might now like China to have a booming economy and become a massive source of demand for American products. Yet, he has imposed even more tariffs since July, most of which remain in place.
Fourth, Beijing’s tariff retaliation during the trade war led some Chinese buyers to sever ties with American suppliers and establish new arrangements with companies or farmers elsewhere. Switching back would be costly, and private Chinese companies may not be willing to do it.
Finally, intervening events have reduced Chinese demand for certain key American products. China’s need for American soybeans will be down, given the impact of the African swine fever outbreak on China’s supply of pigs, a major consumer of soybeans as feed.
Failure to manage trade expectations
Trump’s phase one deal with China was extraordinarily incomplete. US tariffs on hundreds of billions of dollars of trade remain in place. The agreement did not even touch major systemic issues behind the trade war, such as China’s subsidies and state-owned enterprises.
But even the most heralded part of the agreement—China committing to purchase an additional $200 billion of US exports over the next two years—appears precarious. Unrealistic export targets may imperil not only trade peace but also the trade progress required to tackle the legitimate concerns in the US-China trade relationship.
1. The terms of “Expanding Trade” are legally codified in Chapter 6 with the specific coverage of product categories and Harmonized Tariff Schedule (HTS) codes listed in Annex 6.1.
2. These are preliminary estimates as, at the time of writing, 2019 data were still missing for December (US goods exports to China) and the fourth quarter (US services exports to China).
3. While Annex 6.1 indicates US exports of the covered products to China should increase by $123.3 billion in 2021 from 2017 levels, the reality is China is expected to increase purchases by $143.3 billion in 2021 from 2019 levels. This is therefore an increase of 105 percent over two years, not a 92 percent increase over four years.
4. See Chapter 7 on Bilateral Evaluation and Dispute Resolution.
5. Chinese demand growth is measured by Chinese GDP growth, where growth for 2020 and 2021 is based on the International Monetary Fund’s projections in the World Economic Outlook Database (October 2019). Both Chinese GDP and US export growth for these projections are in nominal terms. The other model is based on a linear US export trend from 2012 to 2017.
6. Chapter 6, Article 6.2.6 states, “Official Chinese trade data and official U.S. trade data shall be used to determine whether this Chapter has been implemented. If an analysis of the respective trade data gives rise to conflicting assessments of whether this Chapter has been implemented, the Parties shall engage in consultations.”
7. While not shown here, estimates of figure 2 based on Chinese import data are illustrated in the Excel underlying data file posted at the bottom of this blog post.
8. In contrast, Chapter 3 on Trade in Food and Agricultural Products describes specific Chinese commitments to address certain regulatory problems in agricultural trade, including allowing in American poultry and beef products, long shut out of the Chinese market despite long-resolved concerns about American outbreaks of avian influenza and mad cow disease. Chapter 4 on Financial Services describes specific Chinese commitments to improve US exports of electronic payment services (Visa, Mastercard, and American Express), banking services, credit rating agencies, financial asset management, insurance, securities, fund management, and futures services. Chapters 1 and 2 focus on Chinese protection of intellectual property and treatment of foreign investment.
9. Chinese data on services imports from third countries are not sufficiently disaggregated to construct this analysis for the covered services subsectors described in the agreement.
10. A similar type of dispute arose in the 1980s under the General Agreement on Tariffs and Trade. The European Economic Communities filed a case against Japan after it agreed to undertake a voluntary import expansion of American semiconductors demanded by the United States.
11. See Scott Baker, Nicholas Bloom, and Steven Davis, “The extraordinary rise in trade policy uncertainty,” VoxEU.org, September 17, 2019.