Reflections on Treaties’ End

In her forthcoming article in the Yale Law Journal, Oona Hathaway argues, not only that congressional-executive agreements (CEAs) are constitutionally permissible alternatives to treaties, but also that such statutes have a richer and broader historical pedigree than is commonly assumed, and that, with minor exceptions, such CEAs are superior along several dimensions to traditional treaties as a mechanism for making international agreements. Hathaway therefore proposes that the political branches should presumptively use the mechanism of statutory approval, rather than treaty advice and consent, in order to effect almost all international agreements. (Treaties would, under Hathaway’s proposal, be used only where Congress lacks any Article I power to enact the substantive laws that the agreement requires.)

We are generally sympathetic with Professor Hathaway’s project, and applaud her important contribution to the ongoing debate over the relative constitutionality and virtues of CEAs and treaties. We are skeptical, however, about some of the crucial arguments Hathaway makes in support of her central claim that CEAs create more reliable U.S. commitments than treaties. We will address those arguments in greater detail in subsequent posts. In this initial post, we will focus on those areas where we agree with Hathaway (at least in part) with respect to the desirability of CEAs.

In Part III of her article, Professor Hathaway identifies several reasons why, in her view, the political branches should in almost all cases presumptively choose to use statutes, rather than the Senate’s treaty-consent mechanism, in order to commit the United States to international agreements. We agree with Hathaway that CEAs offer a superior alternative to the treaty in at least two important respects.

First, as one of us (Golove) developed with Bruce Ackerman at more length some years ago, concluding agreements by use of a CEA is the more democratic alternative. This is true in part due to two simple facts: A CEA requires the assent of the more democratic chamber of Congress, and it conditions approval on bare majorities rather than creating a minority veto. There are, however, not insubstantial counterarguments about the democratic value of supermajoritarian rules such as the requirement of a two-thirds vote of the Senate for treaties. More importantly, the argument from democracy is based on a qualitative assessment of the impact of the Treaty Clause’s two-thirds requirement on the particular political institutions created by the Constitution (as well as the unexpected developments with respect to political parties that emerged after its adoption), and, most importantly, on the long, concrete historical experience the nation has had with the veto that can be wielded by one-third of Senators, representing a very tiny percentage of the nation. Indeed, it was that very history – and the terrible dilemmas that it created – that led to the development of the modern congressional-executive agreement.

The second, and more intriguing though ultimately less important, respect in which the congressional-executive agreement is superior to the treaty – which Professor Hathaway also discusses – is that it eliminates a potentially problematic anomaly created by the constitutionally prescribed Senate advice and consent procedure in some cases. The anomaly is that the institutions that make treaties in the first instance (the President and Senate) are differently constituted than the institutions that might be responsible for implementing them and maintaining adherence to them over time (namely, both houses of Congress and the President). If, for example, a treaty requires new appropriations, or the establishment of a new federal criminal law, the United States cannot fulfill its obligations without the House’s involvement. This disjuncture introduces the potential for one set of institutions (President/Senate) to bind us internationally to obligations that another institution (the House), after the fact, may prefer to avoid. By aligning the institutions that decide to make an agreement in the first instance with those that must implement and maintain adherence to them over time, the CEA option avoids potential structural pressures that may lead to defaults on international treaty obligations, thereby making them more reliable in the sense that Professor Hathaway stresses: a House of Representatives that agrees to authorize or approve such an agreement will more than likely also agree to take the steps necessary to ensure compliance with it. Indeed, as Professor Hathaway notes, it is possible, and desirable, for the Congress to take both steps simultaneously, i.e., for a single enactment to both approve the compact and to provide for its implementation (e.g., by making appropriations). This sort of “one-stop shopping,” in Hathaway’s words, ensures our negotiating partners that the agreement we are about to make will (almost surely) not be undercut by subsequent House recalcitrance.

Nevertheless, this problem, and the advantage of using CEAs in order to avoid it, is much less significant than it appears at first, for at least two reasons. First, the President and Senate early on recognized that the problem could be solved, even with respect to treaties, simply by seeking the adoption of implementing legislation in advance of ratification, or by making U.S. obligations under a ratified treaty contingent on the adoption of the necessary legislation. That practice simply makes the misalignment disappear. This “solution,” however, is not ideal, because it exacerbates the democracy problem: It becomes necessary not only to obtain two-thirds consent in the Senate, which gives a relatively small minority an effective veto over treaties, but also to obtain the consent of a majority of the House. That seems in strong tension with the constitutional design, and, more importantly, it ratchets up the difficulty of making treaties an extra notch, a development that even more dramatically tips the playing field against effective treaty-making.

More significantly, however, the problem does not have to arise at all with respect to most international agreements, because the participation of the House of Representatives is simply unnecessary from a constitutional point of view in most instances. It is true, of course, that there are some areas in which the Constitution requires House involvement, such as where agreements compel appropriations or the enactment of criminal laws. But those areas are very limited, and thus the concern Professor Hathaway raises is relatively less significant. It is true that the House has historically sought to claim a required role over more territory, including treaty provisions concerning tariffs, but most leading authorities have not recognized that claim as constitutionally viable. In any case, moreover, to the extent that the House has been able to insist on that claim as a political matter in certain narrow areas, such as with respect to trade agreements, the CEA has, in fact, tended to supersede the treaty form in practice. That makes sense, but it does not amount to a general reason for abandoning the treaty form in other areas in which there is no constitutional obligation to involve the House.

* * * *

In her article, Professor Hathaway focuses primarily on two other ways in which CEAs are said to establish more reliable commitments than treaties: (i) that they can provide for judicial enforcement; and (ii) that they can limit the circumstances of unilateral executive withdrawal. We are much more skeptical of these claims, principally because it is not clear that such commitments cannot be just as easily made in the context of treaties. We will address those arguments in our next post.