« Managing the Vacancy Crisis: An Example of Resource Dependence in Federal Courts | Main | CFP: 2d Annual Ad Law New Scholarship Roundtable »

Friday, February 10, 2017

Do immigrants benefit from lack of due process? The pro-immigrant case for anti-immigrant procedures

Is the lack of procedural due process a bad thing for immigrants? Not necessarily: The more back-end process we afford to immigrants, the more reluctant Congress may become to authorize front-end admission. If it becomes too difficult to deport, or to deny re-entry to, a particular type of visa holder, then Congress may simply issue far fewer visas, leaving immigrants worse off. As Adam Cox and Eric Posner have argued, the American immigration system can be understood as essentially probationary: Congress lets a relatively large number of people into the United States on the theory that the President enjoys enormous discretion to kick them out or deny re-entry at the back-end. By a kind of political hydraulic pressure, the courts' imposing elaborate procedures at the back-end could induce Congress to increase enforcement resources ("build the Wall") or cut back on categories of visas, making more process an uncertain boon. In a thoughtful comment to one of my earlier posts, Chris Enloe, a former foreign service officer, made this case for expedited decisions: "...there is no way embassies and consulates are staffed for [elaborate hearings on visas]. And [such process] would result long term in [our] issuing many fewer visas in the first place....".

Enloe's point about ex ante effects is one reason why I remain agnostic about the value of judicially imposing a lot of procedures, let alone substantive limits, to protect immigrants from executive action. Fortunately, I do not believe that this worry about ex ante effects has much application to the Ninth Circuit's decision in Washington v. Trump. My reasons also go a ways toward answering some objections raised by Jonathan Adler and others to my earlier posts.

1. How a non-delegation canon and procedural due process limit the President's power under 8 U.S.C. Section 1182(f)

Although the Ninth Circuit's opinion rested mostly on procedural due process rights of aliens, its argument was difficult to follow because, despite DOJ's citation and discussion of Bimetallic on page 9 of their Reply brief, the Ninth Circuit ignored Bi-Metallic v Bd of Equalization and never explained how any individual could have a due process interest in participating in a legislative decision. Obviously, individual visa holders do not have some sort of procedural right to visit the White House and consult with President Trump before he issued his EO. As Jonathan Adler noted in response to my earlier post, the President is not an "agency" under APA section 551(1), and Bi-Metallic eliminates any constitutional right to participate in the making of such a legislative decision.

Moreover, if the process by which the EO itself was issued is constitutionally valid insofar as procedural due process is concerned, then individual aliens cannot complain that they are denied an individual hearing before CBP officers at the ports of entry. Jonathan is plainly correct that, under Heckler v Campbell, agencies need not provide hearings to re-litigate causal connections already determined by a prior legislative decision. Trump's EO states that, in order to prevent domestic violence, honor killings, and terrorist attacks, it is barring aliens from seven named nations from entering the United States. If President Trump was indeed authorized to draw such a causal connection by 8 U.S.C. Section 1182(f), then individuals cannot re-litigate the connection before an individual CBP officer at a port of entry. The only fact relevant to the application of Trump's EO is the alien's nationality, a fact that can be readily determined by the officer's inspection of documents.

So how can a general legislative rule like an EO deprive anyone of procedural due process? The argument that might sustain the Ninth Circuit's conclusion rests on the Non-Delegation Doctrine -- or rather the canon of statutory interpretation that bears its name. Under decisions such as Kent v. Dulles and the Benzene Case, broad congressional delegations to executive officers should be construed narrowly not to authorize those officers to impose extraordinary and possibly unconstitutional burdens. 8 U.S.C. Section 1182(f) on its face gives the President sweeping power to suspend the entry of aliens into the United States "[w]henever [he] finds that the entry of any aliens or of any class of aliens into the United States would be detrimental to the interests of the United States." If read literally, this power allows the President to sweep away visa holders'entry rights, protected by a dense network of complex statutory and regulatory procedures, at the stroke of a pen. If those aliens have a liberty interest in such statutory rights, then any congressional delegation allowing the President to toss aside such interests should, under Kent and Benzene, be narrowly construed. The President's EO, irrebuttably presumes that all aliens from a few countries are sufficiently prone to committing domestic abuse, terrorism, or honor killings that all visa holders from such nations can be barred from entering the United States. One can plausibly argue that such a burden on the liberty interests of such aliens is so dramatic that a court might find that some (judicially inferred) limit on the section 1182(f) delegation has been crossed. Such a decision is indeed "legislative" under Bi-Metallic -- too "legislative," because it comes up against the outer limits of the sort of power that Congress can constitutionally give away.

I will not repeat my arguments from earlier posts that statutes and rules defining substantive and procedural rights for visa holders create some sort of liberty interest. Landon v. Plasencia held as much for green-card holders, and I see no bright-line distinction between permanent residents and those visa holders who have been authorized by statute to make their home here for definite periods of months or years. Bracketing that question, it seems to me not a huge stretch to argue that, if non-immigrant visa holders have such a liberty interest, then EOs that trash their statutory rights need a special dose of justification under Kent v. Dulles and Benzene. If Congress should be presumed not to give OSHA the power to burden petroleum companies with a lot of costly safety precautions that produce small health benefits, then Congress equally should not be presumed to confer on the President a blank check to sweep aside visa holders' access to their jobs, homes, families, and schools, when such access has been elaborately defined by procedures and standards ratified by Congress itself. That network of process and standards creates expectations that we should presume Congress wanted to safeguard even as they gave the President otherwise apparently unlimited power over entry.

2. Why limiting the President's section 1182(f) powers with visa holders' procedural rights will not likely induce Congress to limit visas

This brings me back to the point with which I began the post: Will limiting section 1182(f) with these sorts of procedural rights cause Congress to cut back on the issuance of visas to compensate for the extra processes that the court is imposing on the executive?

The big advantage of such procedural limits is that they do not spook Congress into limiting the issuance of visas by disturbing the statutory status quo. Congress has already created a system conferring enormous discretion on CBP officers to review visas and deny entry to visa holders suspected of violating their visa's terms. (Ask any student holding an F-1 visa who has shivered in a port of entry's conference room waiting for "secondary inspection"). There is no reason to believe that Congress applauds Trump's trampling through this network of rules and customs like a T-Rex trampling through Jurassic Park: Congress has never relied on any history of Presidents' using EOs to sweep aside hundreds of thousands of existing visa holders' statutory rights. Protecting visa holders' procedural rights likewise sidesteps tougher questions about (for instance) Article I habeas rights of immigrants who lack prior connection to the United States By limiting the President's section 1182(f) powers with merely procedural rights, the courts would also preserve the President use crude proxies -- even nationality-based proxies -- just so long as the EO preserved some opportunity for the visa holder to make an individualized case that their entry does not present any of the risks that the EO was intended to address.

The procedural due process argument defended here is, in short, is a perhaps disappointingly minimal entitlement. Call it a "defensive crouch" of an immigration libertarian in honor of Mark Tushnet. Given Congress' power practically to eliminate any more robust entitlement, however, a defensive crouch might be precisely the right posture for courts to assume.

Posted by Rick Hills on February 10, 2017 at 06:05 PM | Permalink


This is an interesting and at least doctrinally plausible proposal, although I don't know if you're still in procedural due process land anymore. One concern I have though. On reading the post I went looking for something to read about the Benzene Case and very happily came on Scalia's "A Note on the Benzene Case," which he wrote in his AEI days. Scalia, unsurprisingly, is very critical of the plurality's reading of the statute, and at this stage in his career is more sympathetic to judicial enforcement of non-delegation than he'd later become. One point he makes in defense of a real non-delegation doctrine, as opposed to non-delegation avoidance as practiced by the Benzene plurality, is that actually invalidating vague delegations would reduce judicial policy-making power, rather than, as one might think, enhancing it by freeing them to invalidate statutes they don't like under a vague doctrine. The reason being, he argues, that it's ultimately courts that get to pass (at least in the pre-Chevron world in which he's writing) on the meaning of vague delegations. And here the example he gives is the Benzene plurality itself, which in the guise of avoiding a delegation to OSHA and finding some constraining principle in the statute, really just "ended up doing legislator's work" instead of letting OSHA do it.

I think that's right, and is right about delegation-avoidance generally (or proposals to overrule Chevron); if a court avoids finding a delegation by saying a statute contains some determinate standard it probably doesn't contain, it's really just redirecting the delegation to itself at the point when it crafts the standard, and I'm inclined to think that's not a good thing because as between delegates of unguided discretion, why choose the courts? So, I'm also inclined to think that if the courts aren't going to strike down 1182(f) altogether, finding some "judicially inferred" limit on the President's discretion, or perhaps claiming that 1152 limits 1182 by way of reaching for some semi-plausible off-the-rack statutory limit (which is pretty close to what happened in Benzene, and which I think in some ways would be much worse than simply creating a limit themselves), aren't great alternatives. Besides invalidating this particular EO, what sort of limit could they infer? That the "classes" of aliens 1182 contemplates the President excluding can never be defined as broadly as a country of origin absent hostilities between ourselves and the country? (If that were the rule, by the way, could we still exclude Syrians on the ground that we are at war with a state that controls part of Syria?) Some judicially reviewable risk threshold? What remotely qualifies the courts to decide on that threshold, or review its application in particular cases? Besides invalidating the EO, which would be good policy, why would we expect the limit they arrive at to be good policy?

I also should add, though this isn't at all my field, that I'm a little skeptical there's a serious delegation problem to avoid here, given that we're talking about immigration and national security, where unguided presidential discretion seems a lot less exceptionable than in workplace safety. At least, reading 1182(f)'s "detrimental to the interests of the United States" to authorize exclusion of aliens from countries that contain significant numbers of terrorists doesn't strike me as an overbroad delegation, something that Congress couldn't permissibly authorize if it said just that; perhaps I would be on board with you, even as a purely statutory matter, if the President read "detrimental to [our] interests to authorize exclusion of a country's aliens on the ground that people from that country, in his view, commit a lot of garden-variety violent crime or are too good at math. Anyway, I've really enjoyed this series of posts, and here's the Scalia piece:


Posted by: Asher Steinberg | Feb 11, 2017 2:03:18 PM

Asher, when I teach Benzene, I always emphasize the point that non-delegation canons slight one purpose of the non-delegation doctrine (forcing Congress to make tough policy calls) in favor of another purpose (limiting the policy-making power of the executive branch). One cannot maximize two variables simultaneously, but the choice to cabin executive power often seems attractive judges, and not (just) because they might be empire-builders (a criticism I rather doubt: most judges are pretty risk averse and even lazy). Grants of power like section 1182(f) are so ridiculously broad that they just rub people steeped in a certain brand of Madisonian separation of powers the wrong way.

Richard Epstein offered another narrowing reading of section 1182(f) over at Ricochet (I don't have the URL handy that was more focused on the text of "any class." Whatever the particular rationale or route, all such strategies start from the presupposition that no sensible republic would hand off such immense discretion to a single person, except in dire emergencies or in discrete areas where the republic's constituents are insulated from abuse. The rights of visa holders to re-enter the USA is not one of those areas.

So when judges see such a blank check, my guess is that they will, by hook or crook, try to put an upper limit on the amount that can be entered in the box. This is especially true if the grant has been used for a half-century in much more narrow ways by previous Presidents. That's the gloss of history to which Frankfurter refers in his Youngstown concurrence. Taking this sort of language literally is just not sensible.

In writing the above, I am aware that New-Baconian royalists of the Adrian Vermeule variety might question the premise. I am just banking on the rarity of that type -- and a very fine, upstanding, intellectually rigorous type they are, btw, although I am not of their number -- on the federal bench.

Posted by: Rick Hills | Feb 12, 2017 9:32:46 AM



Posted by: Marty Lederman | Feb 12, 2017 5:37:14 PM

Post a comment