Can the World Trade Organization Act as a Bulwark Against Deglobalization?

Simon J. EVENETT†

University of St. Gallen and CEPR

Whether existing multilateral trade commitments really deter larger trading nations from taking steps that further weaken cross-border commercial ties is assessed here. Evidence from salient commercial policy episodes of recent years is combined with information on the actual leeway available to G-20 members under extant World Trade Organization rules. The upshot is a bleak assessment of the capacity of the existing multilateral trade rule book to rein in any attempt by larger trading nations to “deglobalize” the world economy.

Key words: multilateralism, protectionism, trade policy, unilateralism, WTO

JEL codes: F13, F52, F53

Accepted: 7 September 2023

1. Introduction

After the Global Financial Crisis analysts debated whether globalization – the evergreater integration of national markets – was slowing down. Recently, the debate has shifted to whether the world economy has entered a phase of deglobalization. In light of growing geopolitical tensions and the so-called Polycrisis, some have gone so far as to argue that deglobalization – taken here to mean the conscious thinning of cross-border commercial ties through state action – is necessary.

My goal here is not to assess whether the world economy is or has been deglobalizing (Baldwin, 2022a,b,c,d; Evenett, 2023; Goldberg & Reed, 2023). Nor is my purpose to assess whether deglobalization is desirable or what its end point might be (Evenett, 2023). Nor will I consider the factors that have triggered discussions of deglobalization (Leonard, 2021). Here, I ask a different question: If a government contemplated trade policy measures that markedly thinned cross-border commercial ties, would the current corpus of multilateral trade rules prevent them from doing so? In short, I will examine what contemporary evidence reveals about the degree to which

群内每日免费分享5份+最新资料

300T网盘资源+40万份行业报告

为您的创业、职场、商业、投资、亲子、网赚、艺术、健身、心理、个人成长 …… 全面赋能！

关注公众号获取更多资料

添加微信，备
注“入群”
立刻免费领取
200套知识地图+最新研报

致终身学习者社群

行业报告/思维导图/电子书/资讯情报

收钱文案、增长黑客、产品运营、品牌企划、营销战略、办公软件、会计财务、广告设计、摄影修图、视频剪辑、直播带货、电商运营、投资理财、汽车房产、餐饮烹饪、职场经验、演讲口才、风水命理、心理思维、恋爱情趣、美妆护肤、健身瘦身、格斗搏击、漫画手绘、声乐训练、自媒体打造、效率软件工具、游戏影音……

扫码先加好友，以备不时之需

World Trade Organization (WTO) rules can really act as a bulwark against deglobalization.

The approach taken in this paper is not to assume or contend that existing multilateral trade rules have no bite. Demonstrating such a strong proposition is unnecessary given my research objective. Rather, I will marshal enough evidence on actual contemporary commercial policy decisions and on the leeway afforded to governments in current WTO accords that might give a reader inclined to believe that existing multilateral trade rules will act as an effective bulwark against deglobalization second thoughts. If this argument is correct then, as geopolitical rivalries intensify, we should moderate our expectations as to what WTO accords can deliver and look elsewhere for centripetal forces that might hold the world trading system together.

The evidence presented in this paper can be found in the following three sections. Section 2 recounts the scale and duration of the “trade war” between China and the USA as well as two other salient trade policy episodes. Section 3 looks back further to commercial policy developments since the Global Financial Crisis. Section 4 examines the degree to which the largest economies of the world – the G-20 members – can raise their import tariffs without violating their multilateral trade obligations. The discussion here speaks to how much restraint the provisions of the multilateral trading system actually provide when the government contemplates setting import tariffs, the quintessential discriminatory trade policy instrument. A final section reflects on the evidence presented here and draws out implications for policymaking and for those with a stake in an open world trading system.

2. Evidence from High-Profile Recent Commercial Policy Episodes

In this section, evidence from three salient trade policy episodes is presented. The first episode is the so-called trade war between the USA and China.

Perhaps the most basic multilateral trade obligations relate to the imposition of taxes on imported goods, the so-called import tariffs. For each tariff line, every WTO member has committed whether or not to have a maximum allowed (or bound) tariff rate and, if so, what that maximum rate is. Furthermore, ever since GATT was formed in 1947, the Most Favored Nation (MFN) principle applies. Indeed, that principle was so important it is listed in the first article of the agreed texts, known by their shorthand GATT 1947. That article states,

‘any advantage, favour, privilege or immunity granted by any contracting party to any product originating in or destined for any other country shall be accorded immediately and unconditionally to the like product originating in or destined for the territories of all other contracting parties.’

The USA was a founding member of the GATT and became a contracting party to the WTO. China acceded to the WTO in 2001. Both parties are therefore obligated to treat imports from the other no worse than from any other party. Moreover, whatever tariffs are charged on imports must not exceed their bound or maximum levels.

As has been meticulously documented by Bown and Kolb (2023), the USA unilaterally decided to investigate China’s “laws, policies, practices, or actions that may be unreasonable or discriminatory and that may be harming American intellectual property rights, innovation, or technology development” in August 2017. The USA did not bring a case to the WTO DSU (Dispute Settlement Understanding); instead, the Trump Administration chose to investigate China’s practices itself.

This American investigation resulted in import tariffs of 25% being imposed on 1333 products sourced from China on April 3, 2018. The next day China threatened to retaliate with 25% import tariffs on 106 products. On April 5, 2018 President Trump instructed his officials to consider imposing higher tariffs on \$100 billion more of Chinese imports. This cycle of action and reaction was repeated at different points throughout 2018. Further tariff increases were imposed by both sides during 2019. Ultimately, a “phase one deal” was signed in January 2020. That deal did not include unwinding the import tariffs imposed but it did stave off the imposition of even more tariffs. As part of this deal, China is said to have agreed to quantitative targets that increase their imports of American goods and services, where the stated goal was to reduce the size of the bilateral trade surplus China enjoys with respect to the USA.

Figure 1 reports the evolution of the share of Chinese goods exports to the USA that are covered by import tariff increases from 2009 through to 2023, that is, from before the so-called US-China trade war broke out and since the “phase one deal” in 2020. The comparable share of US goods exports to China facing higher import tariffs is shown as well. The annual estimates presented are based only on import tariff increases in effect that year (that were implemented since the start of 2009).

Only import tariff increases that were implemented count toward these totals; threats to raise import tariffs do not. The annual estimates are duration-adjusted so tariff increases imposed later in the year are given less weight than those imposed earlier in the year and that remain in force. It is, therefore, possible to check whether the share of each country’s goods exports to the other that face higher import tariffs increased significantly once the trade war began and whether those tariffs were unwound; that is, whether this outbreak of import tariff increases was transitory.

Figure 1 Sustained revocation of Most Favored Nation tariff treatment. ( ) All measures by China; ( ) all measures by USA; ( ) all tariffs by China; ( ) all tariffs by USA.

Figure 1 shows that the shares of bilateral US-Chinese goods exports facing higher import tariffs increased sharply in 2018 to one third.1 The shares rise further, plateauing at around 0.6 in 2020. Since then the import shares facing tariff hikes have not reduced – in short, the import tariffs on both sides have not been removed. What is remarkable is the almost identical movement in these import shares once the trade war began, which is consistent with classic tit-for-tat behavior.

It is important to bear in mind that the tariff increases described above were not applied across-the-board to all of China’s or the USAs trading partners. Moreover, the size of the import tariffs imposed exceeded the bound rates for many of the goods imported by both parties. I conclude from this episode:

• Membership of the WTO did not stop the USA from revoking MFN tariff treatment for China.

• Membership of the WTO did not stop the USA from imposing tariffs on goods that exceeded relevant bound tariff rates.

• Membership of the WTO did not stop China from retaliating by imposing tariffs on the USA’s goods exports, doing so in a way that essentially revokes MFN tariff treatment for American goods.

• Membership of the WTO did not stop China from raising tariffs above bound rates either.

• That these import tariff changes have not been reversed, may be permanent, and are at minimum, non-transitory.

It should be noted, however, that these revocations of MFN tariff treatment did not spread to other WTO members. No other WTO member followed the lead of the USA and revoked the MFN tariff treatment of China; likewise, no other WTO member revoked MFN tariff treatment on goods originating in the USA. So, there was not a complete breakdown in norm compliance. Nevertheless, governments representing the two largest economies in the world violated a core principle of the WTO. These two governments are still members of the WTO, calling into question whether there is much price to be paid for breaking a cardinal principle of the multilateral trading system. If this can happen once, what prevents similar violations from occurring should governments decide to selectively decouple – or thin commercial ties – with certain trading partners?

The second episode discussed here relates to the apparent “weaponization” of medical goods and food exports in recent years. For all of its strictures, multilateral trade rules allow for exceptions. In the case of export controls on goods, although generally banned under Article XI of the GATT, exceptions are allowed. Article XI.2(a) states,

‘Export prohibitions or restrictions temporarily applied to prevent or relieve critical shortages of foodstuffs or other products essential to the exporting contracting party.’

My understanding is that there is no agreed list of what products are deemed essential. Moreover, the application of this exception is self-judging, in the sense that a government can determine when the exception is invoked.

The outbreak of the COVID-19 pandemic witnessed a significant increase in demand for a range of medical equipment and medical consumables (such as personal protective equipment.) The vulnerability of populations to COVID-19, potential treatment, duration, and other key factors were not known at the start of the pandemic. Many governments resorted to export controls on medical kit. Figure 2 shows the total number of such controls in force peaked in April 2020 and then fell back slightly. Of the 329 export restrictions documented by the Global Trade Alert (GTA) team in its Essential Goods Monitoring initiative, a total of 198 have stated removal dates. This implies that 131 export controls have no phase-out dates and, therefore, may not be temporary. In fact, for 31% of all export controls on medical equipment and consumables implemented in 2020 there is no evidence to suggest they have been removed.

Concerns about security of supply of food are long-standing in the world trading system. When COVID-19 broke out concerns were flagged then; those concerns were to intensify once the Russian invasion of Ukraine threatened the delivery of wheat, sunflower oil, and other foodstuffs in 2022. Some governments invoked Article XI exceptions and restricted the export of foodstuffs. As in the case of medical goods, were these exceptions temporary? The GTA team documented 427 export bans, quotas, taxes, and other restrictions on food products since the start of 2020. Of those 427 measures, only 276 had known or recorded termination dates. Matters have not improved since the invasion of Ukraine. A total of 221 export controls on food have been recorded and only two thirds (141 to be precise) have known phase-out dates, a necessary condition for a measure to be temporary. The weekly totals of the number of export controls in force on food and agri-food products is shown in Figure 3.

Build-up of new trade policies in force in the medical sector since 1 January 2020

Figure 2 Many “temporary” export bans for medical goods remain in force. Source: Global Trade Alert, April 2023 release.

Build-up of new trade policies in force in the food sector since 1 January 2020

Figure 3 The fallout from attempts to weaponize food trade. Source: Global Trade Alert, April 2023 release.

Failures to unwind temporary export restrictions and the fact that there is no multilateral mechanism to ensure removal raise the possibility that a sequence of systemic crises could result in a permanent reduction of cross-border trade in “essential goods.” Is discrimination in the form of export controls the exception or has it become the norm? This matter is significant nowadays as the wisdom of sourcing all manner of so-called essential goods and critical minerals from geopolitical rivals has been challenged by some officials and analysts (Inkster, 2020).

In light of discussions about security of supply and the desire of some to reduce dependence on imports, Figures 2,3 are revealing in one other critical respect. In both the food and medical sectors since the beginning of 2022 a growing number of import restrictions have been imposed. Even more striking is the huge number of subsidies awarded to local producers. Both of these measures increase the incentive for domestic production, expanding the potential for imports to be substituted by domestic sourcing.2 While few tout a return to the Import Substitution Industrialization policies of yesteryear, it is striking that current multilateral rules have not prevented governments from taking steps that reduce “dependence” on imports.

The third episode relates to invoking national security as a rationale for trade discrimination. Given the concern that intensifying geopolitical rivalries may result in state measures that decouple economies or that induce deglobalization, the following episode is germane. Soon after taking office, in April 2017, President Trump instructed his Secretary of Commerce to investigate whether imports of aluminum and steel products constitute a threat to the national security of the USA. Section 232 of the Trade Expansion Act of 1962 was invoked to initiate these investigations, one for aluminum and one for steel.

In February 2018, the US Commerce Department ruled that these imports harmed national security and on March 1, 2018 additional import tariffs of 25% on steel and 10% on aluminum were imposed on approximately \$48 billion of trade. What followed were threats of retaliation from trading partners and some exemptions were granted by the Trump Administration. Still, significant tariff increases went into effect on March 23, 2018 (Bown & Kolb, 2023). Ultimately, in the 15 months that followed Canada, China, the European Union (EU), India, and Türkiye retaliated in one form or another, typically raising import tariffs on some US exports.

Meanwhile, China, Norway, Switzerland, and Türkiye started WTO dispute settlement proceedings against the USA. The USA brought countersuits against those trading partners that have retaliated against its initial tariffs. On December 9, 2022 a WTO panel of adjudicators found against the USA. One ground of that finding is that the US measures did not meet the required standard of being “taken in time of war or other emergency in international relations.”

These panel reports prompted a sharp retort from the USA. At a January 27, 2023 meeting of the WTO Dispute Settlement Body, the US Permanent Representative to the WTO, Ambassador Pagan, made a wide-ranging statement that included the following statement “For over 70 years, the United States has held the clear and unequivocal position that issues of national security cannot be reviewed in WTO dispute settlement and the WTO has no authority to second-guess the ability of a WTO Member to respond to a wide-range of threats to its security.” While this robust stance was taken by the Biden Administration, completeness requires reporting that during 2022 the USA took steps to partially roll back the import tariff increases imposed by the Trump Administration on steel and aluminum imports from certain trading partners.

In this section, I have provided examples of outright unpunished and nontransitory rule violations by the governments of the two largest economies on Earth. In addition, I have shown that compliance with rules permitting exceptions to non-discrimination has been shaky as well – a finding that implicates many more governments. Still, it would be wrong to conclude that every WTO member defied multilateral trade rules.

3. Evidence on Commercial Policy Choice since the Global Financial Crisis

Some readers may be tempted to dismiss the evidence presented in the last section as relying too much on high-profile exceptional cases. On this view, for sure, multilateral trade rules may not work in extremis but in terms of day-to-day operation; these rules constrain resort to discriminatory or selective policy intervention that disadvantages commercial interests of trading partners. Sadly, this argument is at odds with the evidence, as I will show in this section.

By and large, officially collected information on unilateral commercial policy changes outside of crises is confined to import tariffs, trade defense actions, and subsidies in the agricultural sector. To this should be added, information on new health and safety regulations, which can apply to agricultural and manufactured goods.3 In principle, if governments made complete and meaningful notifications on the subsidy schemes and their awards to corporations, then the WTO Secretariat would have information about that public policy instrument. In practice, however, the amount of information that the WTO Secretariat can draw upon is limited and often includes gaps, as shown by its periodic monitoring reports. Matters are no better at the International Monetary Fund (IMF), where some governments either refuse to supply or refuse to allow publication of information on the subsidies they grant to companies. For researchers, who understandably prefer to download data than incur the costs of collecting it, this data deficit narrows the assessments of both trade policy stance and of compliance with the norms and rules of the multilateral trading system.

Determined to fill in this data gap, with colleagues, I created the GTA. At the time of writing, the GTA database contains information on over 55,000 commercial policy interventions implemented by 196 customs territories since November 1, 2008. A total of 8864 such interventions improve the competitive position of foreign commercial interests, amounting to 17.3% of the sample of recorded policy interventions. This means that for every liberalizing measure recorded on average there are five measures that disadvantage foreign commercial interests. Over 28,000 entries in the GTA database refer to some form of subsidization of firms competing in national or overseas markets.

Figure 4 traces out the changes over time in the share of world goods trade that are covered by (a) an import tariff increase in force in a given year, (b) a subsidy award to one or more local firms in force in a given year, (c) a state-provided export incentive in force in a given year, and (d) any measure that favors domestic commercial interests over foreign rivals. Given the GTA database started collecting information on in-scope commercial policy intervention since November 1, 2008, annual totals from 2009 on are provided. By construction, Figure 4 does not include any policy intervention in force before the Global Financial Crisis. As such, the evidence in Figure 4 will understate the degree to which goods trade is distorted by selective policy intervention in favor of domestic firms.

Figure 4 A significant build-up of corporate subsidies despite World Trade Organization rules. ( ) All included MAST chapters; ( ) export-related measures (incl. subsidies); ( ) subsidies (excl. export subsidies); ( ) tariff increases.

The import coverage shares reported in Figure 4 speak to the medium-term capacity of multilateral trade rules to discourage resorting to discriminatory commercial policies. The first point to note is the developments during the Global Financial Crisis. By 2010 half of world goods trade was in products and to markets where competitive conditions had been skewed toward favored firms. Critically, as pointed out at the time by many economists and officials, governments did not repeat the mistakes of the 1930s and raise import tariffs. Less than 5% of world goods trade faced a tariff increase by 2010 that was not in force before November 1, 2008. Instead, governments resorted to subsidies to import-competing local firms and to exporters, in particular to the latter. Since Figure 4 refers to trade in goods none of the bank bailouts (with their huge price tags) count toward the calculations reported there. Moreover, the export incentives provided were mainly furnished through national tax systems and, by and large, evaded detection by the media and international organizations.

The second finding from Figure 4 is that the share of world goods trade covered by discriminatory unilateral policy grew over time and now stands at approximately 80%. This is multiples of the 28% of world goods trade that has benefited from liberalizing measures that is still in force. If the share of world goods trade where there is a level playing field (no discriminatory measures in force) is an indicator of how well current multilateral trade rules have protected the world trading system, then that share has fallen to 0.2.4

For sure, that share has been stable since 2018, so its low level cannot be attributed to the COVID-19 pandemic or to the fallout from the conflict in Ukraine. That share did fall a little when compared to 2017, implying the US-China trade war contributed – but the damage was largely done well before that. This is significant because the salient trade policy events of the past decade have tended to pile trade discrimination on top of a world trading system that was already thoroughly distorted (by a sustained build-up of corporate subsidies witnessed since the onset of the Global Financial Crisis).

There are compelling political and economic reasons for the resort to firm-specific corporate subsidies as opposed to the import tariff increases. The latter are easier to detect and therefore more likely to irk trading partners. In contrast, many governments publish information on subsidy awards with a considerable lag – the favoritism is dispensed before a trading partner finds out about it.

Moreover, the firm-specific nature of corporate subsidies makes them particularly attractive to recipient firms. Such subsidies give recipients a commercial advantage over other local firms as well as foreign rivals. Plus, governments may find it convenient to “play favorites,” rewarding some local firms and not others. In principle, subsidies are a call on the public finances, but even here governments have found creative ways to shift subsidy awards off the central government’s fiscal accounts.

For sure, the WTO has an agreement covering subsidies. However, it is incomplete in scope (e.g. subsidies for exporters are explicitly banned), it is contested which public authorities are subject to the disciplines, and rules to allow certain potentially welfareimproving subsidies are uncertain (indeed a key provision in this regard, Article VIII of the relevant WTO accord lapsed). Consequently, this agreement is widely regarded as not fit for purpose, hence the current proposals to start discussions on this matter in Geneva.

To sum up, perhaps the best that can be said about the influence of multilateral trade rules on its members’ resort to discrimination is that pressure to favor local or national firms has, by and large, been channeled away from salient policy instruments, such as tariff increases. The consequence, however, is a world trading system where subsidies have become pervasive. Governments have substituted away from a transparent policy tool where the rules are relatively stronger to an instrument that is easier to hide and where existing disciplines are weaker. This dynamic was well underway before American patience with Chinese commercial policies snapped.

4. Evidence from the WTO Tariff Schedules of the G-20 Members

In this section, I focus on the extent to which the governments of the largest economies of the world can raise their taxes on imports without breaking their WTO commitments. The tariff schedules that members have lodged at the WTO have many facets that influence how much leeway a government has to raise taxes on imported goods. First, there may be some goods for which tariffs have not been bound at all. Second, there is the average size of the tariff binding overhang. Third, there is the distribution of that overhang. Fourth, there is the room to increase import tariffs legally to 15% or above (so-called tariff peaks). Fifth, there is number of times the applied tariff rates exceed the bound tariff rates. To be sure the last metric is not a measure of the amount of leeway to legally raise import tariffs. Rather, assuming there are no averaging and reporting errors, it is a potential indication of the degree to which a G-20 member has already departed from its multilateral tariff obligations. Each of these matters is examined in turn for both manufactured and agricultural products.

The latest World Tariff Profiles 2022, published by the WTO Secretariat, reports the extent to which members have bound their import tariffs on manufactured goods. As the first numerical column in Table 1 shows, over 7% of South Africa and South Korea’s tariff lines are unbound, just under 30% of India’s tariff lines have no cap, and a remarkable 57% of Türkiye’s tariff lines are not subject to maximum rates.

With respect to the average tariff binding overhang, the difference between the mean bound tariff rate and the mean applied tariff rate, eight G-20 members could legally raise import tariff rates on manufactured goods by more than the Smoot Hawley tariff. Those eight nations are Argentina, Brazil, Australia, India, Indonesia, Mexico, South Africa, and Türkiye (see Table 1). For agricultural goods eight G-20 members could also legally raise average imports tariffs by 7%, the Smoot Hawley amount. For agricultural goods Australia drops out of the list above and is replaced by Saudi Arabia (see Table 2).

Table 1 G-20 members import tariff regimes vary significantly

	Binding	Simple mean duties Number of			Duty-free		Non-ad valorem duties		Duties > 15%		Maximum duty		Number of MFN
Country	coverage		MFN	tariffs where Bound applied applied >bound Bound applied Bound applied Bound applied Bound applied		MFN		MFN		MFN		MFN	applied tariff lines
Argentina	100	31.6	13.9	16	0	6.3	0	0	98.2	38.3	35	35	9713
Australia	96.8	10.5	2.6	15	22	48	0.9	0	14.5	0	55	5	5628
Brazil	100	30.7	13.8	17	0.7	4.8	0	0	96.5	38.2	35	35	9237
Canada	99.7	5.1	2.1	25	39	79.3	0.3	0	6.7	6.2	20	25	5560
China	100	9.1	6.5	79	6.6	8.6	0	0.3	13.3	1.4	50	50	7465
EU	100	4.1	4.1	332	30	28.6	0.5	0.5	1.6	1.5	26	26	7392
India	70.1	36	14.9	32	2.6	1.8	5.7	5.5	67	28.6	150	328	10,383
Indonesia	95.5	35.8	8	29	2.2	13.8	0	0	89.4	9.8	60	150	9443
Japan	99.6	2.5	2.5	216	55.9	55.9	1.7	1.9	0.8	0.7	473	297	7716
Mexico	100	34.8	6	0	0.2	52.3	0	0.4	99.2	7.3	156	50	6896
Russia	100	7.1	6.1	45	3.4	16.7	7	6.9	0.8	0.8	121	121	9600
Saudi Arabia	100	10.4	5.6	91	8	8.5	0.1	0.4	0.4	0.4	20	20	13,032
South Africa	92.9	15.6	7.6	5	12.8	62.5	0	0.9	32.9	20.3	50	744	6874
South Korea	92.5	9.9	6.6	42	18	18.9	0	0	11.9	3.9	50	50	10,745
Turkiye	43	17.2	5.8	72	3.3	23.8	0.1	0.6	20.4	6.1	100	82	9229
United Kingdom	100	4.1	3	NA	29.7	54.6	0.5	0.3	1.6	1.4	26	25	7401
USA	100	3.2	3.1	154	49.8	50	3.2	3.1	2.3	2.2	57	57	9906

Average measures sometimes hide interesting variation in the underlying distribution. It turns out that this is the case for import tariff binding overhangs for manufactured goods. In researching this paper each G-20 member’s distribution at the tariff line level of the import tariff binding overhang was prepared. Only the EU, Japan, and the USA have distributions that are concentrated around zero. That is, these three G-20 members have exhausted their capacity to legally raise import tariffs. Every other G-20 members, including some higher-income members of that group, have substantial upper tails to their distributions of tariff binding overhangs. This means that for non-trivial shares of their tariff schedules they can raise their tariff rates markedly.

With respect to the leeway to raise tariffs to 15% or above (thereby creating socalled tariff peaks), for manufactured goods a total of nine G-20 members could do so legally on 10% or more of the goods in their multilateral tariff schedules. With respect to agricultural goods, seven G-20 members have the room to do likewise.

Here is a recap. What does this evidence imply? Suppose your mental model of multilateral tariff bindings is that they have prevented governments of the largest economies to raise import tariffs. Well, this mental model only applies to the EU, Japan, the United Kingdom, and the USA. All of the other G-20 members have room to significantly increase some, if not all, of their import tariffs in a manner perfectly consistent with WTO rules. Strictly speaking, the WTO tariff bindings for these countries are not a constraint on their tariff setting. Whether governments, in fact, decide to raise import tariffs to bound levels is a matter of domestic political economy. Formally, the only role played by the tariff binding is to bound from above the import tariff increase.

If the seven percentage point Smoot Hawley tariff increase is the right benchmark for undesirable behavior, then eight G-20 members could implement such a tariff hike and still comply with their multilateral trade obligations. For these WTO members how can anyone credibly argue that extant multilateral tariff bindings prevent a government hell-bent on decoupling? Or are analysts prepared to wager that if enough large economies hike their import tariffs then governments will find their way quickly back to the multilateral bargaining table? Note that conjecture, if it came to pass, speaks only to the restoration of multilateral trade discipline and not to the failure to deter a departure from such norms in the first place.

The last piece of evidence in this section may be a function of how certain averages are calculated or how information is reported in the WTO Tariff Download Facility. Still, completeness requires that I report these results. As the fifth column of Table 1 reveals, with the exception of Mexico and the United Kingdom, every other G-20 member appears to have some applied tariffs on manufactured goods that exceed their bound rates. This phenomenon appears to be most prevalent in the three G-20 members that have the tightest distributions of import tariff binding overhang and low average bound import tariffs on manufactured goods – namely, the EU, Japan, and the

Table 2 Tariff overhangs are larger in agricultural products

	Simple mean duties Number of			Duty-free		Non-ad valorem duties		Duties > 15%		Maximum duty
Country		MFN Bound applied	tariffs where applied >bound Bound applied Bound applied Bound applied Bound applied		MFN		MFN		MFN		MFN	Number of MFN applied tarff lines
Argentina	32.5	10.3	16	0.2	7.3	0	0	95.2	15.1	35	35	1080
Australia	3.2	1.2	4	33.1	76.8	1.6	0.9	3.1	0.3	29	22	856
Brazil	35.5	10.1	6	2.7	7.1	0	0	95.7	14.4	55	35	1064
Canada	16.1	15.9	25	48	68.5	18.4	11.1	9.1	9	559	559	1266
China	15.7	13.8	57	5.9	7.2	0	0.3	35.5	27.3	65	65	1175
EU	12.6	11.7	119	30.4	31.7	33.2	31.6	25.1	23.5	360	160	2101
India	113.1	39.2	1	0	2.9	0.3	0.3	98.3	94.7	300	150	1524
Indonesia	47.3	8.7	0	0	8.5	0	1.4	99.5	9.6	210	150	1369
Japan	18.4	14.9	172	34.1	35.8	15.1	13.2	23.5	21.6	628	628	1866
Mexico	46.4	13.5	3	0.4	22	7.9	4.8	95.6	38.7	254	75	905
Russia	10.8	9.7	65	3	10.3	22.9	24	7.6	7.3	106	106	2709
Saudi Arabia	17.8	10.7	4	0.1	22.5	8.9	9.8	6.3	6	>1000	>1000	1998
South Africa	39	8.8	2	22.5	46.1	0	13.7	71.8	22.6	597	110	1185
South Korea	64.9	56.8	20	2	5.5	5.3	3.1	76.5	53.1	887	887	1731
Turkiye	61.8	41.1	13	0	18.1	0	0.5	87.7	64	225	225	2272
United Kingdom	12.5	10	NA	30.4	39.5	33.2	26.4	25.6	21.6	331	143	2127
USA	4.5	5.2	33	30.7	30.6	42	41.7	4.9	5.5	350	350	1705

USA. If these three G-20 members are breaking their tariff bindings as often as the data presented in Table 1 implies, then it casts further doubt on whether existing multilateral trade obligations actually curtail unilateral commercial policy choice. Since tariff obligations are the clearest cut, easiest to understand multilateral trade obligation, if there are concerns about their efficacy then what does that imply about the likely efficacy of other more complicated, potentially more contested, and harder to monitor multilateral trade obligations?

5. Assessment: Current WTO Rules cannot Contain Pressures to Deglobalize

The evidence presented in this paper implies that we should moderate our expectations as to the extent to which existing multilateral trade rules can limit state-induced deglobalization. Earlier, it was argued the WTO rule book is incomplete – for example, there are no rules on digital trade. Rules on foreign direct investment are patchy. Even where multilateral trade rules have been agreed, they allow many governments significant leeway to fragment markets (recall the size of the import tariff binding overhangs), or contain exceptions (recall the Article XI exceptions and those pertaining to national security) or they can be honored more in the breach (recall the evidence on the build-up of corporate subsidies) seemingly with impunity. Even when the rules were subject to dispute settlement, foundational principles – such as the MFN tariff treatment – have been simply ignored by major trading nations.

The best that can be said about the current multilateral trade rule book is that, to date, many governments (of mainly small- and medium-sized economies) have chosen not to openly violate their commitments. Pressures to favor national interests have been channeled away from transparent forms of trade discrimination (such as import tariffs) to less transparent forms of selective intervention (corporate subsidies, government procurement measures, export controls, and the like.) This shift toward harder to detect and less closely observed commercial policy intervention may have created the impression of a system capable of deterring government steps that fragment markets. If that was the case, the public revocation of MFN tariff treatment for most of China’s exports by the USA shattered that illusion. China’s retaliation and the enduring breach of the MFN principle have not been lost on observers either. Furthermore, the manner in which governments reacted to the crises of recent years and to intensifying geopolitical rivalry suggest little interest among the larger trading nations in returning to rules of the game agreed at the end of the Uruguay Round 30 years ago.

That multilateral trade rules are not likely to be an effective bulwark against conscious efforts to fragment markets or to offers of inducements to move factories across borders does not imply that these are desirable policy interventions. Indeed, that so many governments have chosen not to abandon the MFN principle, and that dozens of WTO members chose to lower trade barriers on medical goods and food during the crises of recent years, suggests that unilateral commercial policy is not always moving in a discriminatory direction. My conclusion here is that, for better or for worse, we have reached a point where governments have options that are no longer meaningfully constrained by multilateral trade rules. Of course, these findings make a mockery of the mantra that governments have insufficient policy space.

In principle, groups of like-minded nations could deter state-led deglobalization within the group – although that proposition ought to be evaluated on a case-by-case basis. Perhaps the expansion of membership of the Comprehensive and Progressive Agreement for Trans-Pacific Partnership (CPTPP) can secure an oasis of less-distorted trade among its members? But creating an oasis does not address the root causes of the growing impotence of multilateral trade rules evidenced in this paper. Still, should compliance with multinational trade rules weaken further, one option available to governments is to seek to substitute lost multilateral market access with an expanded and deeper network of regional trade agreements.

If multilateral trade rules cannot do the job of discouraging policy-induced deglobalization, what can? The decisive battles over unilateral policy will take place in national capitals. With intensifying geopolitical rivalry, we can reasonably expect that national security and foreign policy officials are likely to play a greater role in shaping commercial policymaking. Too many of the former cross-border commercial ties are said to be a source of risk, so little discouragement to deglobalize can be expected from that quarter (Inkster, 2020; Leonard, 2021). For these officials, trade policy is, at best, part of second order diplomacy, to be invoked symbolically and not because of any expectation that curbing trade will actually change a foreign government’s actions. One potential source of countervailing power comes from that part of the business community which has a significant stake in an open global trading system. Whether enough senior executives are willing to do so is an open question.

Notes

1 The UN COMTRADE database is used to calculate these import coverage shares. Bilateral trade at the six digit level of product classification is used. The dates each tranche of tariff increases went into effect are taken from records in the Global Trade Alert database. Annual import coverage estimates are adjusted for the length of time a tariff increase was in effect (not when it was threatened, only imposed tariffs count). As a benchmark, which will be referred to in a later section of this paper, information on the import shares covered by tariff increases is supplemented by calculations of the import shares covered by all policy interventions that favored domestic firms. The latter import shares equal or exceed the former import shares by construction and are therefore represented by a different plotted line.

2 Of course, if demand for these goods is increasing faster than domestic supply then total imports need not fall.

3 So called sanitary and phyto-sanitary measures (SPS) and technical barriers to trade (TBT), respectively.

4 In Figure 4 the share of undistorted global goods trade in a given year is shown by the vertical distance between the upper most line in that year and horizontal line representing one on the vertical axis.

References

Baldwin, R. (2022a). The Peak Globalisation Myth: Part 1. Voxeu.org. Accessed 13 July 2023. Available from URL: https://cepr.org/voxeu/columns/peak-globalisation-myth-part-1

Baldwin R. (2022b). The Peak Globalisation Myth: Part 2 – Why the Goods Trade Ratio Declined. Voxeu.org. Accessed 13 July 2023. Available from URL: https://cepr.org/voxeu/columns/peakglobalisation-myth-part-2-why-goods-trade-ratio-declined

Baldwin R. (2022c). The Peak Globalisation Myth: Part 3 – How Global Supply Chains are Unwinding. Voxeu.org. Accessed 13 July 2023. Available from URL: https://cepr.org/voxeu/ columns/peak-globalisation-myth-part-3-how-global-supply-chains-are-unwinding

Baldwin R. (2022d). The Peak Globalisation Myth: Part 4 – Services Trade Did Not Peak. Accessed 13 July 2023. Available from URL: https://cepr.org/voxeu/columns/peakglobalisation-myth-part-4-services-trade-did-not-peak

Bown C. & Kolb M. (2023). Trump’s Trade War Timeline: An up-to-Date Guide. Washington, DC: Peterson Institute of International Economics. Trade and Investment Policy Watch.

Evenett S. (2023). What endgame for the deglobalisation narrative? Intereconomics, 57 (6), 343–351.

Goldberg P. & Reed T. (2023). Is the global economy deglobalizing? And, if so, why? And what is next? Brookings Papers on Economic Activity Conference Drafts.

Inkster N. (2020). The Great Decoupling: China, America, and the Struggle for Technological Supremacy. London: Hurst Publishing.