Can Trump Defy Congress and Pull Out of the North Atlantic Treaty?
The law is unclear, but he probably can.
The North Atlantic Treaty Organization (NATO) is a cornerstone of American foreign and security policy and enjoys long-held bipartisan support. As Republican Senator and nominee for Secretary of State Marco Rubio stated last year, NATO serves as an essential military alliance that protects shared national interests and enhances America’s international presence.”
Rubio’s boss-to-be, President-elect Donald Trump, has a different view. During the 2016 presidential campaign he complained that allies were not paying their fair share and that NATO was “obsolete.” Trump in his first term often publicly criticized NATO and privately threatened to withdraw from the organization. He continued to bash NATO during the 2024 campaign and said he would encourage Russia “to do whatever the hell they want” to NATO countries behind on payments. In a recent interview, Trump suggested he would withdraw from the North Atlantic Treaty in his second term if NATO countries don’t “pay their bills.”
Senator Rubio has long worried about Trump’s NATO threats and has since 2019 worked to limit a president’s authority to withdraw from the North Atlantic Treaty that created the organization. Late last year these efforts culminated in a statute co-sponsored with Democratic Senator Tim Kaine—Section 1250A of the 2024 National Defense Authorization Act. The law provides: “The President shall not suspend, terminate, denounce, or withdraw the United States from the North Atlantic Treaty” except with two-third Senate consent or an Act of Congress, and it further bars use of appropriated funds for these ends.
“With Ukraine under siege, it is critical to reaffirm the importance of this alliance by ensuring that no U.S. President can withdraw from NATO without the Senate’s advice and consent,” Rubio explained after the bill cleared the Senate Foreign Relations Committee. We will see if Rubio stands by that position during his confirmation hearings. And we will see if Trump claims the power to withdraw from the North Atlantic Treaty, and does so, even in the face of what I will playfully call the Rubio restriction.
If Trump decides to leave NATO, he will probably get away with it. Executive branch lawyers will almost certainly sign off on the withdrawal despite the congressional restriction, and there is a good chance that federal courts won’t get involved due to legal limits on judicial review. Despite the Rubio restriction, in short, the fate of United States’ 75-year-old NATO commitment likely turns entirely on Trump’s impulsive and mercurial decisionmaking calculus.
Background
The United States ratified the North Atlantic Treaty that established NATO in July 1949. Article 5 contains the core NATO principle of collective self-defense. Article 13 contains the less-well-known withdrawal provision. It provides that after 20 years, “any Party may cease to be a Party one year after its notice of denunciation.” The notice of denunciation must be given “to the Government of the United States of America”—an indication of the inconceivability in 1949 of U.S. withdrawal.
There is little doubt that absent the Rubio restriction, Trump could unilaterally withdraw from the North Atlantic Treaty with one year’s notice. It was not always so. As Curt Bradley shows in his new book, Historical Gloss and Foreign Affairs, the treaty termination power was “shared” before the twentieth between the executive and legislative branches. Congress or the Senate typically directed or approved treaty terminations or withdrawals and presidents typically didn’t terminate absent congressional participation. This practice “shifted” in the twentieth century to permit treaty termination to be “exercised solely by the president.”
Today a president’s power to terminate a treaty in accord with its terms is widely accepted and uncontroversial. Trump in his first term unilaterally withdrew from five Senate-approved treaties, including the Intermediate-Range Nuclear Forces Treaty. President Biden unilaterally withdrew from at least one treaty—a tax treaty with Hungary. This accepted presidential authority to terminate a treaty has developed “in absence of either a congressional grant or denial of authority,” as Justice Jackson put it in his famous Youngstown Steel concurrence. It is at least an independent Article II power, which means that a president can exercise it when Congress expresses no view on the matter.
But the question raised by the Rubio restriction is whether the presidential power to withdraw from a treaty is exclusive in the sense that it prevails even if, as is the case with the North Atlantic Treaty, Congress has enacted a limitation on withdrawal. As Trump v. United States, drawing on the Jackson framework and Marbury v. Madison, recently explained: An “exclusive constitutional authority of the President ‘disabl[es] the Congress from acting upon the subject,’” and “courts have ‘no power to control [the President’s] discretion’ when he acts pursuant to the powers invested exclusively in him by the Constitution.”
OLC on the President’s Treaty Termination Power
To answer the question whether the Rubio restriction is a valid legal constraint on the president, Trump will likely turn to lawyers in the Department of Justice’s Office of Legal Counsel (OLC), where I used to work. It seems pretty clear that the Trump 2.0 OLC will advise the president that he can withdraw from the North Atlantic Treaty despite the Rubio restriction. My confidence in this conclusion is based on a Trump 1.0 OLC opinion issued in September 2020 by then-OLC head Steven Engel.
The opinion concerned the Open Skies Treaty, which regulates unarmed aerial overflight over the territory of treaty participants to gather intelligence and promote mutual confidence. That treaty has a six-month notice of withdrawal provision. A 2019 statute required the president to notify Congress 120 days before sending the notice of withdrawal under the Open Skies Treaty. OLC ruled that the president could disregard the congressional notice requirement based on the constitutional principle that (with emphasis added) Congress cannot “regulat[e] the President’s decision to exercise a right of the United States to withdraw from a treaty.” Under this principle, the Rubio restriction would be unconstitutional.
In support of this principle, OLC recounted the historical practice of presidents unilaterally notifying foreign nations beginning in the twentieth century. OLC grounded the unilateral termination power in what it described as the president’s “exclusive authority to execute treaties and to conduct diplomacy,” which in turn it rested in Article II’s vesting of the “executive Power” in the president, its designation of the president as “Commander in Chief,” and its conferral on the president of the power to make treaties (with Senate consent) and to appoint and receive ambassadors.
The basic OLC view is that since the president alone conducts diplomacy and communicates officially with foreign sovereigns, the president alone gets to “execute” a withdrawal or termination provision in a treaty. It is important to OLC’s analysis that several powers related to treaties in addition to treaty termination—including their negotiation and signature, and especially their ratification (the formal act that makes the treaty binding)—are undoubtedly exclusive to the president even though not explicitly conferred in Article II.
OLC also relied on Zivotofsky v. Kerry, which held that the president’s power to recognize foreign sovereigns is exclusive. OLC claimed that Zivotofsky’s functional arguments in support of exclusivity in the recognition context, especially the need for the nation to “speak with one voice,” support the president’s exclusive power to terminate treaties.
Will the Supreme Court Agree?
It is very far from certain that the Supreme Court would agree with OLC that the president has an “exclusive power to withdraw from a treaty” that defeats the Rubio restriction. OLC opinions use different interpretive principles and precedents in constitutional analysis than do federal courts. As I explained in an article on Zivotofsky, these factors “lead OLC (and the executive branch generally) to take a broader, and perhaps much broader, view of presidential power than the Supreme Court.” And of course the Court is in no way bound by OLC opinions in its decisionmaking.
There are several counterpoints to the OLC view that the Supreme Court would consider. Here are a few.
The originalist-minded Court may give special weight (as indicative of original meaning) to early constitutional practice that supports a dominant role for Congress in treaty termination. President John Adams abided by a 1798 statute that proclaimed during the Quasi-War with France that several U.S. treaties with France “shall not henceforth be regarded as legally obligatory on the government or citizens of the United States.” A congressional power to order termination of a treaty might be thought to entail a power to order the president to not terminate absent congressional consent. And Congress has continued to assert a treaty termination power even in modern times. As late as 1986, the Comprehensive Anti-Apartheid Act of 1986 provided that the “[t]he Secretary of State shall terminate” an air services agreement with South Africa.” The Reagan administration complained about the constitutionality of this law but complied with it.
I doubt the Court would allow the more recent dominant practice of presidential unilateralism in the face of congressional silence to both supercede the earlier constitutional understanding that contemplated at least joint action and to establish an exclusive presidential power that is immune from congressional regulation. To be sure, Zivotofsky inferred presidential exclusivity from a similar pattern of modern unilateral presidential action in the face of congressional silence. But historical practice cuts less decisively for an exclusive presidential power to terminate treaties compared to the exclusive recognition power. Bradley notes that the executive branch in the 19th century “maintained … that its recognition power was exclusive, which is not true of treaty termination,” and that “congressional involvement in treaty terminations has been more persistent and direct, and has generated less historical controversy.”
The many examples of congressional regulation of treaty withdrawal and termination, moreover, underscore that Congress too has foreign relations powers, and that its later-in-time statutory enactments prevail over prior treaties as a matter of domestic law. In light of Congress’s long history of participation in treaty termination, one can see the president’s concurrent authority over treaty withdrawal as an issue of institutional allocation for treaty termination under domestic law that is subject to the last-in-time rule reflected in the Rubio restriction.
OLC discounted the relevance of Congress’s power over foreign commerce, war, trade, foreign aid, and neutrality, all of which are arguably implicated by the Open Skies Treaty. But it neglected to mention that Congress also has the constitutional authority to make “all Laws which shall be necessary and proper for carrying into Execution … all … Powers vested by this Constitution in the Government of the United States,” including the treaty-making clause. The regulation of treaty termination can be seen as “carrying into Execution” treaty-making under the Necessary and Proper Clause.
Justiciability
There are many more arguments, and greater complexity, both for and against an exclusive presidential power over treaty termination. I will examine them in depth in due course if the issue becomes ripe. Meanwhile, for an excellent discussion of the issue, see this panel discussion at the 2019 meeting of the American Society of International Law.
In that discussion, former Solicitor General Don Verrilli—who argued Zivotofsky for the government—expressed “a serious doubt whether this issue could get to court.” I agree. The ultimate constitutional merits of the president’s exclusive power to terminate are irrelevant unless a court can review a president’s determination to withdraw from a treaty in disregard of a congressional restriction. Absent judicial review, the executive branch wins when it disregards a congressional restriction.
There are at least two potential justiciability hurdles to judicial review: the political question doctrine and standing.
Four Justices in Goldwater v. Carter ruled that a challenge by legislators to Jimmy Carter’s unilateral decision to withdraw the United States from a mutual defense treaty with Taiwan was a nonjusticiable political question—that is, a question that the Constitution assigns to the political branches rather than courts for resolution. Justice Powell provided the fifth vote for dismissal on the ground that the Court should not address the treaty termination issue “unless and until each branch has taken action asserting its constitutional authority.” That predicate was absent in Goldwater but is satisfied with respect to the Rubio restriction. And the Court more recently suggested that the political question doctrine is inapplicable “when an Act of Congress is alleged to conflict with the Constitution.” There is a good chance that the political question doctrine will not bar review here.
The standing question of who if anyone can sue is a larger hurdle. It is hard to imagine an individual plaintiff who will have “suffered an injury in fact that is concrete, particularized, and actual or imminent” as a result of a U.S. withdrawal from NATO. Members of Congress could try to sue, but for reasons that Scott Anderson has well explained in the context of the Rubio restriction, they likely lack standing here as well. As Anderson notes, the Court has suggested that congressional standing may be more likely when Congress authorizes legislative plaintiffs “to represent their respective Houses of Congress.” Anderson explains that an earlier version of the Rubio restriction “preauthorize[ed] litigation on behalf of Congress in the event a president attempted to exit NATO without congressional consent,” but the provision was later removed. For these and other reasons, congressional standing will be an uphill battle.
Conclusion
Amid the concern about Trump’s threats to engage in unlawful action, it is easy to forget that the president of the United States has enormous lawful discretion to move the country in a very different direction with the stroke of a pen. The presidential power to withdraw from treaties by their terms is one such power. And while the Rubio restriction throws up a roadblock, the validity of a congressional barrier of this kind is unclear and the ability to test it in court is uncertain. Trump thus may have all he needs to pull out of NATO—unless his NATO-loving Secretary of State can convince him otherwise.
______________
Odds and Ends:
Jack’s interview with Bill Kristol on Trump 2.0
Bob and Jack discuss Trump 2.0 on Inside with Jen Psaki
Jack on the Trump Brief in the Tik Tok Case (Twitter/Bluesky)
Jack in NYT with Kate Shaw and David French on Trump 2.0