« Justice Jackson's Concurrence in the Steel Seizure Case
Mike Rappaport
| Main | Originalism in the Blogs
Michael Ramsey


Arizona v. United States and Executive Policy Preemption
Michael Ramsey

Regarding my prior post on Arizona v. United States, reader Brett Bellmore comments: 

[T]he complication here is that there's this huge disconnect between federal law and federal policy when it comes to immigration. The law is popular, the policy not. So Congress can't bring the two into agreement by changing the law. OTOH, the policy is popular with major campaign contributors, so the President won't bring the two into agreement by changing the policy, and Congress won't force him to.

Because the law is popular, and the policy is to not enforce it, the states attempt to bolster it by writing laws which, far from conflicting with federal law, actually support that law. And the federal government objects because this conflicts with the policy. If they were actually concerned with the supremacy of federal law, it's the sanctuary cities they'd be going after, not Arizona.

So, IMO, what this case really boils down to is a test of whether the supremacy clause applies to federal law, or federal policy, when the two disagree. Or rather, which the Court is going to apply the clause to, because we both know which of those words is actually in the supremacy clause.

I think that’s a fair point, to the extent the argument against the Arizona law is that it conflicts with the President’s approach to enforcing federal law, rather than that it conflicts with federal law itself.  (I haven’t studied the laws closely, but my impression is that in the lower courts the Obama administration put a good bit of weight on the President’s policy). 

First, consider a situation in which there is no federal law, only executive policy.  My view (detailed here) is that as an original matter presidential policies have no power to displace conflicting state law under Article VI: the supremacy clause refers only to law; the executive is not a lawmaker; and it’s an aspect of separation of powers that if the President doesn’t like a state law, he has to go to Congress to get rid of it.  Unfortunately the Supreme Court has been of (at least) two minds on the subject.  In American Insurance Association v. Garamendi in 2003 the Court appeared to hold that a presidential policy in favor of settling Holocaust-era insurance claims (for a fraction of their value) preempted a California law favoring litigation of the claims.  (Justices Scalia and Thomas joined the dissent).  More recently, in Medellin v. Texas, the Court held that a presidential policy of complying with a judgment of the International Court of Justice did not preempt a Texas state law in conflict with the ICJ's ruling.  Chief Justice Roberts, writing for the majority, quoted Justice Black's opinion in the Steel Seizure case ("the President's power to see that the laws are faithfully executed refutes the idea that he is to be a lawmaker") and Madison's Federalist 47 ("The magistrate in whom the whole executive power resides cannot of himself make a law").  The basic point is that overriding state law is a legislative act, requiring an act of Congress (or a self-executing treaty).  Medellin had it exactly right on this point, I think, and substantially narrowed the implications of the Garamendi case. 

(For those with interest in this line of cases, the Ninth Circuit last week reheard en banc a closely related issue: the question is whether the President’s policy of not referring to the destruction of the Armenian community in Turkey as a “genocide” preempts a California state law that calls it genocide.  More here, including video, from Roger Alford at Opinio Juris.  My position on the case as an original matter should be obvious; it's outlined here.) 

The Arizona case is more difficult, though, because – unlike in Garamendi and Medellin – a federal statute is involved.  If the purpose of the federal statute is to give the President substantial flexibility and discretion in enforcement, then the state’s less-flexible approach likely does conflict with federal law and so is preempted.  The analogy would not be to Medellin but rather to Crosby v. National Foreign Trade Council.  In Crosby, a federal statute gave the President discretion in imposing sanctions on the Burmese government; the Court (rightly, I think) concluded that this law preempted a Massachusetts law imposing less flexible sanctions. 

But I’m not aware of language in the federal laws at issue in the Arizona case that gives the President enforcement discretion, in the sense that the law did in Crosby.  Rather, although I haven’t studied them closely, they seem not to contemplate flexibility in enforcement.  If that’s right, then as a general matter the state’s program to enforce them strictly doesn’t conflict with the federal laws themselves, only with the President’s policies.  And if that’s right, then the case is more like Medellin than Crosby.  Again, presidential policies not reflected in a federal statute shouldn’t displace state law, because per Article VI only superior law prevails over state law.  Of course, if particular aspects of the Arizona law conflict with federal law then they should be preempted.  And if this analysis is correct, the key to the Arizona case is (or should be) assessing what parts of the state law actually conflict with text and purpose of the federal law and what aspects are objectionable only because they interfere with the President's approach to the law.