2. Excessive Policy Space and the Lack of Discipline on Protectionism

The Framework allows excessive policy space and lacks sufficient discipline over protectionist measures. It distinguishes between beggar-thy-neighbor policies and other policies. As defined in the Statement, a beggar-thy-neighbor policy ‘seek[s] to increase domestic economic welfare at the expense of other countries’ welfare’. The Statement further clarifies that such policies include only those ‘adopted with express beggar-thy-neighbor purpose in mind’ (emphasis added). Purposeful beggar-thy-neighbor policies are prohibited under Bucket 1, whereas other policies are allowed under Buckets 2 and 3, which are at the core of the Framework according to the Statement. These ‘non-prohibited’ policies clearly include protectionist measures defined broadly as those adopted to protect domestic industries through measures conferring a competitive advantage to local industries or undermining the competitive condition of foreign competitors. Accordingly, in this paper, protectionist measures or protectionism are not merely concerned about border measures but include all trade and domestic policies and measures which have the effect of altering the condition of competition in favor of domestic actors vis-à-vis foreign competitors regardless of the presence of a protectionist intent. The examples envisaged under Buckets 2 and 3 – i.e. unfair and harmful industrial policies, domestic regulations in pursuit of certain policy objectives, and ‘trade barriers to protect communities adversely affected by exports’ – are broad enough to encompass all such protectionist measures. Instead of prohibiting such measures, the Framework essentially recommends that the parties negotiate whether they may be reduced or removed, and, if they cannot, how their economic impact on the counterparty may be rebalanced. Accordingly, the Framework provides wide latitude for the US and China to exclude protectionist measures from a trade deal.

The Framework is unlikely to facilitate the US–China trade talks, despite the wide regulatory freedom it provides. In failing to impose any hard rules on protectionism, the Framework fails to tackle the core problems that trade negotiations are concerned about in reality. For eight decades, international trade cooperation has been predominantly devoted to constraining protectionism.Footnote ⁷ WTO rules, the final product of multilateral trade negotiations, cover more than policies with express beggar-thy-neighbor purposes. For example, the Preamble of the Marrakesh Agreement Establishing the WTOFootnote ⁸ targets barriers to trade and discriminatory treatment that may or may not be motivated by beggar-thy-neighbor purposes. Examples that are more specific include some of the most fundamental rules of the WTO such as: (1) Article III of the GATT, the national treatment principle, which inhibits the use of discriminatory measures ‘so as to afford protection to domestic production’;Footnote ⁹ (2) Article XI of the GATT, which prohibits import or export quantitative restrictions irrespective of whether such restrictions are motivated by a beggar-thy-neighbor purpose; (3) Article II of the GATT, which allows Member governments to impose tariffs (as long as they do not exceed the committed/bound levels), even though such tariffs may have an express beggar-thy-neighbor purpose. Under the WTO's dispute settlement mechanism, the trade rules have been interpreted and applied in a way that seeks to discipline protectionist measures that adversely affect trade or the competitive condition of foreign companies vis-à-vis domestic industries of regulating members.Footnote ¹⁰ The above is not to suggest that the US–China negotiations must follow exactly the WTO model. Rather, the point is that the Framework does not address the core problems that exist in reality and the governments’ behavior behind these problems.Footnote ¹¹

Most of the issues that have provoked the US–China trade tensions arise from various trade barriers, which may or may not be motivated by a beggar-thy-neighbor purpose. Indeed, the ongoing bilateral trade war has been dominated by tit-for-tat tariffs which are widely regarded as beggar-thy-neighbor policies.Footnote ¹² However, the tariffs are just one of the means that the Trump Administration employed to push China to yield to US interest on certain systemic issues. These issues, identified in the joint statement of the US, the EU, and Japan (amongst other public announcements and official documents), mainly include China's non-market economic system, SOEs, industrial policies and subsidies, and so-called ‘forced’ technology transfer.Footnote ¹³ These issues are also central to US complaints about China's compliance with WTO obligations.Footnote ¹⁴ Although the exact demand of the US on these issues remains ambiguous, US concerns are loud and clear. To put it plainly, the US is unhappy with the various barriers and market-distorting policies and practices that China maintains to afford protection or competitive advantage to its domestic industries in pursuit of broader goals for economic development.Footnote ¹⁵ Accordingly, the US's central requests, in broad terms, are that China removes state intervention in commercial activities, as well as trade-restrictive and distortive policies and practices, to enhance its level of liberalization and maintain a level playing field for US companies. It is evident that these concerns and requests are not confined to policies with an express beggar-thy-neighbor intent but are so extensive as to capture all policies and practices having a protective, trade-restrictive, and distortive effect that jeopardizes the competitive position of US companies vis-à-vis their Chinese counterparts. While the US's demand that China ‘fundamentally change[s] its approach to the economy and trade’Footnote ¹⁶ would be too ambitious and intrusive to be fulfilled,Footnote ¹⁷ a framework which essentially leaves all protectionist and trade distortive policies without an express beggar-thy-neighbor purpose untouched will not be acceptable to either party.

Furthermore, the Framework lacks a clear and workable criterion. While ‘beggar-thy-neighbor policies’ are generally defined as those raising the welfare of the home country at the cost of worsening economic conditions abroad,Footnote ¹⁸ there seems to be no agreement on what may constitute such policies. Arguably, a protectionist measure is not necessarily a beggar-thy-neighbor policy. A tariff, for example, causes deadweight losses to the home country and hence does not produce the welfare gains that a beggar-thy-neighbor policy would produce. An ‘optimal tariff’ imposed by a big country having the capacity to influence terms-of-trade would be welfare enhancing at the cost of other nations.Footnote ¹⁹ Such a beggar-thy-neighbor effect, however, would need to be established by robust empirical evidence.Footnote ²⁰ Even though a policy is shown to have a beggar-thy-neighbor effect, it will not automatically fall within the prohibited category of the Framework unless it is clearly motivated to have that effect. Such motivation is extremely difficult, if not impossible, to establish. A protectionist measure, in whatever form, may be motivated by a mix of protectionist and non-protectionist policy objectives without an intent to enhance national welfare at others’ costs. In the context of the US–China negotiations, it is unlikely that either party would voluntarily admit or accept a claim (by the other party) that its existing policies were aimed at harming the other's economy. Towards this end, the only measure that both parties would be willing to surrender is likely to be the beggar-thy-neighbor tariffs if the negotiations were to follow the Framework. However, to bring long-term peace to this trade war, the systemic issues (which are mostly non-tariff measures) must be resolved in a balanced way by disciplining the relevant protectionist policy instruments without restricting the parties’ freedom in their choice of non-protectionist regulatory goals.

In this regard, the Framework fails to distinguish between protectionist and non-protectionist policy objectives as well as between policy objectives and policy instruments used to achieve these objectives. It provides too much policy space for governments to adopt measures that do not genuinely serve any non-protective regulatory purpose. As will be discussed in Section 4, this would lead to a failure to achieve the overarching goal envisaged in the Statement, that is, to maximize the gains from trade.Footnote ²¹ For international trade to be welfare enhancing, countries must have the freedom to pursue any non-protectionist policy goals. At the same time, their choice of policy instruments to achieve such goals needs to be regulated through a balanced approach. Practically speaking, the Framework fails to recognize that excessive policy space may not be conducive to a trade deal. What governments usually need in trade negotiations is merely sufficient policy space so that they could secure the political support needed to reach a deal that reins in protectionist measures.

We do not have a crystal ball to foresee the ultimate outcome of the US–China negotiations, or more specifically whether and to what extent the negotiations will address protectionist and trade distortive policies. Our core criticism is that the Statement fails to provide a clear and viable framework for the ongoing negotiations. It grants too much policy space to the two governments to adopt protectionist and trade distortive policies which are precisely the major issues that this bilateral negotiation, and indeed trade negotiations in general, must address.

3. Failure to Respect WTO Rules

In addition to the problem discussed in the last Section, the Statement also ignores the fact that WTO law has already provided sufficient policy space for Member governments. In essence, WTO rules impose different degrees of restraints on a range of policy instruments arguably according to the level of adverse effects of these instruments on trade.Footnote ²² Such restraints include, for instance, the prohibition of import and export restrictions (the most trade-restrictive and distortive measures), and to a lesser degree the imposition of upper limits on (rather than prohibition of) tariffs, and the mere non-discrimination requirement on domestic measures leaving the flexibility for governments to use such measures to address regulatory concerns. In particular, for example, governments may maintain domestic subsidies although they need to remove the adverse effects on the interests of other Members or face countervailing measures. Non-specific subsidies are generally permitted regardless of their impact on trade. Furthermore, through instruments such as the Agreements on Technical Barriers to Trade and Sanitary and Phyto-Sanitary Measures, WTO Members are allowed considerable leeway to choose their own policy objectives and the level of protection and may adopt measures, which otherwise would be WTO-inconsistent, to meet these objectives. To achieve a proper balance between trade liberalization and regulatory autonomy, the Appellate Body has taken efforts to address the protectionist elements of a contested measure without impairing Members’ capacity in pursuing chosen non-protectionist objectives.Footnote ²³ The Statement does not explain why existing WTO rules are insufficient to provide the policy space needed by governments and hence why excessive policy space is required to facilitate the US–China negotiations.

More importantly, the rebalancing approach embedded in Buckets 2 and 3 of the Framework is likely to lead to inconsistencies with WTO rules. Generally speaking, the wide coverage of Buckets 2 and 3 may well leave protectionist instruments unregulated or under-regulated, falling short of WTO requirements. For example, under Bucket 2, the parties may agree that China shall reduce and confine industrial subsidies to high-tech firms to certain levels but may maintain a limited list of discriminatory policies within an agreed timeframe to provide space for the production and sale of domestically made high-tech products (e.g. electric cars). In return, the US may agree to remove or reduce a selected range of antidumping and countervailing duties on Chinese imports. It is evident that these discriminatory measures would breach the national treatment rule. To use this example the other way around, the parties may agree that China maintains its current level of industrial subsidies to high-tech firms for a period of time during which the US may increase tariffs on a selected range of Chinese imports. Such a tariff increase would not be allowed under GATT Article II if it goes beyond the corresponding bound level. If the tariff increase is based on an existing antidumping and/or countervailing duty, then it may not be compatible with the substantive and procedural requirements for the imposition of such duties under the relevant WTO agreements. This would be an example of a Bucket 3 policy. Notably, such deals are not necessarily ‘mutually beneficial’, as characterized in the Statement, for the discriminatory measures and increased tariffs would be detrimental to the welfare of both countries.

Finally, the Statement does not consider what new rules are needed to address the so-called China-specific issues. Many scholars believe that existing WTO rules are insufficient in handling the challenge of China, which as a non-market economy does not fit with the multilateral trading system built on market economy principles.Footnote ²⁴ As we have argued in detail elsewhere, however, existing WTO rules, including China-specific ones, may have provided sufficient tools to deal with the trade-distorting conduct of Chinese SOEsFootnote ²⁵ or the alleged ‘forced’ technology transfer.Footnote ²⁶ The issue, therefore, may not be the lack of rules but the lack of utilization of the existing rules. As demonstrated by China's experience during and after its accession to the WTO, China has been willing to accept restraints on its national policies according to multilateral rules and to comply with unfavorable rulings of WTO tribunals.Footnote ²⁷ Thus, it is much preferable to use WTO law and litigation to discipline China's trade policies. But this approach has been abandoned by the current US Administration, which chose to resort to unilateralism rather than multilateralism to resolve trade tensions. One argument for the need of new rules would be that the existing rules may be further elaborated to enhance their efficacy. Examples of this include Section 9.1 of the Protocol on the Accession of ChinaFootnote ²⁸ and Paragraph 46 of the Report of the Working Party on the Accession of China.Footnote ²⁹ The former requires China to ensure all prices for goods and services in all sectors are determined by market forces; and the latter prevents the Chinese government from influencing commercial decisions of SOEs and state-invested enterprises. While these obligations provide abundant flexibility for WTO Members to challenge a wide range of protectionist and trade distortive policies and practices of the Chinese government including via state entities, they could still be improved through the elaboration of more detailed legal criteria for Members and WTO tribunals to apply, such as specific factors for determining whether Chinese prices are market prices or whether an SOE made a particular decision purely on a commercial basis.Footnote ³⁰ Furthermore, new rules may be needed to address new situations or practices such as variations and developments in China's industrial policies and subsidization and the lack of information on these measures. We will provide some illustrations on these issues in Section 5.

In short, the Statement fails to recognize that the existing WTO rules, as interpreted and applied by WTO tribunals, may have already provided the policy space needed by governments, both to pursue chosen non-protectionist objectives as well as to counter the effects of foreign measures. It fails to account for the efficacy of the existing rules in dealing with the China challenge. Instead, by encouraging the parties to negotiate bilateral deals in total disregard of existing WTO rules, the Statement only works to further undermine the multilateral trading system.

4. An Alternative Framework

An alternative framework is needed to guide the parties to negotiate specific disciplines on all sorts of protectionist measures and the policy space needed especially in the areas where the systemic issues arise. Our framework is not intended to impose undue restrictions on the regulatory autonomy of the parties. Instead, we seek to strike a proper balance between disciplining protectionism and preserving policy space in line with the approach taken by WTO rules and jurisprudence. This balance could be achieved by drawing on the theoretical guidance emanating from the Theory of Distortions and Welfare (Theory).Footnote ³¹ Ultimately, we believe the bilateral negotiation should lay the groundwork for broader negotiations under the WTO (rather than create a significant deviation from it).

The Theory addresses one of the longest-standing criticisms of free trade, that is, trade liberalization may not enhance national welfare when a domestic externality or non-economic objective exists.Footnote ³² It establishes two important propositions: (1) trade liberalization would remain welfare enhancing if the externality or objective is tackled separately with an optimal or first-best policy instrument; and (2) the most economically efficient policy instrument is one which strikes most directly at the source of the externality or objective. Therefore, in the pursuit of a domestic non-protectionist policy objective, the first-best policy would generally be a domestic instrument (e.g. internal tax or subsidy) which strikes directly at the locus of the problem, as opposed to a trade measure (e.g. an import tariff or quota). In other words, the more directly an instrument addresses the root of the problem, the more economically efficient it tends to be. When there is more than one externality/objective, the first-best policy would be the use of two or more domestic policy instruments tackling specifically the source of each externality/objective. A major qualification of the Theory concerns the side effects associated with the application of a first-best policy instrument. However, whether the existence of such side effects would render the optimal instrument economically inferior to other instruments requires careful assessment.Footnote ³³

The Theory provides a framework for trade negotiations with two general steps:

• • first, trade negotiations must identify the externalities/objectives that governments need to address and allow the flexibility for governments to choose their own non-protectionist regulatory goals; and

• • second, trade negotiations should focus on governments’ choice of policy instruments in pursuit of their chosen goals taking into account the economic efficiency of the chosen instruments and their possible alternatives (as per the Theory), their effectiveness in fulfilling the goals, and the side effects of using a more optimal policy instrument.

Based on this framework, trade negotiations will continue to further trade liberalization in a way that allows nations to gain from trade and enhance economic welfare. The reason why trade negotiation/cooperation is needed to achieve an economically preferable outcome is that governments may well succumb to the pressure of powerful interest groups (i.e. import-competing producers) to maximize short-term political gains (at the expense of the long-term national economic interest). As a result, the political choice of trade policies and policy instruments would not necessarily follow the economic rationale; and the more economically efficient measures might just be less attractive politically.Footnote ³⁴ This suggests that in choosing policy instruments to pursue a given regulatory goal, the political preference would likely be compromised by protectionist interests. Trade negotiations provide an effective way for governments to resist protectionist pressure by incorporating the interests of anti-protectionist groups (e.g. exporters) into domestic political calculus.Footnote ³⁵

Compared with the Statement, our framework offers a better mechanism to facilitate the US–China trade negotiation. First and foremost, we draw a distinction between protectionist and non-protectionist objectives and between policy objectives and policy instruments. We specifically target protectionism, the key problem that trade negotiators seek to address in reality. As discussed in Section 2, we define ‘protectionism’ in this paper to have a broad meaning encompassing all policies and policy instruments that tilt the conditions of competition in favor of domestic actors vis-à-vis their foreign competitors. Our focus is on the protective and trade distortive effect rather than the intent of such measures. Meanwhile, our framework recognizes the regulatory freedom of both countries in their choice of policy objectives and provides a platform for them to negotiate all sorts of policy instruments they may adopt to pursue the desired regulatory goals. It encourages the two governments to use the most economically efficient instruments while allowing the flexibility for them to justify the use of second-best means. In doing so, our framework would be more balanced by allowing sufficient but not excessive policy space and would lead the negotiation in a direction whereby both countries would gain from further trade liberalization and less disguised protectionism.

Furthermore, our framework provides more substantive guidance for the bilateral negotiation. Instead of leaving the parties with unfettered latitude whereby they may agree to the use of a wide spectrum of protectionist instruments (subject to rebalancing), our framework offers economic guidance for them to determine how to restrain protectionism. Such guidance may be applied, for example, to all industrial policies that China employs to foster its high-tech industries. As demonstrated above, the steps of negotiation would involve (1) the identification of self-selected non-protectionist policy goals, and then (2) an assessment of whether the policy instrument(s) currently employed to pursue a given goal is the first-best means. More specifically, this latter step would involve sophisticated negotiations on why a sub-optimal measure has been applied, whether it could be replaced by an alternative means, etc. taking into account the impacts on trade and the capacity of the government concerned in accomplishing its chosen objectives.

Finally, our framework would avoid WTO-illegal outcomes. As observed in Section 3, the design of the WTO rules is consistent with the economic rationale of the Theory by regulating policy instruments according to their respective welfare costs. In cases involving trade and non-trade values, the judicial approaches and decisions of WTO tribunals, particularly those of the Appellate Body, have also been largely consistent with the economic rationale.Footnote ³⁶ Our framework is intended to achieve the same balance that the WTO rules are designed to achieve and therefore would be inherently WTO-consistent. To minimize any incidental WTO-inconsistencies in practice, the negotiations must also involve consideration of whether the policy instruments that the parties agree to use for chosen non-protectionist regulatory goals may breach WTO rules.

5. A Way Out for the Trade War

To illustrate the practical application of our framework, we use the recently concluded US–China Phase One trade deal as an example. The deal covers many areas, which can be largely grouped under two categories: market access and rules issues. The market access commitments are supposed to solve the trade imbalance problem, which is what prompted President Trump to launch the trade war in the first place. Technically speaking, however, all the additional tariffs imposed by the US over the past two years were triggered by the rules issues. This is because their justifications were provided by the Section 301 Report into China's Acts, Policies, and Practices Related to Technology Transfer, Intellectual Property, and Innovation, which resulted from the investigation launched in August 2017 to address Chinese measures requiring ‘the transfer of American technology and intellectual property to enterprises in China’.Footnote ³⁷

According to the Phase One trade deal, China has agreed to end its practice of ‘forced’ technology transfer. In essence, this issue concerns Chinese laws and practice which mandate, explicitly or implicitly, the transfer of technology by foreign companies to their Chinese partners in the course of foreign direct investment.Footnote ³⁸ However, China's commitment in the deal does not necessarily mean that it has to abandon its overarching goal for technological development. To the contrary, China still retains considerable freedom to adopt policy instruments to pursue that goal so long as it does not have protectionist effects. To do this properly, we need to first identify the externalities retarding technological development. If the externality lies in the failure of the capital market to provide sufficient finance for research and development (R&D), then China could correct such market failures through measures such as higher tax incentives for R&D expenses deduction, more government investment in fundamental research, etc. If the externality lies in the lack of access to information and knowledge, then the Chinese government may take steps to enhance flow of information and knowledge without undermining the legitimate rights of other stakeholders. Both measures would address the chosen objective(s) more directly than ‘forced’ technology transfer and tend to be more efficient and less-trade-restrictive policy instruments. Indeed, China's existing WTO commitments have included a general obligation to avoid ‘forced’ technology transferFootnote ³⁹ and recently China has revised its Foreign Investment LawFootnote ⁴⁰ (FIL) to better implement that obligation. However, as flagged in Section 3, future negotiations could add more details to the existing commitment by imposing specific obligations on, for example, transparency to address the longstanding concern about implementation and enforcement, and a restriction on the abuse of national security as a justification for ‘forcing’ technology transfer. The Phase One deal has included a provision to ensure due process and transparency for US companies involved in administrative proceedings related to matters of technology transfer in China. However, in view of China's past experience with its WTO transparency obligations,Footnote ⁴¹ a more comprehensive and detailed transparency provision would probably be needed to tackle other potential issues in China's implementation of the FIL.Footnote ⁴² The final outcome of the negotiation would be a compromise which strikes a balance between China's pursuit of technological development and the protection of the legitimate interests of US firms. Where the Chinese subsidies have affected the competitive position of US firms, rebalancing could be achieved through the application of countervailing measures in accordance with relevant WTO rules.

Another example is the alleged lack of access to the Chinese market by American farmers. This is why the deal includes substantive commitments by China to greatly expand purchase of US manufactured goods, agricultural goods, energy products, and services for a total of at least $200 billion over the 2017 baseline amount in 2020 and 2021. It remains to be seen whether China would fulfill this commitment and if so, how.Footnote ⁴³ It also remains unclear how China may implement this commitment without violating WTO rules such as the prohibition on quota under GATT Article XI, the MFN principle under Article I, and the non-discriminatory administration of quota under Article XIII.Footnote ⁴⁴ What is clear is that this commitment is non-reciprocal and one-sided against China, and hence does not provide a balanced and long-term approach to the resolution of the bilateral trade tensions. Under our framework, instead of setting explicit import quotas, the two sides should first try to identify the problems under the current system that affect market access. For example, whether the lack of market access was due to high tariffs, restrictive non-tariff barriers such as arbitrary sanitary and phytosanitary measures without sufficient scientific justifications, or simply the result of pure market forces of supply and demand. Once the source of the problem is properly identified, the two sides can then try to negotiate a proper solution. For example, if high tariffs are the reason, then the US could request China to reduce its tariffs. Of course, this would need to be done on an MFN basis. At the same time, the US should also recognize legitimate Chinese interests. For example, China's high tariffs could be due to the concern of the government over the livelihood of the farmers. In this case, the US could help China to design an appropriate farmers-support program, such as direct cash payment by government to farmers that is not tied with agricultural outputs, which would be less trade restrictive and more efficient than collecting tariffs from American farmers and then transferring the proceeds to the Chinese farmers. If China prefers to use trade measures, it could negotiate for the use of tariff-rate quotas or even special safeguard measures reserved for sensitive products, which would be more flexible and have minimal impacts on trade.Footnote ⁴⁵

A final example would be China's industrial policies which involve the use of a mix of policy instruments such as the provision of subsidies to and preferential treatment of strategic sectors.Footnote ⁴⁶ These policies have not been covered in the Phase One deal and will be subject to further negotiations. As flagged in Section 3, new rules may be needed to address variations and developments in these policies. This may involve, inter alia, a more nuanced definition of SOEs, an expanded scope of subsidies, and enhanced transparency and access to information to reduce the difficulties for complaining countries to discharge the burden of proof on the above and other related issues.Footnote ⁴⁷ Under our framework, the bilateral talk should start by identifying the exact policy objectives that China seeks to pursue under the overall industrial development plan and then focus on those existing instruments of concern to the US by exploring alternative means to attain these objectives. In this article, we do not intend to offer a detailed analysis of all these systemic issues behind the US–China trade tensions and all the policy instruments that may need to be negotiated. Such an analysis will need to be deferred to another article. Here, our objective is to set out the general framework and to demonstrate how it may work in reality.

In the longer term, we believe that the US–China negotiations should aim to lay the groundwork for more inclusive trade negotiations which would eventually contribute to re-invigorating the multilateral negotiations under the WTO. This is achievable if one considers the history of China's negotiations for entry into the WTO, which shows that a US–China bilateral deal is key to accomplishing a multilateral deal.Footnote ⁴⁸ The fact that both the bilateral and the multilateral negotiations now face new challenges does not mean that they are insurmountable. Our framework provides a starting point for the bilateral negotiation which should, in our view, contribute to, and form an integral part of, further negotiations on a multilateral basis.