Imperial Presidency Versus Fragmented Executive? Unilateral Trade Measures and Executive Accountability in the European Union and the United States

Abstract

Faced with a changing geopolitical environment, the European Union has embarked on a legislative program to upgrade its unilateral trade instruments toolbox. By reforming existing instruments—for example, anti-dumping—and by adding new instruments to the European Commission’s toolbox (foreign subsidies instrument, international procurement instrument, anti-coercion instrument, and others), the EU legislature is significantly strengthening the position of the Commission in the governance of unilateral trade policy in the EU. This development raises accountability questions. By means of a comparative analysis of democratic accountability in unilateral trade policy in the United States and the EU, I describe this transformation of executive power in the EU and I argue that a further strengthening of democratic accountability mechanisms is needed to match the Commission’s growing responsibilities in this underexamined corner of EU trade policy.

A. Introduction

Faced with a changing geopolitical environment, the European Union started upgrading its unilateral trade instruments toolbox. ¹ In 2017, the EU modernized its anti-dumping and anti-subsidy rules to better equip the European Commission to deal with state-induced trade distortions. ² In 2019, the EU established a framework to screen foreign direct investment in the EU. ³ In 2021, the EU revised its Enforcement Regulation to enable the Commission to retaliate against violations of international trade rules in the event of a so-called appeal “into the void,” wherein a World Trade Organization (WTO) member appeals against a panel report even though the WTO Appellate Body is currently dysfunctional. ⁴ In December 2021, the Commission proposed an “anti-coercion” instrument, which will grant the Commission broad powers to retaliate against acts of “coercion” targeted at EU Member States or institutions. ⁵ In June 2022, the European Parliament and the Council adopted a regulation that empowers the Commission to restrict the access of economic operators, goods or services from third countries to EU public procurement procedures when it is in the “Union[’s] interest” to do so. ⁶ And in December 2022, Parliament and Council adopted an instrument that would enable it to tackle subsidies issued by third country governments to companies established in the EU—thereby filling an alleged gap between the EU’s state aid, public procurement, and trade defense rules. ⁷ All of these initiatives fit in the Commission’s ambition to strengthen the EU’s “open strategic autonomy” ⁸ and to put in place a trade policy that is “open, sustainable, and assertive.” ⁹

The list of proposed and adopted measures is impressive. ¹⁰ The combined effect of these initiatives is a significant strengthening of executive power in EU trade policy—an area in which the EU holds exclusive competence and in which Member States are thus constitutionally precluded from acting on their own. ¹¹ Moreover, these legislative initiatives have to be looked at in conjunction with the increased use in recent years of sectoral economic sanctions adopted by the Council within the framework of the EU’s Common Foreign and Security Policy (CFSP). Most recently, in response to the Russian invasion of Ukraine, the Council restricted imports of a wide range of Russian and Belarussian goods into the EU, including coal, iron, steel, and wood. ¹² This return to sectoral economic sanctions represents a policy change, as in recent decades CFSP restrictive measures have typically been targeted at individuals rather than economic sectors or entire economies. ¹³ Taken together, this makes for a picture of a heavily-armed European Union trade power—a Union that is increasingly capable of deploying and leveraging its large internal market as an instrument of economic statecraft. ¹⁴ And a Union that is increasingly willing to deploy trade policy instruments not only in pursuit of trade liberalization, but also in pursuit of non-trade related objectives, including, most significantly in this moment: Security. ¹⁵

A growing toolbox of unilateral trade policy instruments empowers the executive—in this case, primarily the European Commission. As executive power grows, it is harder for the other branches of the trias politica—the legislature and the judiciary—to hold the executive accountable. To get a sense of the accountability challenges that executive empowerment brings, it is useful to look at the experience of the United States. The U.S. President has important independent constitutional responsibilities over foreign affairs, including the power to control the military. In the trade policy sphere, the President’s independent constitutional responsibilities are more limited. However, Congress has delegated much of its trade policy-related responsibility to the President. ¹⁶ These powers include important unilateral trade powers: the President can impose economic sanctions in response to perceived threats to U.S. national security, or when sanctions are deemed necessary to preserve U.S. commercial interests. The experience under the Trump administration has demonstrated how difficult it is for Congress to operate as a check over the President. This, in turn, has raised the question of whether the existing division of labor between Congress and the President should be reconsidered.

In this Article I look at how executive power in the unilateral trade policy sphere is held to account in the European Union. I will do so in light of the abovementioned reforms to the EU’s trade defense toolbox, and I will do so from a comparative perspective by contrasting the allocation of executive power in the EU in the area of unilateral trade policy with that in the United States. All knowledge is comparative. In that spirit, I look at the U.S. experience in this area to reflect on how that experience may inform how EU policy makers should think about executive accountability in unilateral trade policy. ¹⁷

The argument I put forward has two components: One descriptive, one normative. In descriptive terms, I argue that executive power in unilateral trade policy in the EU is going through a transformation. Taken together, the various instruments I referred to in the above significantly strengthen the position of the European Commission. This is the case as most of these instruments will be—or already are—deployed by the Commission. The Commission both proposes and adopts measures, and in the process exercises a growing amount of discretion. At the same time, it remains difficult for the Member States, and impossible for the European Parliament, to oppose measures. As I will describe, this state of affairs is not as different from that in existence in the United States as one might expect at first glance given the important structural differences between the EU and U.S. executives. Indeed, the U.S. experience in this area can be understood as a cautionary tale of how growing executive power leads to accountability challenges that over time become increasingly difficult to surmount.

In normative terms, I argue that this process of growing executive power calls for a concomitant strengthening of accountability mechanisms in the EU. Crucially, as a polity that has both states and individual citizens as co-equal sources of legitimacy, ¹⁸ and which is constitutionally committed to representative democracy, ¹⁹ accountability lines must run from the Commission to both the Member States and the EU citizens. This means, in institutional terms, that both the Council and the European Parliament should be meaningfully involved in the executive decision-making process whereby the Commission proposes and adopts unilateral trade measures. Importantly, as trade policy becomes increasingly “geopoliticized” ²⁰ and the amount of discretion that is involved in decision making grows, ex post legal accountability through judicial review is not sufficient to ensure an acceptable level of legitimacy.

The Article has two parts. First, I take a comparative detour to look at how executive accountability works in the United States, and at how the perceived shortcomings of the U.S. arrangements have led to calls for reform. Second, I then turn to the EU and further unpack how executive power is allocated in the EU when it comes to unilateral trade instruments—a category that covers the instruments adopted in the context of both the EU’s Common Commercial Policy (CCP) and the CFSP. In a final section, I develop the normative argument that more and better democratic accountability mechanisms are called for in this area.

B. Unilateral Trade Measures in the United States: Taming the Imperial Presidency

I. The Imperial Presidency in Unilateral Trade Policy

In his 1973 book, Arthur Schlesinger referred to the US Presidency as “imperial” in nature. Writing in the aftermath of the Watergate scandal—President Nixon’s infamous scheme to wiretap the headquarters of the Democratic Party—Schlesinger wrote:

The all-purpose invocation of “national security,” the insistence on executive secrecy, the withholding of information from Congress, the refusal to spend funds appropriated by Congress, the attempted intimidation of the press, the use of the White House itself as a base for espionage and sabotage directed against the political opposition—all signified the extension of the imperial Presidency from foreign to domestic affairs. Underneath such developments there could be discerned a revolutionary challenge to the separation of powers itself. ²¹

As Schlesinger himself noted in this quote, the Imperial Presidency was not a novel phenomenon in the United States of the early 1970s. ²² What was new, he argued, was its extension from foreign to domestic affairs. ²³ For indeed, in the foreign affairs realm, the President’s powers have traditionally been expansive. Domestically, the President is bound by numerous checks and balances—not in the least the fact that the President does not control Congress and depends on it to pass legislation. By contrast, in the foreign affairs realm the President has more leeway to act independently from the other branches of government, be it by making executive agreements that do not require Congressional approval, ²⁴ by making use of his own constitutional powers as Commander-in-Chief, or by making use of the unilateral trade powers delegated to him by Congress. ²⁵

In the trade policy sphere broadly considered, Congress has delegated important responsibilities to the President. ²⁶ As is well known, the United States Trade Representative (USTR) negotiates trade agreements on behalf of the President. ²⁷ In the narrower context of unilateral trade policy, Congress delegated important powers to both the President and the USTR. For example, under Section 232 of the Trade Expansion Act of 1962, the President can undertake “action . . . to adjust the imports of [an] article and its derivatives so that such imports will not threaten to impair the national security.” ²⁸ This includes the raising of import tariffs or the introduction of import quotas. ²⁹ “National security” is left undefined by Congress, leaving it to the President to give meaning to the concept. And perhaps unsurprisingly, the President has interpreted the concept expansively as essentially encompassing the entire national economy. ³⁰ In other words: Congress empowered the President to take trade measures against imports that threaten to impair U.S. national security, and it left the assessment of when measures do so to the President’s own discretion. Similar points can be made about other unilateral trade powers, such as the President’s authority to take measures—including import bans—under the International Emergency Economic Powers Act (IEEPA), ³¹ or the USTR’s power to take measures under Section 301 of the Trade Act of 1974. ³² For each of these instruments, Congress has delegated a power to the President or an agency or cabinet department that the President controls, such as the USTR. This power is subject to relatively undemanding substantive checks in the form of broadly framed, open-ended conditions. ³³

Crucially, the executive has the final say on whether those conditions are met in a specific case. Relevant in this regard is that federal courts in the U.S. have in the recent past been deferential towards the executive in their own assessment of whether the executive had complied with the conditions. ³⁴ And further upstream, U.S. courts have not questioned the constitutionality of the broad delegations of powers. ³⁵ In a similar vein, while Section 232 measures are reviewable in light of the General Agreement on Tariffs and Trade’s (GATT) ³⁶ security exception following the WTO panel report in Russia—Measures Concerning Traffic in Transit, ³⁷ the paralysis of the WTO Appellate Body undermines the effectiveness of the WTO dispute settlement mechanism and thus of the international legal disciplines imposed on the discretion of the President in this area. ³⁸ In Russia—Measures Concerning Traffic in Transit, a WTO panel held that WTO panels have jurisdiction to review aspects of a WTO Member's invocation of the GATT’s security exception. ³⁹ Following this report, a dispute on the scope of the GATT security exception could be brought before a WTO panel. Ever since the Appellate Body became dysfunctional in December 2019, however, the panel could issue a report but each of the parties can appeal “into the void.” As a result, the initial panel report will not be adopted by the Dispute Settlement Body and will thus not become binding. ⁴⁰

II. Congressional Checks on Presidential Power

When delegating powers to the President or any other part of the executive branch, Congress tries to retain a degree of control over how the executive exercises those powers. Historically, the legislative veto was an important instrument for Congress to do so. Under the legislative veto, depending on how the veto was set up, either both houses of Congress or a single house could veto a Presidential order taken on the basis of such delegated authority. It could do so by means of, respectively, a concurrent or a simple resolution. Each chamber could act by an ordinary majority. In contrast to the ordinary legislative process, no presidential signature was required to veto an executive measure. Commentators have argued that this was an effective tool for Congress to exercise oversight over the executive. ⁴¹ The legislative veto was effective particularly because, in the field of trade policy, many of the administrative safeguards of the Administrative Procedure Act (APA)—for example, the requirement that the public has the opportunity to comment—do not apply. ⁴² In this area, some delegations of authority to the President—notably, IEEPA sanctions and Section 232 sanctions on oil and oil-related products—are subject to the legislative veto. ⁴³ However, in the 1983 in the case of INS v. Chadha ⁴⁴ the Supreme Court declared the one-house and the two-house legislative veto without presentment to the President unconstitutional. ⁴⁵ As a consequence of this ruling, it was no longer possible for a single house of Congress to terminate unilateral trade measures taken by the President on the basis of, for example, Section 232 by means of an ordinary majority. ⁴⁶ Instead, a two-thirds majority in both houses was required to insulate the resolution against a potential presidential veto. As a result, to the regret of some ⁴⁷ and the delight of others, ⁴⁸ the legislative veto became ineffective as a means to keep a check on the President’s exercise of delegated trade powers. ⁴⁹

The fate of the legislative veto is symptomatic of the wider state of executive accountability in unilateral trade policy in the United States. Congress has other instruments at its disposal to exercise oversight over the President, but it faces collective action problems in deploying those instruments effectively. ⁵⁰ To be clear, existing trade delegations are subject to some procedural checks. Sections 232 and 301 impose specific executive decision-making procedures, ⁵¹ and Section 232 imposes time limits on Presidential action ⁵² —limits which federal courts have enforced in response to President Trump’s failure to comply with them when imposing tariffs on steel imports. ⁵³ Section 232 also still contains a legislative veto provision, although the statute now expressly provides that Congress can only exercise the veto by means of a joint resolution, which must be presented to the President. ⁵⁴ Similarly, IEEPA requires the President to first declare a national emergency before taking action; the President must consult with Congress for as long as the national emergency continues; ⁵⁵ the President must submit a report to Congress whenever he or she adopts a measure in the framework of IEEPA; ⁵⁶ and the President must periodically report to Congress about the actions the President has undertaken and changes that have occurred since the previous report. ⁵⁷ Similar consultation requirements exist for Section 232 measures, ⁵⁸ but not for Section 301 measures. ⁵⁹ However, more far-reaching mechanisms such as sunset provisions or report-and-wait requirements have not been used in the unilateral trade policy context. ⁶⁰ Likewise, Congress has not tried to frame existing delegations more narrowly. ⁶¹ Bills have been introduced to grant Congress an ex ante veto power over individual measures as opposed to an ex post power to terminate measures as currently exists under the legislative veto, but such bills have not been adopted. ⁶²

The scarcity of alternative procedural checks coupled with the demise of the pre-Chadha legislative veto makes it difficult for Congress to meaningfully impact Presidential unilateral trade policy. It is difficult for Congress, a collective body, to act—even by ordinary majority. ⁶³ To raise the bar further and to require a two-thirds majority to oppose Presidential actions means Congress can act only if there is a quasi-complete bipartisan consensus that the President’s actions were ill-advised. In a two-party system in which the President is a member of one of the two parties represented in Congress, it is difficult to assemble that broad of a coalition. At least when it comes to unilateral trade capabilities, then, procedural checks on an Imperial Presidency endowed with significant trade powers remain limited. While this may have been appropriate in an era in which Presidents felt constrained not only by legal limits but also by non-legal conventions and informal codes of conduct, ⁶⁴ it is worth considering whether the Trump experience does not call for legal guardrails to be introduced to ensure executive accountability moving forward. ⁶⁵

C. Unilateral Trade Measures in the European Union: A Fragmented, Yet Increasingly Powerful, Executive

I. A Fragmented Executive to Engage in Unilateral Trade Policy

How does the European Union compare to the United States when it comes to accountability in the realm of unilateral trade policy? It is important to acknowledge at the outset how very different the EU is structured compared to the U.S. Contrary to the constitutional arrangements in the U.S., the EU Treaty framers—the Member States—did not establish a directly-elected executive with important constitutional powers of its own, including the power to deploy the military. Instead, as described by Deirdre Curtin, executive power in the EU is more dispersed and more fragmented. ⁶⁶ The European Commission exercises important administrative executive powers, ⁶⁷ but so does the Council. ⁶⁸ This is the case in particular in the framework of the CFSP, where the Council can impose restrictive measures. As far as the political aspects of executive power are concerned—that is, the power to exercise political leadership and to determine the course of the “ship of state”—the European Council is the main player. ⁶⁹ Similarly, the Commission is also trying to steer a course between loyalty towards the European Council—of which the Commission president is a member—and steering its own independent course as the EU institution charged with the responsibility of defending the general EU interest. ⁷⁰ That being said, however, executive dominance is very much a feature of the EU decision-making landscape, and the question of how to ensure that executive power is held accountable is a key challenge for the EU moving forward. ⁷¹ This certainly holds true for unilateral trade policy, where EU executive bodies—in particular, the Council and the Commission—exercise meaningful discretionary authority and are likely to do so to an even greater extent as the EU legislature upgrades the EU’s unilateral trade instruments toolbox. ⁷²

How then, is the accountability of the executive ensured in this area of EU policy making? A first observation: The executive power to deploy unilateral trade instruments in the EU is fragmented. Two decision-making set-ups live side by side. On the one hand, the Council can adopt restrictive measures with little Commission involvement. As a general rule, it must do so by unanimity. On the other hand, the Commission plays a central role in the deployment of the CCP unilateral trade instruments listed in the introduction to this Article. Most of these instruments are adopted by means of implementing acts. Bar exceptional cases in which the EU legislature empowers the Council rather than the Commission to enact implementing acts, ⁷³ such acts are adopted on the basis of the so-called “Comitology” system. ⁷⁴ In this system, the Commission proposes measures, and a committee consisting of Member States—technically not a Council committee—has the opportunity to oppose measures the Commission proposes. However, it can only do so by means of a qualified majority of its members. ⁷⁵ This is a high threshold, which makes it hard for Member States to stop the Commission in its tracks. Crucially, in neither decision-making set-up does the European Parliament play a meaningful role. The Parliament only needs to be informed of the general developments in the CFSP, and in the context of CCP-implementing acts, the Parliament can let the Commission know that it believes a proposed measure to be ultra vires the Commission’s powers. ⁷⁶ In such a case, the Commission has to reconsider the measure, but it cannot be stopped from adopting the measure the second time around. ⁷⁷

CCP unilateral trade instruments have traditionally been subject to more demanding substantive conditions compared to their US counterparts. Classic trade defense measures such as anti-dumping and anti-subsidy measures need to comply with the requirements set out in the WTO agreements on the same topic as implemented in the EU’s legal order. ⁷⁸ Many of the proposed instruments are—or will be, if they get adopted—subject to fairly stringent substantive conditions as well. For example, under the Foreign Subsidies Regulation, redressive measures can be adopted only in the presence of foreign subsidies that “distort the internal market.” ⁷⁹ Foreign subsidies distort the internal market “where a foreign subsidy is liable to improve the competitive position of the undertaking concerned in the internal market and where, in doing so, [it] actually or potentially negatively affects competition on the internal market.” ⁸⁰ To make that determination, the Commission proposal lists a number of criteria that the Commission may take into account, including:

1. a) The amount of the subsidy;

2. b) The nature of the subsidy;

3. c) The situation of the undertaking and the markets concerned;

4. d) The level and evolution of economic activity of the undertaking concerned on the internal market;

5. e) The purpose and conditions attached to the foreign subsidy as well as its use on the internal market. ⁸¹

In similar fashion, under the International Procurement Instrument, measures can be adopted in the presence of “third-country measures or practices” that “result in a serious and recurrent impairment of access of Union goods, services and/or economic operators to the public procurement or concession markets.” ⁸² The substantive conditions in both of these instruments reflect the defensive finalities of the CCP unilateral trade instruments. As the Commission emphasizes, they aim to better equip the EU to withstand unfair trade practices by third countries; not to empower the EU to actively pursue geo-political objectives other than trade liberalization. In other words, they are presented as instruments of deterrence rather than economic statecraft. ⁸³

II. Procedural Checks on the European Commission

That said, the abovementioned substantive conditions will only have teeth—and abuse will only be avoided—if those conditions are embedded in an institutional framework that provides for meaningful procedural checks on the decision-maker. At this level, the differences between the European Union and the United States are less pronounced than they appear at first glance. In both systems the power to propose and the power to adopt often rest in the hands of a single institution: The President in the U.S., the Commission in the EU. Certainly, anti-dumping and anti-subsidy investigations are typically launched following a complaint brought by industry. The same will hold true for the International Procurement Instrument. ⁸⁴ However, both of these instruments also empower the Commission to launch investigations ex officio. ⁸⁵ By contrast, under the Foreign Subsidies Regulation, there will not be a complaint mechanism, and the Commission will always start investigations on its own initiative. ⁸⁶ The same will hold true for the Anti-Coercion Instrument if it is adopted in its current form. ⁸⁷ For all of these instruments, the Commission can thus propose measures and also adopt them. Coupled with the reverse qualified majority voting rule in the Comitology committee, this makes for a particularly weak procedural check on the Commission.

Furthermore, if the Anti-Coercion Instrument is adopted in its current form, the Commission will find itself in the peculiar position of being able to circumvent the unanimity requirement that governs the adoption of restrictive measures in the CFSP. Under the Anti-Coercion Instrument, the Commission would be able to propose and adopt measures “where a third country seeks, through measures affecting trade or investment, to coerce the Union or a Member State into adopting or refraining from adopting a particular act.” ⁸⁸ This would be the case when that third country “interferes in the legitimate sovereign choices of the Union or a Member State by seeking to prevent or obtain the cessation, modification or adoption of a particular act by the Union or a Member State by applying or threatening to apply measures affecting trade or investment.” ⁸⁹ Here, as well, the Commission proposal lists a number of indicators to guide that assessment. These indicators include:

1. a) The intensity, severity, frequency, duration, breadth and magnitude of the third country’s measure and the pressure arising from it;

2. b) Whether the third country is engaging in a pattern of interference seeking to obtain from the Union or from Member States or other countries particular acts;

3. c) The extent to which the third-country measure encroaches upon an area of the Union’s or Member States’ sovereignty;

4. d) Whether the third country is acting based on a legitimate concern that is internationally recognized;

5. e) Whether and in what manner the third country, before the imposition of its measures, has made serious attempts, in good faith, to settle the matter by way of international coordination or adjudication, either bilaterally or within an international forum. ⁹⁰

Despite these efforts at further determining what constitutes “economic coercion,” the notion inevitably remains nebulous. The concept of “coercion” is known in international law, but it remains unclear whether coercion can exist outside of the context of the use of military force. ⁹¹ The concept as defined in the proposed Anti-Coercion Regulation also contains a subjective element: For there to be economic coercion, there must be an intention on behalf of the third country to practice coercion. This subjective element raises evidentiary hurdles as it will be difficult—if not impossible—for the Commission to get watertight evidence of what the third country’s intentions really are. ⁹² At the same time, there may be pressure on the Commission to act, especially in the face of gridlock within the Council making it impossible for the EU to adopt restrictive measures.

Indeed, absent procedural checks, and despite the Commission’s insistence that the tool will be used first and foremost for purposes of deterrence, ⁹³ the Commission may in practice be quick to conclude that economic coercion is indeed present, as drawing this conclusion would enable the EU to impose measures very similar to restrictive measures, but without the need for unanimity among the Member States. The situation would be somewhat different if the EU legislature were to empower the Council, rather than the Commission, to adopt the necessary implementing acts to retaliate against coercive practices by third countries. In this scenario, a qualified majority of Member States would have to approve the measure before it gets enacted. ⁹⁴ This is a higher threshold compared to the negative qualified majority requirement in the Comitology process. It is, however, still a lower threshold compared to CFSP restrictive measures, which are adopted by unanimity in the Council. The risk that unanimity in CFSP decision-making will be circumvented through a broad interpretation of the substantive conditions to impose anti-coercion measures thus exists also, albeit to a lesser extent, in the scenario that the Council rather than the Commission has the final say on such measures.

Such “slippery slope” dynamics have played out in the United States where Congress has found it difficult to operate as a check on Presidential power. One example outside of the unilateral trade policy context is the practice of concluding executive agreements. The U.S. Constitution prescribes that the Senate is to give its “advice and consent” to treaties by means of two-thirds of its members. Yet since the late 1930s, over 90% of international agreements concluded by the U.S. are made by the President, often in tandem with Congressional legislation adopted by ordinary majorities. ⁹⁵ This practice is premised on a distinction maintained by the President between “treaties” in the meaning of Article II of the U.S. Constitution and other international agreements. Yet there is reason to question the distinction, as nowhere in the text of the Constitution is any mention made of executive agreements as an alternative way of concluding international agreements. ⁹⁶ A similar phenomenon has occurred in the context of the use of military force. In 2011, President Obama insisted that military operations such as imposing a no-fly zone over Libya did not constitute “war” in the meaning of the War Powers Resolution, and that therefore no Congressional approval of the operation was needed. ⁹⁷ Similarly, in the unilateral trade policy context, “national security” has been interpreted so broadly that it encompasses the entire national economy, ⁹⁸ and “national emergencies”—the presence of which are required for the President to be able to adopt IEPPA sanctions—have become so common that in July 2020 there were no fewer than 33 national emergencies ongoing. ⁹⁹

Could similar erosions of the substantive conditions governing the use of unilateral trade instruments materialize in the European Union? It is important to acknowledge that similar risks do exist in the EU context as well. For various reasons, EU institutions, including the Commission, may feel compelled to stretch the text of the secondary legislation empowering it to act. Examples of such efforts are not hard to find. For example, in response to President Trump’s unilateral Section 232 import tariffs, the Commission imposed safeguard measures. ¹⁰⁰ The legality of such measures under WTO law was questionable, and, in April 2022, a WTO panel indeed concluded that the EU had breached its obligations under the WTO’s Safeguards Agreement and the GATT by unilaterally imposing safeguard measures rather than starting WTO proceedings in reply to the U.S. Section 232 tariffs. ¹⁰¹

Consider a second example: In 2020, the Commission imposed countervailing duties on imports of glass fiber fabrics from Egypt. Duties were imposed following an investigation that established that a Chinese state-owned company had set up shop in Egypt, from where it subsequently exported subsidized fabrics to EU markets. ¹⁰² The Commission held that financial contributions the Chinese company had received from the Chinese government could legally be attributed to Egypt. ¹⁰³ In so doing, the Commission bypassed the thorny question of whether the SCM Agreement covers subsidies granted by a WTO member to a company outside of its own territory. A privileged observer argued in tempore non suspecto that the drafters of the Agreement on Subsidies and Countervailing Measures (the SCM Agreement) had deliberately excluded so-called ‘transnational subsidies’ from the scope of the agreement as including them would bring the U.S. in trouble, which at the time was the world’s largest donor of overseas development assistance. ¹⁰⁴ Yet, whether or not WTO law allows conduct of one member to be attributed to another—here, China and Egypt respectively—is perhaps just as controversial as the issue of the SCM Agreement’s territorial scope. ¹⁰⁵ By imposing countervailing duties, the Commission may very well have misapplied the EU regulation. Yet, there is little that other EU institutions could do to prevent the measures from being adopted as these measures too are subject to the abovementioned reverse qualified majority requirement.

III. Why Judicial Review is Insufficient to Ensure Accountability

To be clear, the Court of Justice as well as international bodies, such as the WTO dispute settlement mechanism, have played and continue to play an important role in holding the Commission to account. At the time of writing this Article in the Fall of 2022, a challenge against the Commission’s decision in the abovementioned trade defense case involving glass fiber fabrics was pending before the General Court of the EU. ¹⁰⁶ Additionally, in other recent trade defense cases as well, the General Court has not shied away from taking a deep dive into the Commission’s calculations and from ultimately concluding that those calculations were not in conformity with the applicable legislation. ¹⁰⁷ If the trade defense context is of any guidance, the General Court and the Court of Justice likely will not shy away from scrutinizing Commission decisions under the Foreign Subsidies Regulation, the International Procurement Instrument or the Anti-Coercion Instrument either. In so doing, both EU courts will ensure that the limits of the delegations of authority are complied with and the Commission does not act ultra vires.

However, judicial review may not be sufficient to ensure accountability. For starters, judicial review operates ex post. For this reason, judicial review may not provide effective relief for the affected industries. ¹⁰⁸ This is all the more the case because measures are presumed to be legal until the Court of Justice establishes they are not. ¹⁰⁹ Moreover, only rarely are duties suspended pending judicial review. As a consequence, affected businesses may very well have left the market by the time the illegality is established by the Court.

But there is a more fundamental point to be made. Despite the comparatively stringent substantive conditions attached to many of the unilateral trade instruments I look at in this Article, all of them endow—or are likely to endow—the Commission with a degree of discretion. This discretion can take many forms. It may consist of granting the Commission the power to open or not open an investigation. Or it may empower the Commission to assess whether imposing measures is in the Union’s general interest—the “Union interest” ¹¹⁰ —and to impose or not impose measures depending on the outcome of that assessment. A degree of discretion can also be found in the Commission’s fact-finding role. Which facts the Commission deems relevant and which facts it deems irrelevant is an assessment that the Commission is best placed to make, and, as Joanna Mendes has argued, “[t]he choice between alternative views on technical issues and the choice between the various significations that value (undetermined) concepts entail is also a discretionary choice . . . insofar as it will determine a certain course of action and entail normative consequences.” ¹¹¹ The Court of Justice recognizes this reality where it accepts that the Commission disposes of a margin of discretion that covers both factual and policy appraisals. ¹¹² As the General Court put it in Hansol, ¹¹³ an anti-dumping case:

In the sphere of the common commercial policy and, most particularly, in the realm of measures to protect trade, the EU institutions enjoy broad discretion because of the complexity of the economic, political and legal situations they have to examine. Since the application of Article 2(11) of the basic regulation requires an appraisal of complex economic situations, the judicial review of such an appraisal is limited to verifying whether the procedural rules have been complied with, whether the facts on which the contested choice is based have been accurately stated, and whether there has been a manifest error in the appraisal of those facts or a misuse of powers. ¹¹⁴

In complex administrative regimes such as the EU, there will always be discretion involved in executive decision-making. ¹¹⁵ The Court of Justice plays an important role in ensuring that the executive—here: The Commission—does not abuse that discretion and stays within the limits of the powers delegated to it and respects fundamental rights. However, within those legal limits, it is not for the Court of Justice to second-guess the Commission’s decisions. ¹¹⁶

At this juncture, separation of powers considerations call for alternative accountability mechanisms to kick in. If we take seriously the Treaty requirement that the “functioning of the Union shall be founded on representative democracy” ¹¹⁷ the Commission must be answerable for its actions not only to the Court of Justice, but also to those institutions that endow the EU decision-making process with input legitimacy. This will ensure that all discretionary actions undertaken by the EU are subject to democratic control. ¹¹⁸ This imperative becomes all the more salient as EU trade policy involves increasingly complex trade-offs between public and private interests—often with redistributive implications. ¹¹⁹ To name but one example, in the context of the proposed Anti-Coercion Instrument: Would it be in the Union’s interest to take measures to tackle economic coercion practiced by China against Lithuania if, by doing so, German exports to China will suffer? As discussed, today such input legitimacy is ensured primarily through the involvement of the Member States in the adoption of implementing acts. In the next section, I argue that Member State oversight is not sufficient to ensure proper accountability in the EU’s unilateral trade policy. European Parliament involvement is needed as well.

D. In Search of Democratic Accountability in EU Unilateral Trade Policy

In the previous Section I explained how risks of executive overreach and abuse of power are not unique to the United States Presidency. They arise also in the area of unilateral trade policy in the EU. Such risks will further increase in the future as the EU legislature endows the Commission with additional powers. To avoid abuses of power and to ensure that the Commission is answerable for its actions, checks and balances must be put in place.

It is interesting to observe, in this regard, that the Lisbon Treaty, which entered into force in 2009, did much to upgrade the role of the European Parliament in EU foreign relations. Since Lisbon, the Parliament has a veto power over most international agreements that the Council concludes. ¹²⁰ However, the Lisbon Treaty did little to democratize executive decision-making in the EU. ¹²¹ On the contrary, it reduced European Parliament involvement in the Comitology process, through which the Commission adopts many of the instruments discussed in this Article. Prior to Lisbon, Council and Parliament were sometimes able to oppose proposed implementing acts. ¹²² Following Lisbon, the Parliament lost that ability for implementing acts, while it acquired it for the newly established category of delegated acts. However, none of the instruments explored in this Article are or are likely to be adopted by a delegated act. Instead, the implementing act has been the instrument of choice of the EU legislature: all of the instruments looked at in this Article provide for decision-making by implementing act. As a consequence, the European Parliament plays no meaningful role in the process of adopting any of the unilateral trade measures that are or soon may be at the disposal of the Commission.

As mentioned, lack of parliamentary involvement in this process of executive decision-making is difficult to justify in an organization that is constitutionally committed to representative democracy. ¹²³ Others have argued persuasively that the distinction between implementing acts and delegated acts is unclear and, to an important extent, a political as opposed to a legal issue. ¹²⁴ Insofar as the distinction is devoid of legal meaning, the rationale for excluding the Parliament from the Comitology process according to which only the Member States should be involved as Member States hold primary responsibility for enforcing and implementing EU law, fails to persuade. ¹²⁵ Both implementing acts and delegated acts are cases of delegation of authority by the legislature to the executive, whereby the former establishes a principal-agent relation with the latter and which endows the latter with a degree of discretionary decision-making authority. ¹²⁶ In both scenarios, the legislature empowers the executive to take decisions which the legislature, as a collective body, is institutionally ill-equipped to take. However, as the U.S. experience shows, the legislature should think carefully not only about the substantive conditions it attaches to the exercise of delegated authority, but also about the procedural checks to which the exercise of that authority should be subject.

The foregoing leads me to suggest that, as the EU endows itself with important new unilateral trade instruments, the lack of European Parliament involvement in the process of deploying unilateral trade measures should be reconsidered. To be clear, the power to deploy instruments should not be fully shared between Commission, Council, and Parliament. Such an arrangement would lead to gridlock. Requiring both Council and Parliament to positively agree to each proposed Commission measure would also run counter to the rationale of delegated decision-making, which is precisely to make it easier to adopt individual decisions. From this perspective, suggestions to endow the Council instead of the Commission with the necessary implementing powers to adopt anti-coercion measures are ill-conceived. ¹²⁷ A more sensible approach consists of endowing the European Parliament with a power to oppose draft implementing acts in a similar fashion to the power already held by the Member States in the Comitology process. In that process, Member States can oppose a draft measure by means of a qualified majority. Similarly, the European Parliament should be able to oppose a draft measure by an absolute majority of its members. Under this arrangement, reminiscent of the procedure to adopt delegated acts, it is difficult, but not impossible, for the European Parliament to oppose a Commission proposal. The mere possibility that the Parliament may veto a draft measure will, in turn, incentivize the Commission to take into account the Parliament’s preferences and interests as it deploys unilateral trade instruments.

In the above set-up, Parliament and the Member States would have the ability to oppose measures before they have been enacted and thus before they enter into force. This avoids a scenario that plagues Congressional efforts to hold the President accountable in the United States: The legislative veto, mentioned earlier, is also subject to high decision-making thresholds. However, unless the Congressional statute delegating authority to the President contains a report-and-wait requirement—which none of the statutes looked at in this Article do—the President does not have to give Congress time to consider issuing a veto. The President can act immediately and thereby create facts on the ground that make recourse to the legislative veto useless. Not so in the EU, where the Commission has to submit draft measures to the responsible Comitology committee. In a reformed Comitology procedure, the Commission would have to submit the same proposal to the responsible European Parliament committee, give the committee time to consider the proposal, and adopt the proposal only after the committee did not oppose the proposal within a given timeframe. This could be achieved by amending the Comitology regulation, by concluding an interinstitutional agreement, or by providing for ad hoc mechanisms in the individual legislative acts empowering the Commission to adopt measures. ¹²⁸

By bringing the European Parliament back into the room, the democratic legitimacy of the Commission’s decisions increases. The Commission would answer to the Council, which represents one, indirect source of democratic legitimacy in EU decision-making, and it would answer to the European Parliament, which represents another, more direct source. ¹²⁹ In this setup, it would remain difficult for either Council or Parliament to veto a draft measure. Conversely, the Commission will in many—and most likely most instances be able to get measures adopted. In this sense, my earlier criticism that the strengthening of the EU’s unilateral trade toolbox also strengthens the Commission’s institutional position remains unaddressed. However, European Parliament involvement may lead to more attention by Members of the European Parliament to what the Commission is up to in its trade policy beyond those high-profile moments on which the Parliament is called on to vote—for example, on a free trade agreement with Canada, or on any of the initial legislative acts that empower the Commission to take the measures I looked at in this Article. If anything, European Parliament involvement in the Comitology process will bring new voices into the Commission’s decision-making process—voices that express concerns and preferences that may not be those of any of the Member State executives, but which are shared by broad segments of the European population.

E. Conclusion

In this Article, I looked at the allocation of executive power in unilateral trade policy in the United States and the European Union. I highlighted how growing executive power leads to accountability challenges. In particular, with reference to the U.S. experience, I showed how also in the European Union, a risk exists for abuse of power by the executive as it deploys an increasingly powerful set of unilateral trade instruments. The existence of such risk should be taken into consideration in the design of the governance framework of unilateral trade instruments. As trade policy becomes more political and indeed geopolitical, the stakes increase and the balancing of interests and values becomes more difficult. Such decision-making should be subject to meaningful procedural checks. At present, and in addition to judicial review by the Court of Justice, Member States can operate as a check on the Commission through their involvement in the Comitology process. In a polity that draws its legitimacy both from states and individual citizens, it is important, however, that the European Parliament also has a role to play in the process of adopting unilateral trade measures as it is that Parliament that represents EU citizens within the EU political process.