Harvard Law Review Harvard Law Review Harvard Law Review

International Agreements

The Failed Transparency Regime for Executive Agreements: An Empirical and Normative Analysis

The full text of this Article may be found by clicking on the PDF link to the left. This Article is accompanied by a set of data visualizations, developed in collaboration with TWO-N, that are available here.

The Constitution specifies only one process for making international agreements. Article II states that the President “shall have Power, by and with the Advice and Consent of the Senate, to make Treaties, provided two thirds of the Senators present concur.” The treaty process has long been on a path to obsolescence, however, with fewer and fewer treaties being made in each presidential administration. Nevertheless, the United States has not stopped making international agreements. Even as Article II treaties have come to a near halt, the United States has concluded hundreds of binding international agreements each year. These agreements, known as “executive agreements,” are made by the President without submitting them to the Senate, or to Congress, at all. Congress has responded to the rise of executive agreements by imposing a transparency regime — requiring that all the binding executive agreements be reported to Congress and that important agreements be published for the public to see.

Until now, however, there has been no systematic assessment of how well the transparency regime has been working.  This Article seeks to fill that gap. Through a Freedom of Information Act lawsuit, we obtained thousands of documents relating to the agreements reported to Congress and the legal authority on which the executive branch has relied for these agreements.  Together with a series of interviews with lawyers directly involved in the process, this new information has given us an unprecedented look inside the system of concluding, publicizing, and reporting executive agreements.  For the first time, we can describe how the system for making and scrutinizing executive agreements actually works — and when and how it fails to work.  The overall picture that emerges is one of dysfunction and nonaccountability.  In brief: there is reason to believe that the executive branch’s reporting to Congress has been incomplete; the entire publication and reporting process is opaque to everyone involved, including executive branch officials and congressional staffers; and Congress is failing in its oversight role.  The “system” is badly in need of repair if we are going to preserve the integrity and legality of the United States’ primary means of making international commitments.


The Constitution specifies one process for making international agreements. Article II states that the President “shall have Power, by and with the Advice and Consent of the Senate, to make Treaties, provided two thirds of the Senators present concur.”1×1. U.S. Const. art. I, § 2, cl.2. The treaty process, however, is on a path to obsolescence. During President Trump’s first three and a half years in office, he submitted only five treaties for Senate consent. This number represents a sharp drop compared to the Obama Administration (which submitted about five per year). The Obama Administration, in turn, submitted fewer treaties than the Bush Administration (which submitted around twelve per year). And the Bush Administration submitted many fewer treaties than the Clinton Administration (which submitted around twenty-three per year).2×2. On President Trump’s submissions, see Oona Hathaway, Reengaging on Treaties and Other International Agreements (Part I): President Donald Trump’s Rejection of International Law, Just Sec. (Oct. 2, 2020), https://www.justsecurity.org/72656/reengaging-on-treaties-and-other-internationalagreements-part-i-president-donald-trumps-rejection-of-international-law [https://perma.cc/6DBUUBBP]. On the other figures, see Jeffrey S. Peake, The Decline of Treaties? Obama, Trump, and the Politics of International Agreements 40 tbl.1 (Apr. 6, 2018) (unpublished manuscript), https:// papers.ssrn.com/sol3/papers.cfm?abstract_id=3153840 [https://perma.cc/9PQQ-M3EL].

The near collapse of the treaty process does not mean that the United States has stopped making international agreements. The Trump Administration, like prior administrations, has concluded hundreds of binding international agreements for the United States without submitting them to the Senate, or to Congress, at all. 3×3. See Treaties and Other International Acts Series (TIAS), U.S. Dept. of State, https://www.state.gov/tias [https://perma.cc/CFC8-BDSU] (follow hyperlinks for each year listed). These agreements, known as “executive agreements,” have long been the primary mechanism through which the United States makes international commitments. Since the late 1930s, well over ninety percent of all international agreements concluded on behalf of the United States have been executive agreements rather than treaties. 4×4. See Curtis A. Bradley & Jack L. Goldsmith, Presidential Control over International Law, 131 HARV. L. REV. 1201, 1210 & n.10 (2018). There has also been some decline in the average number of executive agreements concluded per year since the Clinton Administration (including under President Trump), but the decline is not as sharp as for treaties. See Peake, supra note 2, at 40 tbl.1. Many of these agreements involve minor or routine commitments, but some are quite consequential, involving matters such as security arrangements and alliances, the stationing of troops abroad, outlays of foreign aid, arrangements for border security, free trade accords, and weapons programs.

The main reason for the heavy reliance on executive agreements is that they are much easier to make than treaties. The Article II process requires a President to submit a negotiated treaty to the Senate for supermajority consent before the President can ratify and conclude it. By design, this is a very difficult process. In contrast, the primary form of executive agreement, known as an ex ante congressional-executive agreement, requires none of these steps. Instead, Congress in advance authorizes the President to make the agreement, usually in a statute that provides little concrete guidance. The authorization counts as full consent to the agreement prior to the President’s negotiation of it. This means that the President can make and conclude the agreement, and thus render it legally binding on the United States, before Congress or the public even knows of its existence. This is also true of most other forms of executive agreements, including ones based on an authorization in a prior agreement and ones based on the President’s constitutional powers alone.

Congress has gone along with these alternatives to the Article II treaty process with practically no substantive restrictions. Instead, it has imposed two general transparency requirements. First, since the late nineteenth century, it has required the executive branch to publish U.S. international agreements. The precise contours of this obligation have varied over time, but it has never resulted in anything close to complete public disclosure. Second, beginning with the 1972 Case Act,5×5. Pub. L. No. 92-403, 86 Stat. 619 (1972) (codified as amended at 1 U.S.C. § 112b). Congress has imposed a reporting requirement with broader reach: the Secretary of State must transmit to Congress, but need not make public, almost every international agreement other than an Article II treaty within sixty days after entry into force.

For a time, Congress also often included legislative veto provisions when delegating authority to conclude executive agreements. These provisions required the executive branch to send the negotiated agreement to Congress, prior to entry into force, so that Congress could examine the agreement and decide whether to disapprove it through one-house or two-house vetoes. But this check ended with the Supreme Court’s 1983 decision in INS v. Chadha,6×6. 462 U.S. 919 (1983). in which the Court declared such vetoes to be unconstitutional.7×7. Id. at 959. In response to Chadha, Congress amended many of the relevant statutes to strip out the legislative veto provisions, leaving just the bare delegations behind.8×8. See Oona A. Hathaway, Presidential Power over International Law: Restoring the Balance, 119 Yale L.J. 140, 201 (2009) (“Congress responded in most cases either by eliminating the veto provision altogether, as it had in the case of the Foreign Assistance Act of 1961, or by rewriting it to require the full legislative process in place of the veto.”).

As a result of Chadha, the only forms of accountability for the vast majority of executive agreements are a publication requirement for some agreements and a more comprehensive reporting requirement to Congress. These are very thin forms of accountability for such a consequential form of presidential lawmaking, especially as compared to the other major form of delegated executive branch lawmaking: administrative regulations. In contrast to regulations, executive agreements are only rarely publicized by the executive branch before they are concluded. This means that the public cannot consider the merits and potential effects of such an agreement before it becomes law. Another contrast is that the rules governing executive agreements do not require the executive branch to inform the public about their legal basis. There is also no institutionalized judicial review after agreements are concluded to ensure that they are properly authorized and justified.9×9. Contrast this with the judicial review afforded agency actions. For example, in Department of Homeland Security v. Regents of the University of California, 140 S. Ct. 1891 (2020), the Court blocked the Trump Administration’s effort to end a major immigration program because the Court found the effort to be “arbitrary and capricious” under the Administrative Procedure Act (APA), 5 U.S.C. §§ 551, 553–559, 701–706; 140 S. Ct. at 1915, despite the executive branch’s substantial authority to act in this area. And many of the agreements are never made public, which means that citizens cannot monitor their content and operation even after the fact.

A vital but unanswered question of separation of powers and foreign relations law is how well this limited regime of ex post monitoring actually works. Does the executive branch publish and report as it should? When the President claims to be acting pursuant to congressional delegation, are such claims of authority persuasive? Does Congress pay attention to what the executive branch is doing, and perform its oversight role, in this realm?

The three authors of this Article are among the relatively small cadre of scholars who have written about the executive agreement process. One of us wrote an article more than a decade ago that emphasized that executive agreements had become the key mechanism for making U.S. international agreements — and therefore that the President acting practically alone had come to dominate international agreement making in the United States.10×10. Hathaway, supra note 8, at 144. That article attempted to understand when and how Congress had delegated authority to the President to make international agreements. But it had to guess at many important questions about how the process worked, including which agreements were reported to Congress and the actual legal basis for the international agreements. Two of us wrote an article more recently that analyzed international agreement making in the broader context of the President’s other international lawmaking, law-interpreting, and law-terminating powers.11×11. See Bradley & Goldsmith, supra note 4, at 1204–06. This article too tried to understand how the accountability regime for executive agreements worked, and it included an extensive analysis of one of the nonpublic cover memos that the executive branch includes when it reports executive agreements to Congress under the Case Act.12×12. See, e.g., id. at 1250–52. Since 1981, the State Department has been required by regulation to include with each agreement that it reports to Congress under the Case Act a “background statement” that “shall include information explaining the agreement and a precise citation of legal authority.” 22 C.F.R. § 181.7(c) (2019); see infra p. 650. But this article also had to guess about how the reporting and transparency process worked in practice, as well as the legal basis for agreements generally. The lack of an empirical grounding left a significant gap in both articles — and is a serious problem in the entire literature on executive agreements.

We decided to address this empirical uncertainty by filing a Freedom of Information Act13×13. 5 U.S.C. § 552. (FOIA) request with the State Department seeking the Case Act cover memos that the Department is required to send to Congress with every reported agreement. We hoped that these memos would give us a better sense of the Department’s reporting practices under the Act. We also hoped that they would help us understand the claimed legal basis for the agreements, since almost every memo has a short section entitled “Legal Authority.” After a lawsuit and a negotiated settlement, the State Department produced 5,689 cover memos reported from January 20, 1989, through January 20, 2017 — that is, during the administrations of Presidents George H.W. Bush, William Clinton, George W. Bush, and Barack Obama.

The information that we have been able to glean from the cover memos — and from a series of interviews with lawyers directly involved in the executive agreement process — has given us an unprecedented look inside the system of concluding, publicizing, and reporting executive agreements. For the first time, we can describe how the system for making and scrutinizing executive agreements works — and when and how it fails to work. The overall picture that emerges is one of dysfunction and nonaccountability: there is reason to believe that the executive branch’s reporting to Congress has been incomplete; the entire process is opaque to everyone involved, including executive branch officials and congressional staffers; and Congress is failing in its oversight role. There are also substantial questions about the underlying legal authority for an appreciable number of the agreements.14×14. It bears emphasizing that our study focuses exclusively on unclassified executive agreements. The existence of classified agreements compounds the transparency and legality issues. See Cong. Rsch. Serv., 106th Cong., Treaties and Other International Agreements: The Role of the United States Senate 226 tbl.X-1 (Comm. Print 2001) [hereinafter CRS Treaty Study] (showing the number of executive agreements reported to Congress from 1978–1999; approximately 1.65% are listed as classified); Ashley S. Deeks, A (Qualified) Defense of Secret Agreements, 49 Ariz. St. L.J. 713, 724 (2017) (estimating that “the United States is probably party to approximately 1,000–1,800 secret agreements”); Ryan Scoville, New Evidence of Secret International Agreements, Lawfare (Feb. 19, 2020, 10:23 AM), https://www.lawfareblog.com/new-evidence-secret-international-agreements [https://perma.cc/H42G-3SJH] (finding that “the United States continues to enter into classified agreements with considerable frequency”).

Part I of this Article describes the legal regime governing executive agreements. It begins by describing the various types of executive agreements and the scope of the executive branch’s legal authority to conclude them. Next, it reviews what we call the “transparency regime” for executive agreements — the way in which Congress has regulated executive agreements not by requiring that the agreements be individually approved after they are negotiated, but instead by mandating disclosure to Congress itself and to the public after the agreement is concluded. This review makes clear that while Congress has been content to delegate authority to make executive agreements to the executive branch, it has repeatedly intervened in an effort to effectively observe and monitor how that authority is used.

Part II considers how the process of concluding executive agreements works in practice. It begins with information obtained from a dozen interviews with current and former government lawyers about how the process of concluding executive agreements actually takes place. These interviews were necessary because the process occurs entirely out of public view, and the regulatory guidance reveals only the basic outlines. That qualitative information then informs our quantitative assessment of the process of making executive agreements. We use data gleaned from the more than 5,000 cover memos obtained from the State Department to compare what is made public with what is reported to Congress. This comparison reveals that much of what is reported to Congress is never made public. More troublingly, if we treat the cover memos as evidence of the unclassified agreements that the executive branch reported to Congress, it seems likely that the executive branch has failed to comply fully with the reporting obligations that Congress has imposed.

The cover memos also permit us for the first time to discern the legal authorities cited in support of the agreements. By studying the thousands of citations to legal authorities in the cover memos based on the strength of the authority granted for concluding agreements, we can assess how the executive branch invokes delegated authority from Congress to make agreements. Here we find that less than half of the cover memos cite an authority that gives the executive branch clear and express authority to conclude a binding international agreement.15×15. See infra section II.B.2, pp. 677–91. Almost one-fifth of the memos cite legal authorities that in our assessment offer no support for concluding an agreement.16×16. See infra section II.B.2, pp. 677–91. Moreover, the cover memos frequently cite Article II of the U.S. Constitution in addition to other legal authorities; there is very little distinction made between agreements that rely entirely on the President’s sole constitutional authority and those that rely on authority granted by Congress or by virtue of a prior treaty. Indeed, perhaps the most remarkable thing that we discovered in the process of attempting to analyze and describe how the transparency regime works in practice is how disorganized the system is. Multiple legal authorities are cited without any clarity about the central legal basis for concluding an agreement, the legal authorities are not cited systematically or carefully, public databases of agreements are woefully lacking, and reporting to Congress is slow and likely incomplete.

Part III turns to proposed solutions. These fall into two categories. First, our proposals aim to remedy problems in publication and reporting. It is time, we contend, for the revolution of organization and transparency that took place for federal regulations at the dawn of the New Deal to extend to executive agreements. To that end, we propose to do away with the bifurcated regime and simply impose a comprehensive publication requirement, with minor exceptions, akin to the one that has long been in place for federal regulations. We also propose to reorganize the internal executive branch process for collecting and publishing along the lines of the process that governs federal regulations. As we explain, these reforms should be normatively uncontroversial, and they are dramatically easier to carry out today than they would have been in the twentieth century due to the digitalization of information.

Second, we address the uncertain legality of many of the agreements. We propose that Congress require the executive branch to offer greater transparency regarding the legal basis for international agreements — and clarity about the source of legal authority on which the government relies. A simple solution would be to publish the cover memos themselves — rather than keep them hidden from public view — and require that they cite a primary legal authority. These and related reforms will, we believe, make it more likely that agreements in the future will possess an adequate legal basis.

We should emphasize at the outset that while we are very critical of the current reporting and publication system for executive agreements, we do not believe that the blame for this dysfunction lays at the feet of any particular branch of government or particular agencies or individuals inside the executive branch. The dysfunction is the product of decades of neglect, secrecy, disorganization, resource constraints, and misplaced priorities, in both Congress and the executive branch. The current system is not admired by anyone in the government or serving any institution’s interest. Reform is clearly needed, as many of our executive branch and congressional interview subjects themselves acknowledged. While some of our proposed reforms will be more difficult to implement than others, there should be broad agreement, across parties and across the political branches of government, that the system is badly in need of repair.

† The Harvard Law Review has not independently reviewed the data and analyses described herein.
* Gerard C. and Bernice Latrobe Smith Professor of International Law, Yale Law School.
** William Van Alstyne Professor, Duke Law School.
*** Learned Hand Professor, Harvard Law School.
For helpful comments and suggestions, we thank Sue Biniaz, Kathy Bradley, Kathleen Claussen, Ashley Deeks, Kristen Eichensehr, Jean Galbraith, Monica Hakimi, Duncan Hollis, Mark Iozzi, Jamil Jaffer, Michael Mattler, David Pozen, Ryan Scoville, Margaret Taylor, and participants in a faculty workshop at Duke Law School and in the 2019 Duke-Virginia Foreign Relations Law Roundtable. For excellent research assistance, we thank Benjamin Fleshman, Michelle Lou, Eli Nachmany, Sam Rebo, and Abraham Sutton. For assistance with data collection and statutory analysis, we thank Andrea de Sa, Sasha Dudding, Pardis Gheibi, Katrin Marquez, and Danielle Zucker — who worked nearly two years on this project. We also thank Natalie Giotta, Ariq Hatibie, Kelsea Jeon, Christopher Kim, Tobias Kuehne, Julianna Lai, Sam Larkin, Jared LeBrun, Preston Lim, Randi Michel, Nicole Ng, Max Obmascik, John Park, Kate Pundyk, and Tiana Wang for their assistance in generating and quality controlling the data, and Andrea de Sa and Ayoub Ouederni for their outstanding assistance with analyzing the data. We thank the many government lawyers past and present who provided us with insights into the process for making executive agreements. We are grateful to Michelle Pearse, of the Harvard Law School Library, for assistance with creating the online repository of the data in this Article and to Evelyn Ma, Lucie Olejnikova, and the entire team at the Yale Law Library for their help on a number of research projects, including tracking down obscure sources. For assistance with the FOIA request and lawsuit against the Department of State (Yale Law School Center for Global Legal Challenges v. U.S. Department of State, No. 17-cv-02042 (D. Conn. Dec. 7, 2017)), we thank Hannah Bloch-Wehba, Charles Crain, Diana Lee, Paulina Perlin, and David Schultz of the Media Freedom and Information Access Clinic at Yale Law School. For their work on the data visualizations, see Oona A. Hathaway, Curtis A. Bradley & Jack L. Goldsmith, The Failed Transparency Regime for Executive Agreements: An Empirical and Normative Analysis: Data Visualizations, HARV. L. REV. (Dec. 10, 2020), https://harvardlawreview.org/executive-agreements-visualizations, and Figures 1 and 3 in this Article, we thank Aucher Serr and Natalie Erdem, of Two-N.