18
Examination of witnesses
Curtis Bradley and Andrew Keller.
Q26 The Chair: Good afternoon. This is an evidence session of the International Agreements Committee. We are delighted to have today Mr Andrew Keller, former General Counsel for the US Senate Committee on Foreign Relations, and Professor Curtis Bradley, who is the Allen M. Singer Distinguished Service Professor of Law at the University of Chicago. Both of you join us virtually and I thank you very much. We are grateful for your assistance in this inquiry, particularly on non-binding agreements.
I will start the questioning. I understand you have decided between yourselves in advance who will answer which question, so, cleverly, I am able to address each to the right person. Professor Bradley, would you explain briefly the process for congressional scrutiny of treaties and other legally binding international agreements?
Curtis Bradley: It is my pleasure and it is an honour to testify before this committee. The US constitution has a process for approving treaties, which provides that the President has to obtain the consent of two-thirds of the Senate. That is a difficult hurdle. Obviously, for those agreements there is a lot of scrutiny. The Senate will hold hearings and then decide whether to approve the treaty. The complication in the US is that most agreements concluded in the last 50 or more years have not gone through the two-thirds Senate consent process and are concluded as what we call “executive agreements”. There are many types of those. Most are based on some kind of statutory delegation from Congress. The majority of the two Houses of Congress said, in statute, that the President can go out and make certain kinds of agreements. The President does, and those agreements are not usually brought back to the legislature for review or approval. Some are, but not very many. Major trade agreements tend to go back to Congress for approval; Congress has insisted on that. But most agreements authorised by Congress are then done and Congress does not formally approve them.
The President sometimes makes agreements on his own constitutional authority in certain areas. Obviously, those are not presented to the legislative branch either. As time went on in the 20th century, there were more and more of these so-called executive agreements and they obviously did not go to the Senate. Congress became concerned, particularly about not knowing what the Executive were doing and the kinds of commitments being made on behalf of the United States. A very important law was enacted in the early 1970s, the Case-Zablocki Act, requiring that whenever the Executive make these agreements they must be reported within a certain time to Congress and eventually published for the public at large. That regime has not worked perfectly since the 1970s, but thousands of agreements have been reported since the enactment of that statute. It certainly improved transparency, which was very limited before that.
One limit of that law applies to this hearing: it was interpreted to apply only to binding agreements that the Executive made, not to non-binding ones. Eventually, Congress decided that that was a big omission and needed to be addressed. Recent legislation attempted to address that omission in part.
The Chair: We will come on to talk about that.
Q27 Lord Anderson of Swansea: Gentlemen, welcome. There must be a temptation on the part of the Executive to resort to the use of these non-binding instruments more and more. Have you carried out a comparative study, in both the US and other comparable jurisdictions, to see how the trend has moved over the past few years?
Curtis Bradley: We have. Thank you for that question. Two other professors and I, about two years ago, did a limited survey of a mix of countries in Europe plus Canada and some others. There have been a couple of other studies in Latin America and Europe as well. All the findings were that in recent years, as I suppose you would predict, use by Executives of these non-binding instruments has gone up. There might be good reasons for that and maybe some more questionable reasons. One potential positive reason is that there is just more regulatory co-operation among countries at administrative level. They co-operate agency to agency on sharing information, having common standards or what have you. Those are often done through non-binding, more flexible memoranda of understanding or other arrangements.
Part of the change is a positive one, probably, of increased, informal modes of co-operation, but it is also likely that there are some more questionable reasons for using these instruments instead of binding agreements. We found in our survey of other countries that, for the most part, they have not regulated the use of non-binding instruments domestically in terms of reporting or review—often, even when they have regulated executive use of binding instruments. As you noted, there may be a temptation—I think there is—to conclude these agreements without some of those legal review requirements. Our survey respondents thought that that was happening.
Lord Howell of Guildford: On this subject, one does not hear so much, in either your country or ours, about the consumer or the changing nature of the consumer market. That is now very hungry for the simple reason that three-quarters of consumers have mobile telephones, iPads, Apple Watches, social media and a vast flow of information which enables them to find out about things that they probably knew nothing about before. How much is it taken into account that you now deal with electronic populism of vast proportions, 10 times the size of even 20 years ago, which changes the need for Congress or, in our case, Parliament to be much better informed and better at informing the hungry, outside world?
Curtis Bradley: Yes, I definitely agree that the external environment has changed. That includes the fact that the media is much more disaggregated than it was. People in general have information—sometimes imperfect information—about what is happening, in not only their own country but maybe partner countries and the like. That can increase pressure on the legislature to be not only informed but involved.
To give a very recent example in the United States that we have all read about, we understand that our Administration made an agreement of some sort with El Salvador to house detainees and migrants—a controversial issue in the United States—but we do not have all the details. The public have some knowledge that this exists, partly because of some statements from El Salvador which often make their way into this media that you mentioned. A number of Members of Congress think that they should get more information and educate themselves, and maybe have some role in interacting with the Executive about the policy decisions here.
I think it is harder to keep things completely secret. That is in many ways probably a good thing for transparency, but it often means that the public have limited or imperfect information. One goal of our statute in the US was to regularise publication so that it is not just down to the legislative branch, which is very busy and would have trouble monitoring every one of these agreements, but to some extent to use the public as a monitoring body. There are lots of stakeholders and interested parties out there in the world. It is probably important not to just use the legislature to watch these issues, but interested stakeholders and the public could sometimes alert the legislature about particular issues. The legislature and public scrutiny could work together. That would require some kind of publication and notice system that the public can access beyond just rumours or other pieces of information that they have been getting. If we care about transparency, I think there will probably be a mix of reporting to the legislature and some kind of reliable, public database or registry that allows others to review this information.
The Chair: We may come back to that. Before we do that, we want to focus a little more on non-binding instruments as well.
Q28 Lord Hannay of Chiswick: Good afternoon and thank you for lending us your wisdom to understand a bit better how things are done on your side of the pond. Of course, that affects other people. Perhaps we could look into that a little now.
Could you identify the main reasons why the United States and party X, Y or Z are persuaded to have a non-binding instrument rather than a full, binding treaty? Is it, in most cases, the United States simply saying, “We’re not prepared to do a binding treaty”, or is it more by common agreement? Is it just for the convenience of the third country that they are not forced to submit it to a process, although they may do so? I understand that an instrument agreed between the United States and a third country might be a non-binding instrument for the United States but regarded as binding by the country that enters into it. We may find that the agreement between the UK and US on trade will fall into that category, because I do not quite see how else we will change our tariff schedule—but that is not for this afternoon; it is a different case.
What is the underlying reasoning behind the use of non-binding instruments? To what extent is that driven by the United States rather than by third countries? You gave us one example, of El Salvador, where I infer that perhaps you do not think the right processes were used. Are there other examples where, in your view, the processes were not used properly by making it non-binding? It would help us a lot if you gave one or two specific examples like that.
Curtis Bradley: Of course. Two other professors and I, as part of a project we did on this topic recently, interviewed a number of officials throughout the US Government to try get at this question of what the reasons are for these non-binding agreements. That included some former officials, who were probably more comfortable about being candid. We found a mixed set of reasons. Sometimes—and probably often—it is not some nefarious reason; it is just because it is quicker and allows for more flexibility. Many administrative agencies find they can share information or co-operate in other ways with their partners more easily by making non-binding arrangements. They do not particularly need them to be legally binding. It is in both parties’ interests to co-operate. They do not expect to resort to some kind of international enforcement mechanism. It probably allows them to change the agreement more easily, which might be in both their interests and not just that of the United States as conditions develop or their interests change. In many ways, this seems an unproblematic reason. It is a quicker and more flexible process.
We also sensed that sometimes the reason was more questionable in allowing either party to make it non-binding. Interestingly, it was not always the US. Sometimes we heard that the US agreed to make it non-binding because of a request from the other party who did not have as many options. In the United States, there are all these different tracks for making agreements, many of which do not have to go to the legislature. Some treaty partner countries do not have those tracks and if the agreement was binding it would have to go to the legislature for approval, which would be a cumbersome procedure. Those countries sometimes requested that the US make it non-binding to make it easier on them, whereas the United States could probably have done it as one of these more simple executive agreements.
I can give you an example where it appeared that the United States did it as non-binding to avoid some legislative involvement. One of the most discussed examples, to give you a concrete example, is one that the UK was of course involved in: the Iran nuclear deal, the Joint Comprehensive Plan of Action in 2015. The US insisted that that was a non-binding arrangement—I do not think Iran necessarily saw it that way—even though it had lots of obligations about relaxing sanctions and supporting a reduction of sanctions in the Security Council. It appeared that the main reason why the US did it as a non-binding arrangement was to avoid the requirement that Congress approve it. At that point, it was very unlikely that that would have got approval in Congress. The Executive—the Obama Administration—viewed it as very important to get it done and, they hoped, prevent Iran from developing nuclear weapons. It was done unilaterally, quite controversially, by the Obama Administration. Now, in that particular case, Congress was so exercised about this that it got involved to some extent, but it looked like it was done as a non-binding arrangement, whereas, in a different time, it probably would have been done as a treaty or binding executive agreement. That is a widely discussed example.
A much more complicated one is the Paris Agreement on climate change, done near the same time. That is a treaty, but Presidents Obama and Biden did not submit the US agreement to it to Congress or the Senate. They said that much of it was authorised by a prior treaty, which may or may not be the case, and that the core provision about reducing omissions and having a target to do that was purposefully made as more aspirational and non-binding—“should” rather than “shall”. By doing it that way, they said they avoided the need to go to the Senate or Congress. Again, they would have been unlikely to get approval there. There is a lot of debate in Congress and the US about climate change and what role the US should play in helping to reduce omissions.
Those are two widely discussed examples where it looked like we would have done them as binding agreements in the past and the Administration, partly because of difficulties of getting approval, sought a more non-binding arrangement. That generated some criticism. Part of the greater discussion in the United States about regulating non-binding agreements was in part a reaction to such important agreements being done in this other way.
The Chair: You said earlier that the reasons for not going for treaties but non-binding instruments were flexibility and speed. From what you have just said, it is also to avoid scrutiny by the legislature. Perhaps that is what you meant by quickness and flexibility anyway. Is that right?
Curtis Bradley: I did, yes. I am not saying that it is always a problem. Many of these agreements are pretty routine, administrative arrangements between agencies and not of the importance of, say, the Iran nuclear deal. It is not clear, at least in the US, that the legislature would want to have to approve all of those. Logistically, it would be difficult to imagine that happening. Many of these do not raise high levels of public policy concern; they are really just about co-operation of various kinds. However, even for those, there is a desire to have them be transparent so that, if one is of concern or raises big issues of policy, it is not hidden from view.
One concern that emerged in the US was that we had this big transparency law with a huge blind spot in it. Many important arrangements are not necessarily made as binding treaties, and those were largely omitted from even requirements about publication and notifying Congress that they were happening. As we have seen those types of agreements grow over time, a number of countries—not just the US—have started to realise that their practice for their laws about notice and review have this kind of blind spot. These arrangements can affect countries’ foreign policy quite significantly in some instances. It is hard to know why those should not be at least notified to the legislative branch. Often, unless they are classified and secret material, it is hard to see why the public should not be aware of what the Government are doing as well. Just because they are not binding does not mean they are not important.
Lord Hannay of Chiswick: The issues of transparency and approval are quite separate, are they not? I think that is what you are saying.
Curtis Bradley: Yes, exactly.
Lord Hannay of Chiswick: On transparency, is it not the case that the United States quite frequently invites committees of Congress to send representatives to the negotiations so that there can be increased transparency? I do not think that occurs in practically any other country in the world that I know of. That has perhaps been a way of softening this separation—the keeping at arm’s length of the legislature. Could you comment on that, too?
Curtis Bradley: I invite my friend Andrew Keller to answer that since he has had direct experience in the Senate of some of those issues.
Andrew Keller: Thanks so much, Curt. It is a pleasure to be here today, so thank you to the committee. This is a great question. It is accurate that the executive branch in the United States, the State Department or others, sometimes invites either Members of Congress or congressional staff to join in treaty negotiation delegations. I think that that practice was more frequent several decades ago, even up to probably a few decades ago, and it has been several years since that type of invitation has been extended. I agree with your point to the extent that where Members of Congress or staff can be part of a delegation it certainly helps with congressional oversight and accountability. It does not necessarily check the box with regard to any level of public transparency. You are right that there are different angles from which to look at this question. One is whether legislative or parliamentary approval is required. Another is the level of involvement and understanding from the legislative perspective. Finally, there is transparency for the public. All those things are very relevant to look at.
Lord Hannay of Chiswick: Thank you very much.
Lord Anderson of Swansea: If a Member of Congress or a staffer attends a negotiation, what role do they play? Are they full members of the delegation? Are they expected to specialise in certain areas? Or are they meant to be just sleeping observers?
Andrew Keller: That is a great question, too. In general, it is the latter category. They are there as observers; they are able to attend negotiating sessions but they are not there as negotiators. They do not play a substantive role. Generally, they would come back after the negotiating session. Let us say these were members of the Senate Foreign Relations Committee where I previously served as General Counsel: the notion would be that staff would write up a report and submit it to the members. If it were a Senator on the committee, she or he might hold a small meeting to give a read-out of the negotiating session to other members of the committee.
There are also some situations where the executive branch has been very active throughout the course of negotiations in giving Congress information. Professor Bradley mentioned the Joint Comprehensive Plan of Action. There were dozens of member and staff-level briefings on that instrument while it was being negotiated with Iran, European partners, Russia and China. That was the case even though the JCPOA was a non-binding instrument. Another example I can think of was the New START Treaty with Russia. That was treated as an Article II treaty under the US constitution and was subject to formal scrutiny and a vote in the Senate. That is another example of a treaty on which the executive branch was very proactive in briefing Members of Congress and staff throughout negotiations.
At the end of the day though, that proactive and intensive level of briefing is really the exception to the rule. In general, what ends up happening is that, from a congressional perspective, members of staff will find out about an instrument—at this point whether binding or not—at best through the transparency provisions enacted into law. Sometimes, they first hear about it in the press, which is never a satisfying development from a congressional perspective.
The Chair: Or from a parliamentary perspective here, I am bound to say.
Q29 Lord Howell of Guildford: Listening to what you said, in many ways the United States is ahead of the game with the Case-Zablocki Act and all the amendments to it, and your insistence that Congress should know more. On the other side, the Executive now face in the information age an avalanche of agreements of every kind and shape. If they are ever to deal with them, the natural tendency is to hide them away as other instruments—NBIs, MoUs and all sorts of other things. There is a huge and growing tension there. We have an ancient constitution that has been around various places and not written down very clearly. How do we move into this utterly new world where the volume of agreements is vastly greater, as is the demand of the public to know a lot more and the Executive struggles under more and more complexity of the kind that we see coming up every day in the media? Have you any idea what other countries are doing, following your example? Do you liaise with them?
Curtis Bradley: I can say a few words about that. As you noted, I think the US at the moment is probably ahead of most other countries on this topic because we happened to get this legislative reform a couple of years ago that Andrew was heavily involved in. When I talk to other countries about this, it is clearly a topic of great discussion—as it is in the UK.
There has been some movement in other countries. The most common reform has been at the executive level to make sure there is co-ordination when different parts of departments and agencies make these arrangements, to maybe get approval in the central part of the Foreign Ministry and certainly some kind of database within the Executive so they actually know what they are doing across things. Not knowing what different parts of the Executive are doing has been a problem, including in the US. It needs more reform internal to the Executive.
A number of countries have provided guidance throughout their Executives about how to draft or not draft non-binding arrangements, partly to avoid the danger that they may create binding arrangements by accident. A number of countries, including the US, put out guidance on that. Neither of those things really helps with respect to legislative involvement or transparency, although there has been a little bit of movement on that. In Switzerland, after a big controversy not too long ago over a migration compact, the Executive Council made a kind of informal agreement with Parliament that it would, on a more regular basis, advise it of significant non-binding arrangements. I think we have seen that in a couple of other countries as well, but not as far as I know through binding legislation that we now have in the US. It is much more of a voluntary co-operation between Parliaments and the Executive as these have become more prominent or more controversial.
A number of countries I have talked to have said they are thinking about more reform, including more publication requirements so that the public can access these. That has been discussed in the Czech Republic, France and some other places. Of course, one possibility is that US law may lead to more discussion of those topics, which has been going on for the last five years or so. I am not yet aware of significant legislative developments like the ones we saw recently in the US.
Lord Howell of Guildford: I think it was Mao Tse-Tung who said about Switzerland that it is easier to polish a small diamond. Thank you anyway very much.
Q30 Lord German: Could I take you back in history to when the transparency legislation was put in place on non-binding instruments? What were the concerns that Congress had at that time that led to it and were there any specific things that triggered the action that it took?
Curtis Bradley: Andrew, did you want to talk about the legislation?
Andrew Keller: Sure. The Case-Zablocki Act was first enacted in the 1970s. In general, the reason for implementing that transparency regime was a recognition that there had been a shift over many decades from doing binding international agreements as Article II treaties, from the US domestic perspective, to what Curt described as executive agreements. Again, Article II treaties require the highest level of congressional scrutiny, while executive agreements really do not in general require much, if any. At the end of the day—many years after the fact—the law was about recognising that there had been this shift to executive agreements. It recognised that hundreds, if not thousands, of executive agreements had been concluded with little to no congressional or public awareness. I think it was a very important effort at the time to try to put a framework around the executive agreement practice.
Some elements of the original Case-Zablocki Act reflected what an earlier question raised regarding significant differences in the pressure of doing business and how the public—frankly, how Congress and the public—consume and expect to consume information. The example I have in mind is that the original Case Act required that the executive branch would submit binding instruments only to Congress 60 days after they entered into force. As you know, sometimes an agreement can be concluded and then not enter into force for months or even years afterwards. What we saw over time is that the executive branch began to use the Case Act as the ceiling for its obligations. Sometimes, Members of Congress would understand that an agreement had been concluded and say, “We’d like to see a copy of the text of that agreement”. They might get a response from the executive branch that would say something like, “Well, the agreement has been concluded but it hasn’t entered into force yet and, under the Case Act, we’re not legally obligated to provide a copy of that agreement to you until it enters into force”. Those types of things—having to wait two months after entry into force and then the treatment of the transparency regime really as the ceiling, rather than perhaps what I would hope to see: that it was treated as the floor or the baseline obligation—were some of the factors that led Congress to want to modernise the provisions of the Case Act that applied to binding agreements, as well as adding accountability and transparency provisions to address non-bindings.
Lord German: Do you think that behind the logic of transparency there was also a feeling that more scrutiny might emerge from it, or was that never the expectation? Was it that simply putting it in the public arena would be sufficient?
Andrew Keller: That is a great question on that subtext. To take a step back, the 2022 amendments to the Case-Zablocki Act were very intensely negotiated between Congress and the executive branch, and the most sensitive part of those negotiations was the decision by Congress to pursue the addition of provisions that addressed non-bindings. It was so sensitive in part because the executive branch had become accustomed to operating in a non-binding space, without any friction from Congress or really any attention from the public, for probably 99% of the non-bindings. There were some high-profile exceptions, such as the Joint Comprehensive Plan of Action, but in general at least a few dozen agencies of the US Government were involved in negotiating, concluding and implementing these non-bindings without congressional or public involvement, so it was quite a shock to the system for the executive branch to accept it. Even though there was no effort to require congressional approval of non-bindings, it was still a shock to the system for them to accept the idea that they would have to submit them to Congress and, in many cases, publish them on a publicly available website.
The Chair: Even notification was a shock to the system for them?
Andrew Keller: Even notification was a shock to the system.
The Chair: Can you tell us a bit more, please, about what changes were then made to the Case-Zablocki Act?
Andrew Keller: Particularly with regard to non-bindings?
The Chair: Particularly with regard to those and transparency, although I would like to ask you this at some point, if I may. You talked about the transparency of the notification and so forth, but is there anything to do with scrutiny as well? Could you start, in any event, by telling us what the changes were in relation to transparency on NBIs?
Andrew Keller: The bottom line, as Curt and I have explained, is that prior to 2022 the Case-Zablocki Act covered only binding instruments; there was no treatment of non-bindings. The 2022 amendments changed that so that, for the first time, there is a law of general applicability with regard to non-bindings. The way they are addressed in the law is that a new category was created that is referred to as qualifying non-bindings, or QNIs. I am sorry if the acronyms do not quite match up. Those qualifying non-bindings have a two-part definition.
The first part we think of as creating a push mechanism or a proactive requirement on the executive branch. If there is a reasonable expectation that a non-binding could have a significant impact on US foreign policy, the executive branch is under an affirmative obligation to submit that non-binding to Congress and, unless it is classified for national security reasons or falls into a few other potential statutory exceptions, it has to publish the text of the non-binding within 120 days of it becoming operative.
The second element of the definition of qualifying non-binding is what we think about as the pull mechanism. That allows certain members of congressional leadership, basically through a letter from them to the Secretary of State, to request a specific non-binding—or a group or class of them. On receipt of such a communication, the Secretary of State is obligated, within a certain amount of time, to provide that non-binding or non-bindings to the relevant members of Congress, as well as ultimately publishing them on the State Department website.
I think there is an interrelationship between the push and the pull mechanisms. To the extent that the executive branch is interpreting the significant foreign policy impact standard broadly, and proactively pushing out to Congress and the public a number of non-bindings in faithfully implementing that aspect of the law, I suspect there will be less use of the pull mechanism. But to the extent that Members of Congress feel that the executive branch is interpreting the significant impact standard too narrowly, the expectation is reasonable that the pull mechanism will be used more actively.
The last point I would mention on this is that there is a relevant exception to the definition of qualifying non-binding because it applies to at least one agency that does a lot of non-bindings. During the course of the negotiations between Congress and the executive branch, there was an insistence by the executive branch that any non-binding associated with the US Department of Defense, the armed services or the intelligence community be carved out from the definition of qualifying non-binding and, therefore, from coverage under the Case-Zablocki Act. That does not mean that there could not be other statutes enacted; in fact, there is an example of one in relation to AUKUS—the Australia-UK-US defence partnership—in which a specific provision says that, regardless of the carve-out in the Case Act, non-bindings related to AUKUS still need, by law, to be reported to Congress. That is the basic gist, so I hope that explanation made sense.
The Chair: That is helpful. Before I turn to Lord Boateng and his question, you have been talking about notification. As I said earlier, I am interested in whether that gives rise to any form of scrutiny. Is it right that this is really providing just for notification?
Andrew Keller: In the law, it really provides only for notification. First, I would say that the theory on Case, as Congress was moving forward, was one of what was probably politically feasible at the time. Secondly, even though it does not give rise to formal scrutiny—meaning necessarily a vote in Congress—it does give rise to informal scrutiny, under the notion that if you do not know about the agreement or do not have its text, it is very hard to know even what questions to ask the executive branch. Even to have a staff-level briefing or to consider having a hearing on the non-binding is extremely difficult. The feeling was that this level of notification and transparency, while not leading to formal scrutiny, would lead to additional engagement and, as needed, informal scrutiny over non-binding practice.
The Chair: Thank you very much, that is helpful.
Q31 Lord Boateng: You may have partially touched on this, Mr Keller, but in his article in the University of Chicago Law Review Professor Bradley refers to tensions between the congressional branch and the Executive on this issue of non-binding instruments that go back to the Boxer rebellion and, indeed, to the Spanish withdrawal from Cuba. I can understand the tensions there, but the definition of “qualifying non-binding instrument” gives primacy to the US Government to determine which instruments have a significant impact on US foreign policy and are published, and which do not. Was there a reason for this approach rather putting Congress in the driving seat from the beginning, in terms of setting criteria for publication?
Andrew Keller: I really appreciate that question. I would say—now that I am a private citizen—that one approach is that if you had a Congress and an executive branch that were aligned in trying to resolve these questions of scrutiny and transparency with regard to non-bindings in a mutually acceptable way, you could move much more into a zone where the executive branch gives the Congress almost all non-bindings. As you likely know from Curt’s article, it is very hard to define the class, but you could say that it was all non-bindings with qualities that are similar in some ways to binding agreements—it could be something along those lines—and you would have to give it to Congress quickly: within 10 days or two weeks. I think that would be doable.
The No. 1 issue was that there was no alignment between Congress and the executive branch for many months of that negotiation and it was very difficult. We recognised from a congressional standpoint that, regardless of what formula Congress landed on for the criteria of identifying non-bindings, there would be an element of discretion for the executive branch because that branch, in its role of implementing and executing the laws, has to be able to interpret and apply them in the way that it sees fit. So, No. 1, whatever we did, there would be that element of discretion for the executive branch.
The No. 2 issue, as we thought through the significant foreign policy impact standard, was that there was a question of whether we should try to define that standard in the statute. The feeling at the time was that that would have been too difficult to do, because it would have been too difficult to reach agreement with the executive branch on an appropriate list of factors. What Congress did instead was that the Senate Foreign Relations Committee published a report on the amendments to the Case Act that was contemporaneous with enactment of the new law. That report laid out what I think were fairly authoritative expectations for how Congress thought that the foreign policy standard should be implemented, as well as a set of factors that the committee felt the executive branch and, particularly, the State Department should include as considerations in implementing the statute.
Happily, the State Department then promulgated regulations that went out to the entire federal Government in the United States and that department included a set of factors that it expected it would use, and that other agencies should use, in implementing the foreign policy standard. Those factors were almost identical, or substantively identical, to what was in the Foreign Relations Committee report. That is really how Congress arrived at that aspect of the statute, but it is certainly not a one-size-fits-all thing. Had we had a different set of circumstances in the United States, we certainly could have landed on different formulae. I am sure that there are different formulae for the UK or for other countries that may be considering moving forward in this realm. Curt, I know you have thought through this as well, so you may want to jump in here.
The Chair: If you want to answer, please do.
Curtis Bradley: I have just a couple of thoughts, really quickly, to add to what Andrew said. I would emphasise that for any regulatory body for Parliament, the definition issue is serious. It certainly could have been done differently and, for the reasons Andrew said, probably could have been more expansive. But in the US, at least, there is no category in the executive branch called “non-binding agreements”; it is not a formal category. We all know that if we see one that looks like a treaty or something we would probably call it a non-binding agreement, but there is a range of commitments that the Government make through joint statements, or meetings and the like, that will not necessarily have the form of a treaty but are still non-binding arrangements or agreements, in a sense. It is a challenge to know how to define a category that does not include all diplomacy but captures things that, as Andrew said, are the sort of binding agreements that you would already want to have captured.
As Andrew said, the good news is that the State Department implemented a set of criteria similar to what the Senate had wanted. Those mention a bunch of factors that I think we would all appreciate are reasons why the legislature should know about it, but I have looked at the reporting and there is another piece of good news. It has only been since the October a year and a half ago that we have had experience on the reporting of these, and the Executive are reporting about as many non-binding agreements as binding ones. That confirms that this is a really important trend and is pretty frequent. It also looks to me—Andrew may have a different perspective—as if they are not reading that category too narrowly so far. These are agreements that are not necessarily momentous, but they have the form of what would otherwise be a binding agreement. For the most part, it looks like the Executive are sending those on to Congress and then publishing them.
The Chair: So it is workable—thank you very much indeed. Lord Anderson wanted to come in, although it has been covered a bit by what Professor Bradley just said.
Q32 Lord Anderson of Swansea: Professor Bradley referred earlier to co-ordination changes within the Executive, presumably following the passage of the National Defense Authorization Act. These changes within the Executive would include centralisation, presumably within the State Department, and the establishment of compliance officers within each of the regulatory agencies and departments. What can both of you say about the success of those changes?
Curtis Bradley: I will say one thing; I know that Andrew has some direct experience with this. When we looked at the problems with the original statute enacted in the 1970s, there was in particular lots of late or incomplete reporting to Congress. Probably the most significant reason we heard from the Executive about why that had happened was a lack of sufficient co-ordination across different departments and agencies. The State Department might sometimes not be informed by the Department of Defense or some other agency about an agreement until long after the fact, and then it would of course be late in reporting to Congress. That was not usually because they were trying to hide the agreements; it was just the difficulty of co-ordinating. Of course, the problem for Congress was: how do you try to reduce that problem when it is happening throughout the executive branch and it is not clear who exactly is to blame?
I thought that one of the more significant reforms—Andrew can talk about it—was to have more centralisation and co-ordination in the State Department, as well as more accountability at the agency level for when these agreements are made, so that we know who is in charge of them and who is responsible for sending them on to the State Department. If there is a problem, we now at least have a better sense of what went wrong inside the executive branch, where that was really a big difficulty for many years, but I know that Andrew had experience with this.
Andrew Keller: Thanks for that question. I agree with everything that Curt just said. I just add that, on that co-ordination issue, there was an effort to look at the whole life cycle of how the statute would work. For example, there was a requirement that the State Department promulgate a new regulation that would implement the 2022 amendments. There could have been ways to handle that other than a formal regulation, but that was done for two reasons. One is that we wanted to make sure it was clear that it was the State Department that would be engaged; the second is that there was a feeling that a formal regulation was needed to make sure that all the wrangling of the many other agencies involved in doing non-bindings could be successful.
Additionally, a requirement was included in the statute for ongoing consultation between the State Department and the congressional foreign affairs committees on the implementation of the Case-Zablocki Act, including the non-binding provisions. That was considered to be important just to make sure, even though there is often informal engagement between the State Department and Congress, there is a formal statutory hook for those conversations to continue regularly. Additionally, in the statute there are requirements not that each agency which negotiates a non-binding or a binding should provide the text directly to Congress; rather, they are required to go through the State Department with regard to that text, and then the State Department submits it on behalf of the executive branch.
Similarly, throughout the life cycle of either a binding or a non-binding instrument, it is often the case that the US Government will negotiate a framework agreement or arrangement and, then throughout that instrument’s life cycle, they negotiate various implementing arrangements. It was considered important to give Congress the ability to access those implementing arrangements and that was also done through the State Department. So, at every level, there was a conscious effort to create one focal point within the executive branch and to make sure that that focal point was the State Department.
On that latter side, it was not because Congress was trying to empower the State Department versus other agencies. It was, first, frankly because this effort was driven by the congressional foreign affairs committees and, secondly, because on a substantive level the State Department should have this type of involvement because it is the lead agency for international law, which is obviously relevant for binding agreements. It should also be an agency that is heavily involved in any non-binding that has a significant impact on foreign policy.
The Chair: That makes sense. Let me turn to Lord McDonald, to bring his own experience of this to bear on the question.
Q33 Lord McDonald of Salford: I have two questions, please. The answers may be short for different reasons. First, we have heard a lot about the content of the transparency provisions and the expectations of the State Department, but how do you each assess how these transparency provisions have operated in practice since the amendments entered into force?
Curtis Bradley: I am sure we each have thoughts about that and probably, relatedly, what might be improved if we could ever get more changes through. We are clearly in a better situation than we were before the changes. Why? For one thing, we found that before the changes less than half of the binding agreements were even being published, because the State Department had over time just decided not to publish various categories. We are now back to publishing a larger set. In addition, for example, the State Department has to indicate what legal authority it has for different kinds of agreements. It was not making any of that public before these changes. It is also pretty accessible; as I mentioned, when I checked online, I was quickly able to find the text of these agreements and that was not as easy in the past. There clearly are improvements.
Are there things that could be better? Yes. Because some agreements in the US really have to be approved by the legislature, when they are not I would like a little more explanation. The statute refers to a detailed description of legal authority; all they give right now is a citation and basically wishing us good luck to figure out what it means. As somebody who tries to scrutinise these issues, I would like a little more information.
Then, as Andrew mentioned, there is a huge carve-out for the Department of Defense and other parts of our intelligence and armed forces. We know through our work that many important non-binding agreements are being done in those parts of the Government. If they are classified, there are special ways of communicating those, by the way, that keeps them secret. For me, it is not clear why they should have a blanket exemption from some of this transparency. It is just one of those negotiating points that ended up being a compromise in the legislation, as Andrew mentioned. But we clearly have made an improvement. I think the hope—Andrew can speak to this—was that these changes would be made and then we would maybe revisit them as time goes on, to see whether there could be further improvements. That would certainly be my hope. Andrew?
Andrew Keller: Again, this is easy to do because I am in violent agreement with Curt on everything that he just shared. The heart of the 2022 amendments to the Case Act were about trying to get more reporting, meaning on non-bindings in addition to bindings, to Congress and to the public, and to do that quicker. That is happening and it is an enormous improvement. I agree with Curt that there are some things around the edges—not just the edges but things that are important—where compliance could be improved, including with regard to the statements of legal justification.
I agree—this is perhaps relevant as the UK or other countries think about changing transparency regimes—that there could be even faster reporting. The reporting is all done electronically, so I really do not see why publication, for example, would need to be after 120 days. It could be much quicker. I agree with Curt that, in any improvements, it would be important to revisit the carve-out for the Department of Defense.
There is a question about how to evaluate compliance with the significant foreign policy standard, because at the end of the day nobody in Congress or the public has full insight into the universe of non-bindings that the US executive branch is concluding. Without that baseline, while I think there has been reasonable compliance with that standard, it is very difficult to know for sure.
My last point is that, to address that issue, there is a statutory requirement for the US Government Accountability Office to perform an audit every three years through 2032 or 2033 to determine how well the executive branch is complying, or not, with the statute. The first audit is due next year, so if we were having this conversation in a year, we may have an even more fulsome response.
Lord McDonald of Salford: This leads me to my second question, which frankly is trickier. Both of you describe a familiar world: it is rather stately and quite slow but definitely mutually respectful. I know that the new Administration are less than four months old, but are there already signs that this familiar way of doing things is changing? Is it under threat or disappearing?
Curtis Bradley: I am glad that both of us are not in the Government at the moment so we can comment, I suppose. My main response is that it is too early to tell. As you mentioned, they are only four months in. I will say—I wrote about this online—that I have concerns about the El Salvador example, where it is not clear whether there is a binding or non-binding agreement. That is an important one and it ought to be disclosed. Members of Congress are trying to get a copy under the new provisions that Andrew described and there seems to be more resistance to information flow and sharing than we might have seen in the past.
That said, I checked the websites and, so far, they have been updated every month. That does not give us a lot of months under this Administration, but I have not seen a radical change yet on the regular reporting requirements that Andrew described. It may be that a few agreements, such as the El Salvador one, are the outliers and not what we will see as a routine matter of practice.
I have one other observation that goes back to an earlier question. You may have read about the agreement that the US made with Ukraine recently about mineral sharing and the like. We got that right away, but not through the Case Act. It is because Ukraine had a great interest in publishing it immediately on its website, and therefore we were able to digest it pretty quickly. That does not necessarily tell us what would happen if the partner country were quieter about it. I guess that it is too early to tell. I have some worries but I am hopeful that with the statute in place and the practice so far, it will continue.
Andrew Keller: I have exactly the same analysis. I add only that many laws run in the direction of Congress imposing different types of reporting requirements on the executive branch to report back to Congress on its activities. To a certain extent, all those laws rely on the executive branch engaging in good faith compliance. The Case-Zablocki Act does that as well and, so far, I think the data points are positive. My hope is that that continues.
Q34 The Chair: Thank you very much; this has been enormously interesting. I have one further question for you. It is perhaps particularly for Professor Bradley, but Mr Keller may have an observation. We have really been talking about the regulation of the use of non-binding instruments at the state level, as it were. Do you have any views on whether there would be value in somehow having international regulation of non-binding instruments? Does that seem something that could be valuable to you?
Curtis Bradley: I certainly think that there should be international study and discussion, and maybe even some kind of development of best practices. That is actually starting to happen: starting a year or two ago, the United Nations International Law Commission made non-legally binding agreements one of its big working projects. That is, in part, because we have seen a number of instances in which the parties to agreements did not have the same understanding of them, which was not on purpose. Sometimes you might imagine there would purposefully be some ambiguity about that, but there were times when it was just a mistake. The UK’s practices for what it suggests are non-binding are not exactly the same as those of the United States or other countries. More discussion about what you should be doing, if you wanted to make it non-binding, to avoid misunderstandings would be useful.
The other topic that is important for study in the future is that, even if an agreement is non-binding, it might still have indirect effects on international law, by creating customs, estoppel or other things in the law that would be less direct. More study and a kind of agreement about that among countries would help to avoid misunderstandings or friction between countries when they engage but do not see it the same way. I am not sure that I necessarily see the need for a treaty on this, but probably more—
The Chair: It could be a non-binding instrument.
Curtis Bradley: It may be another non-binding instrument, exactly. I love that and it is a good way to close that point.
The Chair: I want to thank you, on behalf of the committee, very much indeed for giving us that extremely useful evidence. The committee will go into private session so we will send you the transcript, as we have said, and you can comment on it if you will.
