At a time when the international trade system faces the most significant challenge in its eighty-year history, it is a pleasure to read Industrial Policy, National Security, and the Perilous Plight of the WTO by Petros Mavroidis, the Edwin B. Parker Professor of Foreign and Comparative Law at Columbia Law School. The book pre-dates the start of the second Trump administration, but it wisely presages many of the trade issues we are currently grappling with, particularly the relevance of the World Trade Organization (WTO) as a legal order and an organization. The book reminds us that the WTO has certainly faced dire circumstances before. During economic shocks like the Great Financial Crisis or the COVID-19 pandemic, legal scholars often questioned whether the international institution could withstand the pressures on national governments to break from trade norms. But this current crisis is a political, not an economic, shock. The United States has abandoned the WTO framework, and the rules that once provided a stable guide for global commerce are unraveling. It is an open question how much law still exists in international trade or if enough support remains to maintain a rule-based system among major economies.
Mavroidis’s book provides an exceptionally good lens to examine these issues, and the author provides a masterful and nuanced analysis of the history of trade law as well as current and future challenges. The book’s principal focus is on exploring the role of industrial policy in advancing national economic and security objectives. It discusses the history of the U.S. semiconductor industry and offers several proposals for how the WTO should regulate subsidies and export controls. The book also examines important and related questions, including how the WTO should address national security concerns and what reforms are necessary to keep the WTO relevant to today’s challenges. This review will follow that structure—primarily focusing on industrial policy but also covering national security adjudication and WTO reform—while additionally considering the challenges posed by Trump 2.0, including its unconventional industrial policy and attacks on multilateralism.
Industrial Policy in the Semiconductor Industry
One of the book’s primary inquiries is how national governments’ increasing focus on industrial policy can be reconciled with multilateral trade rules. Mavroidis references Lane’s definition of industrial policy as “intentional political action meant to shift the industrial structure of an economy” (p. 46). States pursue industrial policy for strategic reasons (establishing or expanding key industries), economic reasons (boosting growth or employment), and national security interests (securing critical supplies or gaining advantages in military or foundational technologies). Assertive industrial policies can conflict with WTO rules designed to promote non-discriminatory international trade among members. The book’s aim is not to persuade government leaders that the concerns behind states’ industrial policies are misguided or exaggerated. Instead, Mavroidis agrees that greater geopolitical concerns (primarily between the United States and China) will persist. As a result, government subsidization or tariff support of strategic national industries will remain part of the landscape. Instead, Mavroidis aims to limit some of the most harmful effects of modern industrial policy through new multilateral agreements. Yet novel risks are on the horizon. Trump 2.0 poses new challenges, including his unorthodox approach to industrial policy, seeking to extract wealth from foreign governments and private firms, both foreign and domestic.
To explore the effects of industrial policy, Mavroidis begins with a case study of the global semiconductor industry. Semiconductors are an excellent choice for a case study, as the industry is long-standing (and thus has faced various types of threats and policy interventions) and yet remains highly salient to high-value commerce today. Semiconductors provide a window into sectoral international agreements, large-scale subsidies, export controls, and, most recently, import and export tariff policy.
Semiconductors are a transformative technology. While early semiconductors had only four transistors on a chip, cutting-edge chips now contain over 208 billion transistors. Semiconductors are the brains behind the era of computing, the rise of digital empires, and the start of the AI revolution. The industry is so vital to technological innovation and future growth that national capitals have actively sought to shape its trajectory for decades.
Mavroidis provides an illuminating narrative of how the focus of industrial policy has shifted from the 1980s and 1990s to the Biden administration’s focus on the CHIPS Act and export controls. In the 1980s, the semiconductor rivalry was mainly between the United States and Japan. Although the United States had developed the first semiconductors in the 1960s and created Silicon Valley in the 1960s–1970s (with critical government support from the Defense Advanced Research Projects Agency (DARPA)),Footnote 1 American companies were lagging their Japanese rivals by the 1980s.
The solution for American policymakers was to demand a managed trade regime through an international sectoral agreement. The 1986 U.S.-Japan agreement divided up the global chip markets by extracting a promise from Japan to raise chip prices in third-party markets (p. 30). The agreement ended as the European Economic Community, whose industries were paying the higher prices, successfully sued Japan under General Agreement on Tariffs and Trade (GATT) rules. But the end of the 1986 agreement was followed by a subsequent agreement in 1991 with specific quantitative targets for chip sales over the next five years. This agreement went unchallenged but arguably did not solve the American competitiveness concerns. The narrative highlights how this episode was not primarily viewed through a national security lens. Japan was part of the American security umbrella and represented an economic rival but not a military threat.
The case study returns to the semiconductor industry in the 2020s. By the beginning of the Biden administration, U.S. companies were overwhelmingly designing, not producing, semiconductors. The strategic role of semiconductors has also shifted. The United States is not only concerned about maintaining its edge in the semiconductor industry but also in other industries where advanced chips are critical, including AI, quantum computing, and advanced weaponry. In addition, the race to maintain semiconductor superiority is viewed explicitly through the national security lens of the American geopolitical rivalry with China. As the case study points out, the Biden administration’s trade policy was being set by National Security Advisor, Jake Sullivan, not the U.S. Trade Representative, Katherine Tai (pp. 59, 227). Sullivan shifted American national security policy regarding computing-related technologies from a sliding-scale approach, where the United States tried to stay one generation ahead of China, to one that advocated “as large of a lead as possible.”Footnote 2 Sullivan’s policy ushered in an era of export controls on chip sales to China as well as other goods. Rather than push for a large-scale decoupling from China, Sullivan advocated a “small yard, high fence” strategy that aimed to curtail Chinese access to a limited set of goods significantly (p. 67).
The Biden administration also passed the CHIPS Act with broad bipartisan support, which provided over $52 billion in subsidies to the semiconductor industry. The subsidies were designed to aid the Western firms’ rate of innovation as well as to bring more production to American shores.Footnote 3 American export controls cut off an important source of revenue for Western firms—the loss of the Chinese market for advanced chips deprived these firms of significant income that was crucial to re-investment in research and development. The CHIPS Act thus compensated the firms and allowed them to speed development, while export controls sought to hold Chinese innovation back.Footnote 4
Trump 2.0 has taken yet another tack in semiconductor policy. The administration opposes the CHIPS subsidies and instead favors the threat of 100 percent tariffs on semiconductor imports. However, tariffs are not a good substitute for subsidies in the semiconductor industry. Even a very high tariff may not encourage domestic investment, given the high costs of chip manufacturing, the subsidies available from other countries, and the breadth of the global chip market. Tariffs also do not compensate domestic firms for their loss of the Chinese market caused by export restrictions and may further reduce their revenues. As a result, Western firms will have fewer resources to invest in research and development, endangering the American technological lead over China that the CHIPS Act was designed to provide.
But, in another policy reversal, Trump 2.0 may not be concerned with this loss of a technological lead over China. It has rejected many of the Biden-era geopolitical security concerns in favor of a more transactional approach, which downplays the national security aspects of its semiconductor policy. Although it has retained many of the Biden administration’s export controls, these restrictions are on the bargaining table, both with China and American firms. The Trump administration appears willing to relax export controls as part of trade talks with China. It has also effectively sold export licenses to domestic chip firms, Nvidia and AMD, in exchange for 15 percent of their Chinese sales. More recently, the administration agreed to allow Nvidia to sell previously restricted chips to China in return for 25 percent of its Chinese sales. With this more transactional stance, the current administration is reframing semiconductors from a national security concern to a commercial concern, opting to allow greater technological parity with China in return for higher sales volume and tax revenue. The three varying approaches to the semiconductor industry demonstrate the diversity of industrial policies that states can take to the semiconductor industry, as well as the shifting role of national security in American policy.
Mavroidis uses the first two stages—the 1980s and the Biden administration—to discuss how national security and economic policy play out in international trade and to explore possible reforms. Mavroidis persuasively argues that the Subsidies and Countervailing Measures (SCM) Agreement should adopt a “rule of reason” that allows subsidies that produce positive externalities, particularly environmental benefits (pp. 255–56). He also advocates for the creation of a WTO Export Licensing Agreement (ELA) to review members’ use of export restrictions. Members would have to share information on their export licensing practices, including reasons for denying licenses, so that the relevant WTO committee can evaluate whether these export restrictions serve national security goals or protect domestic industries (pp. 257–58). Both proposals are thoughtful and would be a positive development, although the ELA might get more support as a standalone agreement outside the WTO. As most export controls are now imposed on other WTO members (such as U.S. restrictions on chip sales to China), states may prefer to cooperate in a mini-lateral forum that excludes the target rather than the more inclusive WTO.
While these reform proposals address a broad range of concerns related to trade and industrial policy, they do not provide an easy off-ramp for the current political dilemmas in semiconductor trade, including the use of discriminatory subsidies and differing views over the legitimacy of national security concerns. Indeed, the most significant challenges today are not due to the substance or scope of international trade law but to states’ failure to comply with existing trade rules. Partly for these reasons, Mavroidis astutely also supports a sectoral agreement on semiconductors that would include a small number of states, most notably the United States and China (pp. 270–71). A sector-specific agreement tailored to the political and security concerns of the parties is the most likely solution, but the devil is in the details. Beyond the willingness of the United States and China to reach an agreement, this agreement could affect third-party states (including Japan, South Korea, and the EU), limiting their access to Chinese or American markets through market-setting arrangements or quantitative restrictions (reinventing the U.S.-Japan agreement). Given Trump 2.0’s demand for bilateral agreements, other interested states may be left out of the bargain, although an agreement could be expanded later.
The forms of industrial policy and their impacts on trade may also be shifting. Some of Trump 2.0’s industrial policies are unique and may reshape how we think about the public purpose of industrial policy. Trump 2.0 has taken a series of unorthodox steps that arguably amount to an industrial policy aimed at accumulating wealth for the government. In addition to the previously mentioned 15 percent and 25 percent export tax on Nvidia and ADM, the Trump administration has also demanded a 10 percent equity stake in Intel. Intel was the recipient of $8.9 billion of CHIPS Act subsidies, which the Trump administration refused to disburse. To secure its awarded subsidy, the company agreed to give the U.S. government a 10 percent equity stake in the firm. President Trump stated that he would continue to try to get “as much as I can” out of firms in the future, touting this development as akin to creating a sovereign wealth fund.Footnote 5 Administration officials have also discussed demanding an equity stake in other semiconductor companies and defense contractors. Foreign firms have also been subject to these demands. For Nippon Steel to complete its purchase of U.S. Steel, the Committee on Foreign Investment in the United States (the U.S. foreign investment review agency) required the Japanese firm to give the U.S. government a golden share, giving the government a veto over most of U.S. Steel’s business planning. These actions are a form of industrial policy but are not done to aid the domestic industry or produce positive externalities for the broader economy. Rather, the actions are aimed at extracting benefits from corporations, domestic and foreign, for the benefit of the government itself.
This form of industrial policy has been characterized as socialism or state capitalism, but it differs from traditional forms of these in that the government does not plan to run companies or produce goods or services itself. It is also not doing so to preserve or favor these industries, but to extract revenue, equity, or control from them. The closest analogy may be systemic corruption, where the state uses its monopoly power to extract rents for government officials acting in their personal, not public, capacity. The current administration does not appear to be deriving rents for themselves as private individuals (in this context) but for the state. However, the state is not dispersing these assets toward publicly oriented goals but holding them as sovereign wealth. This fits with the president’s commercial background: The state is acting as a business, with a mission to accumulate wealth as its standalone goal.
This unique brand of industrial policy raises different concerns for international trade than the traditional subsidy or tariff issues. These policies do not obviously favor American firms (they are currently targets), so they raise fewer immediate concerns about discrimination against foreign producers. However, these policies could soon target foreign firms directly, demanding equity, corporate control, or higher tariffs for access to the American market. Neither Mavroidis’s proposals nor most WTO agreements address this form of industrial policy, which utilizes individual corporate regulatory actions rather than national-level actions. To the extent that this type of industrial policy becomes embedded in the U.S. governance system, it will raise a set of new challenges that will require different models of cooperation to address.
National Security
One of the most original and thoughtful contributions of the book is Mavroidis’s analysis of how to address disputes when members defend their policy as key to the state’s national security. The book offers a comprehensive and superb analysis of the history of the national security exception, a meticulous examination of recent legal cases, and a proposal to resolve the current national security dilemma.
Article XXI of the GATT provides an exception for member state actions that breach GATT rules, but “which [the member] considers necessary for the protection of its essential security interests” and are “taken in time of war or other emergency in international relations.”Footnote 6 For most of the GATT and WTO’s existence, states used significant restraint in not invoking this exception, although the Soviet Union and aligned states were not in the GATT during the Cold War. As Mavroidis notes, “this time it is different” (p. 14): the geopolitical rivalry between the United States and China is characterized by deep economic ties, and other adversarial dyads (e.g., Russia-Ukraine) are both WTO members. Any national security measure that a state enacts regarding a rival is likely to have WTO implications.
Since 2018, there have been a series of cases where the respondent state has relied on national security as a defense. While many states, including the United States and Russia, have argued that Article XXI is self-judging and therefore outside the jurisdiction of the Dispute Settlement Understanding (DSU) system, WTO panels have disagreed and adjudicated the exception’s application (pp. 146–58). There is significant academic debate over the best approach for addressing WTO disputes with national security implications. Nicolas Lamp argues that cases which invoke the national security defense should be treated as nullification and impairment claims and thus do not release a state from its obligation to compensate injured WTO members.Footnote 7 Others argue that the WTO should not be in the difficult position of adjudicating national security concerns. Mona Pinchis-Paulsen suggests that the WTO could require notification of national security claims but otherwise be a forum for discussion, not legal review.Footnote 8
Mavroidis similarly rejects the idea of judicial review of national security claims and provides an intriguing proposal: that the WTO establish a national security committee that could provide external scrutiny of members’ national security claims but not adjudicate their legal consistency with Article XXI. At a conceptual level, this approach seeks to turn highly political disputes into more technical analyses. The committee would employ national security experts who would mediate and help members find a mutually acceptable solution. The committee’s review would focus less on whether there is an emergency in international relations, and instead address the specific security concern, how the state has chosen to address it, and the fit (efficiency/restrictiveness) of the proposal to the concern (p. 261). Members who refuse to participate in the process (and perhaps those who reject any compromise) would have to offer compensation to injured states.
The proposal certainly has advantages over dispute settlement for core national security concerns, where member states are unlikely to accept WTO adjudication. The question is whether states would be more receptive to technical rather than legal review. This affects how the proposal would work in practice. Would every dispute become a national security matter, effectively ending dispute resolution? Would members act in good faith during the committee deliberative process? Would the remedies (or lack of remedies) drive the committee’s process?
First, would the committee have the authority to reject a member’s determination that an issue is a national security concern? If a respondent state could invoke Article XXI for any dispute, shifting the forum from the DSU process to the committee process—which would focus on finding a mutually acceptable solution rather than a legal review—then legal dispute resolution at the WTO might become the exception. Mavroidis acknowledges that members might abuse the process this way but thinks the proposal is worth the risk. That might be correct given the weak state of current legal dispute settlement. However, it erects another barrier to reviving the DSU or maintaining dispute settlement for WTO members that have joined the Multiparty Interim Appeals Arbitration Agreement.Footnote 9
Second, will politics and power drive the outcome regardless of the committee’s processes? If this is a core national security issue, will members be willing to compromise or adjust to external scrutiny, even if that scrutiny is deliberative rather than legal? While the process may offer members insight into others’ views or options, it is hard to see this influencing some major geopolitical players (Russia, China, the United States) when national security matters are at stake. Additionally, the question of remedies might cast a shadow over the entire process. If some members—like the United States, China, or the EU—threaten to retaliate economically against a less economically powerful WTO member over national security measures, this could shape the deliberative process. The most economically powerful states might demand major policy changes if that member wants to avoid retaliation, regardless of what committee experts recommend. Conversely, economically powerful states might refuse to engage the national security committee in good faith, knowing threats of counterretaliation (and ability to escalate retaliation) will discourage smaller members from insisting on compensation.
Despite these open questions, the book’s discussion of the dilemma of national security issues in WTO dispute settlement and its creative committee-based proposal are important contributions. If members show restraint in classifying disputes as national security issues and use the committee to consider alternative views and policy options in good faith, then the national security committee proposal offers a way to resolve some disputes outside of the all-or-nothing approach of the Article XXI adjudication. The dialogue associated with the committee process could also help build trust between parties, easing some of the concerns related to national security.
WTO Reform
Another area where the book is particularly valuable is its consideration of how to reform the WTO to improve its relevance in international politics. In addition to amending the SCM Agreement and creating an export licensing agreement, Mavroidis argues that the WTO needs to resolve its dispute settlement crisis by allowing members to opt out of the DSU’s jurisdiction. He highlights that this is an existential moment for the institution, so extreme options need to be considered to preserve a cooperative core of the trade regime. In his words, the WTO may need to backtrack from multilateralism to save multilateralism (p. 283).
The WTO’s “perilous plight” has become even more perilous since the book’s publication and requires us to consider what the WTO’s goals should be going forward. The prize of the WTO membership has always been market access, particularly to the world’s largest markets—the United States, the EU, and now China. The most favored nation principle, even with preferential trade agreements, offered all WTO members similar access to these markets, and the national treatment principle regulated protection behind the border. The WTO can no longer credibly promise market access, let alone on non-discriminatory terms.
The Trump administration’s International Emergency Economic Powers Act (IEEPA) Section 232 and Section 301 tariffs have completely upended access to the American market. To regain entry, the United States is demanding bilateral negotiations that raise the American tariffs above bound levels, lower foreign tariffs on a discriminatory basis, require the promise of foreign investment, and often necessitate domestic regulatory changes. In addition, the United States is also applying 50 percent tariffs on WTO members to achieve sundry foreign policy goals, from concerns about the state’s criminal justice system (Brazil) to its purchases from Russia (India). Even after this administration, it seems unlikely that the United States will return to the WTO’s negotiated tariff levels, promise of non-discriminatory treatment, or preference for multilateral negotiations.
Similarly, access to the Chinese market is increasingly uncertain. Over the last decade, the Chinese government has become increasingly coercive in its use of market access to achieve its foreign policy goals (pp. 223–25), including restricting market access to Australian and EU exports after disputes over the origins of COVID-19 and Taiwan, respectively.Footnote 10 Although China has to date implemented most of its actions through informal measures, scholars have begun to document how China is increasingly using formal government action to restrict trade and investment, copying the American strategy.Footnote 11
The WTO model lacks champions among the major economic powers. The EU could have taken up that mantle, and Mavroidis argues that the EU could “become the instigator of the WTO renewal” along with some other G20 members (p. 245). The EU did defend that multilateral order in the face of the first Trump administration’s Section 232 tariffs on steel and aluminum.Footnote 12 However, the EU is no longer willing to fill this role. Confronted with the 2025 IEEPA tariffs, the EU quickly formed a bilateral deal with the Trump administration, agreeing to the American 15 percent tariff rate, lower tariffs for U.S. imports, and promises of greater American investment.
What is the purpose of the WTO if it can no longer offer stable and predictable access to the world’s largest markets, and how should this shape reform proposals? One goal could be to lure the United States and China back into the fold by weakening WTO constraints. Reforms in this vein aim to loosen state obligations both substantively—for instance, reforming the rules on subsidies to allow more interventionist industrial policies—and procedurally—for instance, allowing states to opt out of rule-based adjudication. The book reasonably suggests reforms along these lines to keep the WTO policy relevant. However, given the seismic shifts in trade policy with Trump 2.0, we can question whether the United States and China are still invested in the WTO model and whether relaxing WTO constraints will be sufficient to hold the system together.
An alternative goal might be to preserve the WTO as a system of rules that binds states that are still interested in maintaining an open trading regime. There are many states in the G20, the BRICS, and beyond, which desire to maintain something close to the status quo. Their economic growth depends on exports, and while they may now have to make special accommodations for access to the United States and Chinese markets, they want to diversify their export markets and solidify their relationship with one another. For these nations, loosening the WTO substantive obligations or relieving states of the obligation to engage in rule-based dispute settlement may decrease the WTO’s value. It could make sense to allow special exceptions for the United States and China (particularly if this can end the Appellate Body crisis) while preserving robust rules on market access for others. For these states, their preferred reforms may be far more modest, although they would be open to plurilateral agreements as Mavroidis recommends (pp. 282–83).
In both scenarios, the WTO has a much smaller role in global policy than it has in the past. While it is unclear what the longer trajectory for international trade will be, the era where trade rules had a strong constraining effect on most of the world’s economies appears to be closing. There will likely be a lot less law in this system. Instead, major economies seem to be building up their capacity to act unilaterally to fight for the policy they prefer, irrespective of the policy’s consistency with WTO trade rules. Even the EU has enacted its Anti-Coercion Instrument, providing the EU Commission with the power to invoke a broad range of trade policy tools to address any foreign trade policies it finds threatening without multilateral adjudication.
Mavroidis concludes by noting that “[t]he WTO as is probably came too soon, beguiled by the momentary mirage of the late eighties that led some to believe that history had ended, and we could all walk the same path at the same pace” (p. 286). The WTO reflects American and European economic priorities at the end of the Cold War and their hopes for a sustainable global governance system anchored by trade law. Developed countries, particularly the United States, leveraged international law processes to strengthen the authority, legitimacy, and enforceability of their preferred trade rules. However, these rules provided economic benefits to more than a small group of nations. Many WTO members gained, to varying degrees, from the clarity, predictability, and stability of the trading system. While some states might not view the decline of the WTO as an event to mourn, it remains uncertain what the distributional consequences of future trading relations will be. As Monica Hakimi and Jacob Katz Cogan argue, the end of the U.S.-led international order will almost certainly lead to new sources of power in the international system, some liberal and some illiberal.Footnote 13 The rules are unlikely to be substantively better for smaller states, and there is likely to be less predictability and fewer avenues for redress. In addition, Mavroidis is undoubtedly correct that the costs of losing the WTO are tremendous for addressing collective problems, such as climate change or economic shocks. Yet we seem to be headed toward that result: a more decentralized global trade regime where states increasingly leverage their market power to pursue economic and foreign policy objectives without the constraints of legal rules or processes.
