Part IV - Responses of China’s Trading Partners

13 The Politics of Preferential Trade Liberalization with China

Damian Raess

This chapter investigates citizens’ attitudes toward preferential trade liberalization with China using original survey data in an advanced economy. I focus on Switzerland as an empirical case due to data availability and the fact that the landlocked, continental European country is one of the few advanced democracies to have concluded a preferential trade agreement (PTA) with China. I consider the Sino-Swiss PTA as an instance of a North-South PTA between countries with significant differences in factor endowments and social standards, and I will assess individual attitudes in North-South trade relations against the benchmark of North-North preferential trading among similar countries, using the case of the bilateral agreements between Switzerland and the European Union (EU).

I am interested in whether the impact of the distributive effects of international trade on preferences over PTAs is conditional upon the type of trade (North-South vs. North-North). Moreover, I am interested in whether the belief that ‘deep’ economic integration requires the strengthening of compensatory welfare policies – the ‘embedded liberalism’ compromise redux (Ruggie, 1982) – mitigates the uneven distributional effects of North-South vs. North-North preferential trade liberalization among the losers of international trade. Relatedly, I ask what type of compensation policies – belief in passive and protective labor market policies or actual social investment policies – increases support for North-South PTAs among the losers of trade. Lastly, given that North-South trade has strong distributional consequences and raises issues of social standards in developing countries, I focus on the role of individuals’ ideological (i.e., partisan) self-identification on PTA preferences.

The argument is fourfold. First, the losers from international trade in advanced economies (i.e., low-educated, low-skilled, low-status, and poor individuals) will less strongly support North-South PTAs, such as PTAs with China, than they do North-North PTAs, such as PTAs with the EU. Second, the belief that compensation policies legitimize deep integration among losers will more strongly increase their support for North-South than for North-North PTAs. Third, compared to social investment policies (i.e., training), belief in compensatory welfare and protective labor market policies more strongly increases support for North-South PTAs among globalization losers. Finally, North-South PTAs, in particular preferential liberalization involving a developing country with low social and human rights standards such as China, will drive a stronger wedge between left- and right-leaning individuals than North-North PTAs. The empirical analysis corroborates these expectations.

The chapter makes three contributions to the literature in international political economy. Firstly, while scholars have extensively examined individual preferences over free trade, we know relatively little about attitudes toward PTAs. This is among the first studies that investigates the determinants of mass attitudes toward preferential trade liberalization with China, and among the first that places such analysis in a more general context by theorizing about individual preferences over North-South versus North-North PTAs (though see Chiang et al. 2013). Secondly, most studies focus on mega-regional or hypothetical PTAs, and find that political factors such as sympathy/antipathy toward particular countries or security concerns outweigh explanations based on the income effects of trade (Spilker et al., 2018; DiGiuseppe and Kleinberg, 2019; see also Naoi and Urata, 2013; Jungherr et al., 2018; Dür, 2019). Focusing on the more common bilateral PTAs and on real PTAs, I find that respondent characteristics related to the distributional effects of trade liberalization, such as education, skills, financial situation, and social status, strongly explain public attitudes toward trade agreements. Lastly, the findings have noteworthy implications for the backlash against globalization. National-populist reactions to the China shock are a big part of the globalization backlash (Autor et al., 2013; Feigenbaum and Hall, 2015; Colantone and Stanig, 2018b). The results provide strong evidence for the role of domestic compensation policies in the form of passive/protective labor market policies in increasing support for trade liberalization with China among globalization losers.

The chapter is organized as follows. The first section identifies blind spots in the literature that the study begins to fill. The second introduces the argument. The third and fourth sections present the empirical analysis. The fifth provides a discussion while the final section concludes.

I Literature

Over the past twenty years, research on individual preferences over free trade has fast grown in number and sophistication, with seemingly no end in sight. One prominent line of inquiry and debate has been whether economic self-interest explains trade policy preferences (e.g., Scheve and Slaughter, 2001; Mayda and Rodrik, 2005; Mansfield and Mutz, 2009). A second, and related line of inquiry has been whether compensatory welfare policies help legitimize trade openness (e.g., Hays et al., 2005; Ehrlich and Hearn, 2014). These studies have largely focused on individual preferences over free trade (or protectionism) in general. Jungherr et al. (2018) interrogate whether individual-level preferences for the general principle of free trade and for specific trade agreements are similar. Using public opinion data from Germany, the authors show that while the standard economic and non-economic models perform well in explaining public opinion on trade, contextual factors unrelated to trade are more useful in explaining support for the Transatlantic Trade and Investment Partnership (TTIP).

In fact, we know little about the determinants of citizens’ support for and opposition to PTAs in democracies (Baccini, 2019: 82). The few studies on ‘real-world’ PTAs tend to focus on atypical PTAs, namely the mega-regional trade agreements, which, due to their sheer scope and geopolitical considerations, may introduce some bias in the literature’s research findings. For example, Naoi and Urata (2013) examine individual attitudes toward the Trans-Pacific Partnership Agreement (TPP) in Japan and find that partisanship rather than economic self-interest is the most relevant determinant of TPP support. Jungherr et al. (2018) find that postures toward transatlantic cooperation and predispositions toward the role of interest groups in politics as well as toward domestic market regulation correlate with support for TTIP. Dür (2019) demonstrates that the argument that the TTIP would allow foreign firms to sue domestic governments had a large negative effect on public opinion, while the promise of job creation hardly mattered. Rankin (2004) shows that national identity rather than economic interest explains American and Canadian opinion on the North American Free Trade Agreement.

Experimental studies probing preferences for hypothetical PTAs also conclude about the primacy of non-economic over economic factors. DiGiuseppe and Kleinberg (2019) investigate the role of security considerations. Focusing on American respondents, they find that PTAs involving political rivals and those promising diminished international influence reduce support for PTAs, and that security concerns diminish the degree to which information about the projected economic effects influences individuals’ preferences for PTAs. Spilker et al. (2018) study individual attitudes toward PTAs in Costa Rica, Nicaragua, and Vietnam. They show that sympathy/antipathy toward particular countries matters more than economic considerations. Finally, in his study on mass support for potential PTA partners in Canada, India, and the US, Tuxhorn (2019) finds limited support for economic preferences derived from the factor endowment trade model.

II The Argument

North-South trade is based on differences in the factor endowments of countries (Heckscher-Ohlin model). Rich countries have a comparative advantage in the production of goods that make intensive use of capital and skilled labor, their abundant factors. Conversely, poor countries, being well endowed in (semi- and) unskilled labor, will specialize production in goods that make intensive use of low-skilled labor. North-South trade is predominantly inter-industry trade, a type of trade that has sharp distributional consequences. According to Stolper-Samuelson, under the assumption of costless inter-sectoral mobility of production factors, trade benefits the owners of the abundant factors and harms the owners of the scarce factors. In rich countries, the losers are the semi- and low-skilled workers, whereas the winners are capital owners and high-skilled workers. Opposition to trade in advanced countries should therefore be concentrated among low-skilled workers (and unions representing them), particularly opposition to trade liberalization with developing countries richly endowed with unskilled (manual) labor, such as China.

By contrast, North-North trade is predominantly intra-industry trade, driven by customer preferences for differentiated goods. Compared to inter-industry trade, adjustment costs are likely lower as jobs lost due to customers shifting to foreign suppliers may be offset to a large degree by the job-enhancing expansion in foreign demand for similar, differentiated goods produced domestically. A typical example is the case of the Frenchman buying a VW car and the German buying a Renault, whereas prior to trading they bought domestically produced cars. In short, North-North trade has less strong distributional effects than North-South trade and should therefore be less strongly opposed by the losers of international trade in rich countries. Hence:

Hypothesis 1: Among the losers of trade in advanced economies (i.e., low-educated, low-skilled, low-status, and poor individuals), the level of support is lower for North-South PTAs, in particular preferential trade liberalization with China, than for North-North PTAs.

The post-war compromise of “embedded liberalism” was premised on the idea that domestic welfare compensation helps legitimize an open economy (Ruggie, 1982). Governments committed to free trade provided insurance and other transfers to compensate those who lost economically from increased trade. Historical-comparative analysis has demonstrated an association between economic openness and welfare spending in advanced economies (Cameron, 1978; Rodrik, 1998), at least up until the 1990s (Busemeyer, 2009).

Research has corroborated the microfoundations of the ‘compensation’ thesis. For a sample of thirteen advanced economies, Hays et al. (2005) show that individuals employed in import-competing sectors strongly oppose trade, while unemployment insurance and active labor market programs moderate their opposition. Similarly, using a survey experiment in the United States (US), Ehrlich and Hearn (2014) show that knowledge of the Trade Adjustment Assistance (TAA) program, introduced to the experimental group as a federal program providing expanded unemployment insurance and job retraining opportunities to workers who lose their jobs, results in higher support for free trade among low-income individuals. Based on Swiss survey data, Walter (2010) finds that globalization losers, in particular low-skilled workers, are more likely to experience economic insecurity, demand welfare compensation, and vote for social-democratic parties.

While actual compensatory policies or the belief that compensation buys support for globalization should increase the losers’ support for (preferential) trade liberalization, I expect this effect to be stronger in the North-South than in the North-North trade context. North-South trade and trade agreements have strong labor market effects (Hakobyan and McLaren, 2016), stronger than North-North trade. Trade integration with low-wage countries is thus more likely to generate demands for compensation (Burgoon, 2001), and, conversely, compensatory policies are more likely to increase support for North-South than for North-North PTAs and trade.

It is well established that trade with China has strongly affected labor markets in advanced economies. Autor et al. (2013) report negative effects of Chinese import competition on employment levels and wages in local labor markets in the US, but also higher social transfer payments (see also Autor et al., 2016). Thewissen and Van Vliet (2019) generalize the finding of depressing employment effects in sectors facing Chinese imports to eighteen OECD countries while showing that low-skilled workers endure most adjustment costs as production work by these workers is substituted by Chinese exports. Since preferential trade liberalization tends to be associated with more trade flows among trading partners, losers in advanced economies should be particularly prone to oppose PTAs with China, and belief that compensation enables openness should reduce such opposition.

Hypothesis 2: Among the losers of trade in advanced economies, the belief that increased government compensation (i.e., the strengthening of employment protection, unemployment insurance, and the protection of working hours) enables ‘deep’ economic integration more strongly increases support for North-South PTAs, in particular preferential trade liberalization with China, than for North-North PTAs.

Compensation measures take various forms. Perhaps the most obvious government policies to offset job losses due to increased imports are income support measures. In cross-sectional analysis, the generosity of unemployment benefits correlates with support for free trade (Hays et al., 2005). In the US, the losers of international trade are more inclined to support trade-related unemployment insurance than the winners (Ehrlich, 2010).

Globalization is a source of job insecurity (Rodrik, 1998; Scheve and Slaughter, 2004). The losers report higher levels of fear of losing their jobs (Walter, 2010). Rules that make it costly for employers to fire their workers may prevent job losses and therefore reduce actual or perceived economic insecurity associated with trade openness. Employment protection regulation and unemployment insurance may thus be substitutes in how they increase support for trade liberalization among globalization losers. Alternatively, because wages tend to be sticky due to income policies or collective agreements, labor market adjustments to increased trade competition might occur through longer working hours at a given wage level. European labor markets facing rising Chinese imports might be particularly susceptible to responding in this way. Europeans typically work short hours (while caring about work-life balance issues) whereas the Chinese work long hours, not least because many of them have a preference for working overtime hours to increase their income. Rules and regulations protecting standard working hours in advanced economies might thus also condition attitudes towards (preferential) trade liberalization. In short, not just compensatory welfare institutions but also protective labor market policies ought to moderate the losers’ opposition to trade, particularly North-South trade and PTAs.

What about social investment policies such as occupational training? The evidence is mixed. In the cross-national context, Hays et al. (2005) show that spending on active labor market programs is associated with higher individual support for trade, while Hays (2009) finds that it does not increase support for trade among those employed in tradeable sectors. The results are also inconclusive as to whether participation in TAA training programs improves the employment outcomes of participants (Decker and Corson, 1995; Reynolds and Palatucci, 2012). Regarding training, while it might upgrade skills, help career advancement, and/or sustain wage increases, it does not necessarily reduce the risk of job losses due to increased trade competition. With job retraining programs, meanwhile, individuals run the risk of social downgrading, as they might be required to accept employment at lower skill levels. Finally, government spending on active labor market policies remains a small fraction of total social spending in advanced economies. In short, we have the basis for the third hypothesis.

Hypothesis 3: Among the losers of trade in advanced economies, the belief that increased compensatory welfare and protective labor market policies enables ‘deep’ economic integration more strongly increases the support for North-South PTAs, in particular preferential trade liberalization with China, than social investment policies (i.e., training).

Partisanship matters for trade policy with right-wing parties and individuals being more supportive of free trade than their left-wing counterparts are (Scheve and Slaughter, 2001; Milner and Judkins, 2004). North-South trade integration is likely to raise concerns about poor labor and human rights and low environmental protection in developing countries, opening up a cleavage between left parties that are critical of North-South PTAs and right parties that are supportive. PTAs with China are a case in point, given the extensive violations of fundamental labor rights in China. Moreover, Beijing objects to the inclusion of far-reaching labor provisions in its PTAs, which explains why PTAs signed by China include rather shallow provisions (LABPTA dataset; Raess and Sari, 2018, 2021). This constitutes a major obstacle to the conclusion of bilateral trade agreements between China and the big trading powers. In those circumstances, existing labor-related level playing field rules in Chinese trade agreements are unlikely to increase support for trade liberalization with China in advanced economies, because they do not provide the kind of ex-ante reassurance mechanism or the fair trade norms that help legitimize trade openness (Bastiaens and Postnikov, 2020). Left-oriented parties and individuals should be particularly concerned, given their stance on labor and human rights issues. Hence the final hypothesis:

Hypothesis 4: In advanced economies, the effect of partisanship – with left-leaning individuals being less supportive than right-leaning individuals – is stronger for North-South PTAs, in particular preferential trade liberalization with China, than for North-North PTAs.

III Data

I use representative data from Swiss individuals to test my arguments. I use an original dataset combining data from the 2015 wave of the Measurement and Observation of Social Attitudes in Switzerland survey (MOSAiCH; Ernst Stähli et al., 2015) and from my topical module on Switzerland’s foreign economic relations, which, after a nationally competitive bid, was included in the MOSAiCH survey. While the former provided data on most independent and control variables, the latter provided the survey questions for the dependent variables and a few key independent and control variables.

While the selection of the Swiss case is data related, it is well suited to examine public attitudes toward trade cooperation with China. Switzerland’s leading trade partner in Asia and its third largest partner worldwide is China. Importantly, Switzerland is one of the few advanced economies, and the first major European country, to have signed a PTA with China.¹ The Sino-Swiss PTA was concluded in July 2013, after only two years of negotiations, and came into force on July 1, 2014. The Swiss proponents hailed the deal as their most important agreement since the 1972 PTA with the EU (Dadush et al., 2020). While Switzerland was keen to gain preferential market access to the Chinese market before its main competitors (i.e., EU member states, but also the US), “China saw Switzerland as a gateway into Europe and viewed the trade agreement as an important test case, one that might soften the EU’s traditional reluctance to negotiate with China” (Dadush et al., 2020). Given the salience of the Sino-Swiss trade deal and the timing of the data collection (the survey was administered between February and July 2015), choices made by survey respondents are likely to accurately capture choices they would make in real-world situations (Hainmueller et al., 2015).²

The selection of the EU as the benchmark against which to compare China in terms of attitudes toward trade cooperation is equally fitting. The EU is by far Switzerland’s largest trading partner. Bilateral treaties govern the economic relations between Switzerland and the EU. The 1972 PTA created a free trade zone for industrial products. After the Swiss people voted against joining the European Economic Area in 1992, the government proposed bilateral negotiations (Linder, 2011). The Bilaterals I (1999) cover agreements in seven areas (free movement of persons, technical barriers to trade, public procurement, agriculture, research, civil aviation, and overland transport), the Bilaterals II (2004) in nine (processed agricultural products, statistics, pensions, education/vocational training, environment, media, fight against fraud, taxation of savings, and Schengen/Dublin on internal security). So-called ‘flanking measures’ were adopted in June 2004 in connection with the effective introduction of the free movement of persons. These measures aim to prevent the undercutting of wages and social conditions. They include the reinforcement of collectively agreed on minimum wages, the facilitation of extension clauses (making it easier to declare collective labor agreements generally binding), and the hiring of labor inspectors. In the face of opposition from the national-conservative party, the government, and the business community needed the trade unions’ support to win the referendum, which they did with a comfortable majority of two-thirds (Linder, 2011; Oesch, 2011).

Switzerland has historically enjoyed good political relations with both China and the EU. The Swiss government was one of the first Western countries to recognize the People’s Republic of China on January 17, 1950. It has maintained broad-based bilateral relations with Beijing since the 1980s, including a high-level annual human rights dialogue since 1991. Geographic, cultural, and economic reasons explain Switzerland’s good relations with its European neighbors.

In a comparative perspective, Switzerland is a coordinated market economy, although it has hybrid features not least due to liberal labor market institutions (Hall and Soskice, 2001; Mach and Trampusch, 2011). The Swiss tend to work long hours, collective bargaining coverage is moderate, the protection of employment is low, while unemployment benefits are relatively generous (Emmenegger, 2011; Scruggs et al., 2017).

(i) Dependent Variables

Pro-EU PTA is an ordinal variable measuring respondents’ opinions on “The bilateral agreements with the EU have reinforced the exchanges of goods and services between Switzerland and the EU. To what extent are you favorable to this policy led by the Confederation?” Answers are recorded on a 5-point scale, as follows: 1 = “very unfavorable”; 2 = “somewhat unfavorable”; 3 = “neither/nor”; 4 = “somewhat favorable”; and 5 = “very favorable.” Higher values thus indicate greater support for preferential trade liberalization between Switzerland and the EU. The responses by individuals who expressed no opinion or who did not answer were coded as missing.

Pro-CN PTA measures attitudes towards bilateral trade liberalization between Switzerland and China. Survey respondents were asked their opinion on the following question: “In 2013, Switzerland signed a trade agreement with China, reinforcing the exchanges of goods and services. To what extent are you favorable to this policy led by the Confederation?” Answers are recorded on the same 5-point scale, while “don’t knows” and “no responses” were treated as missing values.

Pro-PTA, the main dependent variable in the statistical analysis, is generated by stacking the data, specifically by stacking the variables Pro-EU PTA and Pro-CN PTA. I created a dummy variable labeled China PTA, which takes a score of 1 for all observations of the variable Pro-CN PTA (0 otherwise; that is, 0 if the observations pertain to the variable Pro-EU PTA). I interacted this dummy with the variables for globalization losers to test Hypothesis 1 and, in a triple interaction model, I interacted these interaction terms with variables measuring compensatory welfare and labor market policies to test Hypotheses 2 and 3.

(ii) Independent Variables

1 Losers of International Trade

I considered four groups of losers in advanced economies, defined by their level of education, their skills, their financial situation, and their (self-declared) social status. Operationalizing the losers of international trade by way of individuals’ educational attainment and skill profile is commonplace in the literature. Considering (subjective) measures of individuals’ finances and social status is less common though no less important as globalization has led to stagnating or eroding incomes for some workers and, if not outright downward social mobility, the perception of declining (relative) social status.

Low education equals 1 if the respondent has completed less than twelve years of full-time schooling. According to this measure, low-educated individuals are high-school dropouts. Manual workers are individuals whose job involves primarily physical work such as building, making, carrying, caring, etc. Individuals in the following major groups of the International Standard Classification of Occupations (ISCO-8) are considered manual workers: clerical support workers (group #4), services and sales workers (#5), skilled agricultural, forestry, and fishery workers (#6), craft and related trades workers (#7), plant and machine operators and assemblers (#8), elementary occupations (#9), and armed forces occupations (#0).³

Poor financial situation gets a score of 1 if individuals answer “bad” or “very bad” to the survey question “How do you rate your current financial situation?” (0 otherwise). Finally, Low status equals 1 if individuals self-identify as belonging to groups that tend to be toward the bottom of society (i.e., on the reverted 1–10 scale, individuals who place themselves on the scores 6–10). It should be noted that the bivariate correlations between these variables are moderate at best.⁴

2 Compensatory Welfare and Labor Market Institutions

In the drop-off questionnaire, I asked respondents if they believe that government compensatory policies help legitimize deep economic integration. The survey question probes individual opinions regarding various compensatory welfare and protective labor market policies, as follows (English translation): “Would you be more in favor of strengthening foreign investment in Switzerland and trade in goods and services with other countries, if the Swiss government took measures to (1) discourage companies from laying off their employees? (2) protect the weekly working time? (3) strengthen the right to unemployment?” The questions are framed as support for moving beyond existing levels of compensation and openness, which is why earlier I referred to belief in the ‘embedded liberalism’ compromise redux. Pro-globalization compensation (or simply compensation), a dummy, equals 1 if respondents say they would be “probably more” or “more” in support (scores of 3 and 4 on the 4-point scale) for each of the policies (0 otherwise). While I aggregated the measures for presentational reasons, I consider disaggregated results in the discussion section.

The measure of social investment policies is actual training. Upskilling through training might thwart trade-related job losses. Training is a dummy that equals 1 if the individual received any training that allowed him or her to improve skills in the past 12 months.

3 Individuals’ Party Orientation

Included as a control, Right ideology measures self-placement on the ideological left-right scale (0–10). Right-wing individuals prefer free markets and should therefore be supportive of PTAs. In the model that tests the variegated effect of partisanship on North-South vs. North-North PTAs (Hypothesis 4), I use a trichotomous variable where Left equals the low scores (0–3), Centre the middle (4–6), and Right the high (7–10).

(iii) Control Variables

The baseline model controls for socio-demographic determinants of trade policy. Female is a dummy that equals 1 if respondents are women. If anything, women are more protectionist than men (Mansfield et al., 2015). Age, measured in years, captures inter-generational differences in attitudes toward trade due for instance to socialization processes. Previous studies have included the effect of rural-urban residence in models of trade policy preferences (Mayda and Rodrik, 2005; Mansfield and Mutz, 2009). Urban residence is a categorical variable with five residential types, ranging from a farm or house in the countryside to a big city. Urban residence should positively correlate with support for PTAs.

Next, I include controls for pre-existing cultural and ideological dispositions. Swiss-German is a dummy that equals 1 if the survey was administered in Swiss-German or Romansh (0 if French or Italian). It captures differences in trade opinions and stereotypes of trading partners that may be rooted in different Swiss cultures (and/or related sub-national ideologies of political economy) as well as differences that may exist in the phrasing of the survey questions in each national language. As Swiss-German citizens more strongly embrace economic liberalism than their French-Swiss or Swiss-Italian counterparts, being Swiss German should positively correlate with Pro-PTA attitudes. Nationalism is measured as opinions on the item “open borders and the intermingling of populations endanger important characteristics of Swiss culture,” with higher values on the 5-point scale indicating stronger nationalist sentiment. Nationalism should negatively correlate with Pro-PTA attitudes. Trust in the EU is an ordinal variable measuring respondents’ trust in the EU. Individuals who trust the supranational institution should generally be more supportive of PTAs, particularly PTAs with the EU. As mentioned above, I include a measure of individuals’ party orientation.

I also control for alternative explanations derived from competing trade theories. Foreign business share is the respondent’s answer, measured on a 4-point scale, to the question “How much does your company or employer export its production or engage in economic activities abroad?” Capturing within sector firm heterogeneity (Melitz, 2003), foreign business share ought to positively correlate with Pro-PTA attitudes. Mobility is a dummy that takes the value of 1 if respondents have ever moved to improve their employment prospects. Mobility influences attitudes toward trade policy (Mansfield et al., 2015; Owen, 2013: 729). Specifically, it should be positively associated with support for preferential trade liberalization. Moreover, individuals as consumers may have different preferences over trade than they do as producers of goods and services (Baker, 2005). Trade lowers prices records opinions on “one must open the borders to trade so that prices fall,” measured on a 5-point scale. Individuals who believe that trade openness lowers consumer prices are likely to be more supportive of PTAs.

Finally, Media exposure measures the frequency with which respondents use the media (including television, newspapers, radio, and the internet) to obtain political news or information (measured on a 7-point scale). As a proxy for cognitive capacity, we expect individuals who regularly use media to be more favorable toward PTA.

The baseline models include ten industry dummies that control the respondents’ sector of employment. Industry characteristics, such as export orientation or import competition, correlate with industry-level preferences over trade policy. I restrict the sample to the working-age population in order to include only individuals directly affected by the distributional effects of trade. The models are estimated using an ordered probit estimator, with robust standard errors clustered by industry. Using ordered logistic models instead does not change the results. Descriptive statistics for all the variables are shown in Appendix Table A1.

IV Results

Figure 13.1 shows the average scores in favor of North-North vs. North-South PTAs among globalization losers. As can be seen in Panels A-D, the level of support is consistently higher for the PTA with the EU than for the PTA with China, providing initial support for Hypothesis 1.

Figure 13.1 Globalization losers and support for EU PTA vs. China PTA.

Figure 13.2 displays the difference in support for North-North vs. North-South PTAs for low-status workers as a function of believing in pro-globalization compensation (Panels A and B) and of being recently trained (Panels C and D). The graphs show that belief in compensation more strongly increases support for trade liberalization with China (+14.8%, from 2.98 to 3.42) than with the EU (+7.7%, from 3.66 to 3.94), whereas training does not seem to affect support for PTAs with either the EU or China. This provides initial support for Hypotheses 2 and 3.

Figure 13.2 Support for EU PTA vs. China PTA among low-status individuals.

Figure 13.3, finally, shows the average support for North-North vs. North-South trade liberalization among left-, centre- and right-leaning individuals. Left-wing individuals hold more negative views of preferential trade with China than their right-wing counterparts do, while the same does not hold for preferential trade with the EU, suggesting first evidence for Hypothesis 4.

Figure 13.3 Partisan orientation and support for EU PTA vs. China PTA

Table 13.1 shows the multivariate regression results for the interaction models involving one of the measures of globalization losers at the time and the China PTA dummy with Pro-PTA as the dependent variable. The statistically significant controls perform as expected, raising our confidence in the model specification. Female respondents and individuals expressing nationalist sentiment are less likely to support PTAs. By contrast, individuals living in urban centers (likely due to their more cosmopolitan worldviews), those trusting in the supranational body of the EU, those working in a firm doing business abroad, those who believe trade is good for consumers, and those more exposed to the media are more likely to support PTAs.

Table 13.1 Globalization losers and support for North-North vs. North-South PTAs

	(1)	(2)	(3)	(4)
	DV = Pro-PTA
Low education	−0.294***
	(0.082)
Manual worker		−0.527***
		(0.078)
Poor financial situation			−0.390**
			(0.186)
Low status				−0.295**
				(0.141)
China PTA	−0.803***	−0.808***	−0.734***	−0.763***
	(0.100)	(0.081)	(0.054)	(0.073)
Low edu*China PTA	0.165
	(0.118)
Manual*China PTA		0.206*
		(0.113)
Poor*China PTA			0.613***
			(0.162)
Low stat*China PTA				0.211
				(0.150)
Female	−0.423***	−0.392***	−0.415***	−0.407***
	(0.074)	(0.087)	(0.081)	(0.087)
Age	0.006*	0.004	0.005	0.005*
	(0.003)	(0.003)	(0.003)	(0.003)
Urban residence	0.099**	0.098**	0.114**	0.121**
	(0.044)	(0.050)	(0.050)	(0.051)
Swiss-German	0.216**	0.181	0.194*	0.178
	(0.109)	(0.111)	(0.118)	(0.115)
Right ideology	0.036	0.029	0.038	0.033
	(0.026)	(0.024)	(0.028)	(0.025)
Nationalism	−0.282***	−0.272***	−0.292***	−0.288***
	(0.042)	(0.041)	(0.041)	(0.042)
Trust in the EU	0.124**	0.110**	0.130**	0.115**
	(0.053)	(0.048)	(0.053)	(0.056)
Foreign business share	0.150*	0.146**	0.170**	0.176**
	(0.081)	(0.072)	(0.074)	(0.078)
Mobility	0.072	0.074	0.110	0.092
	(0.077)	(0.083)	(0.073)	(0.076)
Trade lowers prices	0.219***	0.226***	0.214***	0.214***
	(0.038)	(0.038)	(0.037)	(0.035)
Media exposure	0.062**	0.052*	0.073***	0.067**
	(0.028)	(0.028)	(0.026)	(0.028)
Observations	980	978	980	972
Pseudo R-squared	0.130	0.139	0.129	0.130

Robust standard errors in parentheses; *** p < 0.01, **p < 0.05, *p < 0.1.

The main results show a consistent pattern across the various measures of globalization losers. First, the coefficients for globalization losers are all negative and statistically significant. In the presence of the interaction terms, these coefficients indicate the difference in support for EU PTA between globalization losers and winners. In line with the factor model of international trade, we find that low-educated individuals, manual workers, poor individuals and low-status individuals are less supportive of preferential trade liberalization with the EU than their respective counterparts are. Second, the coefficients for China’s PTA are negative and highly significant, indicating that on average individuals are less supportive of the PTA with China compared to the PTA with the EU. Third, the coefficients for the interaction terms are positive and, in two cases out of four, statistically significant. This indicates that the gap in support between globalization losers and winners tends to be narrower for the PTA with China than for the PTA with the EU. While this finding runs against the predictions from international trade theory, it may hide a more complex pattern whereby compensatory policies condition the relationships between being a globalization loser and support for North-North vs. North-South PTAs (see below).

In any event, is the level of support for PTAs among globalization losers lower for PTA with China than for PTA with the EU, as suggested by Hypothesis 1? Answering this question requires post-estimation analysis. Based on Model 4, keeping all other variables are their mean values, low-status workers are approximately 50 per cent less likely to be very favorable (score of 5) to a PTA with China than a PTA with the EU. The corresponding figures for low-skilled individuals, manual workers, and poor individuals are 53, 54, and 13 per cent, respectively. In short, the empirical analysis corroborates Hypothesis 1.

Table 13.2 shows the results for the triple interaction models with belief in compensation as one of the constituent terms. With one exception, the coefficients for the measures of globalization losers remain negative and statistically significant, and so do the coefficients for PTA with China. Interestingly, the coefficients of the interaction terms globalization losers*China PTA are now negative (with one exception) and statistically insignificant, which is more in line with trade theory. The interpretation of this (double) interaction is that the gap in support between globalization losers and winners who do not believe in compensation tends to increase as one moves from PTA with the EU to PTA with China, although the effect is not statistically significant. The triple interactions are with one exception as expected positive and highly significant, suggesting that the narrowing gap in support between losers and winners tends to be reduced more strongly for PTA with China than PTA with the EU as one moves from non-belief to belief in compensation.

Table 13.2 The conditional effect of compensatory welfare policies on support for North-North vs. North-South PTAs

	(5)	(6)	(7)	(8)
	DV = Pro-PTA
Low edu*China PTA*Compensation.	−0.096
	(0.294)
Manual*China PTA*Compensation.		0.567***
		(0.182)
Poor*China PTA*Compensation.			1.366**
			(0.648)
Low stat*China PTA*Compensation.				0.849***
				(0.215)
Low education	−0.554***
	(0.145)
Manual worker		−0.520***
		(0.163)
Poor financial situation			0.388
			(0.741)
Low status				−0.321*
				(0.165)
China PTA	−0.793***	−0.629***	−0.655***	−0.608***
	(0.117)	(0.083)	(0.071)	(0.099)
Low education*China PTA	0.215
	(0.150)
Manual*China PTA		−0.116
		(0.186)
Poor*China PTA			−0.184
			(0.530)
Low stat*China PTA				−0.214
				(0.206)
Compensation	−0.298	0.015	0.032	−0.012
	(0.189)	(0.154)	(0.104)	(0.127)
Low education*Compensation	0.498**
	(0.225)
Manual*Compensation		0.017
		(0.253)
Poor*Compensation			−0.952
			(0.813)
Low stat*Compensation				−0.046
				(0.220)
China PTA*Compensation	−0.124	−0.422***	−0.270*	−0.429**
	(0.281)	(0.143)	(0.158)	(0.172)
Female	−0.424***	−0.389***	−0.421***	−0.405***
	(0.078)	(0.094)	(0.083)	(0.094)
Age	0.003	0.001	0.003	0.003
	(0.003)	(0.003)	(0.004)	(0.003)
Urban residence	0.130**	0.127**	0.140**	0.151**
	(0.053)	(0.058)	(0.057)	(0.061)
Swiss-German	0.156	0.130	0.159	0.129
	(0.103)	(0.107)	(0.104)	(0.104)
Right ideology	0.036*	0.036*	0.041*	0.030
	(0.022)	(0.020)	(0.023)	(0.019)
Nationalism	−0.282***	−0.284***	−0.304***	−0.293***
	(0.050)	(0.049)	(0.048)	(0.052)
Trust in the EU	0.122*	0.122**	0.136**	0.123*
	(0.067)	(0.059)	(0.065)	(0.070)
Foreign business share	0.106	0.107	0.133	0.128
	(0.105)	(0.097)	(0.102)	(0.107)
Mobility	0.084	0.086	0.111	0.102
	(0.078)	(0.091)	(0.084)	(0.081)
Trade lowers prices	0.207***	0.199***	0.196***	0.192***
	(0.039)	(0.036)	(0.040)	(0.038)
Media exposure	0.051*	0.054*	0.062**	0.058**
	(0.030)	(0.030)	(0.026)	(0.029)
Observations	837	837	837	831
Pseudo R-squared	0.143	0.149	0.140	0.146

Robust standard errors in parentheses; ***p < 0.01, **p < 0.05, *p < 0.1.

Post-estimation analysis reveals the magnitude of the hypothesized effects. Based on Model 8, low-status individuals are about 65 per cent more likely to strongly support (score of 5) PTA with China when they believe in compensation than when they do not. By contrast, they are about 5 per cent less likely to strongly support PTA with the EU when they believe in compensatory policies than when they do not. For manual workers (Model 6), the corresponding (rounded) figures are +30 and +3 per cent, respectively, while for poor individuals (Model 7) they are +20 and –55 per cent. Only in the case of low-educated individuals do we observe that belief in compensation less strongly increases support for PTAs with China than with the EU (–3 and +20 per cent, respectively).⁵ Overall, these findings provide support for Hypothesis 2.

In Table 13.3, we replace our measure of compensatory welfare and labor market policies with our measure of social investment policy. The results show that none of the double interaction terms of interest (globalization losers*training) and none of the triple interaction terms are statistically significant. This suggests that training does not condition the support for North-South PTAs (or North-North PTAs for that matter) among globalization losers. The main results from Tables 13.2 and 13.3 lend support for Hypothesis 3.

Table 13.3 The conditional effect of social investment policies on support for North-North vs. North-South PTAs

	(9)	(10)	(11)	(12)
	DV = Pro-PTA
Low edu*China PTA*Training	−0.269
	(0.504)
Manual*China PTA*Training		0.127
		(0.351)
Poor*China PTA*Training			0.140
			(0.603)
Low stat*China PTA*Training				0.240
				(0.229)
Low education	−0.422**
	(0.207)
Manual worker		−0.567***
		(0.147)
Poor financial situation			−0.942
			(0.919)
Low status				−0.220
				(0.238)
China PTA	−1.113***	−0.949***	−0.814***	−0.858***
	(0.307)	(0.174)	(0.080)	(0.095)
Low education*China PTA	0.426
	(0.386)
Manual*China PTA		0.187
		(0.298)
Poor*China PTA			0.113
			(0.482)
Low stat*China PTA				0.141
				(0.217)
Compensation
Training	−0.236	−0.167*	−0.116	−0.114
	(0.183)	(0.101)	(0.083)	(0.092)
Low education*Training	0.108
	(0.286)
Manual*Training		−0.067
		(0.208)
Poor*Training			0.543
			(1.060)
Low stat*Training				−0.158
				(0.225)
China PTA*Training	0.395	0.200	0.166	0.141**
	(0.285)	(0.134)	(0.104)	(0.063)
Female	−0.424***	−0.389***	−0.416***	−0.416***
	(0.081)	(0.083)	(0.093)	(0.087)
Age	0.006***	0.004**	0.006***	0.006***
	(0.002)	(0.002)	(0.002)	(0.002)
Urban residence	0.089*	0.092*	0.111*	0.116*
	(0.054)	(0.056)	(0.060)	(0.063)
Swiss-German	0.307**	0.278*	0.260*	0.271*
	(0.152)	(0.148)	(0.156)	(0.153)
Right ideology	0.016	0.007	0.017	0.014
	(0.038)	(0.033)	(0.039)	(0.036)
Nationalism	−0.279***	−0.267***	−0.291***	−0.288***
	(0.041)	(0.044)	(0.046)	(0.041)
Trust in the EU	0.105**	0.098**	0.118**	0.112**
	(0.054)	(0.047)	(0.049)	(0.049)
Foreign business share	0.181*	0.180**	0.208**	0.198**
	(0.104)	(0.089)	(0.096)	(0.101)
Mobility	0.039	0.036	0.097	0.058
	(0.108)	(0.113)	(0.106)	(0.105)
Trade lowers prices	0.241***	0.248***	0.233***	0.235***
	(0.038)	(0.037)	(0.038)	(0.040)
Media exposure	0.067*	0.056*	0.077**	0.069**
	(0.036)	(0.030)	(0.032)	(0.033)
Observations	830	828	830	823
Pseudo R-squared	0.136	0.146	0.134	0.135

Robust standard errors in parentheses; ***p < 0.01, **p < 0.05, *p < 0.1.

Table 13.4 Individuals’ partisan preferences and support for North-North vs. North-South PTAs

	(13)
	DV = Pro-PTA
Centre	−0.120
	(0.103)
Right	−0.215*
	(0.129)
China PTA	−1.038***
	(0.081)
Centre*China PTA	0.212*
	(0.114)
Right*China PTA	0.747***
	(0.137)
Low education	−0.140*
	(0.081)
Manual worker	−0.400***
	(0.098)
Poor financial situation	−0.017
	(0.191)
Low status	−0.070
	(0.124)
Female	−0.393***
	(0.085)
Age	0.004
	(0.003)
Urban residence	0.094*
	(0.050)
Swiss-German	0.182
	(0.123)
Nationalism	−0.271***
	(0.044)
Trust in the EU	0.105**
	(0.051)
Foreign business share	0.140*
	(0.078)
Mobility	0.039
	(0.094)
Trade lowers prices	0.233***
	(0.037)
Media exposure	0.036
	(0.029)
Observations	970
Pseudo R-squared	0.147

Robust standard errors in parentheses;

***^p < 0.01, **p < 0.05, *p < 0.1.

Finally, Table 13.4 shows the results for the effect of individuals’ partisan orientation. In this model, left-wing orientation is the reference (and thus omitted) category. While the coefficients for Centre and Right are negative, only the latter is (weakly) statistically significant. In the presence of the interaction term, this means that the support for PTA with the EU tends to be lower among right-wing than among left-wing individuals. This finding comports with the national-conservative Swiss People’s Party’s opposition to and the Socialist Party’s support for the bilateral agreements. The interaction terms Centre*China PTA and Right*China PTA are positive and statistically significant, as expected. Taken together, we observe ‘traditional’ partisan effects on individual support for North-South PTAs but not for North-North PTAs, providing support for Hypothesis 4. Indeed, based on Model 13, the predicted probabilities of respondents having a score of 5 (“very favorable”) on the dependent variable indicate that left individuals are 10 per cent more likely to support PTA with the EU than individuals who self-identify as centrists, while centrists are 8 per cent more likely to support PTA with the EU than right individuals. By contrast, right-wing individuals are 70 per cent more likely to support PTA with China than centrists, who in turn are about 15 per cent more likely to support PTA with China compared to left-wing individuals.

(i) Robustness Checks

I ran supplementary models to assess the robustness of the results.⁶ First, I used an alternate operationalization of the dependent variable. By collapsing the categories 1=“very unfavorable” and 2=“somewhat unfavorable,” on the one hand, and 4=“somewhat favorable” and 5=“very favorable,” on the other hand, I obtained a trichotomous variable. The main results hold up.

Second, I tested for omitted variable bias by including additional controls. The survey question on attitudes toward PTA with the EU is framed in relation to deepening trade liberalization within the context of the Swiss-European bilateral agreements. As explained above, the bilateral agreements consist of a series of sectoral agreements among which the agreement on the free movement of people together with the flanking measures features prominently. Therefore, I added a control measuring respondents’ opinion on “the free movement of people with the EU has had positive effects for Switzerland.”⁷ Moreover, I included a measure of self-assessment of economic knowledge (i.e., understanding of international economic relations). Finally, I controlled for a sector’s comparative advantage (i.e., net exports as a share of sector production), to capture the distributional effects as per the sector model of international trade, as well as attitudes toward incoming FDI from the respective countries (EU and China).⁸ If I introduce these variables individually or jointly in the baseline model, the results are very similar.⁹

V Discussion

Perhaps the most interesting finding is that domestic compensation strongly increases globalization losers’ support for PTAs with China in advanced economies. What compensation policies drive the overall result? Breaking down the aggregate compensation measure in its parts, I find that all three components condition globalization losers’ support for the Sino-Swiss PTA.¹⁰ The analysis shows that the literature’s finding that generous unemployment insurance increases support for free trade among losers travels to preferential trade liberalization, particularly in the North-South configuration. Not surprisingly perhaps, the strengthening of employment protection has a similar effect. This might hold particularly in countries, such as Switzerland, with low levels of employment protection. The conditional effect of policies aimed at protecting weekly standard hours is more surprising. Arguably, the economies of China and Switzerland are complementary, dampening the fear of job losses. Nonetheless, actual or perceived increased competition in North-South trade might lead to (the fear of) an intensification of work to keep labor costs down. Concerns about work-life balance issues, which are prevalent across advanced economies and particularly in Europe, might explain this result.

Do individuals’ partisan preferences over the Sino-Swiss PTA bear any resemblance with how political parties positioned themselves in relation to it? The ratification of the trade agreement exposed a classical left-right cleavage. The Swiss National Council (lower house of Parliament) ratified the agreement in December 2013, despite misgivings by the Socialist Party and the Green Party.¹¹ The left parties remained concerned about labor rights, the environment, and human rights. Even though the PTA included labor and environmental provisions, they were considered weak. While the agreement made scant and only indirect references to human rights, the substantive labor commitments only referenced the 1998 ILO Declaration on Fundamental Principles and Rights at Work, not the eight ILO fundamental Conventions themselves as is custom in the PTAs signed by the European Free Trade Association (EFTA) – of which Switzerland is a member – since 2011.¹² The Socialists (and the labor unions) in particular bemoaned the weak monitoring mechanism (Pedrina and Doka, 2014). They brought forward three motions in the National Council to strengthen the scope and stringency of the labor and human rights commitments as well as the means for the effective monitoring and implementation of the commitments. Large majorities defeated the motions in December 2013.¹³ In short, in line with theoretical expectations, non-trade issues such as labor and human rights drove a wedge between left and centre-right parties, and the positions these parties took over the China-Switzerland PTA appear to have affected the preferences of their members and sympathizers.

It is plausible to argue that Swiss citizens have greater sympathy for their neighboring European countries compared to geographically remote and culturally distinct China. This might explain the observed lower level of support for the PTA with China than the PTA with the EU among globalization losers (Hypothesis 1). However, this explanation cannot account for why belief in compensatory policies among the losers tends to more strongly increase support for the former compared to the latter. This finding suggests that it is trade’s distributional effects rather than (or as well as) sympathy/antipathy toward particular countries that drive the results for Hypothesis 1.

VI Conclusion

Standard economic models and the ‘compensation’ hypothesis have considerable explanatory power when it comes to explaining citizens’ attitudes toward trade agreements, particularly preferential trade liberalization with China. The main findings are as follows. First, the losers of international trade are more supportive of PTAs with the EU (North-North PTA) than of PTAs with China (North-South PTA). Second, belief in compensation by losers leads to a larger increase in their support for PTA with China than for PTA with the EU. Third, the increase in support for PTA with China among globalization losers is driven by compensatory welfare and protective labor market institutions, not by social investment policies. Finally, reflecting left parties’ concerns about poor human and labor rights, left-leaning voters are lukewarm at best to strike North-South PTAs, particularly PTAs with China, unless strong human and/or labor rights provisions accompany them, yielding a partisan effect in North-South PTAs, less so in North-North PTAs.

As twenty-first-century globalization is globalization under Chinese influence, not least due to China’s accession to the WTO in 2001, which strongly increased its trade integration into the world economy, policies to compensate the losers are of paramount importance to sustain the process of global economic integration with Chinese characteristics. The evidence presented in this study on the role of domestic compensatory policies is consistent with the rise of national-populist and protectionist electoral responses to China shock in countries such as the US and the United Kingdom with small welfare states and flexible labor markets or trade-related compensation programs that are unresponsive to changing market conditions (Colantone and Stanig, 2018a, 2018b; Kim and Pelc, 2021; see also Hays, 2009; Feigenbaum and Hall, 2015; Che et al., 2016). The core finding is a reminder that in more socially embedded forms of capitalism, demands for more compensation remain popular and are likely the best bulwark against mounting protectionism and nationalism against the background of increased Chinese competition. Clearly, to preempt the collapse of preferential trade liberalization in the North-South context, particularly trade liberalization with China, it is essential to compensate the losers.

14 Trade as a Foreign Policy Issue A Bilateral Micro Perspective

Tanja Schweinberger and Thomas Sattler

I Introduction

The unilateral increase of tariffs on a range of Chinese goods by the former U.S. Trump administration in 2018 fundamentally threatens open-economy politics. By initiating a trade war with China, the U.S. elevation of tariffs has had important political, as well as economic, repercussions (Fajgelbaum et al., 2020; Brutger et al., 2023). The unilateral tariff increase also scathes the multilateral trading system as the U.S. prioritized aggressive protectionism rather than abiding by WTO principles. For instance, the WTO dispute settlement mechanism advocates principles of reciprocity. This procedure helps to ensure that countries only punish others if the latter previously violated WTO rules. This rules-based framework seeks to “mitigate the imbalances between stronger and weaker players by having their disputes settled on the basis of rules rather than having power determine the outcome” (WTO, 2021). Since reciprocity is one of the most promising strategies to induce cooperative behavior (Axelrod, 1984; Keohane, 1986), it is institutionalized in this pivotal liberal international institution.

The tenacity of this American non-cooperative trade policy initiative towards China is striking. Whilst the former Trump administration levied taxes on goods also from a number of European countries, the tariffs towards China still remain in place under the new Biden administration (The Economist, 2021). The continuation of these non-cooperative trade policies is remarkable, as the current and former administrations’ political orientation is different in nearly all policy areas. These developments suggest that China is different from other trading partners. In contrast to European and other trading partners, China is increasingly challenging U.S. power and is considered a political adversary (Nguyen et al., 2021; Schweinberger, 2021; Smeltz and Kafura, 2021). Trading relations with China are currently much more discussed than any other commercial relations between other countries.

Although the perception of China as a trading partner appears distinct, the current literature does not sufficiently account for this potential variation in perceptions across trading partners. Understanding such perceptions within public opinion is important for cooperation because the mass public can serve as a watchdog of international cooperative principles, as long as voters themselves support these principles (Milner and Tingley, 2013; Christenson and Kriner, 2019). This domestic constraint is relaxed if voters evaluate trade policy through a nationalist lens and value international cooperation less, for example, because they increasingly associate international economic relations with concerns of international political competition, as is the case with China. In the worst case, violations of international cooperative principles have their roots in the public itself when nationalist leaders hope to win votes by disregarding these principles.

We, therefore, examine how the mass publics in three large trading nations, the U.S., Germany, and Australia, value reciprocity as a key cooperative principle in international trade towards different trading partners. Thereby, we study to what extent political considerations – as opposed to purely economic concerns – constitute a source of deviations from cooperative trade attitudes. The International Relations (IR) literature has long emphasized that political and economic relations are intertwined, especially in an international system with changing power relations (e.g., Baldwin, 1985; Gilpin, 2001; Gowa and Mansfield, 2004). Following this literature, it is plausible that voters mingle political perceptions and international economic attitudes more than most trade literature (Kleinberg and Fordham, 2010), commonly based on an open-economy politics framework, suggests. The more trade is regarded as a security externality (Gowa and Mansfield, 1993), the more intricate security and economics become.

Such an analysis requires examining trade attitudes bilaterally, rather than unilaterally. We thereby depart from the dominant approach to examine citizens’ trade preferences unilaterally, that is without consideration of the behavior of the trading partner and its political relations with the home country of citizens (Scheve and Slaughter, 2001). Even though some research suggests that the actions of the other actor need to be taken seriously for investment cooperation support (Jensen and Lindstädt, 2013; Chilton et al., 2020; Feng et al., 2021; Raess, 2021), trade attitudes are traditionally conceptualized as general and thus the possibility of attitudes varying across trade partners has not received sufficient attention.¹ Unilateral trade attitudes therefore represent the views towards economic openness per se, but do not capture how reciprocal attitudes and political perceptions vary across actual country pairs, for example between the U.S. and China.

Whilst commonly assumed that Western trade attitudes specifically towards China differ from attitudes towards other countries, most of the literature does not directly test this. Importantly, studying such attitudes requires a benchmark. That is, attitudes towards trading with China need to be regarded in comparison to trade views with other countries. The recently emerging trade wars and the politics, as well as public debates surrounding them, illustrate how urgent such a bilateral analysis of trade attitudes is. After all, such events essentially represent a series of bilateral, uncooperative trade policy interactions among selected explicitly named countries, most prominently China. Our study therefore adopts a bilateral approach that enables us to describe how trade attitudes vary depending on specific other countries. Reducing citizens’ trade attitudes solely to trade in general is hence inadequate to the extent that citizens evaluate trade relations through a foreign policy lens, as an important part of the previous IR literature suggests.

The results from our survey experiments show that reciprocity continues to play an important role in all three examined countries, especially towards traditional allies, such as Canada, Germany, or Japan. Strikingly, however, a significant share of unconditional, non-cooperative attitudes exists towards non-allies such as China and Russia. Interestingly, Russia is perceived more negatively than China in most cases. Variations in these responses are best explained by perceptions of the other country as political adversary and political ideology of the respondent. Whilst citizens consistently support an increase in trade barriers in response to a protectionist initiative by the other country, at the same time, individuals make significant distinctions between countries that they perceive as political allies, such as Canada, Germany, and Japan, and those that they perceive as political adversaries, such as China and Russia. Support for decreasing trade barriers in response to a free-trade initiative by a trading partner is significantly lower for political adversaries than for political allies.

These findings suggest that citizens view trade policy not only as a means to maximize income but also as a foreign policy instrument that can be used to pursue national political goals in the international arena (Baldwin, 1985). International politics, thus, is an important source of trade attitudes in addition to personal material interests (e.g., Margalit, 2011; Jensen et al., 2017). This is consistent with attitudes towards other foreign economic policies, such as bailouts in the Eurozone (Bechtel et al., 2014) or regulation of foreign investment (Chilton et al., 2020; Raess, 2021). The findings also confirm that sociotropic considerations, such as ideology and national or group-specific distributional concerns, play an important role in evaluations of foreign economic policy (Mansfield and Mutz, 2009; Kleinberg and Fordham, 2010; Mutz and Kim, 2017; Nguyen and Bernauer, 2019; cf. Schaffer and Spilker, 2019).

Finally, the results imply that cooperation within the Western bloc finds broad societal support despite current frictions among Western countries over trade policy. Although attitudes towards adversaries like Russia are more severe than towards China, non-cooperative attitudes towards the most important trading nations are consequential. Precisely when examining Chinese-U.S. economic relations, therefore, these concerns for international political competition need to be taken into account (Kirshner, 2014). To the extent that the growing anti-globalization sentiment is rooted in these international political considerations, a revival of solutions proposed by embedded liberalism, that is the moderation of the distributional consequences of openness through compensatory measures (Hays et al., 2005; Nooruddin and Rudra, 2014), is not sufficient to reinvigorate trade cooperation with China. Instead, the threat to trade cooperation and the multilateral trading system is more fundamental than often assumed.

II A Bilateral Approach to Trade Attitudes

The most striking development in modern U.S. trade policy corresponds to its escalation of commercial relations with China. The former U.S. president, Donald Trump, commenced a trade war by unilaterally raising tariffs on Chinese goods in 2018. As tariffs soared, tariffs on Chinese goods were increased up to 21% and still remain high at approximately 19% as of January 2021 (Bown, 2021). These tariffs, in combination with the retaliatory tariffs emanating in response from China, have severe economic and political consequences (Fajgelbaum et al., 2020; Brutger et al., 2023). Strikingly, whilst the trade confrontation with European powers seems to have diminished, the tariffs targeting China continue to be upheld under the new Biden administration (The Economist, 2021).

More broadly, the U.S. opting for unilateral tariff increases represents an affront to established norms of the multilateral trading system embodied by the WTO. By not selecting the WTO dispute settlement mechanism, the U.S. openly disregarded the institution and its principles. The WTO and its rules are based on openness and reciprocity seeking to mitigate the effects of power in the international trading system (WTO, 2021). Reciprocity is defined by Keohane (1986) as “exchanges […] in which the actions of each party are contingent on the prior actions of the others in such a way that good is returned for good, and bad for bad.” (p. 8) has significantly shaped the creation of WTO rules. The institutionalization of reciprocity allows cooperation in trade to arise even in an anarchical international context. The initiation of the trade war against China is notable because the U.S. is commonly seen as a defender of such international institutions and principles and condemns such unilateral policy moves.

This shift and endurance of hostility in trade politics towards China are accompanied by the impression that trade cooperation with China is distinct from trade relations with other countries. Given China’s sheer economic size and trade activity, trade with China is discussed more than with other countries. Cooperation with China represents a case of North-South trade cooperation involving a developing country with lower social and human rights standards (Raess, 2023). As China matters both as the largest domestic market and exporter nation, the redistributive consequences of trade with China are salient. Commonly referred to as “China shocks,” economists focus on losses due to import competition resulting in unemployment, as well as lower wages and income (Autor et al., 2013; Acemoglu et al., 2016; Bisbee, 2021).

The importance of international rivalry with China, however, extends beyond these individual economic considerations. China has risen economically due to its reform and opening policies in the 1980s and thus challenges U.S. power primacy (Naughton, 2007). As discussed in an earlier chapter in this edited volume, China’s rise as a developing country has diminished US’s “institutional power” (Hopewell, 2023). More recently, the U.S. public has begun to increasingly perceive China as a political adversary, rather than an ally (Smeltz and Kafura, 2021). This shift has the potential to throttle important advances in economic cooperation, as political alliances have positive effects on long-term economic interactions and economic exchange (Gowa and Mansfield, 1993, 2004). The salience of adversarial views of other countries, especially China, is thus crucial for the support of trade. These power considerations are inextricably connected and directly threaten principles of economic openness and reciprocity.

Political analyses of international economic relations have highlighted the importance of power and relative gains for trading relations (Viner, 1948; Gilpin, 1987; Grieco, 1988). Economic cooperation often produces unequal gains even if both countries benefit from it in absolute terms. For instance, trade can lead to more efficiency gains, a higher long-term growth rate, or a greater strengthening of critical industries in some countries rather than others. The sensitivity to such unequal, relative gains is particularly high when the possibility exists that the two states will engage in a political conflict in the future (Powell, 1991). If unequal gains from trade can be turned into a military advantage, then these concerns constitute a constraint that inhibits cooperation. Accordingly, some studies find that who the other country is does indeed matter for foreign economic policy attitudes of the mass public (Herrmann et al., 2001; Spilker et al., 2016; Carnegie and Gaikwad, 2022). Such qualms are likely to expand as the rise of China succeeds past decades of uncontested U.S. hegemony and changes the international distribution of power (Kirshner, 2014).

Against this backdrop, how and to what extent does individual support for trade cooperation differ across countries? Whilst current events suggest that attitudes towards trade with China differ from trading with countries, the current trade attitudes literature does not systematically test this notion. Instead, the focus of the literature lies on gauging such attitudes unilaterally, that is economic openness in general, as opposed to trade policy towards specific countries. By commonly relying on questions such as “Do you think the U.S. government should try to encourage or discourage international trade?” or “[…] Do you support or oppose placing new limits on imports?”, the literature largely does not distinguish between attitudes across countries (e.g., Scheve and Slaughter, 2001; Mayda and Rodrik, 2005; Hainmueller and Hiscox, 2006; Mansfield and Mutz, 2009; Baccini et al., 2017; Owen and Johnston, 2017). Even though material and non-material factors have been explored, the literature has omitted conceptualizing trade attitudes as heterogeneous, that is also depending on the other country and geopolitical ties.

In contrast, our analysis adopts a bilateral approach that takes into account the behavior and characteristics of a particular trading partner. This approach is useful because many trade policy initiatives involve specific country groups or pairs, for example, in the form of preferential trade (Manger, 2009; Mansfield and Milner, 2012; Dür et al., 2014; Spilker et al., 2016). Even more importantly, recent reversals in trade openness and threats to launch a trade war occur on a bilateral basis when one country directly targets another country, as exemplified by the ongoing trade war between the U.S. and China. Changes in international openness today, thus, are often the result of bilateral decisions, in which the political relations between individual states matter much more. Equally, in public debates surrounding such policy interactions, the potential trading partner country is highly salient. Citizens also take into account the prior behavior and characteristics of other countries when evaluating their own government’s foreign economic policy (Spilker et al., 2016; Chilton et al., 2020; Feng et al., 2021; Raess, 2021; Schweinberger, 2021).

The bilateral approach is ideal for studying whether principles of reciprocity or geopolitical alliances matter more for trade attitudes. On the one hand, it is plausible that principles of reciprocity, as institutionalized in the WTO, apply to any potential trading partner, including China, as reciprocity has been discovered to be one of the few cross-cultural norms (Simmel, 1950; Gouldner, 1960; Bowles and Gintis, 2011; Gächter et al., 2017). Correspondingly, the IR literature also promotes the institutionalization of reciprocity to achieve cooperation.² On the other hand, attitudes are likely to vary depending on the trading partner. Especially when studying trade attitudes towards China a benchmark is needed, so that Chinese bilateral cooperative and non-cooperative attitudes can be analyzed in relative terms to bilateral attitudes towards other countries considered as political allies and adversaries. By just studying views towards China in isolation, the basis for comparison is unclear. We are therefore especially interested in exploring support for trade cooperation reciprocity with explicitly named countries, such as China.

This discussion yields the following hypotheses we test. First, as a default, we expect views to follow principles of reciprocity. In this view, China as a trading partner is not perceived differently than other countries, as individuals response to the same policy initiative reciprocally regardless of the trading partner, that is a non-cooperative initiative from China, is met with the same response as towards Canada. Second, the competing hypothesis expects variation across trading partners to prevail, that is cooperation will be supported differently depending on the trading partner. Third, this variation is likely to correspond to varying perceptions of political alliances. For instance, more U.S. voters should perceive Canada as an ally compared to China. hose voters who consider a trading partner as an ally (adversary) should also be more (less) likely to support a cooperative, reciprocal response. As a result, the average response to a Canadian trade policy initiative should be more cooperative than the average response to a Chinese trade policy initiative.

H1: Individuals, on average, respond to a cooperative policy initiative with a cooperative policy response and to a non-cooperative policy initiative with a non-cooperative policy response.

H2A: Bilateral trade support varies as the average response to a trade policy initiative by another country differs across countries.

H2B: Citizens perceiving the other country as an ally are more likely to support cooperative responses. Citizens perceiving the other country as an adversary are more likely to support non-cooperative responses.

III Bilateral Trade Attitudes in Three Countries

(i) Research Design

We use a survey experimental design to test these claims. This approach lends itself to effectively examine what difference it makes when the object of study is systematically changed in some way. In our case, on the one hand, the trading partner changes so that we can examine whether there are differences across trading partner countries so that we can test what difference it makes when the trading partner is China or Canada. On the other hand, the foreign economic policy initiative varies.

In our study, we follow a factorial design and include vignette treatments varying the policy characteristics. In our case, this means that cooperative and non-cooperative foreign economic policy initiatives, that is tariff decreases and increases, from different countries are proposed to the respondent. The respondent is then asked which policy their own government should pursue in response to the initiative of the other country. This study allows us to examine to what extent respondents in different countries deviate from cooperative reciprocal principles and how the preferred response varies by the trading partner.

The survey experiments were conducted in three different countries: the U.S., Australia, and Germany.³ We selected this diverse set of countries to examine to what extent respondents in countries that play different roles in the international political and economic system respond differently. All the countries represent important trading nations, but the international political concerns and political relations with other countries vary. Correspondingly, current political rhetoric varies considerably between these countries, particularly in the context of the trade war (Carnegie and Carson, 2019; Schweinberger, 2021). The U.S., for instance, is the global hegemon, even if its hegemony is currently in decline. Australia is a major regional power in the Asia and Pacific area. Moreover, Germany is a European power that is not necessarily a political challenger, however, it is a leading export power.

For our analysis, we conducted experiments embedded in population surveys in August 2018. The surveys were conducted by respondi, a survey company that uses different country-specific online access panels. Respondents were selected from these access panels based on quotas on age and gender. The samples were restricted to voting-age nationals under 70. For each country, the sample size is around 1,100 (valid responses).⁴

After reading a brief introduction to this section of the survey, respondents are presented with a policy initiative from different countries.⁵ These policy choices include either an increase, a decrease, or no changes in tariffs on imports from the home country of the respondent. Respondents are randomly assigned to one policy initiative per trading partner. The policy initiative, thus, is the treatment that a respondent receives.

Additionally, we vary who the trading partner is. We deliberately adopt this bilateral approach because we expect these attitudes to be heterogeneous also across trading partner countries depending on the political alliance. Although this treatment may lead to confounding, we still believe that mentioning concretely who the trading partner is enhances the validity and reliability of our design as we precisely seek to assess what difference China makes. In the real world, trade is with another country that is not abstract. Public debates about trade politics often address the trade partner country and its attributes. So, we seek to understand if attitudes towards trading with China are distinct, or perhaps comparable with commercial exchange evaluations with another political adversary like Russia.

In our experiments, all respondents see a policy initiative from five countries. The sequence, in which the countries were presented, was randomized. For the U.S., we selected Canada, Japan, and Germany as traditional allies. Whilst Canada presents a traditional and proximate ally, Germany is also an important U.S. ally on the European continent. Japan is also a U.S. ally, even though in the 1980s similar accusations were directed towards it as towards China nowadays. For Australia and Germany, the U.S., the UK, and Japan are included in our analysis as allies. For all countries, China and Russia were selected as countries that represent non-allies. Naturally, China differs as a trading partner from these other countries not just in terms of rivalry and power considerations but also with regard to redistributional trade consequences (Hopewell, 2023; Raess, 2023). Whilst we cannot fully control for these different considerations when referring to China, we address this question by asking respondents to place the mentioned countries on an ally-adversary scale, which varies from 0 (adversary) to 10 (ally) before we conduct the experiment (cf. below on the additional variables we collect in the survey).

The outcome variable is the preferred policy response by the respondent. We ask all respondents to choose one out of three possible policy responses by their own government, or to select the option “Don’t know.” Including this option is important as some citizens may not be knowledgeable about trade policy (Rho and Tomz, 2017). These policy responses include either an increase, a decrease, or no changes in tariffs on imports from the trading partner. Together with the policy initiative of the other country, this policy response indicates the preference for a reciprocal or inverse strategy, that is conditionally cooperative/uncooperative or unconditionally cooperative/uncooperative policy response towards other countries. An example of the exact formulation of the vignette for the U.S. can be found in Table 14.1.

Table 14.1 Overview of the experiment for U.S.–Chinese trade policy

Vignette 1: cooperative initiative	Suppose China strongly decreases its tariffs on goods produced in the U.S. that are exported to China. Which of the following policies do you think should the U.S. government pursue when it comes to trading with China?
Vignette 2: no policy initiative	Suppose China does not change its tariffs on goods produced in the U.S. that are exported to China. Which of the following policies do you think should the U.S. government pursue when it comes to trading with China?
Vignette 3: non-cooperative initiative	Suppose China strongly increases its tariffs on goods produced in the U.S. that are exported to China. Which of the following policies do you think should the U.S. government pursue when it comes to trading with China?

We additionally collect the following variables. First, we ask respondents to place the mentioned countries on an ally-adversary scale, which varies from 0 (adversary) to 10 (ally) (cf. Appendix Figures 14.4a–c). Also, we measure the degree of nationalism with the commonly used battery by Rankin (2001). These questions were asked before the experiment together with other common social demographics such as ideology, age, gender, region of origin, and income (in a randomized order). To tap respondents’ level of skill and education, we request them to tick the highest obtained degree.

Figure 14.1a U.S. responses to different countries

Figure 14.1b Australian responses to different countries

Figure 14.1c German responses to different countries

To address whether variations in bilateral economic threats are driving the responses, we also account for the competitiveness of the employment sector as a covariate. We rely on Acemoglu et al. (2016) and the standard SITC codes to categorize the different sectors to examine whether respondents “win” or “lose” from trade with the other country. With Balassa’s (1977) Revealed Comparative Advantage, we can calculate whether the sector of employment (broken down to the SITC code level) has a comparative advantage with regard to the other country.⁶ This bilateral approach takes into account that economic threats vary depending on the respondents’ and the potential trading partner country’s economies.

(ii) Findings

Figures 14.1a–c show the responses for all the treatments in the U.S., Australia, and Germany. The figures clearly show that reciprocal behavior is rather common in all three countries and towards almost all trading partners. The overall pattern is largely symmetric, that is the distribution of the responses changes strongly with the policy initiative of the other country as expected. Moreover, the reciprocal response to tariff increases by the other country is especially pronounced in the U.S. and Australia, particularly when China or Russia is the country that pursues this non-cooperative policy. Most respondents are conditionally cooperative and uncooperative, depending on the behavior of the other country. The overall responses, thus, are reciprocal, whilst negative reciprocity is more pronounced than positive reciprocity. This largely supports H1, which suggests that responses are reciprocal on average.

Nonetheless, at the same time, the figures also clearly show that a substantial number of respondents are unconditionally uncooperative. As the left columns of Figures 14.1a–c show, a cooperative initiative does not result in a reciprocal effect on the policy response attitudes of a significant share of respondents. The same is true for the situation in which the other country proposes no change, as the middle columns of the same figures show. Among the three countries that we examine, German respondents, on average, are the most reciprocal, while U.S. respondents are the least reciprocal. For the U.S. and Australia, we see that the “Decrease” and “Increase” scenarios are not symmetric in their response distribution, particularly towards Russia and China. Instead, the share of respondents who prefer no change or an increase in tariffs after these two countries proposed to decrease tariffs is quite large. And it is considerably larger than for the other three countries that we examine. Notably, however, attitudes towards Russia are overall more negative than towards China.

We further examine the divergent responses for different trading partners in Figures 14.2a–c. These figures show the differences in the average responses across trading partners and treatments for the three countries. In other words, Figure 14.2a shows how the average responses of U.S. citizens to the same policy initiative, for example, “decrease,” differ across countries in the left column of Figure 14.1a. Since for the U.S., Canada is the reference category, Figure 14.2a shows how the estimated responses for Germany, Japan, Russia, and China differ from the response to Canada.

Figure 14.2a Differences U.S. responses Treatment*Partner

Figure 14.2b Differences Australian responses Treatment*Partner

Figure 14.2c Differences German responses Treatment*Partner

The figures confirm that respondents react very differently to different trading partners and therefore hold heterogeneous views. For the U.S., respondents largely respond in the same way to policy initiatives from Germany and Japan as they do to initiatives from Canada. But they respond very differently to policy initiatives from Russia and China.

Whatever these two countries propose, U.S. respondents on average react much more uncooperatively to these initiatives than to the same initiative by the other three countries. The results are similar for Australia, although some details differ. Again, Australian respondents react more uncooperatively to cooperative initiatives by China and Russia, compared to their response to a U.S. initiative (baseline). They respond more cooperatively toward cooperative initiatives from the UK. For no policy changes and tariff increases, they respond similarly across countries, except for Russia. As for U.S. respondents, Russia is perceived as more adversarial than China (cf. Appendix Figure 14.4b). In comparison, German citizens react differently. Generally, they respond less cooperatively to all countries than to Japan. The strong, noncooperative behavior towards the U.S. can be explained by the current frictions in trade policy between the EU and the U.S. Contemporary political rhetoric and media reporting about other countries may shape individuals’ perceptions. Nonetheless, the size of this non-cooperative response is surprising. The UK is also punished more compared to Japan, but primarily when it threatens to increase tariffs. This could be explained by the tensions surrounding Brexit. Clearly, German respondents are the most cooperative towards Japan.

When examining the interaction between policy initiative and alliance perception of the trading partner in Figures 14.3a–c, we find that the responses vary systematically according to the perceived political relationship with the other country. In all three country samples, individuals prefer raising tariffs on goods from an adversary. Although for the decrease scenario, the difference between allies and adversaries does not have a statistically significant effect on U.S. predicted responses, all other effects are heterogeneous according to the relation with the trade partner. This evidence further supports H2b.

Figure 14.3a U.S. predicted responses to Policy*Alliance

Figure 14.3b Australian predicted responses to Policy*Alliance

Figure 14.3c German predicted responses to Policy*Alliance

We also examine how a number of covariates help to explain variation in preferred policy responses among respondents. The results are in the Appendix in Tables 14.2a, 14.2b, and 14.2c. We find that the perception of the other country as an ally or adversary plays an important role.⁷ In all countries, the ally-adversary variable has the expected impact: if an individual perceives the other country as an ally, then this person is less likely to prefer an increase in tariffs towards the other country. In the U.S., this is the case for Canada and Russia. For Australia and Germany, this effect is even stronger and statistically significant for even more trading partners, although the exact effects vary. Russia is generally perceived as more adversarial than China (cf. Appendix Figures 14.4a–c).⁸

Overall, our results show that reciprocity matters in all three countries for average responses to the trade policies of the respective five trading partners. This supports Hypothesis 1. Nonetheless, a substantial share of unconditionally uncooperative attitudes exist, especially towards non-allies, like China. Respondents differ in the policy responses they select by supporting cooperation with some countries but not with others. This is consistent with Hypothesis 2a. Deviations from cooperative reciprocity are best explained by political perceptions. Reciprocity accordingly appears to be more relevant when confronted with a political ally rather than an adversary. This supports Hypothesis 2b.

IV Conclusion

This study analyzes to what extent the mass publics in the U.S., Australia, and Germany deviate from the principle of reciprocity when facing different political adversaries and allies as trading partners. Whilst we find that citizens in all three countries generally support the principles of reciprocity in trading relations, significant variations across trading partners prevail. There is strong support for commensurate retaliation to protectionist initiatives, towards all countries. However, support for a reduction of trade barriers in response to a free-trade initiative is more difficult to sustain among political adversaries than allies.

Hence, whilst reciprocity is supported on average, we observe important deviations from this baseline. These deviations can be explained through variations in the political perceptions of the trading partner. Particularly for the cooperative policy response, we find that depending on whether the other country is perceived as an ally or adversary matters. Whilst the direction of causality can also be reversed, that is that the policy initiative shapes how the foreign country is perceived, the main finding of this chapter supports the assumption that perceptions of trading partners are crucial for trade cooperation policy support. Thereby, the results affirm the view of differing security externalities and expected duration of future interaction between allies or adversaries when confronted with a political adversary. Trade and foreign policy issues, therefore, are more strongly intertwined than the existing trade policy literature assumes.

Our bilateral approach, thus, yields interesting new insights into trade politics in the contemporary world economy. By shifting the attention away from the conventional approach to conceptualizing trade attitudes as general, our paper reveals that international political concerns play a much greater role in trade attitudes than previously assumed. This is particularly important as the overarching context of the international trading system has become more susceptible to bilateral considerations. Bilateral trade policy interactions manifest themselves within the WTO Disputes Settlement Mechanism, through the spread of preferential trade agreements, and with the rise of trade wars during the past years. When trade relations become increasingly bilateral, reciprocity and cooperation in trade relations are more appealing with certain countries than with others, depending on the political and economic relations with the other country (Hopewell, 2023; Raess, 2023). Our findings suggest that the sources for these uncooperative attitudes relate to international political competition. Clearly, trade policy is not just seen as a means to maximize income, but also as a foreign policy instrument that is subordinated to political goals.

These results pose a challenge to the stability of the international liberal order. First, the political constraint that nationalist leaders face is relaxed if voters increasingly mix international political and economic issues. During the uncontested U.S. hegemony of the past decades, these international political concerns may have increasingly moved into the background (Cooper, 2000). But owing to shifts in international power and the rise of China, the increasing politicization of trade issues in the wake of rising populism and economic nationalism might have gained renewed importance (Nguyen et al., 2021). Second, and like other existing research, we show that individuals do not solely base their attitudes on their personal economic well-being. This underlines the difficulties of upholding cooperation in international trading relations if international political concerns are eminent. This poses a crucial obstacle to upholding open-economy politics in the long run and ultimately challenges the functioning of liberal international institutions, like the WTO.

15 The Post-Accession Treatment of Chinese Goods Exports by WTO Members

Simon J. Evenett

I Introduction

Twenty years on China’s accession to the World Trade Organization (WTO) is rightly regarded as a significant milestone in global economic governance. Much analysis has been devoted to the effects of China’s accession on the economies, politics, and trade negotiating stance of its trading partners as evidenced, for example, by the literature on the so-called China Shock. In contrast, apart from papers analysing the impact of U.S. import tariff hikes on Chinese exports imposed by the Trump Administration, there is considerably less analysis on how China’s interests have been affected by policy intervention taken by its trading partners since 2001. This chapter seeks to go some way to remedy that deficit.

The often-expressed concern is that Chinese manufactured exports – supported by Chinese government policies – have caused dislocation and disruption in trading partners. In this chapter the perspective is reversed by asking: how much Chinese exports were at risk from unilateral trade policy actions taken by other governments? The goal is to elaborate the factual base, to offer a tentative assessment and, then, to cautiously draw out implications for policy and future research.

The systemic importance of this analysis is that it sheds light on the extent to which Chinese membership of the WTO protected its goods exports from excessive trade discrimination. Excessive is meant here in relative terms – that is, relative to other trading partners. In turn, this raises the possibility that over time the benefits that China enjoyed from its WTO membership may have been clawed back by trading partners.

Put starkly, if China’s exporters have faced extensive and growing trade discrimination aboard since 2001, has China’s membership of the WTO become a depreciating asset? It is not inconceivable that answers to that question influence Chinese government calculations concerning its future development strategy, in particular the shift away from export-led growth towards a so-called dual circulation strategy. Related policy-relevant questions include: has the trade discrimination faced by Chinese exports influenced Chinese assessments of the wisdom of decoupling from Western economies? And has foreign trade discrimination influenced Chinese incentives to engage in cooperative behaviour in the multilateral trading system?

The approach taken in this chapter is relentlessly empirical. Such an approach does not deny that there may be valuable theoretical and legal perspectives on the foreign treatment of Chinese commercial interests since its accession to the WTO in 2001. Yet, there is value in putting the facts on the table. That is, in documenting what unilateral trade policy actions confront Chinese exports, how long those policy interventions have lasted, and the scale of the market access at risk or, in the case of foreign trade reforms, opportunities. At some point, arguments based on first principles ought to be confronted with the factual record.

The evidence presented in this chapter is relevant to assessments of the impact of China’s WTO accession on its own economy and society and on its trading partners, not least because post-accession trade discrimination by the latter is likely to have reduced the net benefit of the former joining the WTO. To date, however, such assessments have tended to emphasise effects related to educational outcomes, the environment, labour markets, trade, and policy uncertainty (Chen et al., 2020; Dai et al., 2021; Garred, 2018; Imbruno, 2019; Kim and Xin, 2021; Lin and Long, 2020).

This chapter is not the first to assess the impact of foreign trade discrimination on China, as the growing literature on the Sino-U.S. trade war can attest. High profile papers in this regard include Amiti et al. (2019, 2020) and Fajgelbaum et al. (2020). That such trade tensions have been linked in extant literature to the economic, political, and social consequences of China’s accession to the WTO brings the discussion back to the animating theme of this volume. Assessments of the overseas impact of China’s growing exports, in particular of manufactured goods, include Autor et al. (2016, 2019), Autor et al. (2020), Bloom et al. (2019), Dustmann (2021), and Pierce and Schott (2016). While this paper focuses on policy interventions taken by China’s trading partners, it is important not to lose sight of the factors that may have driven the implementation of those foreign unilateral acts in the first place.

The rest of this chapter is organised as follows. The next section briefly describes the rise of China as a trading powerhouse since its accession to the WTO in 2001. Given the sustained growth of China’s exports and the size of those exports at present, it is unsurprising that there has been a reaction from trading partners. One part of that reaction has been to bring cases against China under the WTO’s Dispute Settlement Understanding and, in the interests of completeness, pertinent statistics in this regard are reported.

The third section of the chapter gets to the heart of the matter and reports on the frequency, form, and scale of trade implicated by the unilateral commercial policy acts of China’s trading partners. Given the best data available refers to policy intervention taken since the onset of the Global Financial Crisis, the emphasis is on developments since that systemic episode.

The final section of the paper discusses how best to assess or interpret the empirical findings presented earlier. The case is made for a relative assessment as opposed to an absolute assessment of the degree of Chinese goods export exposure to foreign trade discrimination. Four implications for policy and future research are drawn from that relative assessment.

II China’s Emergence as a Trading Powerhouse and Push Back from Trading Partners

Since its accession to the WTO, China has become an exporting and importing powerhouse, especially in relation to manufacturing goods. Figure 15.1 shows that China’s share of world exports and world imports now exceeds 10%. The growth in China’s manufacturing trade is even more impressive. Before its accession China’s manufacturing exports accounted for less than 3% of the world total; now nearly one-sixth of world manufacturing exports originate in China. China’s share of world manufacturing imports has risen three-fold and reflects extensive sourcing of parts and components from abroad.¹

Figure 15.1 China has become an exporting and importing powerhouse

The upshot of these manufacturing trade dynamics is that China has earned sizeable trade surpluses (see Figure 15.2). Those surpluses peaked before the Global Financial Crisis at around 7.5% of current GDP and subsequently have halved. At no time since China’s accession to the WTO has it run a trade deficit.

Figure 15.2 China has consistently generated trade surpluses this century

The findings in Figures 15.1 and 15.2 shed some light on why Chinese manufactured goods became a lightening rod for criticism of Chinese government policy. First, for some such unusually high rates of export growth were considered unlikely to have been due to competition on the merits and, on this logic, must have been supported by government policy.

Second, while China’s trading partners benefited from greater variety and lower prices, the quantum of Chinese manufacturing exports that was absorbed by trading partners after its WTO accession has been linked to disruption to local labour markets, plant closures, reduced profits, and the like. Third, some policymakers associate large trade surpluses with unfair or malign commercial and macroeconomic policies. Under these circumstances, it is unsurprising that Chinese manufacturing exports became a target for foreign trade discrimination.

Having written this, immediate resort to trade discrimination was not the only option available to foreign governments. WTO members could avail themselves of the Dispute Settlement Understanding, bringing cases against Chinese policies considered to be in violation of multilateral trade rules. Many governments did just that. As Figure 15.3 shows, since the middle of the last decade between two to six new cases were typically brought against China every year.

Figure 15.3 A steady stream of WTO dispute settlement cases have been brought against China

The European Union and the United States initiated a significant number of WTO dispute settlement cases against Chinese policies, see Figure 15.4 which identifies the eight nations that have brought the largest number of cases. It is noteworthy that the governments of countries on the East Asian land mass do not feature in the top eight most litigious nations.

Figure 15.4 By a large margin, the United States has brought the most dispute settlement cases against China

The contention that resort to WTO dispute settlement did not induce China to accelerate broad-based market reforms may have ultimately accelerated the resort to unilateral measures against Chinese manufacturing exports. Perhaps the most obvious manifestation of the latter was the de facto revocation of China’s Most Favoured Nation status by the Trump Administration. But were these high-profile measures, taken more than fifteen years after China joined the WTO, the only relevant acts of foreign trade discrimination? Could the accumulation over time of less salient discrimination against Chinese goods cover more of that nation’s exports? This question and others are answered in the two remaining sections of this chapter.

III Unilateral Trade Policy Responses by China’s Trading Partners

Although much tends to be written about trade policy developments in regional and multilateral fora, the focus of this paper is on unilateral trade policy actions taken by governments that implicate Chinese goods exports. This focus is justified on the grounds, as previous research has shown, that for more than a decade the majority of global goods trade has been covered² by discriminatory unilateral acts taken since the onset of the Global Financial Crisis of 2008 and 2009 (Evenett, 2019).

Put simply, while multilateral trade processes have stalled, unilateral trade policy is where the action is. Such is the accumulation over time of unilateral acts in force – both liberalising and trade distorting – that any credible assessment of the treatment of Chinese exports after its WTO accession needs to take account of them.

The challenge for analysts is that official sources of trade policy information provide only a fraction of the information required. For sure, not all is lost. The WTO secretariat maintains a comprehensive database of antidumping and countervailing duty investigations and this will be referred to below. Beyond that, however, matters deteriorate quickly. The WTO’s Trade Monitoring Database (TMDB) does include entries where China is an affected trading partner for goods measures – but not for service sector policy interventions and policy measures affecting the protection of intellectual property rights. UNCTAD’s TRAINS database on non-tariff measures contains some valuable information but is widely regarded as incomplete.

For these reasons the principal data source used in this chapter is the Global Trade Alert database.³ This database includes information on 61 different types of policy intervention whose implementation can affect the relative treatment of domestic firms vis-à-vis foreign rivals.⁴ These policy interventions include traditional border barriers (except those associated with Technical Barriers to Trade and Sanitary and Phytosanitary Standards), behind-the-border measures that affect imports (such as government procurement measures), subsidies to import-competing firms, and policy interventions affecting exports, foreign direct investment, labour migration (not refugees), data flows, and intellectual property.

Policy interventions announced or implemented since November 2008 (the month when G20 Leaders first stated they would eschew protectionism) are in scope as the GTA is concerned These policy interventions are at all levels of government in any nation. As of this writing, the GTA database contains reports on over 43,000 distinct policy interventions.

The GTA database includes 13,563 unilateral policy interventions⁵ by foreign governments that implicate the commercial interests of China. Of that total, 12,651 policy interventions relate to trade in goods.⁶ This contrasts with the 1,397 entries in the WTO’s TMDB that list China as an affected trading partner. Three-quarters of the total number of policy interventions affecting Chinese trade in goods in the GTA database were still in force on 24 March 2022.⁷

Of the 12,651 total mentioned in the last paragraph, 9,416 policy interventions by foreign governments tilted the commercial playing field away from Chinese goods exporters. Over three thousand subsidies to import-competing firms have been issued by foreign governments in markets where Chinese firms export. Export measures by foreign governments (both restrictions and state-furnished incentives) that worsen the competitive conditions faced by Chinese exporters have occurred just under two thousand times. Chinese goods exporters have faced 1,631 import tariff increases and have been investigated for dumping, subsidisation, or for causing serious injury a total of 1,231 times since 1 November 2008.

On the other side of the ledger, Chinese firms have benefited from import tariff cuts implemented by trading partners on 2,005 occasions. Overall, though, for every implemented foreign unilateral act that benefited Chinese exporters since 1 November 2008 there are three that have harmed them. Still, while such counts are informative, the quantum of trade covered reveals more about the exports at risk from foreign trade discrimination. Much of the remainder of this section is devoted to reporting trade coverage calculations based on the policy interventions recorded in the GTA database.⁸

One drawback to using the GTA data is that it starts reporting on policy intervention seven years after China joined the WTO. Quite reasonably, some may want to see if foreign trade discrimination against Chinese exports changed after China joined the WTO. The only known time series of sufficient length relates to antidumping and countervailing duty actions and is collected by the WTO. Figure 15.5 reports the number of such investigations into Chinese exports launched each year from 1995 to 2020.

Figure 15.5 Over time a growing number of dumping and countervailing duty investigations have been initiated on products exported from China

Before China joined the WTO approximately 30–35 antidumping investigations into Chinese goods exports were launched each year. After accession, that number has progressively risen and in some years the total number of new investigations exceeded 75. The number of investigations fell sharply in 2010, potentially as a result of the Global Financial Crisis. Similar dynamics, but starting from a much lower base from 2005, can be found with respect to countervailing duty investigations of Chinese exports.

To put the findings in Figure 15.5 in context, it may be useful to compare the growth in the annual totals in that figure with the growth of total Chinese manufacturing exports (as revealed in Figure 15.3). It should be evident that the latter grew faster than the former, implying the propensity to launch dumping and subsidy investigations into Chinese manufactured exports has fallen over time. Put differently, the annual totals of new investigations did not grow in line with the growth of Chinese manufacturing exports.

Turning now to the foreign trade discrimination and trade reform faced by Chinese exporters of manufactures found in the GTA database, Figure 15.6 reports the shares of overall exports where market access gains (potentially from WTO accession) are at risk from subsequent trade discrimination and the shares potentially benefiting from foreign trade liberalisation. The former shares always exceeded the latter and the margin grew sharply from 2013 on.

Figure 15.6 The growing shares of Chinese manufactured exports facing unilateral measures in trading partners

By the time the COVID-19 pandemic hit, over 60% of Chinese manufacturing exports faced one or more policy-induced competitive disadvantages in overseas markets. The calculations reported in Figure 15.6 take account of the duration that a foreign trade barrier is in force and, when a measure lapses, it no longer counts towards the total in subsequent years.⁹ As a result, there is no inherent reason why these export exposure shares must rise over time – still they tend to. Having written that, a plateau was reached in 2019 in Chinese export exposure to foreign trade discrimination, although it should be noted that the estimated share for this and subsequent years may be revised upward as more foreign trade measures are documented by the Global Trade Alert team.¹⁰

The share of Chinese exports of products in foreign markets where trade reforms were introduced also rises until 2019 when it plateaus at around 0.4. Any comprehensive assessment of the impact of unilateral policy choice by trading partners on Chinese exports ought to take account of the fact these reforms, although it is an open question as to whether China might have benefited from some of these reforms had she not joined the WTO.

Any particular trade route that Chinese manufacturers operate in could be affected by multiple unilateral trade actions by trading partners. One might plausibly assume that the more such harmful (liberalising) actions along a trade route the greater the likely size of the adverse (beneficial) effects on Chinese exports. To examine this matter, the share of Chinese manufacturing exports facing three or more harmful unilateral foreign measures is reported for each year from 2009 to 2022 in Figure 15.7. Likewise, the share faces three or more unilateral trade reforms. The gap between the shares facing multiple harmful and liberalising measures widens, with the latter share never exceeding 0.15. Meanwhile, by 2020 the former exceeded 0.4.¹¹ This reinforces the finding that the weight of foreign unilateral action was to disadvantage Chinese manufacturing exports after its accession to the WTO.

Figure 15.7 Thirty percent of Chinese manufactured exports face three or more trade-related hurdles in foreign markets

In principle, resorting to discrimination against Chinese manufacturing exports can differ across China’s trading partners. To investigate this matter, the equivalent of Figure 15.6 was produced for unilateral actions (both harmful to Chinese interests and beneficial) by the United States, by the European Union,¹² and by the other countries in the East Asia and Pacific (EAP) region.¹³ The results can be found in Figures 15.8–15.10 respectively.

Figure 15.8 Large shares of Chinese exports faced hurdles in the U.S. market before President Trump took office

Figure 15.9 Market access impairment in the EU is longstanding

Figure 15.10 The hurdles erected to Chinese exports in its own region cover proportionally less trade than those erected in the EU and the U.S.

There are several noteworthy findings. First, while almost all Chinese manufacturing exports now face policy-induced hurdles in the American market, the share was rising and had reached 0.74 in 2016, that is, before President Trump took office. As is evident from the relevant figure, the share of Chinese exports facing worse market access conditions in the United States deteriorated sharply during the two terms of the Obama presidency (see Figure 15.8).

Second, American public bodies have undertaken enough trade reforms in the products that China exports to the United States that the share of Chinese exports facing improved market conditions is significant. It is, of course, possible that a particular product shipped to the American market faces both unilaterally imposed trade hurdles as well as reforms.

Third, the withdrawal of China from the EU’s Generalised System of Preferences (GSP) and the extensive award of free CO₂ trading permits¹⁴ to EU firms in import-competing sectors account for the very high shares of Chinese exports facing harmful unilateral acts in the European Union (as shown in Figure 15.9). Ever since 2013 over 90% of Chinese manufacturing exports faced one or more unilateral policy-induced hurdles to competing in the European Union market. In contrast, at no point between 2009 and 2021 does 30% or more of Chinese manufacturing exports benefit from policy-induced market access improvements in the European Union.

Fourth, the shares of Chinese exports facing harmful unilateral acts in the East Asia and Pacific region are smaller than the comparable shares found in the United States and the European Union (compare the upper lines in Figures 15.8–15.10). Similar findings arise also with respect to exposure to liberalising measures implemented in the region and this may reflect the fact that China has signed regional trade agreements with many neighbouring countries.

That the United States chose to target Chinese exports with higher tariffs begs the question of whether other governments have done so and how much Chinese manufacturing exports were at stake. To address this matter, those market access-impairing unilateral trade measures where China was the sole affected exporter were identified in the Global Trade Alert database. Refer to these measures as those that “target” Chinese exports.

The shares of Chinese manufacturing exports facing targeted harmful measures in each year from 2009 to 2022 were calculated and contrasted to the shares of Chinese manufacturing exports facing targeted or non-targeted unilateral harmful measures. By construction, the latter share exceeds the former share in any given year; the gap between them reveals the extent to which Chinese exports face untargeted foreign trade discrimination.¹⁵ The comparison is presented graphically in Figure 15.11.

Figure 15.11 Targeting Chinese exports has become more common during the past decade

The share of Chinese manufacturing exports that are targeted by foreign governments for discrimination grows slowly over time but jumps twice; the first time with the exclusion of China from the EU’s GSP regime and the second time with the Trump Administration’s four rounds of tariff hikes issued under Section 301 of the Trade Act of 1974, as amended. By the time the COVID-19 pandemic hit, approximately 42% of Chinese manufacturing exports were singled out for worse treatment by its trading partners.

To conclude, evidence was marshalled in this section that demonstrates that the treatment of Chinese exports by trading partners evolved markedly after China’s accession to the WTO. The goal here has been to report the variation over time, across trading partners, and along other meaningful dimensions in Chinese export exposure to better and worse market access conditions abroad. Rather than focus exclusively on high-profile episodes, such as the Sino-U.S. trade war, the approach taken here also includes less salience unilateral trade policy acts by China’s trading partners and therefore presents a more comprehensive picture of the global commercial policy landscape facing Chinese exporters.

IV Assessment and Policy Implications

How should analysts and policymakers interpret the scale – reported in the last section – of Chinese goods exports facing foreign trade distortions implemented since China joined the WTO reported? What do these findings imply about the degree of protection afforded by WTO rules to one of the world’s major trading powers?

Care is needed in interpreting the empirical findings presented here. After all, they demonstrate that significant shares of Chinese goods exports were exposed to foreign trade distortions. They are silent on the effect of those trade distortions, a topic that should be taken up in subsequent research. Still, the former finding is of interest as, plausibly, it is a necessary condition for finding adverse effects.

On reflection, employing an absolute standard to judge the documented levels of export exposure is not sensible. There are reasons why a WTO member’s goods exports can legitimately be exposed to discrimination by foreign governments. The WTO rule book is not complete nor is it uncontested (the policy domain of subsidies being a case in point). Furthermore, certain types of trade policy discrimination are allowed under multilateral trade rules. Surely, no one with a basic understanding of the WTO agreements would reasonably expect that any member of that organisation is guaranteed that its exports be completely protected from foreign discrimination. Zero export exposure is not the right benchmark.

Moreover, growing absolute levels of export exposure to foreign trade distortions is not necessarily evidence of violations of WTO rules. Even so, this may offer little comfort to officials having to explain why their nation’s exporters face deteriorating market access conditions in trading partners. Just because discrimination is perfectly legal under WTO rules does not mean it cannot alter the political economy of support for multilateralism in the affected trading partner. Seen through this particular lens, the rising shares of Chinese export exposure to foreign discrimination reported earlier might cast a shadow over support in some Chinese quarters for its membership in the WTO.

Perhaps a better way to assess the findings of this paper is to employ a relative criterion. That is, to ask if the exposure of China’s goods exports to foreign trade distortions found here is larger than other WTO members. In terms of the figures discussed earlier, one might ask what Figures 15.6 and 15.7 look like for other WTO members or groups of those members.

Before exploring this line of inquiry further one might consider two hypotheses. First, that Chinese export exposure to foreign trade discrimination be lower than smaller and less influential WTO members on account of Beijing’s growing clout in world affairs. Second, that the fast pace of China’s export growth since its WTO accession made it a lightning rod for foreign trade discrimination and so the opposite outcome would be expected.

To add empirical flesh to this comparison, the exposure of Chinese goods exports to foreign trade discrimination that discourages imports was compared over the same timeframe with that the aggregate for every other nation. The focus on market access impairment from policy interventions affecting domestic conditions of competition in the implementing jurisdiction is deliberate. This design choice excludes state-provided export incentives from the calculations. China has implemented several changes in such export incentives since the onset of the Global Financial Crisis and to include them in the calculations would have increased the estimated export exposure to foreign trade distortions of other nations. In turn, this would inappropriately skew the comparison in favour of finding that China’s goods were better protected by its WTO membership.

Figure 15.12 reports the changes over time in Chinese and in non-Chinese export exposure to policy interventions by trading partners that discourage importation.¹⁶ Recall those policies include subsidies to import-competing firms, other behind-the-border policies (such as local sourcing incentives and requirements in government procurement policies) as well as import restrictions including import-related non-tariff barriers.

Figure 15.12 Chinese goods exports are more exposed to market access curbs in trading partners than exports from other nations

A comparison is provided between the annual exposure to any trade distortion abroad (represented in Figure 15.12 by line one or more trade distortions) and the annual exposure of exports to three or more foreign trade distortions. Exposure to three or more trade distortions could be thought of as being associated with exposure to greater trade discrimination abroad.

As far as the overall exposure to foreign trade distortions is concerned, since 2013 the share of Chinese goods exports exposed has been greater than for other nations. By 2021 such export exposure for China was more than 10 percentage points higher than for non-Chinese exporters.

Similarly, from 2019 Chinese export exposure to three or more trade distortions exceeded that of other nations; with the gap reaching at least five percentage points by 2021. Overall, then, it is difficult to argue that WTO rules have shielded Chinese goods exporters from foreign trade distortions. Put differently, existing multilateral trade rules have not prevented WTO members from taking steps that once added up discriminated against more Chinese exports than the goods exports of other WTO members.

In assessing the policy implications of this relative finding and others in this chapter, four thoughts come to mind. First, China’s growing clout – economic and otherwise – does not appear to have spared it from high absolute and relative levels of foreign trade discrimination. Although the finding of lower levels of Chinese export exposure to trade discrimination in the Asia-Pacific region suggests a nuanced clout-based argument might still apply. Further research might usefully investigate whether the carrots and sticks that China deployed had a greater effect in the Asia-Pacific region. The Belt and Road Initiative comes to mind.

Second, that overall exposure to foreign trade discrimination has risen for China and for other nations is a source of concern. It begs the question as to whether and why multilateral trade cooperation has broken down. As argued earlier, any breakdown cannot be solely pinned on the trade wars of the Trump Administration – the rot set in years before. The extent to which China’s accession to the WTO influenced observed levels of multilateral trade cooperation could be further researched. What needs to be explained is the gradual breakdown of cooperation (as manifested by greater shares of trade exposed to discrimination), not just the rhetorical fireworks of the past five or six years.

Third, that there is so much export exposure to foreign trade discrimination strongly suggests that either multilateral trade rules afford governments lots of policy space to influence trade flows or that violations of those rules are now widespread. This is important as many analysts and trade diplomats still cling to the notion that existing multilateral trade rules are a first-order constraint on governmental room for manoeuvre. One hypothesis worth further exploration is that existing WTO rules channel the impulse to favour local firms away from certain policy instruments towards others – rather than reducing the pressures to discriminate in the first place.

Fourth, since policy space is alive and kicking and has been deployed extensively to discriminate against Chinese goods exports, then surely ex-post assessments of China’s WTO accession ought to take into account the degree to which the benefits of accession have been eroded since 2001, in particular, since the onset of the Global Financial Crisis.

If future research were to establish that a significant share of those benefits was subsequently eroded on account of foreign trade discrimination then, other things being equal, this should moderate expectations of China’s willingness to take on additional future multilateral trade obligations in the absence of meaningful binding commitments by other WTO members to temper their resort to discrimination. As far as the potential for progress at the WTO is concerned, one implication of this line of argument is that the next twenty years of Chinese WTO membership is increasingly likely to be an “all or nothing” proposition.

16 Small Steps toward the Next Phase of U.S.-China Trade Relations

Simon Lester and Huan Zhu

I Introduction

The Trump administration’s four years in power were tumultuous and confrontational for trade policy in general, and for U.S.-China trade relations in particular. Trump’s trade policy legacy presents a challenge for the Biden administration. While the Biden trade leadership may want to focus on other policy areas, it cannot avoid making some difficult choices on China trade policy: It will either have to pick a new direction or stay the course.

Staying the course would mean keeping the Trump administration’s policies mostly intact. The key aspects of these policies are the Section 301 investigation and tariffs; the Phase One agreement; and the questions about China’s role in the WTO. Biden administration officials might not have followed the same approach to these issues if they had been in power in 2017, but having inherited these policies in 2021, they may be difficult to undo.

U.S. Trade Representative Katherine Tai has begun to lay out her vision for a U.S. trade policy as it relates to China. We are still in the early stages, with more words than actions so far. But Tai’s words do tell us a bit about where things might go. This paper examines the general guidance and specific details Tai has offered, in an effort to understand the direction of U.S. policy in this area.

The paper begins by reviewing Tai’s account of the recent history of U.S.-China trade relations. It then turns to the Trump administration’s actions and the Phase One agreement, including its flawed enforcement mechanism; and finally, it examines the Biden administration’s apparent decision to stick with the Phase One agreement as the framework of its policy rather than break from it in a significant way.

The paper also considers some broader themes that may inform the Biden administration’s approach to trade relations with China. While there are many criticisms that can be offered of the Biden administration’s words and actions so far, the role of economic realities and domestic politics helps explain why the Biden administration has adopted the approach that it has. The small steps it has taken will eventually lead somewhere bigger, but for the time being they may be all there is.

II Alternative Versions of the History of U.S.-China Trade Relations

In a major speech at the think tank CSIS in early October of 2021 (Tai, 2021), U.S. Trade Representative Katherine Tai said that she would “lay out the starting point of our administration’s strategic vision for realigning our trade policies towards China to defend the interests of America’s workers, businesses, farmers, and producers and strengthen our middle class.” In the process of doing so, Tai began by “reflect[ing] on how the U.S.-China trade relationship has evolved in recent decades and how we got to where we are today.”

In Tai’s version of events, “[f]rom the late 1970s to mid-1980s, China went from the world’s 11th-largest economy to the eighth largest,” with U.S. exports to China increasing “approximately fourfold, while imports grew 14 times in less than 10 years.” This economic growth, she said, “set the stage for China’s efforts to join the WTO.” This created “an important challenge,” which was “how to integrate a state-led economy into a trade institution created by those dedicated to open market-oriented principles.”

Over the next decade and a half, Tai explained, “the United States pursued a dual-track approach with Beijing.” One track involved “annual high-level dialogues between U.S. and Chinese officials over three successive presidential administrations,” while the other track “focused on dispute settlement cases at the WTO.” But both approaches, she argued, came up short, with “meaningful reforms by China remain[ing] elusive.”

In recent years, she said, “China’s leaders have doubled down on their state-centric economic model.” Facing a “reality that neither the dialogue nor the enforcement tracks were producing meaningful changes,” the Trump administration “decided to use a different paradigm – unilateral U.S. pressure – to try to change Beijing’s practices.” This led to “substantial U.S. tariffs on imports from China, and retaliation by China,” and then later to the phase-one agreement.

There is some truth to Tai’s version of history, but it also leaves out some key details, as described by Lester and Zhu (2020). China’s WTO accession was mainly negotiated during the Clinton era, but the first president to have to deal with China as a WTO member was George W. Bush. China’s economy had already been growing quickly in the pre‐WTO era, and its rise continued after entry into the WTO. The continued high growth and the shift to the production of more sophisticated industrial products put Chinese companies in competition with American companies to a degree not seen before. The Bush administration faced a difficult decision on how to respond.

Trade journalist Paul Blustein (2019) describes the Bush administration’s trade policy response as “sluggish,” and says: “It is reasonable to wonder why a more forceful approach wasn’t taken.” He offers the following explanations for why more was not done about Chinese trade practices that violated the letter or spirit of WTO rules: optimism that China would continue moving toward freer markets on its own; fear of a U.S.-China trade war; U.S. companies were making money in China and wanted to avoid disruptions, and thus did not complain much; the administration needed Chinese support on its “anti‐terrorism” policies; and finally, the global financial crisis weakened the ability of the Bush administration to make demands.

In terms of actions not taken, Blustein focuses on the Bush administration’s rejection of domestic industry complaints under Section 421, which provides for the possibility of a product‐specific “safeguard” tariff/quota on Chinese imports. But there is also the option of filing WTO complaints, which the administration was slow to pursue at first, although the complaints picked up in later years: one complaint in 2004, one in 2006, three in 2007, and two in 2008. According to U.S. trade officials from this era, there was a sense initially that China deserved a chance to settle in at the WTO before complaints were brought.¹ By 2005, it was clear that complaints were needed. However, U.S. companies were not pressing the U.S. Trade Representative (USTR) to bring claims, and without the evidence they could provide, the cases were unlikely to be successful. As a result, cases emerged slowly.

The Bush administration also found a diplomatic way to pursue these issues, with an approach called the Strategic Economic Dialogue and the Senior Dialogue. This led to some minor successes, but when the financial crisis hit in 2008, the administration became consumed with domestic issues and was not in a position to make demands of China.

President Obama then took office in the middle of that financial crisis, and his initial focus was on domestic policy. Eventually, he turned to trade and foreign policy, and Asia and China were a big part of that. Obama’s “pivot to Asia” involved giving greater prominence to the Pacific region, with the Trans‐Pacific Partnership (TPP) as a key element. The TPP had several goals, but one of them was to respond to China’s rise.² While Obama and others in his administration spoke mostly of “writing the rules” of trade in the region, many commentators emphasized that the TPP would “contain” China. As law professor Daniel Chow (2016) put it: “The U.S. led the TPP negotiations and deliberately excluded China from the negotiations. This ploy by the U.S. was a calculated effort to contain China and to shift power in trade in the Asia‐Pacific from China to the U.S.” But as is well known, the Obama administration could not get the TPP through Congress, and President Trump formally withdrew the United States from the pact, whose other members have now gone ahead with a modified version of it.

In addition to the TPP as a way to address concerns with China, the Obama administration imposed tariffs on Chinese tires under Section 421.³ It was also a frequent user of the WTO dispute settlement mechanism: during his eight years in office, his administration brought 14 complaints against China.

At the same time, the Obama administration also tried to engage with China through negotiations. It continued the bilateral negotiating approach started by the Bush administration, replacing the Strategic Economic Dialogue and Senior Dialogue with the U.S.-China Strategic and Economic Dialogue. The Obama administration also carried out a bilateral investment treaty negotiation with China, but the talks were never completed.

Thus, Tai’s version of history is not so much wrong as it is incomplete. Many of the wounds U.S. politicians feel in relation to trade with China are self-inflicted. If the U.S. government had not been distracted by the War on Terror or domestic crises, it might have made more progress in its efforts with China.⁴ And if the TPP’s domestic political strategy had been better conceived, the United States might have been able to use it as a tool to address concerns with China. Furthermore, the United States decided, for various reasons, not to use the tools that it did have, including a broader range of WTO complaints that made use of the various WTO-plus provisions in China’s accession protocol, as described in Zhou et al. (2019).

III Trump’s Trade War and Phase One Agreement

Trump and his trade team accused many U.S. trading partners of unfair practices and used a variety of U.S. statutes to do it. Section 301 became the vehicle for the tariff war with China, with an investigation by the U.S. Trade Representative’s Office under Section 301 providing the factual and legal basis of the U.S. actions against China.

Section 301 provides a mechanism for the U.S. government to take action against a wide range of broadly defined behavior by foreign governments, including an “act, policy, or practice” of a foreign country that “violates, or is inconsistent with, the provisions of, or otherwise denies benefits to the United States under, any trade agreement,” or “is unjustifiable and burdens or restricts United States commerce.”⁵ In this case, the focus of the investigation was on China’s laws and policies related to intellectual property, technology transfer, and innovation. The Section 301 investigation was launched in August 2017, soon after Trump took office.

After eight months of a USTR investigation, hearings, and comments from interested parties, USTR reached the following conclusions in March of 2018: China pressures foreign companies to transfer technology to Chinese partners; certain Chinese licensing regulations discriminate against U.S. firms; China directs foreign investment in order to acquire U.S. technology and intellectual property; and China conducts and supports intrusions into U.S. companies’ computer networks.⁶ A range of tools would be used to address these issues, including filing a WTO complaint for one issue as well as imposing tariffs on Chinese imports immediately.⁷ According to USTR, certain issues were not covered by WTO rules, so unilateral tariff action was the only possibility.⁸

The resulting tariff war began in July 2018. After many months of tariff escalation, today both countries face steep tariffs on the goods they trade with each other. According to Bown (2021a), China’s average tariffs applied to U.S. exports have risen from 8.0 per cent in January 2018 to 20.7 per cent by January 2021. This is more than triple the average 6.1 per cent tariff rate applied to other countries after China unilaterally cut its tariffs in recent years. On the other side, the average U.S. tariff on Chinese goods has soared from 3.1 per cent in 2017 to 19.3 per cent in 2021. As a result, 66.4 per cent of Chinese imports are subject to additional U.S. tariffs, and 58.3 per cent of U.S. goods face retaliatory tariffs from China.

This tariff war provided the background for the trade negotiations that ultimately led to the Phase One agreement, which was signed on January 15, 2020, and took effect on February 14, 2020. Under this agreement, China made a number of commitments, the most high profile of which was to substantially increase imports from the United States of agricultural products, industrial products, natural resources, and services. As of October 2021, however, China was on track to come up nearly 40% short of the US goods it promised to buy over 2020–21, according to Bown (2021b). This outcome is not surprising, as the purchase targets were set at a level that many people considered to be unrealistic, and on top of that the pandemic undermined trade flows in general.

China also took on other obligations in the Phase One deal, including in relation to intellectual property protection, forced technology transfer, and regulatory trade barriers for various U.S. goods and services. While China has addressed many of these obligations in its recent legislative and regulatory actions, the implementation of these rules in China is still a bit uncertain. However, enforcement of these obligations will be difficult due to the flaws in the agreement’s dispute resolution provisions.

IV The Flawed Phase One Enforcement Mechanism

The problem with enforcing the structural obligations under the Phase One agreement is that its dispute resolution section does not have the traditional neutral adjudication mechanism found in most trade agreements. Trade enforcement typically works as follows. If one government thinks another is not complying with the obligations in a trade agreement, the complaining government can raise its concerns through a request for consultations. If the consultations do not resolve the issue, the complaining government can ask for a neutral panel of experts to consider whether the other government’s actions violate the terms of the agreement. That panel will issue a ruling on the legal question of whether the respondent government is in compliance.

The WTO has the most advanced version of this process, with 606 complaints since it was established in 1995, and hundreds of panel reports and appellate reports reviewing those complaints. During her CSIS speech, Tai noted that over the years, the United States “brought 27 cases against China … . We secured victories in every case that was decided.” (Currently, the United States has blocked appointments to the WTO’s Appellate Body, which has caused significant problems for the functioning of WTO dispute settlement.) Bilateral and regional trade agreements have their own version of panels, without appellate review.

The neutral adjudication provided through this kind of process helps with the enforcement of these agreements. One government’s view that another is in violation is not seen as objective: It is simply the position of that government, rather than an impartial conclusion. An unbiased adjudicator, by contrast, has the credibility to determine whether a violation exists in a way that can be persuasive to all parties. This process helps bring the rule of law to international trade disputes.

In contrast, the Phase One agreement does not have the typical neutral adjudication mechanism, but rather has a mechanism under which either side can determine on its own if the other is not in compliance, and can then – after a consultations process – take what it considers to be the appropriate action in response (most likely, this will take the form of tariffs).⁹

The Trump administration may have seen this as a tough enforcement mechanism because it would be a quick way for the United States to impose tariffs. The problem is, if China believes it is in compliance, but the United States does not, these unilateral tariffs are unlikely to induce China to take any action to come into compliance. That is especially true in a situation like the current one when significant tariffs are already in place. By contrast, if there were a ruling by a neutral adjudicator that China is not in compliance, China might take some action. It has done so in response to WTO rulings, and it might do so in the context of Phase One disputes as well.

V The Biden Administration’s Decision to Stick with Phase One

Nevertheless, despite these flaws, the Biden administration has made clear that it sees the Phase One deal as the framework for governing the U.S.-China trade relationship. While Biden administration officials have been critical of the Trump administration’s approach to China,¹⁰ they are sticking with its set of rules in this area. As Tai put it, “I think that the structure, the architecture of this [Phase One] agreement, is where we have to start, … .” But how exactly she plans to use it is a bit unclear.

In her CSIS speech, she emphasized that the Biden administration would take enforcement actions under Phase One (Tai, 2021). She said that “we will discuss with China its performance under the phase-one agreement. China made commitments that benefit certain American industries, including agriculture, that we must enforce.” She did not, however, provide much in the way of details of the specific areas of enforcement, which makes it difficult to evaluate the likelihood of success here. When pressed after the speech, Tai did not seem willing to offer any clarity.

In thinking about what actions they might take, it is useful to examine the different categories of obligations in the Phase One agreement. Broadly speaking, they can be grouped into two categories: Purchase commitments for specific products, and substantive obligations related to structural issues in the Chinese economy. In response to questions, senior administration officials indicated that both kinds of obligations are on the table for enforcement: “We intend to raise all elements of Phase One with China where we think they have not lived up to their commitments. We’re not going to shy away from that, we want to make sure that we’re discussing kind of the full breadth of obligations there. The engagement with China will determine which ones become the focal point of discussions, … .” Given the problems noted earlier with regard to purchase commitments, however, it is not clear how China’s failure to comply could be addressed, and thus what the value of these commitments really is.

The bigger compliance concern is the structural rules, such as on forced technology transfer, for which there are detailed provisions in the agreement and genuine concerns about China’s practices. If these obligations could be enforced, the Phase One agreement could provide an important means of achieving greater liberalization in the Chinese economy. As noted above, however, the problem with enforcing these kinds of rules under the agreement is that its dispute resolution section does not have the traditional neutral adjudication mechanism found in most trade agreements.

Nevertheless, the Biden administration appears to want to give the Phase One deal’s dispute provisions a try. There is no history of using this sort of mechanism to enforce trade agreements, but the Biden administration seems to be indicating that they will test it out.

If this is their plan, it could be helpful if the administration were transparent about its actions. The agreement itself does not offer guarantees of transparency (which is an additional problem with the approach to dispute resolution taken here). However, the Biden administration could push for more of the details related to its complaints about Chinese trade practices to be made public. For example, if the administration files a “Request for Information” under Article 7.3, it could make that document publicly available. The situation relating to an “Appeal” made in writing under Article 7.4, paragraph 1 is more complicated. This provision states that “[t]he Appeal and any information and matters related to it are confidential and shall not be shared beyond the Bilateral Evaluation and Dispute Resolution Office, absent the agreement of the Parties.” While the default approach to these appeals is confidentiality, there is the possibility of transparency if the parties agree. The United States has traditionally pushed for more transparency in trade disputes and could follow the same approach here.

Beyond pure enforcement measures, Tai indicated that the Biden administration had broader concerns about China’s policies that require engagement but did not specify how she would approach them: “we continue to have serious concerns with China’s state-centered and non-market trade practices that were not addressed in the Phase One deal. As we work to enforce the terms of Phase One, we will raise these broader policy concerns with Beijing. And we will use the full range of tools we have, and develop new tools as needed, to defend American economic interests from harmful policies and practices.” She later said: “we will also directly engage with China on its industrial policies.”

This direct engagement could come in a number of forms: Another Section 301 investigation (which although confrontational could in theory lead to negotiations), which has been rumored in the area of subsidies; trilateral work as has been taking place with Japan and the EU; at the WTO; or new bilateral talks, whether classified as Phase Two or not. This last possibility could even be carried out through Phase One agreement mechanisms. Article 7.2 of the Phase One agreement talks about “high-level engagement” and in this context refers to “arrangements for future work between the Parties.” It is not clear whether this formal structure could or should be used here. Ideally, there would be some transparency in the discussions taking place in this context, but the agreement does not provide for that and the public may not get much of a sense of what is happening.

It is worth noting that whatever this engagement is, it will probably not be identified as “Phase Two” of the U.S.-China trade agreement, as Tai seemed opposed to using that terminology.¹¹ The name of the next stage is not particularly important though.

A few days after Tai’s speech at CSIS, she had a call with Chinese Vice Premier Liu He. Very few concrete details were released publicly, but the USTR readout of the call indicates that the two sides “reviewed implementation of the U.S.-China Economic and Trade Agreement and agreed that the two sides would consult on certain outstanding issues,” and that “Ambassador Tai emphasized U.S. concerns relating to China’s state-led, non-market policies and practices that harm American workers, farmers and businesses.” Press reports contain details of Biden administration officials briefing reporters on the call, which repeat many of the points made during the speech: “The main principle is that China needs to live up to its commitments, and we are going to engage with them to make that point,” one official said. “And it’s up to China to demonstrate whether they’re willing to do that”¹²; “We recognise that Beijing is increasingly explicit that it is doubling down on its authoritarian state-centric approach and is resistant to addressing our structural concerns. … Therefore our primary focus will continue to be on building resilience and competitiveness, diversifying markets, and limiting the impact of Beijing’s harmful practices.”¹³ The United States would base future engagement with China on “how China responds to tonight’s call,” and the call is “a test of whether or not this type of engagement will help to secure the outcomes that we’re looking for, and we’re going in with the hopes that China will respond positively.”¹⁴

VI The Initial Reaction in China to Tai’s Statements on U.S.-China Trade Relations

The early reaction from Chinese officials, scholars, and media was both muted and mixed. The statements by Tai and other officials did not set out a clear new path for U.S. policy here, but the absence of a confrontational tone was probably a relief for people in China.

When asked about Tai’s remarks at an October 8 press conference, the spokesperson for China’s Ministry of Foreign Affairs provided only very general thoughts on these issues.¹⁵ The Chinese ambassador to the United States offered more in-depth comments. He noted that Ambassador Tai mentioned that the United States is now seeking to “recouple” with China, which, he said, “has some positivity in it,” as “[t]he two sides can sit down and sort out the areas of ‘decoupling’ and how to get them ‘recoupled.’” More generally, he suggested that “the two countries’ trade frictions over the past few years have once again proved that China and the US both stand to gain from cooperation and lose from confrontation,” and “[t]here is no winner in a trade war or tariff war.” While “[i]t is … normal for us to have economic competition and trade frictions … [t]he key is how to deal with them.” What China advocates is that “we should pursue solutions acceptable to both sides through communication and consultation, based on the principles of mutual respect and mutual benefit.”¹⁶

In the media, the state-run Global Times, by contrast, published a commentary in which it noted that Tai presented “a tough attitude towards China,” and responded with the following statement¹⁷: “if the US intends to shake China’s foundations, prevent China from formulating plans to develop its technological innovation capacity, and change China’s national policies conducive to promoting competitiveness, China will never permit it. … The China-US trade war has lasted for more than three and a half years. Instead of being weakened, China’s economy has taken a step forward in comparison with the scale of the US. The Chinese people are more confident and their stamina continues to increase. We are clearly aware that all this is the basis for the US to consider using non-trade war coercive methods to discuss issues with China.” A Xinhua opinion piece emphasized the importance of U.S.-China cooperation: “Both nations as well as the whole world will benefit from China-US cooperation, and both countries and the world will suffer from China-US confrontation. It is hoped that the United States will change its course, respect the principles of market economy and international trade rules, and meet China halfway, so as to promote the healthy and stable development of China-US economic and trade relations, and further benefit the people of the two countries and around the world.”¹⁸

Scholarly reaction to the speech was mixed. Some Chinese scholars reacted somewhat positively to Tai’s statements. “Unlike his predecessor Donald Trump’s aggressiveness, the Biden administration aims to maintain negotiations while mounting trade restrictions. In a way, it accords with the stance of China, which seeks to solve disputes through dialogue,” Huo Jianguo, former president of the research institute of the Ministry of Commerce, told the Global Times. However, it is unlikely that China-US trade relations will go back to the pre-trade war period, Huo warned. “The US should drop its confrontational mentality toward China and facilitate competitive cooperation via dialogue and negotiations,” Huo added.¹⁹

Tu Xinquan, Dean of the China Institute for WTO Studies at the University of International Business and Economics, also expressed positive views on the Biden administration’s statement at a CSIS event. “I think, generally speaking, my impression is, it’s positive, her remarks and statements, especially if she does not support decoupling. I think it’s a very great concern for China. And she used the word like durable coexistence, and recouple. These words, kind of new, but basically I think they are positive for the US-China trade relationship.”²⁰

In resolving the differences between the two nations, Tu noted that “bilateral conversations and dialogues are important” but “international rules are even more important, because the two [powerful economies] have the capability to hurt each other.” Hence, “if we can have the same set of rules, we follow the same set of rules, then it would be easier to deal with conflicts between each other.”

Wang Yong, Director of the Center for International Political Economy at Peking University, also praised Ambassador Tai at the same event for “recognize[ing] the value of the US-China commercial relations,” and that “she’s very right in trying to come back to the dialogue with the Chinese counterparts to settle the differences of interests and positions.” When commenting on China’s enforcement under the Phase One Agreement – in particular, the purchase commitments – Wang said that “it’s very important to recognize … all these factors, including the impact of pandemic and rising cost of cargo …unfortunately, influence the implementations of the Phase One agreement.” Wang also called for both sides to “de-politicize or de-securitize the trade,” which is “very important.”

At the same time, some Chinese scholars were more cautious towards the view that this is a turning point in U.S.-China relations. Zhao Dingxin, professor at the Department of Sociology at the University of Chicago and Zhejiang University, stated in a recent article (Zhao, 2021) that “China, as the world’s second-largest economy that has a vast territory, huge population, military strength, and a cultural and political system that is very different from the United States and the West, will inevitably bear the brunt of the United States’ destructive spillover effects” and the key to China’s handling of China-US relations is to avoid falling into the “scapegoat trap” which means China becoming the scapegoat for U.S. domestic problems.

In general, Chinese government officials and Chinese scholars are probably waiting to see what the new Biden administration policy looks like in practice before developing strong views. Tai’s speech left a lot of questions unanswered, making it hard to know how the upcoming months and years of the U.S.-China trade relationship will unfold.

VII Broader Themes Guiding the U.S.-China Relationship

Beyond the specific details of the Biden administration’s recent statements, there are several important themes lurking in the background that can help inform the issue of the future of U.S.-China trade relations: The calls for “decoupling” of the U.S. and Chinese economies; the role of the state in the economy, in both countries and in the Phase One agreement itself; and the internal political debates in the Democratic party on trade.

(i) Decoupling vs. Recoupling

There has been a great deal of recent talk among foreign policy and trade policy commentators about “decoupling” of the U.S. and Chinese economies. The Biden administration does not appear sympathetic to the idea of decoupling, with Commerce Secretary Gina Raimondo seeming skeptical of it,²¹ and Katherine Tai in her CSIS speech characterizing it as not “a realistic outcome.”²² For Tai, the issue is, “what are the goals we’re looking for in a kind of re-coupling?” And “[h]ow can we have a trade relationship with China where we are occupying strong and robust positions within the supply chain and that there is a trade that’s happening as opposed to a dependency?” In a subsequent interview, she made clear that the United States and China are not, in her view, in a “Cold War.”²³

For Tai, then, the issue seems to be how the United States and China can have an economic relationship that works for both sides politically and economically. The economic concerns are about the impact of trading with China on U.S. workers and companies. The political concerns are more about national security and geopolitical power. The Biden administration does not seem to have an answer to these questions at this point, but that is the goal it is trying to achieve.

(ii) The Role of the State in the Economy

In her CSIS speech, Tai referred to China’s “industrial policies” and its “state-centered and non-market trade practices” as problems that needed to be addressed. However, it cannot be ignored that at this same moment, within U.S. politics and policy, there are many calls for industrial policy and a greater role for the state in the U.S. economy, and the Biden administration seems eager to move the economy in this direction. From Buy American policies to reconfiguring supply chains to calls for “economic resilience,” the Biden administration at times seems to be emulating the same Chinese policies it is criticizing, although to be fair the degree of state involvement in the economy is less.

Along the same lines, at the same time the United States is asking China to be less state-oriented in its approach to the economy, the Phase One agreement adopts a very state-centered vision itself in the form of its purchase commitments. Tai was asked specifically about the purchase commitments but did not acknowledge the contradiction here.²⁴

Part of the problem the Biden administration may be having in formulating a way forward on its China trade policy is the dilemma over what economic policy it wants for itself. In the past, the United States has pushed hard at the WTO on issues related to non-market economies.²⁵ The Biden administration has already picked up on that idea in its work with the EU on the Trade and Technology Council,²⁶ but some of its own actions could undermine its efforts in this regard if it becomes difficult to distinguish U.S. policy from Chinese policy. For example, there are calls in the United States for significant subsidies to the semiconductor industry. Efforts by the Biden administration to challenge Chinese industrial subsidies will come across as hypocritical if the United States is doing the same thing.

(iii) Democrats’ Infighting on Trade

Traditionally, one of the primary goals of U.S. trade policy was “market access,” that is, opening up foreign markets to allow more sales of U.S. goods and services. While the idea of imports may have been controversial, exports were seen as universally positive. However, the battle within the Democratic party on trade has called even this view into question.

After her CSIS speech, Tai was asked about this issue very directly: “Is increasing market access to China one of your goals? And if so, what sectors are you targeting?” Tai’s response was vague and non-committal, and seemed to cast the past focus on market access in a negative light.²⁷ But if market access is being pushed aside, what exactly is U.S. trade policy about? Tai and others in the administration have continuously emphasized “workers,” but that does not say much. Any policy, including lower tariffs, can be marketed as something good for workers. What exactly does the Biden administration have in mind here? Without a clearer picture of the broader trade policy they are advocating, it may be difficult for them to come up with a coherent China trade policy. And it may also be that no such policy is forthcoming any time soon, as domestic policy and other foreign policy issues take precedence on the administration’s agenda.

VIII Conclusions

With all the emphasis on bilateral trade relations, one might ask, where is the multilateral, that is, the WTO, in all of this? Tai was asked directly at the CSIS event “What role does the WTO play in all this?,” and seemed skeptical of its ability to help with U.S. claims about China.²⁸ It remains to be seen how the Biden administration relates to the WTO in general, and how China’s role there continues to progress, but for now, the WTO’s role in the U.S.-China trade conflict may be limited.

In terms of the bilateral side of things, at this point in time, it appears that the Biden administration is willing to just put the ball in China’s court and see how it reacts. This may mean that the status quo stays in place for a while. As much as that prospect aggravates many U.S. business groups and trade policy experts, it does not seem to bother the Biden administration. In part, that may be due to their reluctance to adopt a new policy that will bring criticism from different sides (the progressive left and the nationalist right). They may not like the current deal, but they are not eager to negotiate a new one. There is still plenty to do on the domestic policy agenda, and they can withstand criticism from pro-trade moderates in the meantime. As a result, although the administration has taken tentative first steps on China trade issues, it may end up standing still for a while.

17 China’s Entry into the WTO A Mistake by the United States?

Jennifer Hillman

…it seems clear that the United States erred in supporting China’s entry into the WTO on terms that have proven to be ineffective in securing China’s embrace of an open, market-oriented trade regime

2017 USTR Report to Congress on China’s WTO Compliance

I Introduction

On December 11, 2001, China was admitted to the World Trade Organization (WTO), the culmination of an American-supported process that opened China to global investment and helped make it the workshop of the world. Since then, the halls of Washington D.C. have been wracked by debates over how to understand – and manage – China’s entrance onto the global stage.

By delineating the drivers of U.S. decision-making in the lead-up to China’s accession to the WTO and cataloging actions taken in the years immediately following, this chapter hopes to offer a partial explanation of how and why the United States has grown skeptical of the rules-based trading system’s ability to address concerns with China, despite serving for decades as its chief architect and as a key proponent of China’s entry into the WTO. The chapter highlights the wide range of issues that many politicians thought could be addressed as part of China’s accession processes, contrasting those views with the reality of what was both achievable and included in China’s protocol to join a trade organization with a particular and circumscribed set of rules. While it would be misguided to attribute the current state of the international trading system solely to the choices of U.S. policymakers or the actions of China, U.S. decisions, particularly the decision to prevent new judges from being appointed to the Appellate Body and the decision to pursue unilateral tariffs on Chinese exports to the United States, have clearly had a major impact. These choices, once unimaginable, are best understood as part of Washington’s reaction to China’s rise along with perceived failures by the WTO.

(i) Shifting Views of Engagement with Trade and China

The old China consensus was built on a broad understanding of the benefits of engagement, an optimism about liberalization in China, a desire to avoid Sino-American confrontation as well as a belief that Chinese-American economic ties would provide material and, in some cases, domestic political benefits, for key U.S. stakeholders. These interests and beliefs were the basis of U.S.-China policy and ensured relative stability from administration to administration. It was under the auspices of this coalition that the pre-requisite for China’s admission to the WTO – “permanent normal trading relations” between China and the United States – was established. Now, however, a growing conviction that the People’s Republic of China (PRC) poses a fundamental threat to the United States’ hegemonic role, as well as economic losses and growing income inequality associated with trade with China, has undermined support for the expansive trade policy the United States once regularly pursued.

Policymakers on both sides of the aisle have also become skeptical of the material benefits offered by engagement. Rather than opening markets to U.S. firms, a more cynical view that sees globalization and trade as hollowing out the U.S. manufacturing base and the middle class is increasingly common.¹ As China has moved up the global value chain and become a direct competitor in sectors previously thought to be U.S. strongholds like advanced computing and telecommunications, China’s economic might and state-led practices have begun to pose a direct threat to the United States’ defense industrial base and other influential elements within the U.S. political system.² Both parties have endorsed and continue to explore industrial policies that would have once been taboo.

At the same time, the political benefits of opposing a liberal trade policy have grown. Republican Party leaders, a traditional stronghold of free trade sentiment have, in recent years, grown more skeptical both of China and open trade policies more generally. The Democratic Party, long divided on trade, has been faced with the need to compete for blue-collar voters distributed in strategically important states and has grown skeptical that trade can deliver the broad-based growth imagined by members of the Clinton administration, who championed China’s accession to the WTO.³ Those within the Democratic Party who see an open trade agenda as essential to both economic growth and successful competition with China have become more self-conscious of the political liabilities associated with trade and have tempered their aspirations accordingly.⁴

Washington’s evolution has deep implications for the future of the WTO. Born in an era of American self-confidence, military dominance, and liberalizing ambition, the WTO is perceived in some quarters to be ill-suited for the current moment. In the face of a growing Chinese-American contest, the WTO’s challenge is to avoid being trampled or sidelined while still working to preserve the multilateral, rules-based trading regime. This can only be accomplished if the WTO reorients itself to become a venue where competitors are able to come together to address pressing global problems like climate change, global health, income inequality, digital commerce, and the implications of significant non-market economy actors as members of the WTO.

(ii) Policy of Engagement

From at least the end of the Cold War to the late 2010s, the United States pursued a strategy of “engagement” towards China.⁵ Under the auspices of this strategy, the United States attempted to expand economic and political ties between the two countries. While trade normalization is the most prominent element of this strategy, it also included military-to- military dialogues, educational and scientific contacts, and collaboration in multilateral fora.

As a strategy, engagement was broadly intended as a way of managing the rise of China. The architects of engagement sought to accomplish a wide variety of goals, ranging from increased Sino-American collaboration on non-proliferation and environmental issues to accelerated economic growth in the United States.⁶ Among these goals were the political and economic liberalization of China. By approaching China with an open hand, U.S. officials believed they would be better positioned to accomplish these goals. The engagement was thought to produce comity and trust between the two powers, empower less confrontational elements within the regime, and set the groundwork for liberalization.

In recent years, the policy of engagement has been reconsidered.⁷ This shift has been substantially the result of China’s growing geopolitical power, increasingly assertive behavior on the international stage, and a turn toward a more state-centric, Chinese Communist Party (CCP) controlled economy. The engagement was explicitly intended to prevent a Sino-American confrontation. Today, there is a sense that some sort of Chinese-American competition is inevitable and, perhaps concerningly, may even be desirable. China’s pivot away from market reform has also undercut one of the primary rationalizations for engagement. Unlike their predecessors, policymakers in the United States are increasingly skeptical that reformist elements in China can change China’s economic path or that U.S. actions can empower the reformists. Finally, domestic dynamics have made open trade writ large increasingly toxic and trade with China an especially hot-button issue.

As the U.S. policy towards a more assertive China has come under increased scrutiny, attitudes toward the WTO have shifted. Initial optimism about the power of the WTO to discipline and mold China has been replaced by frustration with the WTO’s apparent inability to confront Chinese abuses. That frustration has itself contributed to a lack of support for the WTO as an institution, particularly for its dispute settlement system and its Appellate Body.

II Implications of the Debate over Granting China “Permanent Normal Trade Relations” Status

As noted below, the process for China joining the WTO involved both the negotiation of China’s WTO Protocol of Accession (Protocol) and its accompanying Working Party Report (Report) and legislation in the United States to grant China “permanent normal trade relations” (PNTR) to replace the annual review of whether Chinese goods could enter the United States under “Most Favored Nation” (MFN) rates of duty. PNTR was necessary to meet the prerequisite required of the United States to “immediately and unconditionally” grant Chinese goods the same tariffs and trade treatment as goods from all other WTO members – called MFN everywhere but the United States, where the term “normal trade relations” is used instead.⁸

(i) Expansive View of the Role of the WTO Accession Process

China’s accession to the WTO was negotiated under the Clinton Administration, taking as their point of departure President Clinton’s view that China could play a positive role in advancing environmental standards, fighting transnational crime, bolstering the international trading system, contributing to an arms control regime, and promoting stability in East Asia.⁹ While “political pluralism” and “free markets” in China were stated goals of the Clinton administration, these long-term objectives were part of a larger agenda that included pressing short-term concerns like the burgeoning North Korean nuclear program and the desire to diffuse tensions over Taiwan. PNTR was viewed internally by advocates, as Ambassador Charlene Barshefsky later recounted, as “the one chance the U.S. and China had to create an enduring foundation for the relationship.”¹⁰ The logic was that “if we treated China as an enemy, we were guaranteeing an enemy in the future. If we treated China as a friend, we could not guarantee friendship, but we could at least keep open the possibility of more benign outcomes.”¹¹

For the Clinton team, the benefits of engagement were not theoretical. They had seen how deeper ties could help diffuse both bilateral and regional stressors. The U.S. Trade Representative at the time, Mickey Kantor, would later argue that trade ties helped the two countries manage the 1995 Taiwan Strait Crisis, commenting that he was “convinced that the trade relationship was a strong connecting bond between the U.S. and China at a time when we needed it, particularly in late ’95 and early ’96.”¹² According to former Secretary of Defense William Perry, the threat posed by North Korea, “provided a pretty strong incentive to see if we could go out and re-establish a reasonable relationship with [China],” which had previously provided useful intelligence about North Korean ambitions.¹³

The Clinton administration also saw their ability to encourage and support reformers within the Chinese regime as essential to accomplishing their varied goals. The contention was the boost to Chinese economic growth contributed by PNTR would “bolster the confidence of Chinese leadership in ways that reduce their fear about political reform.”¹⁴ Enabling “reformers” within China could allow the United States to pursue a wide range of priorities by creating a China that was “more cooperative on such crucial issues as nonproliferation, regional security, peacekeeping, human rights, and arms control.”¹⁵ Rejecting PNTR, however, would empower elements within China “who want to tighten the internal clamps, invest more heavily in the military-industrial complex, and hunker down for the ‘inevitable struggle’ with America bilaterally, regionally, and globally.” The “worst case” scenario was a “US-China confrontation across Asia.”

The United States’ ability to tip the scales in China’s domestic political debates was a core assumption of the engagement strategy. President Clinton suggested that refusing to engage with China “would encourage the Chinese to become hostile,” while National Security Sandy Berger argued that U.S. aggressive U.S. policies would “fuel the very inward-looking forces that trample human rights.”¹⁶ While the administration was also clear that China would ultimately choose its own destiny, there was a sense that the right set of U.S. actions could enable the rise of reformers who would be amicable to the United States and the international system.

(ii) The PNTR Vote

In the summer of 2000, nearly 18 months before China would formally join the WTO, came the Congress’ big moment to weigh in on U.S.-China trade policy. In the end, the vote was not particularly close, especially in the Senate, with the House voting in favor of granting PNTR to China in May 2000, 237–197; and the Senate following in September 2000, approving by a vote of 83–15.¹⁷ But the vote came only after fierce debate before a skeptical Congress. While technically the Congress had been acting on China policy through its annual decision to waive freedom-of-emigration requirements (called Jackson-Vanik provisions) that would have the effect of taking away China’s Normal Trade Relations (NTR) status, the reality was that while the House of Representatives voted to deny NTR to China in 1990, 1991 and 1992, there was no agreement from the Senate, and neither legislative body took any action to change China’s status after 1992.¹⁸ As a result, China had effectively been receiving NTR treatment and access to the U.S. market on terms comparable to other WTO members since 1980. But the lead-up to the 2000 vote to grant China NTR status on a permanent basis underscored deep divisions and concerns over the future economic relationship between the United States and China.

The Clinton Administration put on a full-court press, with most of the cabinet weighing in to support a vote in favor of shifting U.S. policy from one in which China’s access to the U.S. market had to be, at least technically, reviewed each year, with the possibility that it’s “most favored nation/normal trade relations” status could be removed at any time. In urging Congress to pass PNTR legislation, President Clinton noted that supporting China’s entry to the WTO was in the United States’ broader national interest because it represented “the most significant opportunity we have had to create positive change in China since the 1970s.”¹⁹ He made clear, however, that it would also advance U.S. economic interests, describing the U.S. agreement as “the equivalent of a one-way street. It requires China to open its markets – with a fifth of the world’s population, potentially the biggest markets in the world – to both our products and services in unprecedented new ways. All we do is to agree to maintain the present access which China enjoys.”²⁰

President Clinton described the outcome of the affirmative vote in the House as “a historic step toward continued prosperity in America, reform in China, and peace in the world” and for “an America that will be more prosperous and more secure; for a China that is more open to our products and more respectful of the rule of law at home and abroad.”²¹ At the same time, Clinton recognized that nothing about the trajectory of China or the U.S.-China relationship was guaranteed – but represented a chance for the U.S. and China to build a better and different future in the Asia Pacific community.

The American business community for its part believed that the normalization of trade relations with China would offer an economic windfall. The business community had long been advocates of increased trade with China. Their lobbying – supported by the Chinese government – helped reverse Clinton’s initial China policy, which had linked trade access to progress on human rights.²² As Warren Christopher, Clinton’s Secretary of State and an advocate for the linkage policy, later put it, “the business community had convinced the president that trade for America was a higher value, or perhaps to put it more charitably, that nothing would be accomplished in the field of human rights by denial of trade, and so that became the basic policy.”²³

Advocates argued that PNTR would promote U.S. security and economic interests and have little downside. Then-Senator Joe Biden, for instance, argued that it would “help promote stability across the Taiwan Straits,” encourage China to reform its economic system, and “enhance their respect for the rule of law,” while offering the United States “one-way” trade concessions.²⁴ Those supporting PNTR for China focused on the economic gains and the chance to support economic reforms in China, with even Federal Reserve Chairman Alan Greenspan declaring PNTR would “create new opportunities for American businesses and farmers.” Critics, on the other hand, focused on China’s poor human rights record, its continued threats to Taiwan, its contribution to nuclear proliferation, its violation of environmental standards for development, and its labor abuses.

The Clinton administration explicitly pushed PNTR as a way of accelerating Chinese marketization and democratization, suggesting that it would “strengthen reformers” who were “trying to move policy in the right direction.”²⁵ The deal was seen as a potential boon to U.S. exporters, something that excited many Congressional supporters, while USTR Charlene Barshefsky indicated that the administration had “no reason to expect any substantial increases in Chinese imports at all.”²⁶

(iii) The Reality of China’s Accession Commitments

While the arguments in the United States for granting China PNTR status in order to pave the way for its entry into the WTO may have focused on a broad range of geostrategic, political, national security, human rights, and economic issues, along with the empowerment of the reformists in China, the negotiations in Geneva were limited to trade issues and to the trading rules that China was signing up to.

When the accession negotiations were ultimately concluded and China joined the WTO in December 2001,²⁷ it did so only after making a substantial number of commitments and changes to its domestic economic laws, including:

• Major reductions in Chinese tariffs. Average tariffs on industrial products were reduced to 9.4% by 2005; elimination of all tariffs on high-technology products; auto tariffs fell from 80–100% to 25% by 2006; agriculture tariffs fell to an average of 17.5% by 2004.

• Elimination of import quotas and licensing requirements by 2005.

• Granting import and distribution rights to foreign corporations, which allowed them to set up wholly owned distribution, sales (including retail), shipping, and service networks over a three-year phase-in period.

• Financial services and telecommunications – ending the outright ban on any foreign ownership but leaving a number of restrictions and limitations on foreign ownership in place.

• Other services – increased market access for professional services, including accounting, consulting, engineering, medical, and information technology, while maintaining numerous restrictions and limitations.

• Commitment to implement and enforce international standards on the protection of intellectual property; provision for increased access and distribution rights for a specified number of motion pictures, music, and software.

The WTO-Director General at the time, Supachai Panitchpadi, described China’s accession agreement as signaling “China’s willingness to play by international trade rules and to bring its often opaque and cumbersome government apparatus into harmony with a world order that demands clarity and fairness.”²⁸

The USTR report to Congress in 2004 summed up the process and the outcome:

The United States and other WTO members negotiated with China for 15 years over the specific terms pursuant to which China would enter the WTO. As a result of those negotiations, China agreed to extensive, far-reaching, and often complex commitments to change its trade regime, at all levels of government. China committed to implement a set of sweeping reforms that required it to lower trade barriers in virtually every sector of the economy, provide national treatment and improved market access to goods and services imported from the United States and other WTO members, and protect intellectual property rights (IPR). China also agreed to special rules regarding subsidies and the operation of state-owned enterprises, in light of the state’s large role in China’s economy. In accepting China as a fellow WTO member, the United States also secured a number of significant concessions from China that protect U.S. interests during China’s WTO implementation stage. Implementation should be substantially completed – if China fully adheres to the agreed schedule – by December 11, 2007. By contrast, the United States did not make any specific new concessions to China, other than simply to agree to accord China the same treatment it accords the other 146 members of the WTO.²⁹

None of these commitments, however, directly addressed the broader issues and aspirations for change within China that were an integral part of the debate over the passage of PNTR. For example, none of them spoke to issues such as nuclear proliferation or human rights that had been critical points underlying support for PNTR.

III Special Scrutiny for China

The level of concern in some quarters about granting China PNTR and paving the way for its entry into the WTO can be seen in two unusual provisions that were included in the PNTR legislation: (1) the establishment of the U.S.-China Economic and Security Review Commission (“China Commission”) and (2) the statutory requirement that USTR report to Congress every year on China’s compliance with its WTO obligations. No other country has similar provisions or undergoes the level of scrutiny that China does.

(i) The China Economic and Security Review Commission

The China Commission was created with the legislative mandate to monitor, investigate, and submit to Congress an annual report on the national security implications of the bilateral trade and economic relationship between the United States and China, and to provide recommendations, where appropriate, to Congress for legislative and administrative action. Even the title of the Commission with its joint focus on security and economics reflects the desire to use PNTR and China’s accession to the WTO as leverage over security and geostrategic issues as well as trade and economic matters. From its inception, the Commission was skeptical that China would live up to its WTO commitments without constant and extraordinary vigilance from the U.S. In its very first (2002) report issued less than one year after China’s accession, the Commission recommended that: (1) the U.S. shore up its toolbox by renewing the “Super 301” law that identifies priority practices and priority countries for trade liberalization, (2) commence WTO litigation over non-compliance with intellectual property rights, (3) engage in a more intensive examination of WTO compliance, and (4) consider a national security-based case to protect the U.S. steel industry.

The Commission also focused early attention on one of the major issues upsetting the balance of concessions reached through China’s WTO accession – currency manipulation. Starting in the early 2000s and continuing for at least a decade thereafter, China kept the exchange rate of its currency pegged at artificially low levels.³⁰ Such currency manipulation made Chinese exports to the world cheap but exports to China more expensive. The Commission noted that “China’s currency manipulation acts as a subsidy for Chinese exports to the United States and a tax on imports from the United States, and serves as an incentive for U.S. and foreign firms to move production to China.”³¹ The Commission noted in 2005 that most economists believe that the Chinese renminbi (RMB) is undervalued by 15–40 per cent.³² China’s ability to manipulate its currency by buying dollars and other foreign currency in China at a fixed rate contributed to an excessive reliance on export-led growth that deepened distortions in the Chinese economy and in its trading relationships.

By the mid-2000s, the China Commission’s reports were sounding the alarm about the “profound differences between the open-market approach of the United States and the managed trade principles and predatory practices observed by the Chinese government.”³³ It began using a “responsible stakeholder” index to assess whether China not only observes international norms but works to strengthen them, finding that China was far from meeting that standard.³⁴ Following Chinese President Xi Jinping’s first state visit to the United States in 2015, the Commission reports cataloged long and growing lists of grievances the United States has against Chinese behavior, with currency manipulation, forced technology transfer, intellectual property theft, and excessive use of state subsidies often topping the list.³⁵ The most recent (2020) report concludes:

The CCP has launched determined and systematic efforts to hollow out global governance institutions, suppress internal opposition, subjugate free peoples in Hong Kong and around China’s periphery, dominate global economic resources, and project military power. These efforts threaten vital interests of the United States and the security and vitality of an increasing number of countries around the globe. A clear understanding of the CCP’s adversarial national security and economic ambitions is essential as U.S. and allied leaders develop the policies and programs that will define the conditions of global freedom and shape our future.³⁶

(ii) Annual USTR Reports on China’s WTO Compliance

To attempt to hold China to its WTO commitments, the United States used a multi-tracked approach. One track involved a series of annual high-level bilateral talks between U.S. and Chinese officials over three successive presidential administrations. These talks, initially called the Joint Commission on Commerce and Trade (JCCT, started in 2004), the Strategic Economic Dialogue, begun in 2006, and the Strategic and Economic Dialogue (S&ED) begun in 2009, were intended to push China towards complying with and internalizing WTO rules and norms and making other market-oriented changes. The second track involved WTO disputes to challenge China’s compliance with its WTO obligations.

Both tracks for holding China to its WTO commitments have been cataloged over the past 20 years in the annual report that USTR is required to submit to Congress under the mandate included in the legislation granting China PNTR status. The initial report, submitted in 2002, reflects considerable optimism, noting the significant progress China made in implementing its WTO commitments, including “reviewing more than 2,500 trade-related laws and regulations, repealing 830 of them, and amending 325 more.”³⁷ The report also acknowledges the considerable resources devoted to restructuring the various trade-related government ministries and agencies and to the education and training of central and local government officials. It also emphasized the commitments that the U.S. thought it had obtained regarding China’s transition to a market economy: “For much of the past two decades, China had been gradually transitioning toward a market economy from what in the late 1970s was a strict command economy. In acceding to the WTO, China was required by the United States and other WTO members to agree to accelerate this process of market reform in order to comply with WTO requirements.”³⁸

At the same time, the USTR Report raised early concerns over transparency, agriculture, intellectual property rights, and services. Just two years later, the 2004 report was even more hopeful, quoting two trade associations’ view that “China is now substantially in compliance with its WTO obligations – a marked improvement over last year.”³⁹ The report also noted continued areas of concern, including intellectual property rights, services, agriculture, industrial policies, and transparency. It added that the work of one of the many high-level dialogue groups –the Joint Commission on Commerce and Trade (JCCT) – was “highly constructive.”

Over the next number of years, these annual reports continued to report on bilateral dialogues and WTO disputes, coming to the general conclusion that the intensive dialogues with China generated positive outcomes on a number of contentious issues, while U.S. use of WTO dispute settlement continued to generate favorable settlements and favorable WTO dispute settlement decisions. The reports also continued to note ongoing problem areas, including principally intellectual property rights, industrial policies, trading rights, agriculture, and services, including distribution services. Increasingly the reports focused on the Chinese government’s interventionist policies and practices and the large role of state-owned enterprises and other national champions in China’s economy, which generated significant trade distortions giving rise to trade frictions. The final Obama Administration report (2016) noted the major expansion in U.S. goods and services exports to China but concluded that “despite these positive results, the overall picture currently presented by China’s WTO membership remains complex.”⁴⁰

Once the Trump Administration took office, however, the tone and underlying message changed to one of failure. The view shifted, noting that “China largely remains a state-led economy today, and the United States and other trading partners continue to encounter serious problems with China’s trade regime. Meanwhile, China has used the imprimatur of WTO membership to become a dominant player in international trade.” The report concluded: “Given these facts, it seems clear that the United States erred in supporting China’s entry into the WTO on terms that have proven to be ineffective in securing China’s embrace of an open, market-oriented trade regime.”⁴¹

IV Failure to Use the WTO and China Accession Tools

Critical to gaining support for bringing China into the WTO were the tools built into China’s protocol of accession and the WTO rules themselves that many in the United States believed would both protect the American market from any downside risks and hold China to account for its commitments.

Key among the provisions designed to guard against harm to domestic economies around the world were:

1. (a) a selective safeguard provision lasting 12 years permitting countries to impose safeguards (tariffs or quotas or other restraints) if they found that an increase in imports from China alone was causing disruption to their domestic producers of comparable products,⁴²

2. (b) a right to apply for non-market economy status when calculating anti-dumping duty rates to imports from China, at least for a period of 15 years⁴³;

3. (c) a provision making it easier to apply countervailing duties to subsidized imports from China if distorted market conditions in China present “special difficulties” to countries in identifying and measuring subsidies;⁴⁴ and

4. (d) a provision lasting until December 31, 2008, limiting China’s textile and apparel exports to amounts no greater than 7.5% (6% for wool products) above the amount entered in the previous year if Chinese imports were impeding the orderly development of trade.⁴⁵

(i) Failures to Guard Domestic Markets

Among the most highly touted of the provisions designed to guard against any potential harm from China was a product-specific, selective safeguard provision that allowed WTO members to impose safeguards (which normally can only be applied to imports from all sources) on imports from China alone.⁴⁶ The provision included a lower threshold for its application than traditional safeguard measures which require a showing that imports have caused “serious injury” to a domestic industry producing comparable goods. Here, the standard permitted the application of a China-only safeguard if Chinese imports were causing “market disruption.” In addition, a clause in this new safeguard allowed a second country to justify its own imposition of a new import restriction after a first country has implemented a China safeguard on the basis of a “trade deflection” threat alone, without having to carry out its own injury investigation. In the United States, the common answer to Congressional concerns over a potential flood in imports from China was the existence of this special safeguard and the ease in its application. The process to implement it called upon the independent U.S. International Trade Commission (USITC) to investigate all claims of market disruption and to report to the President any affirmative findings, along with recommendations for actions to be taken to address the market disruption caused by Chinese imports.⁴⁷ The President was given the authority to accept, modify or reject the USITC’s recommendations.⁴⁸

In the 12 years that the China-specific safeguard provision was in effect and despite the substantial increase in imports from China in a wide variety of products, the United States imposed a China-specific safeguard only once.⁴⁹ Part of the reluctance stemmed from decisions made by the Bush administration not to impose safeguards despite a recommendation from the USITC in five cases to do so.⁵⁰ Failure to obtain a remedy despite proving market disruption may have deterred potential complainants while concerns over WTO decisions striking down global safeguards may have contributed to reticence in applying for safeguard relief. The one China-specific safeguard that was imposed – on passenger vehicles and light truck tires – came eight years after China’s accession to the WTO. It was also the first such safeguard challenged by China, with the WTO’s Appellate Body upholding the United States’ determination to apply safeguard measures to Chinese tires.⁵¹

Antidumping (AD) (selling goods in foreign markets for less than prices at home) and countervailing duties (CVD) (offsets for government subsidies), however, were more commonly deployed. From 2001 to 2020, WTO members have reported imposing 917 AD measures against Chinese imports, approximately 30% of the total reported, and 129 CVDs against Chinese imports, approximately 46% of the total actions.⁵² For U.S. industries, the primary tool to respond to increased imports from China, particularly in the 2000s, was anti-dumping duties, with the use of the “non-market economy methodology” permitted under Section 15(b) of China’s protocol of accession to calculate the amount of those duties. As of September 2021, the United States has 142 AD orders in place on various goods from China, far more than the number of AD measures against imports from any other country.⁵³

Prior to 2007, the United States did not apply its CVD law to countries considered to be nonmarket economies (NMEs) based in part on a conclusion by the Department of Commerce (Commerce) that it could not determine where government action began or ended and therefore could not specifically identify subsidies. In 1986, the US Court of Appeals for the Federal Circuit in Georgetown Steel Corp. v. United States upheld this interpretation of the CVD statute as reasonable. In 2006, Commerce changed its position, accepting a petition seeking a CVD on imports of coated free-sheet paper from China. Commerce distinguished the current Chinese economy from the Soviet-style economies at issue in Georgetown Steel and found that the imported Chinese paper was subsidized. Numerous CVD cases followed, with 80 CVD orders now in place against imports from China. But the rest of the world has been less willing to use this tool, perhaps in part because it involves a direct challenge to the practices of the Chinese government and the CCP compared to anti-dumping cases, which focus on the behavior of individual companies. Even less clear is how willing countries are to use the “special difficulties” tool provided in Paragraph 15 to overcome evidentiary hurdles in proving the existence of a subsidy.

The textile-specific growth limit was similarly far less utilized than might have been expected, given China’s dominant position as a supplier of textiles and clothing. In the United States, the use of such safeguards became bogged down in protracted legal battles over the application when there was only a threat of market disruption.⁵⁴

The world’s failure to use these tools – at least not early and often enough – was part of what allowed China’s relatively unchecked rise in exports to the world.⁵⁵ In the United States, China’s rise was documented and labeled in a 2016 article titled “The China Shock: Learning from Labor-Market Adjustment in Large Change in Trade,” by David Autor from MIT, David Dorn (University of Zurich) and Gordon Hanson (UC-San Diego).⁵⁶ The article noted that China’s economic size, speed of growth, and import penetration were all of an order of magnitude different from previous waves of imports from Japan or Mexico or others. Imports from China grew from 1.0 per cent of the US GDP in 2000 when China’s accession to the WTO was being negotiated to 2.6 per cent of GDP ten years later. The impact on communities where goods competing directly with Chinese imports (such as furniture, toys, electronics, jewelry, shoes, and clothing) were particularly profound, due in part to the lack of sufficient safety nets or trade adjustment assistance and in part due to the unexpected immobility of labor.

While the U.S. imposed 199 anti-dumping and countervailing tariffs, primarily targeting raw and semi-processed imports with some cases addressing finished goods like furniture and tires, the wave of imports appeared to simply overwhelm many companies, leading to a conclusion that the trade rules were not up the task of coping with the China shock and to the parallel determination that it was a mistake for the United States to have allowed China to enter the WTO on the terms that it did.⁵⁷

(ii) Failure to Hold China to Its Commitments on Transparency, Market Economy Orientation, Rule of Law, and More

A critical aspect of its advocacy for PNTR and China’s accession to the WTO was the assertion that China’s conduct would be “monitored by more than 130 other WTO Members with a common interest in seeing China’s market opened.”⁵⁸ “Unlike our bilateral agreements,” the Clinton administration argued, “we will not be alone in our enforcement efforts if China fails to live up to its commitments.”⁵⁹

The most significant commitments for which compliance was sought involved those that required China to open up its market under numerous specific commitments with respect to trade in goods, agriculture, and particular services commitments;⁶⁰ various commitments with respect to the rule of law and transparency in the Chinese system,⁶¹ intellectual property rights requirements,⁶² and numerous commitments intended to keep China on the path toward becoming a fully market-oriented economy.⁶³

For the first few years following China’s 2001 accession to the WTO, most WTO members took a “wait and see” approach. For its part, the United States filed 23 different cases against specific Chinese practices, with its first case filed in 2004, winning all that was completed, settling eight of them through mutual agreement, with three still pending.⁶⁴ The entire rest of the world combined brought a comparable number of cases, with many countries appearing reluctant, particularly early on, to challenge China for fear of retaliation or for lack of evidence from China’s opaque system. However, each of these cases was somewhat narrowly focused on individual measures or particular sectors. None spoke to the bigger, more systemic issues that are at the heart of U.S. concerns with China or to China’s failure to fulfill its notification and transparency requirements.

What might have been a better approach would have been a “big, bold, coalition-based case” that would have represented an “opportunity to bring together enough of the trading interests in the world to put sufficient pressure on China to make it clear that fundamental reform is required if China is to remain a member in good standing in the WTO.”⁶⁵ Just such a case was recommended by the U.S.-China Economic and Security Commission, based in part on the author’s testimony to the Commission.⁶⁶

In my construction, the case could include claims to address China’s: (1) coercion of technology transfers in light of China’s commitment that it would not condition investments on the transfer of technology,⁶⁷ (2) restrictions on the right of foreign companies to license their technology (or choose not to license it) under the conditions and terms that they would like in violation of China national treatment and MFN commitments;⁶⁸ (3) direction of outbound investment to obtain cutting-edge technology in service of China’s industrial policy, in violation of China’s commitment to treat foreigners on a reciprocity basis;⁶⁹ (4) investment restrictions that preclude or unreasonably delay market entry for foreign companies in violation of China’s commitment not to condition investments on performance requirements or technology transfer,⁷⁰ (5) use of export taxes to restrict or encourage certain exports over others, in violation of China’s commitment not to charge such export taxes other than on a specific list of products,⁷¹ (6) services restrictions that are inconsistent with China’s GATS schedules, (7) restrictions on agriculture imports under non-transparent and non-science based sanitary and phytosanitary measures, (8) lack of transparency and access to China’s laws, regulations and rules on timely basis, (9) failure to establish independent judicial review of trade-related administrative decisions,⁷² and (10) failure to meet the reasonable expectations of WTO members that China’s economy would become a market-oriented one.⁷³

The last claim that I suggested is the one designed to get at the heart of the United States’ concern – that China’s U-turn away from market orientation to an ever more state-controlled economy violates the spirit, if not always the letter, of the WTO.

(iii) China’s Turn Away from Market-Oriented Reforms

The overarching Marrakesh Agreement establishing the WTO declares that the organization was designed as a world trading system “based upon open, market-oriented policies.” China, for example, expressly declared as part of its accession commitments that “that all state-owned and state-invested enterprises would make purchases and sales based solely on commercial considerations, e.g., price, quality, marketability and availability …”⁷⁴

As a result, the United States and all other WTO members had legitimate expectations that China would increasingly behave as a market economy – that it would achieve a discernable separation between its government and its private sector, that private property rights and an understanding of who controls and makes decisions in major enterprises would be clear, that subsidies would be curtailed, that theft of IP rights would be punished and diminished in amount, that SOEs would make purchases based on commercial considerations, that the Communist Party would not, by fiat, occupy critical seats within major “private” enterprises and that standards and regulations would be published for all to see.

But starting in the mid-2000s, China began what has now become a complete U-turn back to becoming a state and Communist Party dominated economy.⁷⁵ While parts of the economy appear to have a thriving private sector, intervention by the government and the CCP has become far more pervasive. Institutions were established giving Beijing tighter control over China’s large and fast-growing SOEs (overseen by the State-owned Assets Supervision and Administration Commission (SASAC)) and its banks (influenced via Central Huijin Investment). The overlapping ways in which China’s economy is unique and state driven results in a phenomenon dubbed “China, Inc.” that is hard to reach with WTO rules.⁷⁶ Concerns about the direction of China’s economy greatly intensified with the rise of Xi Jinping as China’s leader in 2012 and the release of the Made in China 2025 plan designed to create Chinese self-sufficiency in a range of critical technology sectors.

These concerns were succinctly summarized in the statement made by the then U.S. Ambassador to the WTO, Dennis Shea, in a May 8, 2018 statement to the WTO General Council:

China … is consistently acting in ways that undermine the global system of open and fair trade. Market access barriers too numerous to mention; forced technology transfers; intellectual property theft on an unprecedented scale; indigenous innovation policies and the Made in China 2025 program; discriminatory use of technical standards; massive government subsidies that have led to chronic overcapacity in key industrial sectors; and a highly restrictive foreign investment regime.⁷⁷

It is this collective failure by China that underlies the trade friction between the United States and China.

The concerns over the market orientation of China’s economy are shared by many other WTO members. When the G-20 trade ministers met in September 2020, for example, much of the discussion centered on strengthening the WTO, beginning with a reaffirmation of commitment to the objectives and principles in the Marrakesh Agreement. But when it came to affirming that “market-oriented policies” is a principle of the WTO, China objected. It is this fundamental split that has led some leading trade scholars to conclude that “the world is now presented with two conflicting economic systems: (1) a Western-led, market-driven, model based on the rule of law and (2) an authoritarian state-driven model championed by China,” and that the solution is to “establish a ‘compact’ among like-minded developed market economies to agree to new common approaches to counter unfair non-market practices; address critical twenty-first-century economic issues such as the digital economy, climate change and the environment, and labor; and improve economic ties in industries that are key to innovation, economic growth, and national security.”⁷⁸

V Conclusion: Implications of the Failures and Where We Go from Here

The failure of China’s accession to the WTO to meet the very large and arguably unrealistic expectations across the economic, trade, geopolitical, and national security arenas likely portends a permanent shift in the U.S. approach to China. The initial vision of the Clinton administration was that the WTO could serve as the backstop to the Sino-American trade relationship – a facially neutral arbiter that could help enforce the liberal trade rules that Washington preferred. It was created and nurtured in an era where U.S. strategists saw trade with China and the rest of the world as a largely unambiguous good, capable of advancing both their strategic and domestic political ambitions. For the reasons noted above, that view has fundamentally changed.

This new reality means, at a minimum, that the U.S. will need to rely on a far wider array of tools than simply the WTO and its dispute settlement process to address its concerns.⁷⁹ It also means that reforms at the WTO will need to be deeper and more directed at addressing the fundamental schism between market-oriented economies and state-controlled ones.

A fulsome explanation of what the U.S. should do outside of the WTO is beyond the scope of this paper, but it should include at a minimum renewed attention to the work of the U.S.-EU-Japan trilateral cooperative process aimed at developing new rules on subsidies and tech transfer and greater reliance on the deep transatlantic alliance with Europe, the recently reinvigorated Quadrilateral Security Dialogue (Australia, India, Japan, and the United States) and alternative forums such as the G-7, the G-20, and the OECD to develop a more coordinated approach to the China trade challenge. Also included should be an exploration of the U.S. rejoining the Trans-Pacific Partnership (now the Comprehensive and Progressive Agreement for Trans-Pacific Partnership (CP-TPP)) in order to ensure that the United States has a seat at the table while the trade architecture in Asia is put in place and reassure our trading partners that the United States remains committed to the region. The U.S. must also follow through on the Biden administration’s Build Back Better plan to shore up American competitiveness and supply chain resilience, particularly in key sectors and technologies. It must also rely on both bilateral and plurilateral dialogues to address the many geostrategic and national security concerns raised by a more assertive China.

At the WTO, the United States needs to lead the effort to fix the WTO. The WTO is in desperate need of reform and revitalization, but that will not happen absent U.S. leadership and commitment. Reforming the WTO would also allow it to be a more effective tool among many that the United States will need to address its China concerns. The reforms need to focus on the structural flaws at the WTO, including the imbalance between its dispute settlement, negotiating, and executive functions and the unsustainable bifurcation between developed and developing countries that China has exploited despite its immense economic heft and the power of its trade. The reforms will also need to focus on the gaps in the substantive rules, starting with new disciplines on the transfer of technology, the classification of State-Owned Enterprises (SOEs), and the rules on subsidies. Additional efforts should also be made to bring into the WTO the e-commerce and digital trade provisions from the USMCA or the CP-TPP.

While it is clear that the WTO should not and cannot serve as the only forum for working out America’s concerns with the rise of China, the WTO cannot and should not be abandoned. It should be fixed in its own right as the premier forum for bringing together the world’s trading nations to negotiate and enforce rules, exchange information, disseminate best practices, and provide transparency with respect to trading practices and measures. In so doing, the WTO can begin what will be a long process of developing rules or norms to address concerns over China’s industrial policies, its non-tariff barriers, and its abuse of intellectual property and technology rights.

In the end, the United States expected too much of the WTO and the WTO delivered too little. Righting that balance will require a stronger, more responsive, and more inclusive WTO and a more robust tool kit to address the national security and geostrategic concerns that the WTO was never going to be in a position to resolve.