Posted by: Charles Seby, Connor Sonksen, David Schapira, Daniel Rubinov, Tyler Schwenke, and Danielle Solomon

NAFTA was the worst trade deal in the history – it’s like – the history of this country.”

Donald Trump on June 28, 2016


It would not be an overstatement to say that the first few months of the Trump administration included several diplomatic mishaps, including two jumbled iterations of a travel ban, a less-than-polite phone call with Australia, and awkwardly unsubstantiated claims that the UK spied on the President himself. But something the public hasn’t heard much about since the campaign trail is the President’s position on the North American Free Trade Agreement (“NAFTA”). During the election campaign, President Trump called NAFTA the “worst trade deal” ever signed by the United States. Trump’s campaign suggested everything from renegotiating the terms of the agreement to abandoning the deal altogether. As of this writing, his administration has taken no major steps toward fulfilling this campaign promise. But the question remains whether President Trump even has the power to unilaterally fulfill this promise. The simple answer: no. The longer answer: it’s complicated.

Let’s begin with expanding on what precisely NAFTA is and how it differs from the typical implementation of a treaty. NAFTA is a trade agreement between Canada, Mexico, and the United States. The agreement eliminated all prior tariffs between the parties, allowing for the free movement of goods between the countries. What sets NAFTA apart from many other treaties is the manner by which it was passed into law. Under Article II, Section 2 of the Constitution, a standard treaty requires for a two-thirds approval by the Senate after it is made by the executive. Once the treaty is approved (assuming it is non-self-executing, meaning that the treaty itself does not outline the exact codifications of laws that will implement the underlying treaty’s promises), Congress must put laws in place that uphold the conditions of the treaty. This can be a lengthy process. As a practical alternative, NAFTA was passed as a congressional-executive agreement (“CEA”) through what is referred to as “fast track authority.” Fast track authority allows for the President to negotiate trade deals and present them to Congress as laws. The executive branch employed this “fast track authority” to negotiate a number of trade agreements including: the Tokyo Round Agreements in 1979, the 1985 US-Israel Free Trade Agreement, the 1988 US-Canada Free Trade Agreement, and the 1994 Uruguay Round Agreements. This fast track process removes the need for a two-thirds Senate approval and reduces the requirement of congressional approval to a simple majority of each house of Congress. Additionally, no further laws need to be put into place following the passage of a trade deal under fast track authority.

NAFTA and its Statutory Codification

Picture4Because NAFTA is a CEA, and not a self-executing treaty, it required congressional action in order to put the agreement’s terms into force. The laws passed by Congress related to NAFTA are codified in Title 19, Chapter 21 of the United States Code. While not specifically addressing the procedure and protocol for withdrawal from NAFTA, Congress passed specific procedures for entry into the agreement and enforcement of its tariff provisions, which speak to the President’s ability to unilaterally withdraw from the agreement.

Article 2203 of NAFTA provides that the agreement enters into force following an “exchange of written notification” from each party certifying the parties completed the required legal procedures within their nations. While the passage of NAFTA received congressional approval, this approval itself did not formally enter the US into the agreement. Rather, the congressional approval permitted the President to enter the agreement on behalf of the US. For example, 19 U.S.C. § 3311 authorizes the President to “exchange notes” to enter the agreement with Canada and Mexico once he determines that both Canada and Mexico have made the necessary legal changes within their own nations consistent with NAFTA’s terms. Pursuant to this authority, President H.W. Bush negotiated NAFTA, and President Clinton officially signed the agreement into law in 1993.

NAFTA Article 2205 states that “a Party may withdraw from this Agreement six months after it provides written notice of withdrawal to the other Parties.” In this context, “Party” refers to a nation in the agreement (Canada, Mexico, and the United States). As mentioned, Congress specifically delegates power to the President to be the representative of the United States as a “Party” in entering the agreement. However, Congress did not expressly extend this capacity to withdrawing from the agreement under Article 2205.

Further, in passing the statutes pertaining to NAFTA, Congress included two important provisions. The first is 19 U.S.C. § 3317(b)(4)(C), which articulates that for future trade negotiations, the President is obligated to submit to Congress a “written report that contains . . . legislative proposals to ensure adequate consultation with the Congress and the private sector during the negotiations, advance Congressional approval of the negotiations recommended by the President, and Congressional approval of any trade agreement entered into by the President as a result of the negotiations.” (emphasis added). The second is 19 U.S.C. § 3317(b)(1)(D), which states: “The Congress makes the following findings: . . . The system of trade agreements can and should be structured to be consistent with, and complementary to, existing international obligations of the United States and ongoing multilateral efforts to open markets.” (emphasis added). Both of these statutes suggests Congress intended 1) to maintain a level of discretion over future executive activities regarding NAFTA and 2) to encourage a larger system of open markets (which, as the name suggests, are often associated with free trade agreements).

Because much of the publicity surrounding NAFTA concerns the elimination of tariffs and other trade barriers between the three parties of the agreement, many think that the President altering tariffs would be synonymous with an exit from NAFTA. However, those two actions are not identical. First, under 19 U.S.C. § 3314(a)(1), the President is given the discretion to make proclamations that enforce NAFTA. He likewise has latitude in deciding tariff measures. Congress, certainly, could have retained more control to maintain the agreement by codifying specific tariff provisions by statute. Instead, Congress delegated some of its authority to the President. Under 19 U.S.C. § 3331(a)(1)(A), “the President may proclaim” modifications to essentially any of the NAFTA duties as he deems “necessary or appropriate” to carry out the duty-free provisions of NAFTA. In President Trump’s case, his proposed modifications are to increase tariffs, which would not fall under this authority. However, under 19 U.S.C. § 3331(b), the President may also proclaim modifications to any duty to maintain a “general level of reciprocal and mutually advantageous concessions with respect to Canada or Mexico. . . .” Under this provision, the President could argue that, because American industry arguably suffered under the terms of NAFTA, increasing tariffs may be necessary.

Accordingly, if President Trump unilaterally acts under the power delegated to him by 19 U.S.C. § 3331(b), Congress does not appear to have left itself any recourse to prevent the President from unilaterally unraveling certain tariff provisions of the agreement. However, that is not equivalent to Congress delegating to the President the authority to withdraw from NAFTA entirely. To understand further how far executive power extends over the whole existence of a CEA such as NAFTA, one must analyze the source of modern fast track authority–that is, the Trade Act of 1974.


What does the Trade Act of 1974 say?

The Constitution gives express power to Congress in Article I, Section 8 “to lay and collect taxes, duties, imposts and excises,” to “regulate commerce with Foreign Nations,” and to make laws which “shall be necessary and proper” to effectuate the powers of the United States. In the Trade Act of 1974 (particularly §§ 101 and 124), Congress delegated to the President some authority to “proclaim” reductions, increases, eliminations or impositions of duties that the President determined to be required or appropriate in pursuance of provisions of trade agreements. The Act did not expressly delegate to the President a unilateral authority to withdraw the nation from the trade agreements themselves.

Between 1934 and the passage of the Trade Act of 1974, Congress delegated some of its authority to negotiate tariffs, and then later allowed for the President to implement the agreed-upon tariff via a “proclamation.” As economies developed and trade negotiations became more complex (with measures increasingly more complicated than mere tariffs), Congress wanted to ensure the President had the capacity to negotiate trade agreements as a unified voice consistent with his dual role as the head of both state and government. Consequently, Congress passed the Trade Act of 1974, with the intent to provide the President with streamlined Congressional approval (called “fast track”) in exchange for enhanced Congressional oversight. Section 151 of the Act contains many provisions regarding subsequent implementation acts that reflect Congressional intent to retain power delegated to them under the Constitution to manage international trade. Even through subsequent legislation, Congress has not expanded the President’s authority to modify tariff rates outside the confines of trade agreements or through the Trade Act’s fast track authority.

Section 101, reiterated in § 124, of the Trade Act lays the groundwork for § 125 and § 301 executive authority. Section 101(a) states the President, if he finds any duty or import restriction “unduly burdening and restricting the foreign trade of the United States,” may negotiate trade agreements and subsequently “proclaim” modifications or additions that he determines appropriate to carry out the agreement, with some limitations. Section 125(a) requires every trade agreement implemented under the Act to have a withdrawal provision, presumably requiring the President to take this into account when negotiating the agreement. Section 125(b) also allows the President to terminate any “proclamation” made under § 101(a)(2), giving him the authority to modify or terminate proclamations of changes to duties or tariffs. Nowhere in the Act is “proclamation” used to refer to subsequent acts of legislation.

Picture2Section 301 provides the President with greater authority (in cases where foreign governments’ practices are functioning contrary to the United States’ interest) than what is allowed under his ability to make proclamations, but does not give him any express authority to unilaterally withdraw from an entire trade agreement. Section 301 provides broad authority to retaliate against or obtain removal of foreign government behaviors that are “inconsistent with the provisions of, or otherwise denies benefits to the United States under, any trade agreement” or otherwise unjustifiable, unreasonable, or discriminatory. This section allows the President to take actions that “are within the power of the President with respect to trade in any goods or services, or with respect to any other area of pertinent relations with the foreign country,” as articulated in § 301(a)(1)(B)(2). The President’s range of discretion is broad, and authorizes him to take “feasible action” (§§ 102, 201) including, but not limited to, those allowed pursuant to proclamations (§ 101), suspension or elimination of trade agreement concessions (§ 301), and even restricting service sector access authorization (§ 2411(c)-(e) of the Trade and Tariff Act of 1984). In addition, no provision grants a right to judicial review of presidential action under section 301. Thus, theoretically, the President could implement sanctions on a country such as Mexico that sharply depart from the original tariff terms of NAFTA agreement in an effort to reach a fair level of trading reciprocity.

In § 301(c), the Act enumerates what the Office of the United States Trade Representative (USTR) may do if it finds an unfair practice, either on its own or through a petition from private parties. It includes the ability to suspend, withdraw, or prevent the application of benefits of trade agreement concessions and impose duties or other import restrictions for as long as he determines. This section comports with the purpose of the Act as intended by Congress “to provide adequate procedures to safeguard American industry and labor against unfair or injurious import competition, and to assist industries, firm, workers, and communities to adjust to changes in international trade flows.” In § 102, Congress acknowledged the threat of unfair trade practices toward the United States and urged the President to act when necessary. Although §§ 101, 102, 125, and 301 give the President broad authority to take unilateral actions to preserve the interests of the United States under trade agreements, nothing in the Act gives the President authority to, in one swoop, unilaterally withdraw from the entirety of a US commitment enacted by Congress under any implementation act pursuant to § 151.

How would the judiciary evaluate whether Trump could unilaterally withdraw from NAFTA?

Certainly, there would remain a separation of powers question if Trump were to attempt to exit NAFTA without Congressional approval. Some might argue that the issue of unilateral executive withdrawal from a trade agreement may simply be a political question and thus nonjusticiable. Given such concerns, could the Supreme Court hear the case to enjoin a unilateral withdrawal? If so, how would the Court likely rule? The Constitution hints that the executive might lack such unilateral authority under the Foreign Commerce Clause, but this text alone does not refer explicitly to power allocations within CEAs. Although the precedents are complex in this area, several cases from various postwar administrations provide an outline for the instances in which the President’s unilateral power falls short. Even in scenarios of international implication, relevant case law demonstrates that Trump could not unilaterally withdraw the United States from NAFTA and, were he to attempt, the Supreme Court could and would enjoin him from doing so.

Many might look to the case Goldwater v. Carter (1979) and quickly conclude that the Supreme Court would simply dismiss a suit that sought to prevent the President from withdrawing unilaterally from a congressional-executive trade agreement. In that case, the Court refused to hear Senator Barry Goldwater’s suit challenging President Carter’s unilateral withdrawal from the Sino-American Mutual Defense Treaty. However, in the Trump-NAFTA hypothetical, three points must be considered. First, a business (economically affected by Trump withdrawing from NAFTA) and not a collection of senators most likely would be the party suing the Trump administration (thus eliminating concerns of standing, as actual economic injury would provide grounds for suit). Second, the issue of a unilateral withdrawal being a nonjusticiable political question would likely command further analysis because such a situation would warp an express congressional delegation of power and create an executive interference in an arena not exclusive to the executive. Finally, the ultimate decision of the Court would largely depend on the Court’s jurisprudential conception of the separation of federal powers. Let’s examine those conceptions below.

The Formalist Perspective

Under a formalist interpretation of separation of federal powers, one might be tempted to argue that President Trump could unilaterally withdraw from NAFTA. Looking at the conception of the separation of powers under INS v. Chadha (1983), the branches of federal government are found to be markedly distinct and isolated in their roles and responsibilities. In that case, the court found that Congress could not exercise a one-house congressional veto because such action did not comport with the strict outlines of Congress’s constitutional legislative power. As such, Congress had no power in that case to override the will of the executive. Encapsulating such a delineated perspective on the separation of federal powers, Justice Scalia’s hyperformalist dissent in Morrison v. Olson (1988) provided a lone outcry (on even an otherwise formalist Court) that the independent counsel portions of the Ethics in Government Act of 1978 were unconstitutional because they invaded the vested executive power. There, the Act had provided Congress with the ability to appoint independent counsel to investigate executive officials for violations of federal law. While the rest of the Court found the provisions within the constitutional reaches of Congress, Scalia argued that the executive branch’s vested power under Article II, Section 1 of the Constitution is all (not some) of that executive power. Unlike the majority, Scalia reasoned that Congress had zero authority in the investigation of such executive officials and thus could not contradict the President in this area. Under such a formalist interpretation, the separation of powers draws unmistakable lines where one branch’s power begins and another branch’s power ends. Would such an outlook be determinative in the hypothetical of President Trump unilaterally withdrawing from NAFTA?

Not likely. While a Court employing this perspective might be tempted to reason that President Trump’s unilateral withdrawal could be within the executive power so long as it was a power completely outside of legislative interference, the Court would be seriously stretching to overcome such obstacles as Congress’s delegated authority per the Foreign Commerce Clause. That being said, under such an understanding of the separation of powers, the Court may be also tempted to find that Trump unilaterally withdrawing from NAFTA would fit into the sixth category of nonjusticiable political questions in Baker v. Carr (1962). This category includes suits which involve the “potentiality of embarrassment from multifarious pronouncements by various departments on one question.” If Trump submitted a written notice of withdrawal from NAFTA per Article 2205 and litigation to enjoin the withdrawal persisted past the six month point at which the United States would technically exit, a Court might find it simpler to avoid declaring that Trump’s action was improper or otherwise reversible. Formalist interpretations of categorical power distributions aside, such a Court would undoubtedly avoid the embarrassing possibility of adjudicating a decision that informed the international community that Trump had acted outside of his own executive authority. However, the legal implications of such a renunciation of legal decision-making would be disastrous for the Court (if not even more embarrassing). Even a Court that touted formalist judicial restraint would not likely ignore such a blatant executive violation of congressionally delegated power and such an executive overhaul in an arena (foreign commerce) nonexclusive to the executive.  

The Realist Perspective

Youngstown Sheet & Tube Co. v. Sawyer (1952) is one of the dominant authorities regarding the boundaries of executive authority. Following President Truman’s seizure of the steel mills across the United States out of fear of a labor shut down, the Court held that the President lacked such seizing power. The Court explained that any power exercised by the executive must originate from either an act of Congress or from the Constitution itself. Finding no statute even suggested the President had the power to seize steel mills, the opinion then emphasized that Congress had declined to delegate seizure powers to the executive in passage of the Taft-Hartley Act of 1947. Statutory permission and past legislative intent lacking, the Court held the President had overreached the boundaries of his executive authority. In his concurring opinion, Justice Jackson presented three situational categories that outlined potentialities of presidential action. First, the President is at the height of his powers when acting pursuant to congressional authorization. Here, the only way for the President not to succeed in action would be if the executive and the legislative lacked authority in this area entirely. Second, the President is limited to his already-vested executive powers when Congress has been silent on a potential executive action. In this category, the “imperatives of the events and contemporary imponderables” will dictate the test of power. Third, the President is at the lowest point of his powers when Congress has expressly or implicitly established rules stating that the executive may not pursue a particular action. Here, the President may only succeed when such a domain is beyond congressional authority entirely. This third category is where Jackson placed President Truman’s executive order seizing the mills. Given that Congress is not without the authority to dictate matters of interstate commerce, Justice Jackson supported the majority opinion in declaring unconstitutional such an expansion of executive power.

Conversely, the Court held in United States v. Curtiss-Wright Corp (1936) that the President —pursuant to a congressional Joint Resolution expressly authorizing the executive to ban particular sales of firearms—had the authority to issue an executive order banning the sale of guns to Bolivia. The Court stressed the executive’s role as the representative of the nation in matters of foreign affairs. It held that although the Senate is necessary to authorize the ratification of treaties, the President alone interacts with other sovereign entities in matters of foreign diplomacy. Supported by an unequivocal statute granting the executive the power to ban the sale of guns, the President was found by the Court to be within his constitutional authority to issue such an order.

The Court further refined the constitutional limits of executive authority in Dames & Moore v. Regan (1981). Acting pursuant to a federal act that suspended the conveyance of all property of Iran in the wake of the Iranian hostage crisis, President Carter issued an executive order which was later continued by President Reagan that suspended any American prejudgment attachments toward and ended litigation involving Iranian property. Invoking Justice Jackson’s categories from Youngstown, Court held the President was within his constitutional power because the executive order was issued pursuant to congressional legislation (thus placing the President in the first category, at the height of possible executive power). The Court reinforced, however, that the express congressional authorization was “crucial” to their decision to uphold the executive order, and highlighted that nowhere had Congress expressed dissatisfaction of the President issuing such an order.

Recalling the congressional delegation of power to the executive branch per the 1974 Trade Act in CEA affairs, any unilateral attempt by President Trump to withdraw from NAFTA in its entirety would likely not fall into the first or second of Justice Jackson’s categories. Although Congress did delegate the negotiation responsibilities to modify or terminate the proclamation of tariff-provisions within NAFTA per § 125 of the Trade Act of 1974, Congress never expressly or even implicitly delegated to the executive any authority to terminate a CEA as a whole. Similarly, Congress has not been historically silent on whether it must have final discretion in the realm of international commerce, given certain unmistakable provisions within codified NAFTA statute—think back to 19 U.S.C. § 3317(b)(1)(D) and § 3317(b)(4)(C). Thus, Trump attempting to unilaterally pull out of NAFTA would likely fall into the third of Jackson’s categories. To survive as a valid exercise of power within the third category, President Trump would need to be operating in a scope beyond congressional authority entirely–a tough sell, considering executive power within the CEA process requires congressional participation and approval given that Congress delegated the President the authority to make the agreement.

Under a realist perspective, the possibility of President Trump unilaterally withdrawing from NAFTA would be distinguishable from Curtiss-Wright Corp. and Dames & Moore in that both those cases involved express congressional approval of the issuance of both executive orders. By contrast, Congress never approved of unilateral executive termination of an entire CEA in the Trade Act of 1974 or in the codified NAFTA statutes. Instead, like President Truman’s executive order seizing control of domestic steel mills, President Trump unilaterally withdrawing from NAFTA would be a Jacksonian category three operation that is outside of the boundaries of vested executive power. Thus, in such an instance, the Supreme Court operating under a realist jurisprudence would likely enjoin President Trump from unilaterally withdrawing from NAFTA.

So, can President Trump unilaterally pull out of NAFTA?

While such a process would foray into legally messy and politically contentious arenas, President Trump attempting to unilaterally withdraw from NAFTA would not likely be upheld by the Supreme Court. Although the Senate just recently confirmed Neil Gorsuch, the Court has yet to accept the devoutly formalist perspective articulated by Scalia’s dissent in Morrison. That is, the Court would not likely interpret the separation of powers to grant the executive branch a carte blanche to terminate an agreement that arose out of a congressional delegation permitting executive participation. The potential of Trump unilaterally withdrawing from NAFTA is not to be equated with Goldwater v. Carter, as in that case President Carter did not require delegation from Congress as an entity. Instead, he held such power as a vested constitutional ability. In a concurring opinion in that case, Justice Powell argued that the collection of senators who sued over the executive order lacked standing because the Senate as a body had taken no official action. Furthermore, Powell argued that the Court did not know whether there would be “an actual confrontation between the legislative and executive branches” regarding that issue. In the case of Trump withdrawing from NAFTA, a potential suit would likely be brought by either those who suffered actual economic harm (such as a company) or a Congressional body that did have “an actual confrontation” with the executive branch regarding NAFTA.

Picture1Even Justice Gorsuch would likely reject the contention that the Trump administration’s power to terminate NAFTA is inherent. It should be noted that, among multiple formalist personalities, Justice Scalia was the sole dissenter taking such a categorical position over legislative and executive powers in Morrison. As such, the introduction of Neil Gorsuch to the Court will not likely appear to be swing the cards in Trump’s favor. While Gorsuch’s precarious political appointment to the Court might tempt him to simply hold that a unilateral withdrawal from from a CEA would be a nonjusticiable political question in the sixth category of Baker (recalling the some of the Justices’ thoughts in the outcome of Goldwater v. Carter), the staying and striking down of a presidential decision would not frankly be any more “multifarious” than the two instances where federal courts already countered President Trump’s executive orders on immigration. While the Court might find it a nonjusticiable political question were a suit to challenge whether President Trump could modify tariff policies within the NAFTA framework (that is, a dispute over political decisions and not one of constitutionally defined power), a suit seeking to enjoin President Trump from withdrawing is much more so a question of legal and thereby justiciable significance.

Moreover, the potential of a unilateral executive pullout from NAFTA would provide six months before any official withdrawal commenced, which would provide six months for the Court to stay such an order and ultimately render a decision enjoining such an act entirely. Similar to the (roughly) three months between the time President Truman issued his executive order and the Court’s decision striking it down in Youngstown, six months would be an ample amount of time for the Court to reach a decision that could uniformly be understood by our federal government and the members of NAFTA. While Trump might argue that unilateral executive authority is appropriate or even unreviewable in this area (citing cases such as Curtiss-Wright, Chadha, and Goldwater v. Carter) the reality remains that each case he might point to would be distinguishable from the hypothetical of him unilaterally pulling out of NAFTA. Reason being: nowhere, in any statute, did Congress delegate to the President the entire power to withdraw from a congressional-executive agreement. Other cases, furthermore, support the conclusion that the termination of a congressional-executive agreement on international trade is not within the sole power of the President. Notwithstanding the Trade Act of 1974, few formalists would likely squeeze the executive authority over international trade agreements into an exclusive and unchecked box of power. More realistically, Youngstown’s categories admitted, the Court likely would not place Trump’s unilateral withdrawal in the first or second categories, considering that both the language of the Trade Act of 1974 and the codified NAFTA statutes indicate Congress made express congressional disapproval of such executive action. This would force the Court to place a unilateral withdrawal by the President in the third of Justice Jackson’s categories. Given that the executive power to even participate in NAFTA stemmed from a congressional statute, the Court would thereby find the executive power insufficient to withdraw from NAFTA by operating solely within its own authority. Ultimately, it would not be surprising if the Court were to seize the opportunity to demonstrate that its “so-called” decisions carry a weight that even President Trump’s administration must bear.