In a previous post, I argued that the Paris Climate Change Agreement is a binding commitment under international law. As a result, under U.S. law, ordinarily it should be approved by 2/3 of the Senate pursuant to Article II, Section 2 (or by a majority of both houses of Congress as a congressional-executive agreement, to the extent one thinks that is a constitutional alternative).
As also noted in the prior post, however, the President seems unlikely to seek such approval, and Professor Daniel Bodansky -- the leading expert on the Agreement's legal status -- argues that that is constitutionally permissible. In his view, the Agreement can be approved as an executive agreement. That is so, he says, because (a) the material obligations in the agreement are nonbinding, and (b) the binding obligations in the agreement are not material. Specifically, on the latter point, he argues the agreement's binding provisions can be undertaken by the President alone
either because [they are] procedural in nature, because they reiterate[ ] obligations the Senate had already approved in [a prior treaty], or because they reflect[ ] and complement[ ] existing U.S. law.
I agree that the Constitution allows the President to enter into some binding international agreements on his own authority. (See 77 N.C. L. Rev.133, in which I make the argument at length). And in general I agree that the line between treaties and non-treaty agreements turns on whether the agreements impose material obligations (although the original meaning on this matter is challenging to state with precision). But I am dubious of some of Professor Bodanksy's categories.
To begin, I think it is wrong to say that the President can undertake international obligations that "reflect and complement existing U.S. law." Even if we leave aside the ambiguity of what "complements" existing law, converting a domestic law into an international obligation is a material step. Domestic law can be repealed without international repercussions. However, once a provision of domestic law is embedded in international law, it cannot be repealed without violating international law and upsetting international expectations. It is true that, as a matter of U.S. domestic law, Congress can violate international law if it chooses. But it cannot violate international law without repercussions (at least in terms of reputation). Put another way, undertaking an international law obligation that parallels a provision of domestic law is an international promise not to repeal the law. I think it unlikely that the President has unilateral power to promise other nations that a law will not be repealed.
Second, I'm doubtful that calling an obligation "procedural" categorically puts it within the President's sole power. Of course, some procedural aspects of ordinary diplomacy are likely so immaterial that they can be managed by executive agreement. But consider a leading "procedural" provision of the Paris Agreement discussed in my prior post, Article 4.2's requirement that "[e]ach Party shall prepare, communicate and maintain successive nationally determined contributions ["NDCs," that is, emissions targets] that it intends to achieve." It's true that the NDCs themselves are not binding (if they were, Senate approval would clearly be required). And it's true that as a result preparing, communicating and maintaining the NDCs can be described as "procedural" in the sense that it will presumably be done within the executive branch, and appropriately so since that process does not create any law. The commitment nonetheless seems significant: the U.S. must prepare a series of emissions targets that it intends to meet (even if it doesn't have an obligation to actually meet them). If a future President decides that the process isn't worthwhile, that President cannot simply cease submitting U.S. NDCs or declare that the U.S. doesn't intend to meet its NDCs (without violating international law). I would call that a material constraint on the future President's action.
I don't see anything in founding-era history or in subsequent precedents that establishes "procedural" constraints as categorically exempt from the treatymaking requirements. Although the Supreme Court has directly approved executive agreements several times, it has always done so in the context of claims settlements. (As outlined in my North Carolina Law Review article, executive claims settlements date to 1799 and were common in the nineteenth century, so they have a strong historical foundation). The cases did not turn on a category of "procedural" obligations.
Professor Bodansky's remaining category is the most promising: provisions that reiterate obligations the Senate has already approved in a prior treaty. I think it is probably correct that in general mere reiteration of an existing treaty obligation does not require further Senate approval because it is not really making a new treaty. However, it would have to be the case that the new agreement did not expand the terms of the prior agreement. For example, if it extended the duration of the commitment, or if it made the commitment harder to terminate, that would not be a "mere reiteration" but rather an extension. For this reason, I'm doubtful that the Paris Agreement in fact is, for any material terms, a "mere reiteration" of a prior agreement. I take it the likely candidate is the UN Framework Convention on Climate Change [UNFCCC], which received Senate advice and consent in 1992. Without undertaking a detailed analysis of the UNFCCC, I would be surprised if the widely celebrated 2015 agreement merely repeated the obligations of a two-decade-old agreement. To the extent it did, however, I agree it would not need renewed Senate approval.