Reforming the World Trade Organization: Practitioner Perspectives from China, the EU, and the US

estimation


I. Introduction
After 26 years since its establishment in 1995, the World Trade Organization (WTO) undoubtedly needs to renovate both its rules and working practices.Structural weaknesses and gaps in WTO rules were evident right from the start.Over 20 years ago, Sylvia Ostry (1999), one of the architects of the WTO, provided a compelling case for institutional the EU and the US are broadly aligned on the WTO reform agenda, while respondents from China often diverge in the priorities accorded to these subjects.Our premise is that reforming the WTO is a necessary condition for the organization to be a more salient forum for the large economies to address trade tensions, which in turn would help to resolve the problems of the WTO.
The paper is organized as follows.We begin in Section II with a brief discussion of the systemic context of WTO reform.Section III introduces the survey data and the methods used, which form the basis for the analysis in the rest of the paper.Section IV focuses on the negotiation function of the WTO and the initiative to launch plurilateral negotiations among groups of WTO members as a response to the problems created by the WTO consensus working practice.Section V discusses the approach taken to recognizing economic development differences, which has been an important negotiation obstacle and thus a prominent subject in WTO reform debates.Section VI discusses the challenge of improving transparency, another key area of WTO reform, as a common understanding of prevailing policies that is a necessary condition for cooperation.Section VII considers reforms to WTO working practices to enhance the use of WTO bodies for joint deliberation on the operation of the WTO -which is salient for transparency as well as preparing the ground for negotiations.Section VIII turns to a fi nal area where action by WTO members is needed: addressing the Appellate Body crisis and improving disputesettlement processes.Section IX discusses implications of our analysis for fostering cooperation between the three major trade powers in the WTO.Section X concludes.

II. The systemic context of WTO reform
Many of the WTO's problems are due to a general malaise in multilateralism and to structural changes in the world economy.Power matters, it has shifted, and the system has not caught up.Whether or not the US ever was a "hegemon" (Snidal, 1985), it manifestly is not one now.It is uncomfortable in this new situation and is especially uncomfortable with the rise of China.One might normally expect a rising power to respond to shifting power by demanding institutional change (Goddard, 2020), but it is not clear that that is what China wants.Chinese scholars argue that "WTO members need to engage with China in a more positive and constructive way, taking China's rise and 'Chinese characteristics' into consideration" (Li and Tu, 2018).Indeed, some scholars think the challenge of rising powers in the WTO is not to its institutional arrangements but to the dominance of the US (Hopewell, 2016).On the other hand, sometimes it is the established or declining power that seeks institutional change (Kruck and Zangl, 2020), which may be a better description of what we observe.
Generations of Atlanticism, that is, of shared elite perceptions, helped underpin the post-World War II order.Nothing comparable exists for the emerging 21st century order.Every round of multilateral trade negotiation under the General Agreement on Tariffs and Trade (GATT), including the Uruguay Round, only crystallized when the US and Europe had reached a basic accommodation, which then provided the parameters for the rest of the bargains.The EU and the US still disagree about many things, notably how to pursue regulatory cooperation and the role of the Appellate Body, but solving their outstanding conflicts will not remove the central blockage in the WTO.
The US and China have not developed a trans-Pacifi c accommodation comparable to the one painstakingly established across the Atlantic.That accommodation will require a recognition by both parties that China now has a leadership role in global governance.China's institutional role is not yet commensurate with its weight in the world economy.China has indicated it will accept reforms that make the WTO better for all members, but not ones that challenge its identity as a developing country, that deny it scope for how it organizes its economy, and that fail to recognize its status as a major power (Liu, 2019;Gao, 2021;Tan, 2021).
The paradox of the postwar order was that it encompassed commitments to free trade abroad and the administrative state at home (Ruggie, 1982).This paradox has made practical people uncomfortable for a long time whether they favour one arm or the other; the rise of China just makes their discomfort worse.Can the WTO as a normative framework survive if powerful members, such as the US and China, have such divergent conceptions of the role of the state in the economy?The US government has a significant role in the economy, from welfare spending through regulation, to supporting R&D in key sectors and incentivizing investment through subsidies and public procurement.China does too.Although the instruments differ, all WTO members defend their right to manage their economies in their own way.
At its core, the compromise of embedded liberalism was about reconciling domestic differences within a liberal trade order.If a leading state is not liberal, does the compromise of embedded liberalism break down?That is the real question that the US asks in a document tabled at the WTO in July 2018 on the importance of market-oriented conditions.This submission recalled the expectation that when China joined the WTO it would continue the transition to an economy based upon open, market-oriented policies.
Instead, China's "state-led, trade-disruptive economic model … imposes substantial costs on and presents severe challenges to WTO members" (WTO, 2018, p. 2).Among its specific complaints, the US asserted that China's opaque system makes it hard for members to assess China's compliance with its WTO obligations, notably because of inadequate subsidies notifi cations. 2  These types of concerns were already in evidence during China's accession negotiations and made that process long and diffi cult.The Accession Protocol contains various measures permitting members to observe implementation of negotiated domestic reforms and apply measures to address potential competitive distortions during a 15-year transition period.This framework faced new pressures as provisions of China's accession protocol came to an end (Lang, 2019).The concerns regarding China's economic model were reflected in unilateral actions and international initiatives that excluded China.
The US has repeatedly raised concerns on which it thinks WTO reform depends, even going so far as to submit a draft resolution proposing that the General Council express its serious concerns with non-market-oriented policies and practices "that have resulted in damage to the world trading system and lead to severe overcapacity, create unfair competitive conditions for workers and businesses, hinder the development and use of innovative technologies, and undermine the proper functioning of international trade" (WTO, 2020a, p. 1).
Chinese scholars recognize this US view, which they contrast first with the vast number of developing members represented by China who adhere to the basic purposes and principles of the WTO, and second with the compromise position of such major economies as Europe, Canada, and Japan (Liu, 2019).In a 2019 submission to the General Council, China indicated that it supports WTO reform if core values of the multilateral trading system such as non-discrimination and openness, safeguards for the development interests of developing members, and decision-making by consensus are preserved.China made clear in its submission (WTO, 2019c) that it sees the US as the problem, with proposals on: • breaking the impasse of the appointment process of Appellate Body member; • tightening discipline to curb the abuse of the national security exception; • tightening discipline to curb unilateral measures inconsistent with WTO rules.
The suggestions on improving trade remedies rules similarly target areas where developed countries could be said to have abused the existing rules, for example on price comparison in antidumping proceedings, or subsidy identifi cation, and calculation 2 Before China joined the WTO, analysts worried that it would be unable to meet transparency requirements due to domestic politics, institutional capacity, and the nature of the legal system.Although reforms occurred after 2001, not necessarily because of the WTO (Du and Kong, 2020), issues with regulatory transparency, notifi cation of subsidies, and implementation of reforms at sub-central government levels persist.
of benefi ts conferred.In response, Chinese offi cials consistently make the obvious and valid point that the market is not given free rein in OECD countries. 3  The EU is sympathetic to the US analysis but tellingly did not co-sponsor the proposed WTO General Council resolution.It also went ahead on its own in the negotiations for the EU-China Comprehensive Agreement on Investment announced at the end of 2020.At an event celebrating the 25th anniversary of the organization, EU Director-General for Trade Sabine Weyand remarked that "the WTO is not the place to drive systems change.It's not about regime change.This is about dealing with the consequences of certain economic systems and to make sure that these are being dealt with in a manner that everyone can live with.And that requires compromise on all sides" (Monnicken, 2020, p. 2).The EU prefers to focus on concrete spillovers of non-market practices that would be susceptible to action in the WTO.The EU also wants to address issues where its views and interests are not aligned with the US, as in their paper on WTO reform (European Commission, 2021).
This triangular tension is evident in the survey results that are presented in what follows.

III. Data and methods
Our discussion of perspectives on the WTO reform agenda in China, the EU and the US draws on the results of an original survey of trade experts (Fiorini et al., 2021).The survey instrument includes questions on different challenges confronting the WTO and collects information on the priorities accorded to each by members of WTO delegations, government offi cials dealing with the WTO, and other stakeholders.

The survey instrument
The survey ran between 5-21 June 2020 using an online survey platform.Data were compiled on 24 June.The questionnaire was sent by email to all WTO delegations and distributed through e-mail and social media channels to trade practitioners using a contact list developed by the Global Economics Program of the EUI Robert Schuman Centre for Advanced Studies.Recipients were requested to forward the survey to others in their networks working on trade and WTO matters. 43 See, for example, the commentary by Jesse Kreier, a former WTO offi cial, on a November 2020 affi rmative preliminary determination by China regarding alleged dumping of wine from Australia.This discusses Australian market conditions in detail, and based upon the facts available fi nds the existence of "non-market conditions" including supportive measures through, inter alia, R&D funds and investment subsidies.These fi ndings parallel fi ndings of government intervention in the US energy market.Available from: https://ielp.worldtradelaw.net/2020/12/china-preliminary-ad-determination-on-australian-wine.html.

4
The sampling frame is not designed to ensure that respondents will be representative of the WTO stakeholder community and no claim to this effect is made.Responses to all questions used a 5-point scale, with values from 1 to 5. The value 1 corresponds to the response "very low" (or "strongly disagree"), the value 2 to "low" ("disagree"), 3 to "neutral", 4 to "high" ("agree"), and 5 to "very high" ("strongly agree").Respondents were asked to indicate their professional affi liation (more than one allowed) and the nationality of their company, organization, or institution. 5A total of 1,092 people opened the survey; 75 percent (819) responded to at least one question.Around 800 responses were received for most of the questions (the lowest number of respondents for a question is 733; the highest 807).Government offi cials (including the EU) represent the second largest category of respondents (24.6 percent) after academia (25.1 percent), followed by the private sector (companies and business associations) (19 percent), staff of international organizations (18 percent), and nongovernmental organizations, labor unions and think tanks (11 percent of the sample). 6  The geographic distribution of respondents covers 104 WTO members.One third of respondents were female.Of the 616 respondents that specifi ed the nationality of their employer, 37 percent were in Europe, 20 percent in North America, 18 percent in Asia, 12 percent in Africa, and 11 percent in Latin America and the Caribbean.Fiorini et al. (2021) report further details on the survey structure, the empirical population of respondents as well as descriptive statistics on all answers.For the analysis that follows, we extract the results for respondents who identifi ed the nationality of their organization as China, the EU or the US.The resulting sample consists of 215 respondents from the EU, 85 from the US,and 18 from China.We recognize that the small number of respondents from China makes robust inference impossible, but we have some confi dence that results are nonetheless informative because of the design of the sampling frame: respondents are experts, i.e. the responses refl ect the views of sophisticated members of the Chinese trade policy community.

Empirical analysis
We use two analytical instruments to characterize the revealed preferences of respondents in China, the US, and the EU, on key elements of the WTO reform agenda.
The fi rst instrument is a discrete choice model to estimate alignment or nonalignment of revealed preferences between two groups of respondents on a given issue.More precisely, we take responses to a specific issue (e.g.What priority do you assign to providing support for plurilateral agreements?) as the dependent variable y of an 5 Survey responses are anonymous.The software used makes it impossible for the researchers to identify respondents.Any question could be left unanswered. 6Retirees were asked to indicate their professional affi liation during their last working position.
ordered probit model with a unique regressor consisting of a dichotomous variable identifying the two groups whose revealed preferences on that issue we want to compare.
The latent variable model is specifi ed as y* = βD + e, where the dependent variable is given by the answers to a specifi c question, with values ranging from 1 (very low/ strongly disagree) to 5 (very high/strongly agree).The regressor of interest is a dummy variable, D, taking value 1 if the respondent belongs to the relevant group (e.g.China) and 0 if he/she belongs to the respective baseline group (e.g. the US).The latter becomes the baseline group for the comparisons we discuss below.The latent variable y* can be interpreted as the continuous utility from priority being assigned to the issue specifi ed in the specifi c survey question.The error term e is assumed to be normal, i.e. e|D is distributed as Normal (0,1).
The sign of estimated β through the ordered probit model using the observed categorical variable y can be interpreted in terms of the conditional expectation of the associated unobservable latent variable (y*, the continuous utility) that is the focus of analysis. 7The point estimate indicates whether belonging to a given group is associated with smaller utility (if negative) or higher utility (if positive) from priority being assigned to the issue in the question, relatively to the baseline group.In the discussion that follows, point estimates are presented graphically together with their respective 95 percent confidence intervals.If the confidence interval lies completely to the left or completely to the right of the vertical line at 0, the association entailed in the point estimate is statistically signifi cant.
The second empirical instrument used to characterize survey responses is a rank correlation analysis.Individual issue areas are clustered according to thematic areas (e.g.questions pertaining to improving WTO transparency) and for each group of respondents the issues within a cluster are ranked in terms of the total scores assigned by a respondent's group.We then estimate the Spearman rho correlation coefficient8 between group-specifi c rankings for each cluster of issues.These correlations provide empirical evidence on whether the preferences of two groups of respondents reveal alignment (positive correlations), are uncorrelated (correlations close to 0), or show nonalignment (negative correlations) in rankings of relevant issues.

IV. Plurilateral negotiations as a response to the consensus problem
The accession of China at the 2001 Doha ministerial occurred in tandem with the launch of the ill-fated Doha Round at the same ministerial.The backward-looking Doha Round agenda prioritizing tariffs on manufacturing and agricultural support policies became increasingly disconnected from 21st century priorities as the negotiations dragged on, eventually becoming deadlocked.One result was that policies affecting the digital economy, and cross-border data flows and foreign investment, among others, were neglected because a consensus could not be achieved to address issues that were not part of the Doha agenda.Another implication of the deadlock was that the WTO did not play a role as a forum for deliberation on policy differences that gave rise to trade tensions between the US and China, notably the use of industrial subsidies and the activities of state-owned enterprises (SOEs).
In 2017, many countries decided to shift gears and move away from negotiations that span all WTO members and are premised on the working practice of consensus decisionmaking by launching plurilateral talks among groups of WTO members.Underlying the interest in plurilaterals are the trade tensions resulting from the rapid growth in exports of emerging economies.That tension has been growing for decades.The Uruguay Round single undertaking, which forced all developing countries, and all those who subsequently acceded to the WTO, to accept the whole set of agreements exacerbated the problem.The move to plurilaterals is only a partial solution to the diffi culty of concluding negotiations by consensus.Each negotiation can only be concluded if a critical mass of members participates.Plurilateral approaches,therefore, are not a panacea, but they offer a mechanism for large trade powers to cooperate without engaging in negotiations with all WTO members (Hoekman and Sabel, 2021).
The so-called "joint statement initiatives"(JSIs) that are now being pursued in the WTO on a plurilateral basis span e-commerce, domestic regulation of services, investment facilitation, and measures to enhance the ability of micro and small and medium enterprises (MSMEs) to utilize the opportunities offered by the rule-based trading system.These initiatives include a cross-section of the WTO membership.The e-commerce JSI talks involve more than 80 WTO members and focus on a mix of trade restrictive policies such as regulation of cross-border data fl ows and data localization requirements and digital trade facilitation -issues like electronic signatures, e-invoicing, electronic payment for cross-border transactions, and consumer protection. 9Talks on domestic regulation of services span more than 60 WTO members and center on 9 For a summary of the issues that have been tabled by different participants, see Y. Ismail (2020).
matters associated with authorization and certification of foreign services providers with the aim to reduce the trade-impeding effects of domestic regulation.The MSME and investment facilitation groups differ from the other two JSIs in addressing matters that are not covered in existing WTO agreements.The talks on MSMEs involve some 90 WTO members; those on investment facilitation over 100 members. 10China was an initiating co-sponsor of three of the four JSI groups and joined the fourth group -on e-commerce -soon after deliberations commenced.The EU also participates in all four groups.Although the Trump Administration chose to join only one of these initiatives, on e-commerce, this decision may be revisited by the Biden Administration.The point estimates and associated confi dence intervals reported in Figure 1 show that both EU and US respondents to the survey gain far more utility from (attach greater priority to) concluding the JSI talks than Chinese respondents.At the same time Figure 1 also shows that Chinese and US respondents are aligned on the question of whether WTO members (and thus the Secretariat) should provide support for plurilateral agreements.European Union respondents are slightly less supportive than the US and Chinese respondents on this second question, but the differences are much smaller than for the question on successful conclusion of the extant JSIs.The survey responses suggest that Chinese respondents are less supportive than those from the EU and the US, but, as we would expect, are not opposed to the principle of pursuing plurilateral agreements in the WTO. 11

V. Special and differential treatment as a negotiation obstacle
The prospects for plurilateral initiatives to attract wide membership, and for agreement to be possible among the EU, US, and China will depend significantly on whether emerging economies insist on being accorded special and differential treatment and more generally on whether and how such agreements address development differences.Special differential treatment for developing countries has become part of WTO theology, to the point that it was baked into the 2001 Doha Development Agenda.In practice, it does not matter if small developing countries are permitted to provide less than full reciprocity in a negotiation, but unwillingness of larger developing countries matters to major exporting nations.This has long been a bone of contention in the WTO (Hoekman, 2012), but has become a rising source of division in the WTO because of the economic success of large emerging economies.
A lengthy document submitted by the US to the WTO General Council early in 2019 (WTO, 2019b) provides a great deal of data, and commentary on WTO decisions, arguing that self-declared developing country status no longer makes sense and undermines new negotiations.In May and November 2019, the US submitted a proposal for a decision on "Procedures to Strengthen the Negotiating Function of the WTO" with criteria for assessing which countries will not avail themselves of SDT in WTO negotiations (WTO, 2019d).The US asked for this item to be placed on the agenda of one General Council meeting after another in 2019 and 2020, with some support from other members but unrelenting opposition from China and most developing countries.

11
This conclusion is consistent with the fi nding discussed in Section VIII: when it comes to launching talks on new issues (other than the 2017 JSIs), the US and Chinese respondents are aligned.
At the July 2020 meeting of the General Council, the representative of China, echoing the introduction and much of the argumentation of an earlier submission on SDT by China and others (WTO, 2019a), said that in an international organization with developed and developing members, non-reciprocity was a means and a principle to realise equity.He argued that reclassifi cation of WTO members was not a way out.Rather than revisiting the current practice of self-designation of developing country status, he suggested those in a position to do so be encouraged to make a greater contribution to the best of their capabilities, which China was willing to do (WTO, 2020d).The EU has suggested flexibilities be made available to those members who actually need them to be able to fully benefi t from their membership in the organization, an approach advocated by former WTO offi cials (Low et al., 2019).This approach has long been advocated by some scholars as a more effective means of ensuring specifi c capability gaps be identifi ed and addressed (Gao, 2021). 12 Survey respondents from China, the EU and the US are close to each other in the utility they would receive from resolving differences on SDT (Figure 2) but, surprisingly, in results not shown here, all three sets of respondents accord signifi cantly less weight to such a resolution than respondents located in other WTO members.A possible reason is a recognition that these three major traders will need to negotiate rules that apply equally to each of them, with specifi c commitments and exceptions agreed on an issue-specifi c basis.Addressing the spillover effects of subsidies is an example where this applies.
WTO subsidy rules are intertwined with SDT reflecting a presumption that subsidies can play an important role in fostering economic development.This is demonstrated in the ongoing negotiations on fi sheries subsidies.Launched in 2001, these talks have yet to be concluded, missing the latest deadline of end-2020.There are many reasons for the failure, but SDT was a factor.China's position is ambiguous.Echoing Annex D paragraph 9 of the 2005 Hong Kong ministerial declaration (WTO, 2005), China supports disciplines that prohibit fi sheries subsidies that contribute to overcapacity and overfi shing, and eliminate subsidies that contribute to illegal, unreported and unregulated (IUU)-fishing, while "recognizing that appropriate and effective SDT for developing country members and least developed country members should be an integral part of the negotiations" (WTO, 2019f).
Another proposal by the US and others (WTO, 2019e) proposes a sliding scale of commitments that differentiates members by proportion of marine capture production.This would result in differentiated levels of commitment, meeting developing countries' substantive needs for fl exibility.In contrast, an Indian proposal on subsidies for fi shing in a country's exclusive economic zone that contribute to overcapacity (WTO, 2020e) would signifi cantly limit China's access to SDT because of a provision restricting this to developing countries with per capita incomes below US$5,000.Gao (2021) suggests such an approach would allow China to be a developing country in name but not in practice.
Figure 2 makes clear that Chinese respondents to the survey gain much less utility (priority) than those from the EU and the US from addressing this subject, although the EU and the US are also not aligned.The same is true for addressing the negative international spillovers of industrial subsidies, which play a major role in the trade tensions between the US, EU and China.Respondents are even further apart on the utility of revisiting rules on industrial subsidies than they are on fi sh subsidies.A possible reason is that subsidies can help to address market failures and therefore might have a good economic development rationale despite giving rise to potential negative crossborder competitive spillovers.As discussed at greater length in Hoekman and Nelson (2020), this is not simply a "China issue."Subsidies of one type or another constitute the great majority of trade interventions imposed since 2009.To address these tensions, it is necessary to have a good understanding of the use and incidence of subsidies.
Governments can see distortions that look like they were caused by industrial subsidies offered by other countries, but they lack the data to illuminate that state support.Rather than engaging in discussions at the WTO, the trade ministers of the US, Japan, and the EU launched a trilateral process motivated by "concern with the non-market-oriented policies of third countries and […] actions being taken and possible measures that could be undertaken in the near future" (European Commission, 2018, p. 1).In a May 2018 statement, the trilateral ministers endorsed a joint scoping paper defining the basis for the development of stronger rules on industrial subsidies and SOEs.They also mentioned technology transfer and noncompliance by some governments with their WTO transparency obligations (USTR, 2018a).These themes recurred when the trilateral ministers met again in September 2018 when they "called on advanced WTO members claiming developing country status to undertake full commitments in ongoing and future WTO negotiations" (USTR, 2018b, p. 2).
The trilateral approach pursued by the EU, Japan, and the US led to circular discussions in the WTO Subsidies and Countervailing Measures (SCM) committee.
For example, in October 2020, according to a Geneva trade offi cial, each of the three expressed their joint concern about the role of state subsidies in contributing to excess production capacity in sectors such as semiconductors, steel, aluminium and others.
The US and the EU raised concerns about China's lack of information regarding subsidy programs referred to in offi cial communications.China replied that the Ministry of Commerce (MOFCOM) publishes large amounts of information in its trade and economic gazettes as well as on its website and that its national enquiry point fulfils China's WTO transparency obligation.The US responded it had searched the MOFCOM Offi cial Gazette but was unable to fi nd the information it sought.
Survey respondents from China are far apart from those in the US and EU on the priority attached to negotiating stronger rules on industrial subsidies, while the latter two groups are much more aligned.But a revamped subsidy regime requires participation of all three major trading powers.The rules must be seen to support the generalized gains from open trade and global production, not an attempt to isolate or "reform" China.At the same time, China should accept that it has a leading role to play in the regime.In the short term, agreement on binding rules is unlikely to be possible.Work on developing more informal discipline on subsidies based on information, dialogue and peer review may be more feasible, especially if concluded on a plurilateral basis (Li and Tu, 2020).
This applies as much SOEs, another focal point of the trilateral concern, as it does to subsidies.WTO members do not necessarily know enough about SOEs, including in China, to be sure whether there is a systemic problem, and hence what ought to be done. 13The absence of specific WTO disciplines on SOEs, including the absence 13 For a more extended discussion, see Wolfe (2017).
of a definition, leads to the non-existence of transparency requirements about who they are and what they do.On any defi nition, OECD countries also have many SOEs.
Multilateral consensus on new rules might prove elusive, but if all the players involved knew more, and had a space to discuss what they know, they might be able to use the neutral ground provided by the WTO to discuss how to attenuate competitive spillovers caused by SOEs and subsidies more broadly.
The Appellate Body seems to think that, in terms of the Agreement on Subsidies and Countervailing Measures (ASCM), a body exercising "governmental authority" can be identified by what it does, not by who owns it.The distinction may matter in the case of an SOE providing a subsidy but does not help in assessing other ways in which SOEs might affect the commercial interests of foreign fi rms.Moreover, without more information on the universe of SOEs, and what they do, the distinction has little operational utility, although the fact of the false distinction contributed to American unhappiness with the Appellate Body and its jurisprudence (Ahn, 2021).Mavroidis and Sapir (2021) argue that a more appropriate path to addressing concerns about the competitive spillovers of the operation of SOEs is through reform of the ASCM.The behavior of SOEs in the EU is subject to integrated state aid and competition law disciplines, which apply to all fi rms, whatever their ownership structure.
Adopting a similar approach in the WTO may be more feasible and more effective than seeking to discipline the use of SOEs.Here again a necessary condition for cooperation is a common understanding of the extent and spillover effects of contested practices.
As noted by Hoekman and Nelson (2020), calling for a work program on subsidies may be criticized as kicking the can down the road.It is not.World Trade Organization members simply do not have enough information to develop a common understanding of where new rules are needed and the form they should take.That problem goes beyond subsidies.

VI. Improving transparency is central to WTO reform
Transparency of actor behavior and expectations is a core requirement of international regimes.This objective requires high quality information (Wolfe, 2018).The WTO agreements have dozens of formal notification obligations; compliance varies by committee and by member.One-time obligations to notify existing legislation can be simpler than ad hoc ex ante notifications of new regulations, which in turn are often easier to prepare than regular ex post notifi cations of subsidies.In the current context, only some of the COVID-19-related measures reported to the WTO are "notifi cations." The WTO website lists regular (required) notifi cations with a COVID-19 connection, voluntary reporting of COVID-19-related measures, responses to Secretariat requests for information, and the Secretariat's own information derived from web scraping and news reports. 14There are problems with all these sources of WTO information.
Inadequate notifi cation of trade policies is an old issue, but its inclusion on the "WTO reform" agenda only began at the 2017 Ministerial Conference when Robert Lighthizer, then the US Trade Representative, said that "it is impossible to negotiate new rules when many of the current ones are not being followed" (USTR, 2017).The US tabled a detailed proposal that reviewed how compliance with notifi cation obligations under the Trade in Goods agreements is unsatisfactory.The US proposal included punishment for members who are behind in their notifi cations (WTO, 2017).A revised version of that proposal (WTO, 2020b), with several co-sponsors, including the EU, suggests consideration of both systemic and specific improvements that can help members improve compliance with notifi cation obligations.
Although uneven compliance with notifi cation requirements of WTO Agreements is generally recognized, there is no consensus on a diagnosis.Some of the problems with notifi cations may be due to conscious unwillingness to provide the information, or a general lack of transparency in China's governing institutions, but threats to identify the Chinese ambassador as a "member with notifi cation delay" when offered the fl oor in the General Council (WTO, 2020b) will not help address the matter.Some of the problems may have little to do with the Ministry of Commerce in Beijing, let alone the delegation in Geneva, since more powerful domestic ministries may not see the benefi ts to themselves in preparing the information (Gao, 2018).This applies a fortiori to countries that have weaker institutional capacity than China.
If the problem is a lack of capacity, then technical assistance may be needed.If the real diffi culty is outdated and overly complex notifi cation requirements, a thorough review is warranted.Only if the reason for a poor notifi cation record is bad faith would penalties as suggested by the US be appropriate.But China does recognize the notifi cation problem (WTO, 2019c) and is willing to engage in discussion of improvements, starting naturally with developed countries leading by example (Li and Tu, 2020).
The periodic monitoring reports prepared by the Secretariat ought to be able to provide information that supplements notifications.The reports aim to enhance transparency of trade policy developments, consistent with the mandate of the Trade Policy Review Mechanism to aid in understanding members' trade policies but not to assess compliance with formal obligations.The reports therefore do not cover "subsidies," which are defi ned for legal purposes in Article 1.1 of the ASCM, but they should in principle cover the full 14 This is available from: https://www.wto.org/english/tratop_e/covid19_e/covid19_e.htm.extent of the "general economic support" provided by governments.They do not.Central to coverage of general economic support, which goes beyond formal notifications, are responses to periodic questionnaires issued by the WTO Director-General (DG).The overall response is weak, and the response on general economic support is dismal.For the most recent report, 67 WTO members and one Observer volunteered information on 638 COVID-19-related general economic support measures.The EU did so; the US and China did not.The US is less cooperative than China or the EU with the trade monitoring exercise, neither responding to the DG's questionnaire nor verifying information the Secretariat found in other places (WTO, 2020c).
Figure 3 reports the results from the rank correlation analysis described in Section II above.It consists of a heatmap of the Spearman rho coefficients for each relevant pair of survey respondent groups computed on the cluster of questions pertaining to notification and monitoring.EU and US respondents are perfectly aligned with each other in identifying the most pressing issues when it comes to improving transparency, while Chinese respondents expressed different priorities.The results of the discrete choice analysis plotted in Figure 4 suggest that differences in the rank ordering of transparency-related issues observed in Figure 3 are associated with a stronger preferences of Chinese survey respondents for improving transparency than US and EU respondents.The EU respondents to the survey gain more utility on COVID-19 monitoring than those from China or the US (Figure 4).However, Chinese respondents gain much more utility from strengthening regular monitoring than the EU and the US.This is observed both for a question on the priority accorded to improving compliance with notifi cation obligations, and much greater utility associated with strengthening monitoring of trade policies.While the small sample size precludes inference, this result suggests transparency may be an area of reform where there may be scope for convergence.

VII. Improving the operation of WTO deliberative bodies
The WTO committees and councils are deliberative bodies for discussing emerging issues and addressing trade concerns without recourse to the dispute-settlement system.
Or at least they should be (Wolfe, 2020) modifi cation or even withdrawal of a measure that has adverse consequences for trading partners.Discussion of trade concerns is increasing in other bodies.Since 1995, close to 6,000 questions (much like an STC) have been raised in connection with individual notifi cations under the Committee on Agriculture (CoA) review process.Between mid-October 2014 and mid-October 2019, 1,158 issues and concerns were raised in 129 formal meetings of 17 WTO committees and councils, other than SPS, TBT, and CoA (WTO, 2020c).These numbers dwarf the number of formal disputes.
A handful of large traders make most frequent use of procedures to raise trade concerns, notably the US and the EU.China is now number 5 on the list of users and is the target of more trade concerns than any country after India (WTO, 2020c).We see this pattern in questions raised about the response to COVID-19 during the second half of 2020.In the Council for Trade in Services, China accused India of taking a series of discriminative and restrictive measures.In the CoA, Canada questioned the US on a possible US$14 billion Coronavirus Food Assistance Program payment while Australia, EU, India, New Zealand, and Paraguay questioned China's COVID-related border control measures.In the SPS committee, Canada, supported by Australia, Brazil, Mexico, Paraguay, the UK, and the US, raised concerns regarding China's suspension of imports from facilities where cases of COVID-19 are reported.Still, the procedures could be more extensively used, and participation could be enhanced.
One suggestion to this effect is to establish guidelines for all WTO bodies.Tabled by the EU and supported by 19 other members, including China, this proposal (WTO document WT/GC/W/777/Rev.5, which we refer to as "777" in what follows) aims to make better use of the possibility offered by WTO Council and committee meetings to discuss and resolve concerns with trade-related measures by equipping them with horizontal procedural guidelines (WTO, 2020f).At the July 2019 meeting of the General Council, the co-sponsors stressed that the point of improving the discussion of STCs is to avoid escalation to the dispute-settlement system.China shared that objective in its remarks to the Council, but China has resisted some aspects of 777.This is consistent with survey results presented in Section VII below, which reveal Chinese respondents gain considerably less utility than respondents in the US and the EU from using WTO bodies to defuse potential disputes by raising STCs.
The 777 proposal begins by clarifying timelines for convening documents and other meeting arrangements, which are important for making effi cient use of committee time.China shared that objective in its own reform paper (WTO, 2019c).The 777 proposal encourages submission of written questions and answers, which would enhance transparency for other members, or fi rms, having the same concern.Although the US was cool to the 777 proposal for obscure reasons, it made a similar proposal in the SCM Committee for ensuring timely written responses to questions posed by members on the subsidy programs of other members (WTO, 2020g).China has resisted every time the item comes up, arguing that the ASCM does not require members to submit responses to such questions in writing, nor to provide them within a specific time-period, and that setting deadlines as proposed by the US would impose substantial new notifi cation obligations on WTO members and cause difficulties for developing countries in particular.

Enhancing opportunities for dialogue
Policy dialogue in WTO bodies is important to consider what works well under agreements, what is not working, and what should be next on the agenda.Committees also need to hear from stakeholders who use their agreements, including regulators, other international organizations, and the private sector.
The heatmap in Figure 5 shows little correlation on relative priorities accorded to enhancing engagement with capitals and stakeholders by respondents to our survey.
The heatmap pertains to a cluster of deliberation-related reform options presented in the survey: (i) deepening engagement with stakeholders, specifi cally through so-called thematic sessions of WTO bodies; (ii) allowing binding decisions to be made in virtual meetings; (iii) making virtual meetings and video conferencing standard options; and (iv) delivering WTO technical assistance and training online and through video conferencing.The latter three options are all mechanisms to facilitate engagement with and input by officials and stakeholders in national capitals; the first is a mechanism for delegations in Geneva to interact with colleagues based in capitals, including other ministries and sectoral regulators, the private sector, and other international organizations.
Survey responses reveal that Chinese and US respondents gain more utility than EU respondents from greater engagement with stakeholders in WTO bodies (Figure 6).One instrument for such engagement is meetings that are sponsored by or associated with a WTO body in some way, but that are not part of its formal meetings and thus permit (in principle) participation by stakeholders.The WTO held over 100 such "thematic sessions" from 2017 through 2019.Wolfe (2021) reviews the use of thematic sessions by WTO bodies, fi nding that many did not hold any such meetings, and revealing substantial variation in how thematic meetings with stakeholders are organized, how themes are chosen, who speaks, the degree of transparency, and funding.Over 40 percent of speakers in the events analyzed are either Geneva-based delegates or WTO officials, or come from other international organizations.Barely a third of capital-based speakers came from non-G20 members, and only half the  sessions had speakers from non-G20 countries.China was relatively well represented, with nine Geneva-based Chinese government officials and 15 officials coming from national capitals speaking in thematic sessions during the 2017-2019 period, along with seven business and two academic participants.
The use of thematic sessions could be improved if every committee organizes discussion of the systemic issues posed by the operation of regional trade agreements in their respective policy areas, mitigating the absence of horizontal discussion of systemic issues in the Committee on Regional Trade Agreements.For example, the recently

Taking the WTO online
There are lessons for the reform of WTO working practices in how members managed to carry on talking through the pandemic. 15There may be some time before regular meetings can resume, but when they do members should institutionalize some pandemic-related innovations.Beyond practical teething difficulties, adapting the WTO's work to a virtual world posed some special challenges.Ensuring that all of the WTO's diverse members can participate while maintaining an agreed balance of rights and obligations within a reciprocal framework is complicated.Activities centered on learning, deliberation, and transparency have proven more straightforward than negotiating and agreeing to binding commitments.
While it is reasonable to assume that greater inclusion through involvement from capitals would help to reduce differences in the capacity of WTO members to participate fully in the organization's regular business, an important caveat is in order.There is a risk of an aggravated marginalization of some developing countries on account of inadequate connectivity and/or the need for more training for operating in a more virtual environment.Support for a hybrid meeting model is likely to increase if these challenges are addressed.
Reliance on hybrid meeting arrangements involving capitals will not necessarily sit well with Geneva ambassadors, who may fear an erosion of their influence and functions.This concern is also refl ected in our survey of the trade community, which reveals support for an intensified use of videoconferencing in the daily operations of the WTO, but a signifi cant contrast between Geneva-based respondents and others in respect of taking binding decisions in a virtual meeting: Geneva-based respondents were less supportive than other government offi cials (Low and Wolfe, 2020).Figure 6 shows that US and EU respondents are aligned in the priority (utility) attached to making more use of virtual meetings and allowing binding decisions in them, whereas respondents from China are considerably less keen.

VIII. Dispute settlement and the Appellate Body crisis
A central dimension of the "value proposition" offered by the WTO is independent, third-party adjudication of trade disputes reflected in the principle of de-politicized confl ict resolution.An effective dispute-settlement mechanism is critical for existing WTO agreements to remain meaningful, and for the negotiation of new agreements.
The different pillars of the WTO are interdependent.Resolving the Appellate Body crisis and bolstering the dispute-settlement function is critical for the continued relevance of the WTO.
The Trump Administration brought fewer cases against China than would appear warranted by the work the US did to document objectionable Chinese policies -e.g. the 2018 Section 301 United States Trade Representative report.This may in part refl ect a perception that matters objected to are not amenable to WTO disputes.The US has been critical of the system, alleging that the Appellate Body has too frequently overstepped its mandate.The Appellate Body ceased operations in December 2019 because of US refusal to agree to appoint new Appellate Body members and/or re-appoint incumbents.Resolution of the crisis requires reform of how the system works.US concerns are long-standing, and the US is not alone in at least some of its concerns (Fiorini et al., 2020).By the end of 2020, sixteen appeals were pending before the dysfunctional Appellate Body and only fi ve new cases had been fi led, the lowest for any of the WTO's 25 years.If appeal "into the void" remains possible, issued panel reports will have no legal value, unless the disputing parties forego their right to appeal, and accept the panel report as the fi nal word in their dispute.The Multi-Party Interim Appeal Arbitration Arrangement (MPIA), which includes the EU and China, provides a short-term alternative but is not a solution (Hoekman and Mavroidis, 2020).
The heatmap in Figure 7 shows a striking degree of correlation on assigning priority to this cluster of issues across the respondents to our survey.Respondents from all three trade powers are of the view that re-establishing an operational disputeresolution system is a top priority.Figure 8 shows that Chinese respondents to the survey gain more utility than the EU and US respondents from making the Appellate Body operational again and from considering reforms to dispute-settlement processes more broadly.We speculate that China values a functioning system that provides some protection from the US, or at least some recourse if the US does act unilaterally.Chinese scholars see the dispute-settlement system as the fi rst option to reconcile the relationship between China and its trading partners, hence wishing to be seen as a responsible player.China is motivated to comply with dispute-settlement rulings (Li and Tu, 2018).China is not a player in the Appellate Body debate, although it is much closer to the EU position than to the US.This is refl ected in China's decision to join the MPIA to assure the possibility to continue to appeal panel reports in disputes among signatories.Having invested considerable effort in developing trade law expertise in government and academia, China became a sophisticated user of dispute settlement to push back on US and EU use of trade remedy law (Shaffer and Gao, 2018).Although China lost many of the cases brought against it, Appellate Body rulings on key matters such as what constitutes a public body under the ASCM fuelled US frustration (Ahn, 2021).The US would see no point in any kind of WTO reform that did not address this problem.Equally, the US would see no point in any new agreements aimed at Chinese practices if dispute settlement remains slow and ineffective.All three of the major trade powers have an incentive to make dispute settlement operational again.Of the three players, the survey suggests Chinese respondents place even greater priority on addressing the crisis in dispute settlement than EU and US respondents, suggesting not only that this is an area where convergence should be possible, but that it may be one that could be linked to an issue that is less of a priority for China than it is for the EU and/or the US.

IX. Implications for future cooperation
The WTO members face many problems that call for cooperation.Prominent items include ensuring a consistent response to global public health crises, resolution of conflicts regarding the use of industrial-cum-tax-subsidy policies, taxation of digital services, regulation of data privacy and cross-border data flows, and the appropriate role of trade policy in reducing the carbon intensity of economic activity.Revisiting the terms of engagement with China is a necessary condition for revitalizing the WTO as a forum to address these matters and to sustain an open world economy.Refl ecting on the priorities of our respondents suggests that compromise will be needed all around on the design of a negotiating agenda or set of issues to be considered.
Recall that Figure 2 shows that respondents from China are far apart from those in the US and EU on negotiating stronger rules on industrial subsidies, while the latter two groups are close to agreement with each other.But a revamped subsidy regime requires participation of the US, the EU, and China.This suggests that the trilateral approach that has been a focus of EU-Japan-US deliberations is a dead end unless it is expanded into a plurilateral discussion and negotiation that includes China.The same applies to issues such as clarifying the role of trade policy in tackling climate change and promoting the sustainable development goals.Figure 9 reveals alignment among Chinese and American respondents on the value of adding new issues to the agenda, but otherwise divergence between Chinese respondents and those from the EU and the US.Chinese respondents attach less importance to the question of how trade policy should be used to address climate change and put greater weight on the question of using trade to attain the sustainable development goals.
The Doha Round deadlock means plurilateral initiatives and agreements are likely to be the main path for negotiations to address the spillovers of domestic economic policies.Plurilateral agreements permit countries to refrain from engaging in new rulemaking efforts, allowing those who participate to overcome the perceived failures of consensus and the single undertaking, but without China they will not succeed for lack of critical mass.The views of the expert respondents from China, the EU, and the US to our survey reveal substantial differences in views on the relative importance of many areas of international trade policy and potential cooperation.This suggests that getting to yes will not be easy.But the survey results also provide hints regarding the potential for package deals and the issue linkages that will be needed to get to yes, as well as issues where domain-specifi c plurilateral agreements may be feasible.How the three could conclude a package of reforms given the potential for obstruction by some other WTO members should be a major focus of the policy officials and advisors tasked with the design of trade strategy in the three major trade powers.All three major trade powers have to accept that progress on some issues will require leaving some obstructionist members behind -i.e. to negotiate on a plurilateral basis.They should be willing to proceed on a critical mass basis on issues where waiting for the full membership is not necessary.

X. Concluding remarks
The approach of the Trump Administration -relying on unilateral trade measures (notably a huge increase in tariffs) that were not sanctioned by the WTO to change the best alternative to negotiated agreement (BATNA) for China and by insisting on bilateral negotiations -did little if anything to address core differences between the two countries (Economist, 2020).At the time of writing in early 2021, we cannot say if the Biden Administration will adopt a change of direction in addition to a change of tone.
The similarity of views of our EU and US respondents, who do not necessarily represent their governments, suggest reasons for optimism.
signifi cant extent on rules, such as on subsidies and trade remedies, that are supplied multilaterally.Most bilateral or regional agreements do not have strong institutional arrangements in practice, whatever is put on paper.This might not impede successful implementation of their market access provisions, but may reduce the effectiveness of new rules and can be fatal for regulatory cooperation.
Any outcome in international negotiations depends on how negotiators understand their interests and those of their partners.Our survey of the trade policy community provides some insight into the views of expert stakeholders who can be presumed to be part of the epistemic community of negotiators in the three jurisdictions.Their views are likely to infl uence the positions adopted by their governments.The survey responses reveal substantial alignment between EU and US respondents' views, and substantially less alignment with those of respondents in China.This suggests cross-issue linkages will be needed to bridge differences in views and preferences in some policy areas.
Rather than speculate on the content of such potential linkages in negotiations, which are, by definition, endogenous, and not a matter on which our analysis sheds light, our findings highlight the importance of the three trade powers using all possible deliberative opportunities to identify a set of issues that provide a basis for negotiating a Pareto-improving package deal.
An important result from the survey is that it reveals there are areas where respondents from China, the EU, and the US are close to each other in terms of the importance they accord to resolving differences.This is important, as it is a necessary condition for discussions on WTO reform to progress.Such areas include SDT, where all recognize that this is a matter that must be addressed.Two other areas are dispute settlement and transparency.Respondents from all three powers place great weight on resolving the dispute-settlement crisis, with China-based respondents indicating that this is a particularly urgent priority.The survey reveals a striking degree of correlation among respondents from all three major trade powers on assigning a high priority to the cluster of transparency-related issues.Transparency and related accountability mechanisms are likely to remain the most useful for the trading system, which is why reform in this area is so important.The US would see no point in any kind of WTO reform that did not address the enforcement problem.But all WTO members would benefi t from knowing more about each other's trade policies.The fact that Chinese respondents to our survey gain much more utility from strengthening regular monitoring than the EU and the US, and that they gain slightly more utility than the US and the EU from improving compliance with notification obligations, suggests collaborative solutions are possible.

Figure 1
Figure1suggests that US survey respondents are closely aligned with EU respondents on the priority they assign to JSIs.The estimates plotted in Figure1come from the discrete choice model analysis described in Section II to compare responses by European, American and Chinese respondents, respectively.For each of the three comparisons reported in Figure1(and the other graphs in subsequent sections) point estimates to the right (left) of the 0 line reveal higher (lower) utility for the group of respondents of interest with respect to the baseline group.The distance from the line suggests the degree of alignment (divergence) on the priority respondents assign to the issue addressed by a survey question.

Figure 1 .
Figure 1.Plurilateral negotiations as a response to the consensus problem

Figure 2 .
Figure 2. Dealing with traditional problems

Figure 3 .
Figure 3.The importance of improving WTO transparency

Figure 4 .
Figure 4. Improving transparency is central to WTO reform . The most effective WTO bodies in addressing trade concerns are the Technical Barriers to Trade (TBT) Committee and the Sanitary and Phytosanitary Measures (SPS) Committee.Members raise "specifi c trade concerns" (STCs) to seek clarification of already adopted measures, and discussion can lead to Legal statement -This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.© 2021 The Authors.China & World Economy published by John Wiley & Sons Australia, Ltd on behalf of Institute of World Economics and Politics, Chinese Academy of Social Sciences.

Figure 5 .
Figure 5.The importance of deepening engagement with capitals and stakeholders

Figure 6 .
Figure 6.Views on deepening engagement with national capitals and stakeholders

concluded
Regional Comprehensive Economic Partnership (RCEP) has relatively few WTO-plus or WTO-extra provisions, and often relies on the WTO framework and notifications.Other WTO members might want to use an informal thematic session to assess how much China has willingly accepted a WTO framework in RCEP, putting into perspective claims that China's system is incompatible with the WTO.A similar discussion of the Comprehensive and Progressive Agreement on Trans-Pacifi c Partnership (CPTPP), which includes provisions on SOEs and to which Vietnam is a signatory, could help clarify whether it might be feasible to consider such disciplines in the WTO.Such discussions are particularly important in light of the fact that regional trade agreements often have only limited provisions on the use of subsidies, refl ecting free rider concerns.Such matters call for deliberation and dialogue.Thematic sessions offer a means for WTO members to participate in them.

Figure 7 .
Figure 7.The importance of improving WTO confl ict management and decision-making

Figure 8 .
Figure 8. Dispute settlement and confl ict resolution

Figure 9 .
Figure 9. Dealing with novel problems Legal statement -This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.© 2021 The Authors.China & World Economy published by John Wiley & Sons Australia, Ltd on behalf of Institute of World Economics and Politics, Chinese Academy of Social Sciences.
Legal statement -This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.© 2021 The Authors.China & World Economy published by John Wiley & Sons Australia, Ltd on behalf of Institute of World Economics and Politics, Chinese Academy of Social Sciences.