Executive Agreements, Iran, and Missouri v. Holland

So. A bunch of Republican Senators have apparently written an "open letter" to the government of Iran regarding negotiations currently taking place between Iran and various Western countries, including the United States. This is, shall we say, highly improper. It would rather easily answer to the description "unconstitutional," in a vague, amorphous sense (obviously there's nothing for a court to review or invalidate, but the letter is pretty clearly inconsistent with an overall constitutional structure in which the President is given primary responsibility for diplomacy, out of a felt need for the nation to speak with one voice in foreign affairs). Arguably it also at least skirts the line of being illegal, under the Logan Act of 1799, which bans unauthorized citizens from corresponding with foreign governments to try to influence their behavior toward the U.S. (Although obviously the Obama Administration is not going to prosecute these Republican Senators for this offense, even if they're guilty of it: prosecuting your political opponents, generally not great politics. Also the law is of dubious constitutionality (and they could easily have mooted the whole issue by just reading their "open letter" into the Congressional record, invoking the Speech and Debate privilege, which is absolute).)

But that's not the point, of this blog post at least. Rather, what interests me about the letter is the way it helps tie together some thoughts I've been having of late about "executive agreements." Because that's the main thrust of the letter, which purports to explain certain "features of our constitution" which Iran should keep in mind while negotiating. By this they mean the esoteric fact, which surely the leaders of Iran have never encountered before, that the United States has the world's worst treaty ratification system, requiring a two-thirds vote in the Senate that is basically impossible to obtain even on totally uncontroversial things. This means, the Senators accurately note, that any agreement reached in the present negotiations will be a mere "executive agreement," which, they note, a new President could revoke unilaterally. This is, if you take it on its face, a pretty majestic display of concern-trolling: it's not that the Senate Republicans are opposed to the deal (which would make it a little tougher to deny the Logan Act violation), they're just concerned for the poor Iranian government, which might find itself being misled by that nasty Obama into thinking it has a binding agreement, and might do something like act in reliance on that belief, only to find that, in fact, the next President pulls the rug out from under them. We wouldn't want that, now would we.

But of course, they are right: the deal would only be an executive agreement (because we all know that these same 47 Republicans would vote against ratifying the treaty that will never be drafted, let alone presented to the Senate, before they even read its language (which is why it will never be drafted)). So... what? What exactly are executive agreements, in the constitutional structure? How are they different from treaties? Are those differences relevant from Iran's standpoint? (Spoiler alert: the answer to that last one is "no.") Like probably most people who've ever heard the phrase, I first learned about "executive agreements" in my high school civics class (okay, it was AP U.S. Government, which might explain why we learned about executive agreements...), and have always had the general sense that they're just this thing, this informal treaty-like thing, that doesn't require Senate ratification but also therefore in some vague way doesn't have the same kind of force of law. That way of discussing them leads to the feeling that there's something vaguely unsettling or illegitimate about them. But there isn't, and seeing why requires merging the great big lesson of a class I'm taking this semester, Administrative Law, with the great case Missouri v. Holland.



To know what an executive agreement is, we need to know what a treaty is, and particularly how a treaty comes to have that mysterious thing known as the force of law. That is to say, okay, so the Senate has ratified a treaty (by some miracle); now what? How does that treaty go about changing the way the United States government functions, and/or imposing new rights and duties on private parties? There are a number of possible answers. The treaty may be "self-executing," in which case it does those things the same way an ordinary Act of Congress does. These treaties do more than just create international obligations (which all treaties do, inherently); the very act of their having been ratified and signed makes them law, domestic law, in exactly the same way as every other federal law. Not many treaties are self-executing like this, and courts tend not to treat them as being unless they say very explicitly that they are. I used to be skeptical of this but there's good reason for it, which we can see when we consider the next route to attaining the force of law: an implementing statute. After the treaty has been ratified, creating an international obligation but no corresponding change in domestic law to bring about compliance with said obligation, Congress might pass an ordinary statute implementing the treaty and complying with the obligation. The nice thing about an implementing statute, and the reason why we don't like to read treaties as being self-executing, is that it gets the House of Representatives involved; ratifying the treaty in the first place only took the President and the Senate, and it's just a little weird to create domestic federal law without getting both Houses of Congress to sign on.

The final possibility is executive or administrative action. Now, either a self-executing treaty or an implementing statute will likely prompt a number of actions by the executive branch to, y'know, execute them, say by no longer the thing the treaty and/or the implementing statute forbids. But in addition to any such "ministerial" executive actions compelled by the treaty or statute, there may be other things the executive could do to implement the treaty--perhaps even in the absence of an implementing act! Suppose, for instance, that X is a federal crime, but recently-ratified Treaty A forbids the criminalization of X (and Treaty A is not self-executing). Well, X is still a federal crime, nothing about the treaty has changed that if Congress hasn't yet passed an implementing act, but that doesn't remove the international obligation not to criminalize X. But! The President can probably implement this treaty all by himself, by simply declining to prosecute anyone for X, and maybe also by pardoning anyone who's been convicted of X already. That's within his lawful authority, even without an implementing statute; hell, that authority comes straight from the Constitution, and it's doubtful Congress could stop him if it tried. He could have done those things in the absence of a treaty. The treaty just gave him a reason to use the powers he already had in a certain way.

And that's where the central lesson of administrative law comes in: there's an awful lot of stuff that Presidents already have the lawful authority to do. Every U.S. President just walks around at every moment of every day imbued with an amazing amount of policy-making power, most of it basically delegated by Congress to some administrative agency within the executive branch. (Current law and Supreme Court doctrine conspire to keep some of these agencies outside of the President's very direct control, probably unconstitutionally. In theory we might as well just assume a unitary executive here, where the President exercises all executive power, and all delegated legislative authority from Congress.) And the President is actually constantly exercising some tiny portion of this authority, usually through his (many, many) deputies and subordinates. As with the pardon and nonprosecution powers, this vast reservoir of administrative policy-making authority gives the executive branch a lot of tools that it could use in a lot of different ways. Some of those ways might happen to bring the United States into compliance with some international obligation, but that's not why the President can do them. He can do them because American law empowers him to do so. Of course there are things the President can't do unilaterally; probably in most instances he can't make something a crime, unless that thing is plausibly read into some prohibition in a general statute like the Securities Act or whatever. For those things he needs a statute, or a self-executing treaty, at the very least.

And this last point lets us see what an executive agreement is, at a very fundamental level. It's an agreement with some foreign leader that the President will use this vast reservoir of executive and administrative power in just such a way as to happen to comply with the international obligations created by the agreement. And there's a lot of room to do that! So no wonder executive agreements have been pretty much up to the task of replacing formal treaties: it turns out that the executive, in the modern administrative sense of that word, is pretty much the key figure to get on board if you want to shape American public policy.

Now let's consider the other two possible sources of legal authority for a treaty. One of them is plainly lacking in the informal context: there's no way for an executive agreement to be "self-enforcing." As a matter of domestic law, the agreement itself is a nullity;* the executive and administrative actions taken that just so happen to implement it, however, are not nullities. But if there happens to be some tiny little corner of public policy the President doesn't currently have authority to reshape by executive action, he can't acquire that power simply by reaching an executive agreement--whereas he would acquire that power if he got a self-executing treaty ratified by the Senate, or an implementing statute passed by Congress.

Which brings us to implementing statutes. Can Congress pass a statute implementing an executive agreement? Of course! Congress can pass any statute it wants--and the President is hardly likely to veto it. A President would be unwise to promise a foreign leader that he could get an implementing statute from Congress, unless the political calculation is quite a simple one (though the same is true with treaties), but if by some chance Congress wants to cooperate then it can bake the executive agreement into the U.S. Code and, if necessary, augment the President's authority to implement it beyond what the reservoir of administrative power already provides. If, then, two Presidents of the United States, in two parallel worlds, announce on the same day that they've reached a landmark agreement with some foreign nation, but one of them announces an executive agreement while the other proposes a (non-self-executing) treaty to the Senate, which ratifies it, well, Congress is equally at liberty in both worlds to pass a statute implementing the agreement and bringing the country into compliance with its international obligations.

Almost equally.

And this is where Missouri v. Holland comes in. Because, you see, Congress can't pass any statute it wants. It can only pass those statutes it is empowered to pass by the Constitution, which, in addition to placing a number of express restrictions on governments (e.g. the First Amendment's prohibition of laws violating freedom of speech or religion), provides Congress with only limited, enumerated legislative authority in the first place. If a law does not fall within one of the eighteen grants of power in Article I, Section 8, or within one of the grants of power in the various amendments, well, Congress just can't pass it. Which led, as it must, to the following rather curious sequence of events early in the 20th century:

Congress passed a Migratory Birds Act, regulating, well, migratory birds; I think it was an early sort of Endangered Species Act for a particular kind of endangered species. A U.S. District Judge declared this law beyond Congress's enumerated powers and hence unconstitutional. Now, this was probably wrong. No, let me rephrase that: this was wrong. But rather than try to appeal the decision, the government did something clever: it hammered out a treaty with the United Kingdom (i.e. Canada, whose foreign relations the U.K. still handled at the time) creating a framework for the conservation of migratory birds in North America, and then, after ratifying the treaty, passed the Migratory Bird Treaty Act of 1918, implementing the treaty. This was very similar to the earlier Act. Missouri, as apparently seems to be its wont, complained that this violated federalism and states' rights and the Tenth Amendment and all that, apparently on the grounds that, per the earlier lower-court decision, the states owned the migratory birds that happened to be within them, and Congress had no power to legislate otherwise.

Justice Oliver Wendell Holmes begged to differ. Congress did have power to pass the Act, he said, even if we accept the view that the earlier act had been unconstitutional. Because it had the power to implement the treaty with Great Britain, or more specifically to "make all Laws which shall be necessary and proper for carrying into Execution the foregoing Powers, and all other Powers vested by this Constitution in the Government of the United States, or in any Department or Officer thereof." Now, this case has been viewed by many as holding that a treaty can violate the Constitution, which it most emphatically does not say. What it says is simply that Congress can legislate to implement a statute, under the Necessary and Proper Clause, even if it would not have the authority to so legislate in the absence of a treaty. Support for this is drawn from the Supremacy Clause, which makes "[t]his Constitution, and the Laws of the United States which shall be made in pursuance thereof; and all treaties made, or which shall be made, under the authority of the United States ... the supreme law of the land."

Note the difference in wording: ordinary federal laws must be made "in pursuance [of]" the Constitution, but treaties are made simply "under the authority of the United States." The idea here is that the treaty itself does not need to have a basis in one of the enumerated powers of Congress, since it is not an action of Congress; it is an action of the entire country. But once the treaty is in place, implementing it becomes one of the things that Congress can do under its power to make all laws necessary and proper for carrying into execution every other government power--including the treaty power. Hell, they're called implementing acts, a pretty good shorthand for a "law which shall be necessary and proper for carrying into execution" a treaty. And because both the treaty and the implementing act are valid for Supremacy Clause purposes, they supplant any state law to the contrary, e.g. Missouri's claims to ownership of the birds. Nothing in the Tenth Amendment, which reserves to the states** all powers not delegated to the national government, says otherwise, because the treaty power was given to the national government.

So these implementing acts don't violate the Constitution, tho' they might have without the treaty, because the treaty is constitutionally authorized and, once ratified, provides an alternate basis for Congressional action (much as the later amendments have supplemented the grants of power in Article I, Section 8). Note that this says nothing whatsoever about a treaty or an implementing act that violated one of the express prohibitions on government. The (hypothetical) 1951 Act implementing the 1950 Suppression of Communists Treaty with France, Britain, and West Germany by criminalizing all advocacy of Communist ideas would violate the First Amendment; a hypothetical act implementing a treaty declaring that some specific person should be imprisoned by, well, declaring that that person was to be imprisoned would violate the Bills of Attainder Clause. And this is so even under Missouri v. Holland: it is true that in these cases, Congress is acting within the scope of its enumerated powers (well, sort of), but even when acting within its enumerated powers it cannot violate the First Amendment or pass a bill of attainder. (The qualifier signifies that I'm not sure these treaties would themselves be legitimate, since they would oblige the United States to violate the express prohibitions of its constitution. But that's sort of a side issue.) So you see, it's really quite a simple case, once you understand how the Constitution works.

But it's also completely inapposite if we're dealing with an executive agreement. Congress, as noted above, may pass laws implementing executive agreements just as it may pass laws implementing treaties. Except here's the thing. Treaties, which are made under the authority of the United States, are the supreme law of the land, and implementing them is a good enough reason all by itself for Congress to pass a law. Executive agreements are, as noted above, a nullity. They are not the supreme law of the land, and therefore they don't do one whit to add to Congress's legislative authority. So there's our second difference: executive agreements cannot be self-enforcing, and they also cannot provide the basis for their own implementing acts. Of course, limiting Congress to its otherwise-enumerated powers has almost no practical effect, since with just a little creative use of those powers Congress can do basically anything. (The Court recently emphasized and reaffirmed this fact in a little-known case when it held that Congress can very specifically use its taxing power to create a de facto regulatory regime that it could not erect directly using its power to regulate interstate commerce. Oh, wait--the case wasn't little-noticed, it was the ObamaCare case. But this aspect of it did somewhat escape notice, perhaps even by Chief Justice Roberts.)

But as a conceptual matter, these are the differences between executive agreements and treaties. These two features, the power to be self-executing and the power to invoke the Missouri v. Holland rule, together define what it means in our constitutional system to say something is a treaty. And... doesn't that kind of make sense? I mean, is it an accident that the vote in the Senate to ratify a treaty, two-thirds, is the same as to pass a constitutional amendment? Not if the main thing we're doing is creating an independent predicate for Congressional legislative authority. That almost is like a little mini-amendment. And it's two-thirds of the Senate, the organ of the federal government most responsive to the states (though far less so than prior to the 17th Amendment). And everything else that could ever be part of an international agreement is, well, just sort of routine, as a constitutional matter. Hell, in a way it's the creation of federal law without the involvement of the House that's the strangest thing, constitutionally speaking, not the Missouri v. Holland principle. But if an agreement doesn't involve either of those constitutionally unusual--which is not to say unconstitutional--practices, then there's absolutely no reason not to let it be an informal agreement between heads of state.


Oh, and as to that other concern the Senate Republicans raise, that a future President could go back on the deal? Well guess what: that's true of treaties, too. Or at least, it might be. We think it is. No one is quite sure. Back in 1979 there was a case, Goldwater v. Carter, where Jimmy Carter had unilaterally abrogated a certain treaty that had recognized the Republic of China, and instead recognized the People's Republic of China. A bunch of members of Congress sued, saying that the Senate needed to be involved in breaking a treaty just as in making one. The Court refused to rule on the issue, which is to say it let Carter's actions go forward. The case is oft cited for the proposition that this issue, of unilateral Presidential treaty abrogation, is a "political question" into which the courts should not and must not intrude--which, again, would be to say that, as a matter of law, Presidents could do that. But technically the case doesn't say that, it only says that, because the people suing weren't "Congress" but rather just "Barry Goldwater and friends" (almost reminds one of Tom Cotton and friends), well, it's actually a little unclear what the doctrinal consequences was. One way of putting it is that Barry & Co. didn't have standing to challenge Carter's actions; another is that, because Congress hadn't even made up its formal mind yet about whether it disliked what Carter had done, the "political question" of whether Carter's actions had been right had not yet crystallized, ripened, into a proper constitutional dispute between the different branches.

In any event, it remains a theoretical possibility that the Court could rule on this issue some day, but the general sense is that it won't/wouldn't, and that therefore, this is a thing Presidents can do, at least in the sense that no one will stop them. And I'm not sure I think the President doesn't have this authority. After all, only Congress can declare war (yes, yes, I know, ha ha, how quaint), but the President can at any time end a war. And in the Carter case, for instance, the President has the power to receive ambassadors, commonly thought to imply a broader foreign policy and diplomatic power that, at the very least, encompasses the right to recognize one country or not recognize another. Surely that means the President has the unilateral authority to recognize the PRC instead of the Republic of China, right? At the very least it's suggestive, and I don't think a court would be likely to hold otherwise any time soon. Of course, Carter's actions likely put the U.S. in violation of its international obligations, as would Congressional refusal to pass an implementing statute if a treaty could not be effectuated administratively. As, for that matter, might state refusal to change state laws if, say, we somehow ratified a treaty (or even maybe just made an executive agreement) to abolish the death penalty, but Congress didn't pass an implementing act abolishing it at the state level. There are lots of ways for us to violate our international obligations, and the question of which organ of the American government is responsible for seeing that we don't (or that we have good reason when we do) is one of American constitutional law. In the abrogating treaties case, that seems to mean the President, which means that Iran would be no more secure with a formal treaty than with an informal agreement with Obama.

Oh, and the Iranian Foreign Minister seems to know all of this, heheh.



*Well, basically a nullity. I do have one idea for how it might not be, though I don't know if anyone with any actual political power, or even with a platform in a law journal, has ever suggested this. There's a basic standard that governs all administrative actions, basically, which is that they must not be "arbitrary or capricious." Courts read this as meaning that, when an administrative agency promulgates a new regulation through "notice and comment" proceedings, it must respond to any valid concerns about the rule raised in the comments, enough to satisfy the court that it had good enough reason for implementing this rule as opposed to some other rule, or no rule at all. And an issue arises when an agency wants to change a rule, which usually happens because there's a new President in town with a new ideological orientation. Is that simple political change good enough reason to change the rule? There are reasons to think it ought to be, if the new rule would have passed the arbitrary and capricious test had it been writing on a blank slate, but the Courts don't openly accept it as such.

By analogy, it seems to me that "we need to make this rule to fulfill our international obligations" ought to be a valid reason for arbitrary and capricious review--which, note, only comes into play once we've determined that the regulation is within the agency's statutory authority to issue. So this would not in any way undermine the requirement that executive agreements be implemented with tools already at the President's, or at least the executive's, disposal. It would just make it, I should hope, virtually impossible to challenge the President's choice to use the tools at his disposal to comply with our international obligations.

And I also feel like there's at least a bit of a case to be made that, under the same arbitrary or capricious standard, a later President's decision to undo the executive actions implementing the original agreement should be subjected to a heightened standard. Because there is automatically a very strong argument against taking those new actions: they will put us in violation of our international agreements! Now, Presidents certainly do have the authority to abrogate executive agreements, so this wouldn't be an absolute bar on changing course, and I'm not even sure I think there should be a heightened standard at all. But it's a way that executive agreements could be a little more than a nullity as far as domestic law is concerned, and could actually affect what things Presidents did or didn't have the authority to do. (Well, sort of: supposedly arbitrary and capricious review isn't about whether the authority is there to do the thing, but only whether there were decent enough reasons for using the authority in that way. But it adds up to much the same thing, in the end.)


**Or to the people. It's easy to forget that part of the Tenth Amendment, and it's by far the most conceptually interesting part.