This paper examines the economic implications of the tariff increases by the United States and by China during the Trump era trade dispute and the gains from their potential removal. The increases were dramatic, with the US raising tariffs on industrial products by a factor of six – with particularly large tariff increases on intermediate and capital goods – and China increasing its tariffs on US agricultural products more than five-fold. These changes distort trade and production decisions in both countries and undercut the global trading system. They resulted in substantial economic losses to each country, with import volumes reduced by 4.9% in China and 4.5% in the USA, and bilateral trade patterns were massively distorted. Their cost to the United States rose at the end of 2021, when the import expansion provisions of the Trump era Phase One Agreement expired. Negotiating the abolition of these costly and disruptive tariffs would generate substantial real income gains for both countries and help lower US consumer prices.
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited.
Copyright © The Author(s), 2022. Published by Cambridge University PressSince the consolidation of China's trade and market reforms in the early 1990s (World Bank, 1994), China's exports to the world have grown at around 12% per year, with exports to the United States growing at almost exactly the same rate. Footnote 1 After accession to the WTO in 2001, China's growth rate in exports to both the US and the world declined slightly but remained above 11% per year. This extraordinary growth performance created tensions – as well as opportunities – in a world where overall trade grew at 6.5% per year between 1993 and 2020. Part of the motivation for the prior members of the WTO in approving China's entry to the WTO in 2001 was to provide a forum in which the tensions associated with rapid trade growth by such a large country could be resolved through negotiation rather than conflict (Fewsmith, Reference Fewsmith 2001).
The growth in bilateral trade between the US and China generated substantial gains to both countries. The ability to import agricultural products, and particularly soybeans, from the United States greatly mitigated the constraints on land and water facing China (Ali et al., Reference Ali, Huang, Wang and Xie 2017), while providing important market access gains to US farmers. The ability to separate stages of production via global production sharing allowed substantial increases in production efficiency and the variety of goods produced and traded (Baldwin, Reference Baldwin 2016), and reductions in consumer prices (Amiti et al., 2020).
In early 2018, the Trump administration raised a series of arguments about China's trade and trade-related policies, including concerns about requirements for foreign investors to form joint ventures and/or transfer technology and perceived inadequacies in protection of intellectual property rights (Bown and Irwin, Reference Bown and Irwin 2019). Instead of taking these concerns to the WTO, where the US could have worked with other members, or negotiating bilaterally with China on specific issues, the Trump administration set off a tariff dispute, taking many US tariffs against China far above the levels allowed under US tariff bindings at the WTO.
On 15 June 2018, the Trump administration released a list of 1102 tariff lines covering roughly US$50 billion of imports on which the initial tariff would be increased by 25 percentage points (US Trade Representative (USTR), 2018). In response, the Chinese government raised tariffs by 25% on 659 commodities with a total value of about US$50 billion originating in the United States. On 18 June, President Trump asked USTR to identify an additional US$200 billion in imports to be subject to tariff increases of 10 percentage points. Following several rounds of threats, retaliation, and negotiations (Table 1), the United States had imposed additional tariffs on US$370 billion of imports from China, and China had reciprocated on US$140 billion. In the China–US Phase One Economic and Trade Agreement reached in January 2020, the additional US tariffs on approximately US$120 billion of Chinese goods were cut to 7.5%, and the additional tariffs imposed by China on approximately US$30 billion of US goods were cut to 2.5% or 5%. China also agreed to make substantial additional purchases of US goods for two years.
Table 1. US–China trade dispute timeline: major events since June 2018
Source: Based primarily on Bown and Kolb ( Reference Bown and Kolb 2022).
The massive tariff increases remaining after the reductions agreed in the Phase One Agreement have enormous implications for the world trading system as well as for the United States and China. They violate both the WTO's fundamental principle of non-discrimination in Article I of the GATT and each country's Article II commitment to maintain tariffs on each product below the levels permitted by its scheduled tariff bindings.
The primary policy issue addressed by this paper is the economic impacts on the United States and China of ending the Trump era tariffs. This is clearly an important issue for both countries, and for the world trading system. Flagrant violations of international treaties such as the Trump-initiated increases in tariffs above the limits to which each country had agreed risk both emulation by other countries and further escalation once these agreed limits have been breached. The analysis in this paper examines both the economic consequences of the series of tariff increases involved in the China–US trade dispute, and the slight de-escalation under the Phase One Agreement of 2020. It focuses particularly on evaluating the impacts of the tariff increases, and their potential future removal, on real incomes and trade outcomes.
A subsidiary policy question is whether the original tariffs introduced by the United States could have generated benefits to the United States even had China not retaliated. The world trading system is, after all, designed in large measure to deal with situations where individual countries seek to gain by imposing trade barriers that make them better off at the expense of their trading partners (Bagwell and Staiger, Reference Bagwell, Staiger, Bagwell and Staiger 2016).
Most of the simulation experiments examine the effects of changes in US tariffs together with the tariff changes imposed by China in retaliation. In addition, two experiments examine whether the first two increases in US tariffs – imposed when the US President thought that ‘tariff wars are good and easy to win’ – would have increased the welfare of the United States without retaliation by China (Reuters Staff, 2018). This analysis shows that, even in the absence of retaliation by China, the United States would have lost from the second of these tariff increases, which took tariffs too high to allow improvements in the terms of trade to compensate for the associated efficiency losses.
The Global Trade Analysis Project (GTAP) model (Hertel, Reference Hertel 1997; Aguiar et al., Reference Aguiar, Chepeliev, Corong, McDougall and van der Mensbrugghe 2019) is used to assess the consequences of removing the tariffs that will remain after completion of the Phase One Agreement. For several reasons, the analysis does not examine the effect of the import targets in the Phase One Agreement. A key reason is that these non-market measures fell far short of achieving their targets (Bown, Reference Bown 2022), making it particularly difficult to assess the impact of these non-transparent and unenforceable non-tariff measures. Another is that they were only for two years, after which any benefit they may have provided to the United States disappeared and the key policy question became what to do about the remaining WTO illegal Trump era tariffs.
This paper builds on earlier research on the Trump era US–China trade dispute, such as the study by Li et al. ( Reference Li, Balistreri and Zhang 2020), primarily by focusing on the new policy question of what to do about these tariffs once the Phase One import targets have expired. It builds on the prodigious contributions of Bown ( Reference Bown 2022) in documenting the policy changes and proximate impacts of the Trump era tariff impositions and China's policy responses, by estimating the impacts of dismantling these interventions. Like Flaaen and Pierce ( Reference Flaaen and Pierce 2020), it considers three key channels of effect for each country – the effect of tariff increases by its trading partner, the impact of its own tariffs on intermediate inputs, and the effects of its own tariffs on consumer goods. Unlike them, we focus only on the Trump tariff dispute with China, rather than the Trump era trade conflicts with a wide range of trading partners.
The second section of the paper reviews some of the key literature; the third section discusses research methods as well and simulation scenarios, i.e. baseline scenarios and policy simulation scenarios; the fourth section presents results of the simulation analysis of the impact of the China–US trade dispute; the fifth section considers tariff reductions under the China–US Phase One Economic and Trade Agreement; the sixth section presents results from simulations of both countries eliminating the Trump era tariffs; the seventh section presents conclusions and policy implications of the study.
China–US trade issues have attracted considerable analytical attention in recent decades. At the time of China's accession to the WTO, there was considerable concern about the adjustment pressures China's tariff cuts might place on China's economy and particularly in sectors then seen as vulnerable, such as agriculture (Huang et al., Reference Huang, Liu, Martin and Rozelle 2009) and automobiles (Francois and Spinanger, Reference Francois, Spinanger, Bhattasali, Li and Martin 2004). Every single tariff in China's tariff schedule was cut by the accession agreement and weighted average tariffs fell from 40.6% in 1992 to 12% at accession and to 6.8% after all of China's concessions were phased in (Ianchovichina and Martin, Reference Ianchovichina and Martin 2001). By contrast, US tariffs were not changed at all. The only liberalization by the United States was a very slow phase out of quotas on China's exports of textiles and clothing.
Detailed analyses of China's WTO commitments and the economic impacts of the agreement on China in Bhattasali et al. ( Reference Bhattasali, Li and Martin 2004) highlight the importance of China's accession for both China and her trading partners. At the time, there was also great concern about the possible implications for China's agriculture, industry, and services (Anderson et al., Reference Anderson, Huang, Ianchovichina, Bhattasali, Li and Martin 2004; Pangestu and Mrongowius, Reference Pangestu, Mrongowius, Bhattasali, Li and Martin 2004). WTO rules in general, and China's WTO Accession agreement in particular, are designed to reduce the risk of arbitrary changes in policy, such as tariff increases, that can be very costly and ignite multiple rounds of escalation. They also provide access to multilateral mechanisms for consultation and dispassionate resolution of disputes. Each member commits, in particular, not to raise its tariffs above the tariff bindings to which it has agreed at the WTO. Absent such commitments in the pre-WTO period, the result was frequently bursts of trade conflict in which large countries raise tariffs, their trading partners retaliate, and all sides lose (Irwin, Reference Irwin 2011).
China's own liberalization before and after accession to WTO helped to increase efficiency and to expand exports rapidly. Another factor promoting growth in exports to the United States was a reduction in trade policy uncertainty. Prior to accession, China faced a risk each year of US tariffs rising to the much higher levels introduced by Smoot and Hawley during the Great Depression. Believing this risk removed, firms made many investments associated with producing for and supplying the US market (Feng et al., Reference Feng, Li and Swenson 2017). Unfortunately, and surprisingly given the widespread perception that US labor markets are relatively flexible, the rapid growth of Chinese exports appears to have created problems of unemployment and wage reductions in many local labor markets (Autor et al., Reference Autor, Dorn and Hanson 2013). While they estimate that trade accounted for only around a quarter of the total impact and exports from China for only a portion of that, exports from China received enormous attention from researchers and policy makers.
Posen ( Reference Posen 2021) believes that much of the focus of US policy on competition from China (and from Japan a generation previously) in manufacturing reflects nostalgia for an economic past with a larger share of the workforce in manufacturing. Like Rodrik ( Reference Rodrik 2016), Posen notes that a decline in the share of manufacturing in employment is a standard feature of economic development for higher income countries. Recent World Bank data suggest that China has already reached the beginning of this decline, with employment in industry peaking at 30% in 2012 and falling to 27% by 2019. Attempting to reverse such changes through trade protection results in both overall income losses and potentially strong adverse shocks in the industries and locations forced to contract by the adverse impacts of protectionist interventions.
Clearly, another contributing factor to the trade dispute was President Trump's focus on bilateral trade deficits and his – unfortunately all too common – misconception that trade deficits are determined by trade policy (Johnson, Reference Johnson 2020). This failure to understand that the current account deficit is determined by the gap between income and expenditure led to contradictory policies, such as pairing expansionary budgetary policies that increase the current account deficit with trade policy interventions that are ineffective in reducing it.
In terms of quantitative research, the main research tool adopted for this type of issue has been the computable general equilibrium model. Some of the research was conducted before the outbreak of US–China trade dispute, examining stylized shocks such as the 45% tariff increases on Chinese goods proposed by Trump before he took office (Haberman, Reference Haberman 2016). Rosyadi and Widodo ( Reference Rosyadi and Widodo 2018) used the GTAP model to analyze this policy change and concluded that, if 45% additional tariffs were imposed on manufacturing products by both countries, the real incomes of the United States and China would decline by US$80 billion and US$94 billion, respectively. Similarly, Dixon's ( Reference Dixon 2017) research on 45% additional tariffs on manufacturing products by both countries showed that the real incomes of the United States and China would decline by 0.7% and 2.5%, respectively. Research by Guo et al. ( Reference Guo, Lu, Sheng and Yu 2018) shows that, if the US imposed a 45% tariff on all Chinese goods, US real wages would decline by 0.66%; if China imposed a reciprocal tariff, they would decline 0.75%. Bouët and Laborde ( Reference Bouët and Laborde 2018) analyzed the impact of hypothetical trade disputes between the United States and emerging market countries such as China and Mexico and showed that even if China and Mexico did not retaliate, additional 35% tariffs imposed by the US on all goods from China and Mexico (except energy goods) would reduce US social welfare by 0.1%; if reciprocal retaliation by China and Mexico were considered, US social welfare would be reduced by about 0.4%.
Some studies have considered policy simulations closer to the actual outcome of the US–China tariff dispute. Cui et al. ( Reference Cui, Zhu, Song and Zheng 2018) concluded that if the United States imposed a 25% import tariff on Chinese goods of US$50 billion, and China imposed an equivalent tariff, real gross domestic product (GDP) in the United States and China would fall by 0.02% and 0.13%, respectively. If the US imposed 10% tariffs on an additional US$200 billion of Chinese goods and China imposed 10% tariffs on an additional US$60 billion in US goods, the real GDP of the US and China would be reduced by 0.03% and 0.29%, respectively (Zhou and Shi, Reference Zhou and Shi 2019). Zhou et al. ( Reference Zhou, Zheng and Lu 2019) used a policy simulation similar to Cui et al. ( Reference Cui, Zhu, Song and Zheng 2018). They concluded that total imports of agricultural products in China decreased by 6.6% after China and the United States increased their tariffs in 2018, and that US imports and exports of agricultural products would decline by 2% and 5.4%, respectively.
Li et al. ( Reference Li, Balistreri and Zhang 2020) focused on the tariff increases after the China–US Phase One Agreement and found that these additional tariffs would reduce the welfare of the US and China by 0.2% and 1.7%, respectively. Li et al. ( Reference Posen 2021) considered the regional impacts within China and found that if the US had unilaterally imposed 25% tariffs on all products already announced, China's eastern coastal provinces would have been particularly adversely affected. Bellora and Fontagné ( Reference Bellora and Fontagné 2019) considered a much wider set of Trump-initiated trade barriers against the EU, Canada, and Mexico, and other trading partners as well as China. Their study focused on the adverse impacts for US industry given the prevalence of tariff increases on intermediate goods, a topic to which we return.
Freund et al. ( Reference Freund, Maliszewska, Mattoo and Ruta 2020) considered the impacts of China's agreement under the Phase One Agreement to expand imports from the United States. They concluded that these would discriminate against other countries, and particularly other developing countries, unless China used this as an opportunity for broader liberalization. Feenstra and Hong ( Reference Feenstra and Hong 2020) focused on the agricultural import targets under the Phase One Agreement. They estimated a non-homothetic demand system for agricultural imports into China and showed that the most efficient way for China to achieve the Phase One Agreement's import targets would be to mimic the effect of an import subsidy. If China's agricultural imports did not otherwise grow from their 2017 values, they concluded that the subsidies would need to be 42% and 59% to meet the 2020 and 2021 targets, respectively. These subsidies would divert agricultural imports to the USA and away from other suppliers. This trade diversion would be especially strong for Australia and Canada, followed by Brazil, Indonesia, Malaysia, Thailand, and Vietnam. Estimating the impact of these import expansions now seems a low priority since these provisions did not bind (Bown, Reference Bown 2022) and expired at the end of 2021, while the Trump era tariffs remain in the absence of a negotiated agreement.
The existing studies provided important background for this study but leave many important questions unanswered. Most were conducted before the full features of the Trump era trade dispute were known and so are based on potential tariff changes or on tariffs at earlier stages of the negotiations. Many focused on the impact of the trade dispute on macroeconomic aspects such as overall economic welfare but lacked analysis on the gains and losses to specific industrial sectors, and particularly the differential impacts between agriculture and industrial trade. This paper uses detailed information on the actual trade dispute tariffs and traces the impacts for key sectors of removing them.
The Global Trade Analysis Project (GTAP) is a global network of researchers and policy makers conducting quantitative analysis of international policy issues (Walmsley et al., Reference Walmsley, Aguiar and Narayanan 2012). A key component of the GTAP project is the CGE model known as the GTAP Model, documented in the GTAP book (Hertel, Reference Hertel 1997). In this paper, the GTAP model is applied to create baseline scenarios and policy simulations.
The data underlying the model are primarily drawn from the GTAP10 database developed by Purdue University (Aguiar et al., Reference Aguiar, Chepeliev, Corong, McDougall and van der Mensbrugghe 2019). This database includes final expenditures, input–output data, bilateral trade data, trade barriers, and relevant behavioral parameters. Because the China–US trade dispute broke out in 2018, intensified in 2019, and was attenuated by the China–US Phase One Economic and Trade Agreement reached in 2020, it is useful to update the database before assessing the impacts of each of these policy changes. In this study, 2018, 2019, and 2020 are selected as three baseline years. Using the dynamic recursion method proposed by Walmsley et al. ( Reference Walmsley, Dimaranan and McDougall 2000), population, GDP, capital, and labor data in GTAP database are projected and simulations are used to update the endogenous variables in the database.
The 141 countries (regions) and 65 sectors in the GTAP 10 database are aggregated into 10 regions and 16 sectors consistent with the research purpose. The countries/regions are: China, the USA, Korea, Oceania, ASEAN, the Middle East, EU, Brazil, Other South America, and Rest of the World; and the sectors: Rice, Other grains, Oil crops, Fruits/vegetables/nuts, Tobacco, Meat, Aquatic products, Dairy products, Other agricultural products, Fuels, Chemical and rubber products, Transport equipment, Metal and products, Electronics, Other industrial products, and Services.
Because the GTAP database represents the pre-trade dispute situation, we must introduce the trade dispute tariffs, prior to their removal. The simulation scenarios are based on the commodities subject to additional tariffs announced by the Chinese government and US government, including three baseline scenarios and six policy simulation scenarios.
In the policy simulations, we focus primarily on the changes in each round, where the US increased tariffs and China retaliated. But we also consider the outcome where the US imposed tariffs but China did not retaliate, since this allows us to see whether each country's tariff increases were welfare-reducing for itself, as well as for its trading partner (Table 2).
Table 2. Baseline scenarios and policy simulation scenarios
Aguiar's ( Reference Aguiar 2019) correspondence between HS codes and GTAP industry sectors was used with bilateral trade weights to calculate the tariff increases for the GTAP sectors. These were, in turn, aggregated to the 16 industrial sectors used in the simulations (Table 3). The lists of commodities subject to additional tariffs were obtained from the Tariff Policy Commission of the State Council of China and the Office of the US Trade Representative. Bilateral trade data for 2017 were obtained from the General Administration of Customs of China Footnote 2 and the United States Census Bureau Footnote 3 . The tariff data of the two countries and the trade data for China are at the 8-digit level. The trade data of the US are at the 10-digit level, and we aggregated them to the 8-digit level to calculate weighted average tariffs. While trade weighted averages are downward biased when assessing the effects of liberalization because high tariffs reduce the weight on the goods to which they apply (Laborde et al., Reference Laborde, Martin and van der Mensbrugghe 2017), this problem does not apply in this case because the trade weights reflect the low tariffs prior to the tariff dispute.
Table 3. China–US bilateral import tariffs for each industry before and after trade disputes
Note: Trade-weighted average tariffs computed from commodity level tariff and trade data, weighted by each exporting country's exports to the world before trade disputes.
Figure 1 shows that the US increased its average import tariffs on Chinese agricultural and industrial products to 9.2 and 9.5% in 2018, and further increased them to 23% and 19% in 2019. China's increases in bilateral tariffs on industrial products resulted in more than a tripling of the average, from 5.3% to 17.1%. China increased its tariffs on agricultural imports much more than on industrial products, to 26.7% in 2018, and to 32.1% in 2019, taking them to almost six times their initial level. In 2018, the US responded by introducing subsidies to farmers whose products were adversely affected by these tariffs (Bown and Kolb, Reference Bown and Kolb 2022). These payments helped farmers in the short term but added additional economic costs to those measured in this paper and are seen as likely to exacerbate future trade tensions and to precipitate future WTO disputes (Glauber, Reference Glauber 2020).
Figure 1. China–US bilateral import tariff increases, percentage points. Source: GTAP database and the list of additional tariffs published by China and the United States.
A surprising feature of the US tariff increases is that they were much larger on intermediate and capital goods than on final consumer goods – a distinction that reinforces the concerns of Bellora and Fontagné ( Reference Bellora and Fontagné 2019) about potentially adverse impacts on global value chains. Using the Classification by Broad Economic Categories (BEC), bilateral trade between the United States and China was divided into three categories: consumption goods, intermediate goods, and capital goods. Table 4 shows that the focus of US tariff increases on China was on intermediate goods, for which tariffs increased by 22.2%, and capital goods, for which tariffs rose by 20.9%, by contrast with an 11.6 percentage point rise for consumer goods. By contrast, China increased import tariffs on consumption goods, intermediate goods, and capital goods by 23.6%, 14.9%, and 5.4%, respectively.
Table 4. Bilateral tariff increases by broad economic category, % points to 2019
Note: The correspondence between HS and BEC was used with bilateral trade weights to calculate the tariff increases for intermediate and final goods.
The strong focus on intermediate and capital goods in the US tariff increases is unusual because business interests are usually more effective than consumer interests in opposing tariff increases. Consistent with this, it appears that some firms were successful in obtaining ‘exclusions’, but that the large majority of these expired by December 2020 (Flaaen et al., Reference Flaaen, Langemeier and Pierce 2021). The unusual focus of the Trump tariffs on intermediate and capital goods presumably reflected a desire to avoid being blamed for increases in the costs of prominent consumer goods – many of which are made in China for American companies such as Apple. Research by several eminent American trade economists (Amiti et al., Reference Amiti, Dai, Feenstra and Romalis 2020) concluded that the US tariff increases were, in fact, passed through entirely to US importers and customers. Another possible reason for the relatively large increases in tariffs on intermediate and capital goods is presumably that these tariff increases were not a result of pressure from US firms, but rather based on ideological arguments and misperceptions, such as the belief by senior Trump trade advisor Peter Navarro that no country would retaliate against US tariff increases because the US is the world's biggest market (Bown and Irwin, Reference Bown and Irwin 2019, p. 128) or that tariff increases would reduce the US current account deficit. The strong focus of the US tariff increases on intermediate and capital goods has more serious adverse impacts for the competitiveness of US firms than China's pattern of tariff increases.
The estimated impacts of the tariff increases in 2018 and 2019 on the macro-economies of the United States and China are presented in Table 5. The initial tariff increases by the United States and by China reduce the volume of China's GDP by 0.2%, and that of the United States by 0.04%, as increases in the costs of intermediate inputs cause the volume of output to decline. The 2019 intensification of the trade dispute had further negative impacts on economic output in both China and the United States. Compared with baseline scenario 2, the volume of GDP in China and the United States would be reduced by 0.3% and 0.15%, respectively. The real income measures based on Equivalent Variation (EV) decline by more than the GDP volume measures because they incorporate the welfare costs of distortions on consumer goods as well as the impacts on output resulting from higher prices of intermediate goods.
Table 5. The impact of the China–US trade dispute on macroeconomic outcomes
Source: Simulation results from the GTAP model.
