AN ECONOMIC ANALYSIS OF THE US-CHINA TRADE …

Staff Working Paper ERSD-2020-04

19 March 2020

_________________________________________________________________________

World Trade Organization

Economic Research and Statistics Division _________________________________________________________________________

AN ECONOMIC ANALYSIS OF THE US-CHINA TRADE CONFLICT

Eddy Bekkers and Sofia Schroeter Economic Research and Statistics Division

World Trade Organization

Manuscript date: 26 February 2020

_______________________________________________________________________________ Disclaimer: The opinions expressed in this article should be attributed only to its authors. They are not meant to represent the positions or opinions of the WTO and its Members and are without prejudice to Members' rights and obligations under the WTO. Any errors are attributable to the authors.

Address for correspondence: Eddy Bekkers, Rue de Lausanne 154, 1211 Geneve, Switzerland. email: eddy.bekkers@. We want to thank Marc Bacchetta, Coleman Nee, Robert Koopman and Robert Teh for useful comments and suggestions. Disclaimer: The opinions expressed in this article should be attributed only to its authors. They are not meant to represent the positions or opinions of the WTO and its Members and are without prejudice to Members' rights and obligations under the WTO. Any errors are attributable to the authors.

AN ECONOMIC ANALYSIS OF THE US-CHINA TRADE CONFLICT

Eddy Bekkers and Sofia Schroeter Economic Research and Statistics Division

World Trade Organization

ABSTRACT This paper provides an economic analysis of the trade conflict between the US and China, providing an overview of the tariff increases, a discussion of the background of the trade conflict, and an analysis of the economic effects of the trade conflict, based both on empirics (ex post analysis) and on simulations (ex ante analysis). Bilateral tariffs have increased on average to 17% between the US and China, and the Phase One Agreement signed in January 2020 between the two countries only leads to minor reductions in the tariffs to 16%. The trade conflict has led to a sizeable reduction in trade between the US and China in 2019 and is accompanied by considerable trade diversion to imports from other regions, leading to a reorganization of value chains in (East) Asia. The simulation analysis shows that the direct effects of the tariff increases on the global economy are limited (0.1% reduction in global GDP). The impact of the Phase One Agreement on the global economy is even smaller, although the US is projected to turn real income losses into real income gains because of the Chinese commitments to buy additional US goods. The biggest impact of the trade conflict is provoked by rising uncertainty about trade policy and the paper provides a framework to analyze the uncertainty effects. Keywords: Trade conflict, Economic simulations, Trade effects of tariffs JEL-codes: F12, F13, F14, F17

Address for correspondence: Eddy Bekkers, Rue de Lausanne 154, 1211 Geneve, Switzerland. email: eddy.bekkers@. We want to thank Marc Bacchetta, Coleman Nee, Robert Koopman and Robert Teh for useful comments and suggestions. Disclaimer: The opinions expressed in this article should be attributed only to its authors. They are not meant to represent the positions or opinions of the WTO and its Members and are without prejudice to Members' rights and obligations under the WTO. Any errors are attributable to the authors.

1 INTRODUCTION

Since 2018 the United States and China have imposed various restrictive measures on trade flows between the two countries, of which increases in tariffs have been most prominent. Tariff increases by the United States on Chinese imports led to a quick response of China, also raising its tariffs on American imports. Although the trade conflict seemed to escalate in the Fall of 2019 with new tariff increases, at the end of 2019 the two countries agreed on a truce, cancelling some of the announced tariff increases and rolling back some of the earlier increases in tariffs. The truce led to the so-called Phase One Agreement signed in January 2020.

This paper analyzes the economics of the trade conflict, discussing the economic background of the trade conflict, providing an overview of the different measures taken and exploring the economic effects, going into the impact already generated (ex post analysis) and the impact expected in the future (ex ante analysis), paying special attention to the possible effects through uncertainty about trade policy. The economic effects are examined both reviewing the literature and doing own analysis. For the ex post analysis, we explore how trade flows between China and the US have changed over the last two years and we analyze patterns of trade diversion.

For the ex ante analysis, the WTO Global Trade Model (GTM), a recursive dynamic computable general equilibrium (CGE) model, is employed. There is extensive discussion about the potential effects of the trade tensions on the global economy through rising trade policy uncertainty and we incorporate the framework proposed by Handley and Limao (2017) into the Melitz firm heterogeneity version of the GTM in simplified form. In particular, rising trade policy uncertainty is modelled through an increase in the discount rate triggering an increase in fixed export costs for given sunk export costs. In the ex ante analysis, we develop four scenarios to be able to evaluate the impact of the Phase One Agreement and to assess the impact of rising trade policy uncertainty.

Since the start of the trade conflict between the US and China the two countries have raised tariffs substantially on each other's exports, from 2.6% to 17.5% on Chinese imports into the US and from 6.2% to 16.4% on US imports into China. The Phase 1 Agreement between the US and China reduced the tariffs on Chinese imports into the United States to 16%. To limit the scope of the paper, it focuses on the trade tensions between the US and China.1

The tariffs on Chinese imports have been motivated with at least four arguments: (i) address bilateral trade imbalances; (ii) make tariffs more reciprocal; (iii) bring back manufacturing jobs; (iv) address Chinese policies with negative spillovers such as poor IP protection, subsidies of state-owned enterprises, and forced technology transfer. The economic underpinning of the first three arguments will be discussed in detail in the paper.

Although trade flows from China to the US still increased in 2018 because of frontloading by about 7%, exports from China to the US fell substantially in the first three quarters 2019 for tariffed goods, by 13%. US exports to China fell by about 1% in 2018, accelerating to a reduction of more than 25% in the first three quarters of 2019. While Chinese exports to the US still increased in 2018 by 7% because of frontloading, they dropped in the first quarter of 2019 by about 13%.

The ex post analysis also shows that there was significant trade diversion towards imports from other trading partners. Four East Asian countries (Japan, South Korea, Taiwan, and Viet Nam) exported less to China and more to the US, in particular in the electrical equipment sector. This indicates that value chains in East Asia are reorganized in response to the trade conflict. The main insight from the empirical literature so far on the trade conflict is that there has been complete passthrough of higher import tariffs on Chinese goods to tariff inclusive import prices.

The ex ante analysis shows that the direct impact of the tariff increases in the trade conflict are expected to be limited, in the range of a 0.1% reduction in global GDP. Furthermore, the impact of

1 In the quantitative exercise we also take into account the tariffs imposed by the US on steel and aluminum since 2018. Additional US tariff measures are under discussion in 2019. The US government is considering raising tariffs on imports of cars and car parts (about 300 billion dollars) to 25% with the same legal justification as the steel and aluminum tariffs, national security under Section 232 of the Trade Expansion Act. Such tariffs would in turn provoke retaliation. On October 18, 2019 the US also raised tariffs on imports of a value of 7.5 billion on various products from the European Union as retaliation measures for the WTO-inconsistent subsidies in the EU for Airbus. This will likely trigger retaliation by the EU on US imports (on 12 billion import value) because of WTO-inconsistent subsidies to Boeing.

the Phase One Agreement on global GDP is an order of magnitude smaller and hardly changes the direct impact on the global economy. Remarkably, the analysis shows that the real income effects for the US change from negative into positive with the Phase One Agreement, because of the Chinese commitment to buy additional US goods. The positive impact of the trade conflict on other countries because of beneficial trade diversion is projected to turn negative as a result of the Chinese commitments to source more imports from the US.

The analysis discussed so far only takes into account the direct effects and does not consider possible effects through uncertainty. Our next step is to include an analysis of the effects of uncertainty. Determining the uncertainty impact is a complex undertaking and so our results should be interpreted with caution. Nevertheless, our analysis indicates that the impact through uncertainty is much larger than the direct effect. Measures of trade uncertainty have increased manifold since 2018. Taking into account uncertainty effects, the loss in global GDP would be much more considerable and increase to between 0.34% and 0.50%.

The projected effects reported in other ex ante studies are close to the simulated effects with the WTO Global Trade Model, except for the distribution of losses across regions. In the simulations including uncertainty we project larger negative effects for the US because uncertainty about future policy according to the Trade Uncertainty Indicator is mostly related to trade between the US and its trading partners.

The paper is organized as follows. Section 2 gives an overview of the tariff measures taken so far. Section 3 goes into the impact on trade so far (ex post analysis). Section 4 discusses the different arguments given to underpin the tariff measures against China thus providing background of the trade conflict. Section 5 goes into the impact of the trade conflict through trade policy uncertainty. Section 6 then presents the results of simulations with the WTO Global Trade Model and Section 7 concludes by relating the simulation results with the earlier simulation results on the effects of a global trade war in Bekkers and Teh (2019).

2 OVERVIEW OF MEASURES

The United States started to increase tariffs on Chinese imports in March 2018 and the response of the Chinese government followed shortly afterwards. Table 1 shows the value of trade affected by each of the rounds of tariff increases, while Figure 1 shows the evolution of the average tariff rates on US imports from China and Chinese imports from the US. The tariff increasing measure affecting most additional trade was taken on September 24, 2018. The US levied additional tariffs of 10% on roughly $200 billion Chinese imports, which increased further to 25% on May 10, 2019.

The average tariffs imposed by the US on imports from China have increased substantially since the start of the trade conflict, from a 2.6% MFN tariff rate to a tariff rate of 17.5% on the 1st of September 2019. Initially the US announced a further extension of the scope of the tariff increases, which would have raised average tariffs to 24.4% on December 15. However, because of the truce in the trade conflict this increase was never implemented and instead, as part of the Phase 1 Agreement between the US and China, average tariffs fell to 16%. The tariff increases implemented on the 1st of September 2019 on about 120 billion of consumer goods will be halved, from a 15% increase to 7.5% increase.

Figure 1 shows that the tariffs China imposed on imports from the US have increased from 6.2% in January 2018 to 16.4% in September 2019. The planned further increases to 20.7% by December 2019 was not implemented. At the same time China reduced MFN tariffs on other trading partners, corresponding to an average reduction in tariffs of about 5%.

The average tariff rates reported in Figure 1 differ slightly from the numbers reported in Bown (2019), because of differences in weighting schemes. Whereas Bown (2019) weights the tariff averages by US exports to the world and China's exports to the world in 2017, we weight by bilateral imports. Thus, the averages are only weighted by trade that is affected by tariff increases. As a comparison we use a similar reference group weighting method as Bown (2019) (Figure A1 in the Annex) based on US total imports from the world and China's total imports from the world. Figure A1 confirms the sensitivity of the averages to the weighting scheme. The United States started the trade conflict targeting primarily intermediate goods but has moved to the imposition of tariffs on almost all goods, including consumption goods. Since the US imports disproportionately fewer intermediate goods from China than the rest of the world and disproportionately more final goods,

bilaterally weighted average tariff rates tend to be lower for the first tariff measures in 2018 and higher for the tariff measures at the end of 2019 compared to tariff rates weighted by reference groups.

Table 1 ? Trade Coverage of Tariff Increases

US imports from China

Measure

Trade in US$ billion

Mar. 23, 2018

3.60

Jul. 6, 2018

33.44

Aug. 23, 2018

14.31

Sep. 24, 2018

198.87

May 10, 2019

-

Sep. 1, 2019

130.15

Dec. 15, 2019

161.88

Cumulative trade in US$ billion

3.60

35.81

49.52

237.07

237.07

336.36

487.35

Chinese imports from the US

Measure

Trade in US$ billion

Apr. 2, 2018

2.97

Jul. 6, 2018

42.52

Aug. 23, 2018

14.11

Sep. 24, 2018

53.39

Jun. 1, 2019

52.85

Sep. 1, 2019

28.67

Dec. 15, 2019

44.80

Cumulative 2.97

45.37

57.45

107.91

107.74

110.96

113.59

trade in US$

billion

Note: Values of trade coverage are based on trade data in 2017. The numbers in the row "Cumulative trade" are

not equal to the sum of the numbers in the row "Trade", because in some cases additional tariffs cover the same

HS-lines. Cumulative trade of China for June 1, 2019 is smaller than for September 2018, because China has

eliminated some MFN tariffs in between.

The descriptive analysis of the tariff increases in this and the following section is based on two datasets. First, the Trade Data Monitor (TDM) provides quarterly trade data on US imports and Chinese imports with all partners at 8-digit tariff-line level from the first quarter of 2017 to the third quarter of 2019. Second, the WTO-Secretariat constructed a database that includes MFN tariffs, applied tariffs and additional tariffs due to all measures taken between January 2018 and December 2019. Products affected by additional tariffs are identified by information given on governmental websites and notifications to the WTO. 2

As part of the Phase 1 Agreement China committed to raise its imports from the United States by 200 billion over 2020 and 2021. This promised Chinese increase in imports from the US is sizeable. Relative to the pre-trade conflict level of 2017, imports should increase according to the Phase 1 Agreement by 200 billion over two years (32 in agriculture, 52 in energy, 78 in manufacturing, and 38 in services). This increase of 100 billion on average per year constitutes an almost 50% increase compared to the pre-trade conflict imports of about 210 billion in 2017. (153 billion of merchandise imports and 57 billion of services exports from the US to China, both based on WTO data).

China has not committed to a reduction of its additional tariffs on US imports and can choose its own way to realize the 200 billion increase in imports from the US. Although the agreement states "that purchases will be made at market prices based on commercial considerations", an increase in imports by 50% will require intervention by the Chinese government likely leading to sizeable trade diversion away from exports from other countries.

2 The data sources mentioned in Section 2 have been complemented by data from Chad Bown () on China's MFN tariff cuts on 1st of May 2018, 1st of July 2018, 1st of November 2018 and 1st of January 2019. Calculations of the tariff rates in Figure 1 take the MFN tariff cuts into account.

Figure 1 ? Evolution of Average Tariff Rates

25

20

15

10 6.2

5 2.6

6.6 2.7

Average Tariff Rates

24.4

11.1

13.8

14.515.3

17.5 16.4

8.8

20.7 16.4

16.0

4.3

0 Jan-18 Apr-18

Jul-18

Sep-18

Jun-19

Sep-19 Dec-19 Jan-20 (planned)

USA China

Note: average tariff rates on US imports from China and Chinese imports from the US are weighted by total imports from China and total imports from the US in 2017 respectively Source: Author's calculations based on trade data from the Trade Data Monitor and tariff data collected by the WTO secretariat.

3 TRADE EFFECTS 2018-2019 (EX POST ANALYSIS)

3.1 Changes in Imports

This section analyzes the impact of the tariff measures taken by the US and China so far on trade between the two countries. Table 2 gives an overview of the import values and corresponding percentage changes for three categories of products between 2017 and 2019, products subject to tariffs, products not subject to tariffs, and all products. In 2018 imports from China still increased relative to 2017, for tariffed as well as non-tariffed products, which seems to indicate that anticipation effects of higher tariffs played a role. However, in 2019 trade between the US and China has fallen substantially. US imports of products that were affected by tariff measures decreased by up to 13.5% while Chinese imports of tariffed products declined even stronger, namely by 25%.

In order to analyze the role of the tariff increases in the decrease of imports and to determine the importance of anticipation effects, we look at the development of total imports between the US and China over time for different categories of products in greater detail. In Figure 2 changes in imports are calculated as the percentage change compared to the same quarter in the previous year. The changes are calculated for five categories of products: products that are not at all affected by tariffs during the trade conflict and products that are affected in one of the four periods displayed in the graphs.

Table 2 ? Trade between the US and China between 2017 and 2019

US imports from China

Chinese imports from the US

2017

2018

2019

2017

2018

2019

Non-tariffed products

import value perc change

10.78 --

12.47 15.69

13.09 4.99

42.20 --

47.43 12.41

41.42 -12.68

Tariffed products

import value perc change

507.81 --

542.92 6.91

469.68 -13.49

107.46 --

106.62 -0.78

79.39 -25.54

All products

import value

518.59

555.39

485.89

149.66

154.06

123.72

perc change

--

7.10

-12.51

--

2.94

-19.69

Note: Import values are in US$ billion; trade values for 2019 are rescaled by the total value in 2017 over the sum of the first three quarters of 2017 since trade data for the fourth quarter of 2019 were not available yet. Source: Author's calculations based on data sources mentioned in section 2

The graph on the left of Figure 2 shows that imports of Chinese goods into the US that were affected by the first tariff measures in spring 2018 (red dashed line), declined strongly from the second quarter of 2018 onwards. Products affected by the tariff increases in August 2018 (grey dotted) show the same pattern with an almost immediate negative effect on imports. Considering products that faced a tariff increase in September 2018 and further increases in May 2019 (yellow dashed) show a rather delayed effect due to frontloading which means that Chinese exporters sold more in anticipation of even higher tariffs in 2019. Non-tariffed products had overall stable import patterns. Taken as a whole, the strongest effects of the additional tariffs occurred in 2019 with decreases of imports of up to 40%.

The graph on the right of Figure 2 displays the impact on Chinese imports. Effects are more difficult to disentangle compared to US imports, since products are overlapping considerably between the time periods. However, it can be noted that products affected by tariff increases in spring 2018 (red dashed) had the strongest negative decline in imports at the end of 2018 while imports of products that were not affected by tariff increases seem to have increased to compensate for declined imports of affected goods. Alternatively, the increase in imports of non-tariffed goods could be due to anticipation effects of export restrictive measures by the United States.

Overall, the analysis in Figure 2 indicates that tariff increases between the US and China have already had strong effects on bilateral trade between the two countries with the responses displaying intuitive patterns, including anticipation effects.

Figure 2 ? Percentage changes in the value of imports

Source: Author's calculations based on data sources mentioned in section 2

Looking at economic sectors and making the broad distinction between products that were hit by a tariff measure in 2018 and non-tariffed products confirms the impact on imports described above. The left panel of Figure 3 clearly indicates that US imports of tariffed products decreased while imports of almost all non-tariffed products increased. The sector with the strongest decline is petroleum and coal products where imports decreased by up to 75%. In other sectors such as machinery, electrical equipment and agricultural products imports decreased by up to 40%. The picture for Chinese imports (right panel Figure 3) is again more blurred because of the general slowdown of the Chinese economy. There has been an overall decrease of imports from 2018 to 2019. However, the effect on imports of tariffed products is clear: imports in all sectors have faced a negative growth rate with imports of petroleum, metals and agricultural products declining the most. Remarkable is the increase in imports of Computer, Electronic, and Optic Equipment, which can be related to the fact that Chinese companies wanted to make sure to be able to import crucial inputs in the production process before the US would take export restricting measures. Figure 3 ? Percentage changes in imports by sectors

Notes: Percentage changes are calculated as the difference in imports between the first half of 2018 and the first half of 2019; products are first aggregated by GTAP sectors, then further aggregated by 14 broad GTAP sector categories Source: Author's calculations based on data sources mentioned in section 2

 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download