An Eventful Thursday for Immigration Law at the Supreme Court: United States v. Texas, Mathis v. United States, and What’s Next

On Thursday, June 23, the U.S. Supreme Court issued two decisions of significance to immigration law: a 4-4 affirmance without opinion in United States v. Texas, and a 5-3 decision in Mathis v. United States.  The first, which was more obviously immigration-related, is very disappointing and has rightly received a great deal of media attention, but the second is also worth noting and is somewhat more positive.

The Court’s evenly-divided decision – or one might say lack of decision – in United States v. Texas left standing the previous 2-1 decision of a panel of the Court of Appeals for the Fifth Circuit, which had upheld District Judge Andrew Hanen’s preliminary injunction against Deferred Action for Parents of Americans and Lawful Permanent Residents (DAPA) and against the related expansion of Deferred Action for Childhood Arrivals (“DACA+”).  This is quite a momentous outcome to have been reached without explanation.  As former Solicitor General Walter Dellinger has written: “It is hard to know what to say about an immigration opinion affecting 4.3 million people that reads, in its entirety: ‘The judgment is affirmed by an equally divded Court.’  Seldom have so many hopes been crushed by so few words.”

It has long been customary for an evenly divided Supreme Court to affirm the judgment below without offering opinions, as was done here (and has been done since at least 1909), although this is not a custom followed by all multi-member appellate courts in the United States.  The Court of Appeals for the First Circuit sitting en banc, for example, has provided opinions explaining the views of its judges when it has divided 3 to 3 in recent years and so affirmed the district courts below, as in the immigration detention case of Castaneda v. Souza, on which the First Circuit split in December 2015.  If the Supreme Court were to adopt a similar custom, one would at least have the satisfaction of knowing the reasoning behind the Justices’ votes.  In this particular case, one might also have hoped that some of the Justices who voted to affirm the Fifth Circuit could have been convinced to change their minds by a compelling dissenting opinion that they knew they would have to confront publicly, although presumably draft opinions were circulated internally, given the long lapse of time between oral argument in the case on April 16 and the issuance of decision last week.  The 4-4 deadlock, and the Supreme Court’s custom of not issuing opinions in that scenario, has left those Justices who voted to affirm the Fifth Circuit in the position of being able to do so without having to explain formally and publicly why such a position is legally coherent.

While it is frustrating that the injunction in United States v. Texas was affirmed without explanation and without any precedential decision, however, this does have the benefit of leaving the door open for a different outcome in the long run.  United States v. Texas could return to the Supreme Court once a 9th Justice is seated on the Court, and potentially be decided differently, in one of at least two ways.

First, as SCOTUSBlog pointed out soon after the 4-4 decision came down, the government can petition the Supreme Court to rehear the case, and ask that the petition be held until a 9th Justice is seated on the Court.  Former Solicitor General Dellinger also endorsed that approach in his above-quoted post at Slate’s “Supreme Court Breakfast Table”.  This would be one way for U.S. v. Texas to come back before the Supreme Court, potentially quite quickly after a 9th Justice is seated.  Under Supreme Court Rule 44, a petition for rehearing ordinarily will only be granted both “by a majority of the Court” and “at the instance of a Justice who concurred in the judgment or decision”, but it is unclear how this latter requirement could possibly be applied in the case of a 4-4 affirmance without opinion, where the Court has not issued its own judgment and there is no public record of any Justice concurring in the affirmance more than any other Justice.  Thus, it appears that an ordinary majority, presumably composed of the 4 Justices who voted to reverse plus a newly arrived 9th Justice, could grant a petition for rehearing if it were still pending when a 9th Justice were confirmed.

SCOTUSBlog also noted today, however, that the Court had denied one petition in another case seeking such rehearing by a 9-member Court, which may not bode well for the rehearing possibility.  Specifically, the Court denied rehearing in Hawkins v. Community Bank of Raymore, which had been the subject of an affirmance without opinion by an equally divided Court on March 22, 2016.  One might think, though, that there is a significant difference between a case like U.S. v. Texas, decided just before the end of the term and affecting national policy so substantially, and a case like Hawkins, in which the Court failed 3 months earlier to reach a conclusion regarding the questions “(1) Whether “primarily and unconditionally liable” spousal guarantors are unambiguously excluded from being Equal Credit Opportunity Act (ECOA) “applicants” because they are not integrally part of “any aspect of a credit transaction”; and (2) whether the Federal Reserve Board has authority under the ECOA to include by regulation spousal guarantors as “applicants” to further the purposes of eliminating discrimination against married women.”  Without meaning to minimize the significance of discrimination against married women seeking credit, one might reasonably suggest that this sort of garden-variety issue of statutory interpretation does not call for unusual procedural measures to achieve a final, reasoned resolution to the same extent as the issue of the legitimacy of DAPA and DACA+.

Even if rehearing is not granted, it is likely that U.S. v. Texas may ultimately return to the Supreme Court, unless it is rendered moot in the meantime by Congressional enactment of comprehensive immigration reform or recission of DAPA by a hypothetical Republican President.  (Hillary Clinton, the presumptive Democratic Presidential nominee, has made quite clear that she would seek to defend and expand administrative relief such as DAPA, not rescind it.)  The injunction issued by Judge Hanen and upheld by the Fifth Circuit was a preliminary injunction, and the case would now ordinarily be expected to proceed to a trial on the merits, or at least some sort of further proceedings.  At the conclusion of such further proceedings, Judge Hanen may then issue a permanent injunction.  The grant of such a permanent injunction could be appealed back to the Fifth Circuit, and if it were again affirmed by the Fifth Circuit, the government could seek certiorari from the Supreme Court regarding that affirmance.  One hopes that by the time the case worked its way back up through the Fifth Circuit to the Supreme Court in this fashion, there will be a 9th Justice seated on the Supreme Court.

There has also been some speculation that a new case regarding DAPA and DACA+ could be commenced, in a Circuit other than the Fifth Circuit, which might come out differently and create a circuit split.  It may be unlikely that such other litigation, even if deemed feasible, would make it back to the Supreme Court sooner than one of the other two routes discussed above.  If there is a way to launch such other litigation despite the current national scope of the injunction against DAPA and DACA+, however, it could have other benefits: David Leopold, for example, suggests in a recent blog post that further litigation might allow DAPA and DACA+ to go into effect in portions of the United States even if not nationwide.  Achieving such a goal would be difficult, given the current nationwide injunction against DAPA and DACA+ issued by Judge Hanen and upheld by the Fifth Circuit, but it appears that some intelligent and creative attorneys may be looking to see if they can find a way.

Ultimately, however, it appears that the future of DAPA and DACA+ will likely depend on who (if anyone) fills the currently vacant 9th seat on the Supreme Court.  In this, as in many other things, the outcome of this November’s elections will be crucial.

If Hillary Clinton is elected President and the Democrats retake the majority in the Senate, the 9th Justice who would be confirmed, whether that is President Obama’s nominee Chief Judge Merrick Garland of the Court of Appeals for the D.C. Circuit or a new nominee put forward by President Clinton, would likely vote to overturn the injunction against DAPA and DACA+ if and when the case returned to the Supreme Court.  In the tragic event of a Donald Trump Presidency, on the other hand, the issue would be moot, since DAPA and DACA+ would be rescinded anyway.  In the event that Hillary Clinton is elected President and the Republicans maintain control of the Senate, one hopes that they would not attempt to block a Supreme Court nomination indefinitely, but given the current behavior of the Senate Republican majority, one cannot be sure.  Therefore, a Democratic victory in not only the Presidential election, but also a sufficient number of Senate elections to reclaim the majority (requiring a net gain of at least four seats), would give the best hope for a revival of DAPA and DACA+.  If the Democrats can also regain the majority in the House of Representatives, then the issue of DAPA and DACA+ could be rendered moot in a much more pleasant way: comprehensive immigration reform, along the lines of the Senate CIR bill S.744 that was passed by the Democratic Senate with a bipartisan majority in 2013 but denied a vote in 2013 and 2014 in the Republican-controlled House of Representatives, could become law.  Hopefully, many of the U.S. citizen relatives of those who would be affected by DAPA, DACA+, or comprehensive immigration reform will be motivated by these possibilities to turn out and vote in November.

Until comprehensive immigration reform becomes law or DAPA and DACA+ come into effect, however, immigration attorneys will remain on the lookout for other small pieces of good news where we can find them.  The Supreme Court’s decision Thursday in Mathis v. United States, while at first glance not about an immigration case at all, provided just such a piece of good news for noncitizens with certain types of criminal convictions.  (Since most significant criminal convictions would have precluded applying for DAPA and DACA+, the set of noncitizens who will benefit from Mathis has very little overlap with the set of those harmed by U.S. v. Texas, so it may only be from the perspective of immigration attorneys that Thursday was something of a good news / bad news situation; hopefully I do not strike the reader as too insensitive for examining the two decisions in one blog post nonetheless.)

Mathis was primarily a sentencing case, arising under the Armed Career Criminal Act, or ACCA.  That statute provides for harsher criminal sentences against those with certain sorts of prior criminal convictions.  ACCA has been interpreted by the Supreme Court to provide for a “categorical approach”, where what is important is what one can be certain a person has been convicted of, that is, the elements of their crime, and not other facts regarding what they may actually have done in the past.

In its use of the categorical approach, ACCA operates similarly to several provisions of the Immigration and Nationality Act (INA) regarding noncitizens with criminal convictions.  As the Supreme Court explained in Moncrieffe v. Holder, 569 U.S. ____, 133 S.Ct. 1678 (2013), the categorical approach, grounded in the language of immigration statutes that ask what a noncitizen was “convicted” of, “has a long pedigree in our Nation’s immigration law.”  Indeed, it goes back more than 100 years, at least back to the Second Circuit’s decision in United States ex rel. Mylius v. Uhl, 210 F.860 (2d Cir. 1914).  There are some provisions of immigration law that have been interpreted to deviate from the categorical approach, such as the $10,000 loss threshold for a fraud crime to qualify as an aggravated felony under section 101(a)(43)(M)(i) of the INA, addressed in the Supreme Court’s 2009 decision in Nijhawan v. Holder, but they are the exception, not the rule.

Because both ACCA cases and many areas of immigration law rely on the categorical approach, the reasoning of ACCA cases is often found to control in immigration cases.  Moncrieffe, for example, which addressed the immigration consequences of a conviction under Georgia law for possession of marijuana with intent to distribute, cited and relied upon Shepard v. United States, 544 U.S. 13 (2005), and Johnson v. United States, both ACCA cases.  (The particular ACCA provision involved in Johnson was held unconstitutionally vague by the Court, but the principles behind the categorical approach were still usefully elucidated in that case.)  Footnote 2 of the Mathis majority opinion specifically acknowledged the applicability of the categorical approach discussed in Mathis to immigration cases, citing Kawashima v. Holder, 565 U. S. 478, 482–483 (2012).

In both the ACCA context and the immigration context, issues have arisen regarding the application of the categorical approach to what are known as “divisible” statutes.  In effect, such statutes contain multiple separate crimes, and so one can look at the record of the conviction, using what is known as the “modified categorical approach” to determine which of these crimes applied.  The Supreme Court clarified in Mathis that this is only to be done when the difference between the components of the statute of conviction turns on a true element, a fact on which a jury would have to agree to convict, or which a defendant would have to admit in a guilty plea.  It does not apply to alternate means of commission of a crime, even if they are listed in the statute of conviction.

Mr. Mathis had been convicted multiple times of burglary under Iowa law, which covered unlawful entry into “any building, structure, [or] land, water or air vehicle.”  Mathis, slip op. at 5.  For ACCA purposes, on the other hand, the Court had held years earlier that a conviction only counted as “burglary” if it involved unlawful entry into a building or other structure.  The question thus became whether one could look at the record of Mr. Mathis’s conviction to see whether he had been convicted of breaking into a building or other structure, on the one hand, which would qualify as ACCA “burglary”, or breaking into a land, water or air vehicle, which would not so qualify.  Iowa case law made clear that these were merely alternative means of committing a single crime, and that a jury could convict someone of burglary without agreeing on whether the defendant had burgled a building or a vehicle.  The government sought nevertheless, however, to argue based on documents from Mr. Mathis’s prior criminal cases that he had in fact been convicted of burglarizing a house and not a vehicle.

In a decision written by Justice Kagan and joined by Chief Justice Roberts and Justices Kennedy, Thomas, and Sotomayor, the Court held that this was not permissible.  Facts on which a jury need not be unanimous are not elements of a crime, and so the Iowa burglary statute at issue was not truly divisible: it created only one crime, not many.  Because the categorical approach focuses on what the defendant was convicted of doing, and not what he or she may have actually done, one cannot pick through the record of a prior case and speculate regarding whether the jury might have agreed on something that the law did not require it to agree on in order to convict.

Justice Breyer, joined by Justice Ginsburg, dissented, arguing that if the record revealed that evidence supported conviction of a defendant only pursuant to one statutory word or phrase, it should not matter whether that word or phrase was termed an element or a means.  Justice Alito, writing only for himself, compared the Court’s ACCA case law to a Belgian woman who had set out for Brussels and ended up in Zagreb, Croatia, by following her GPS too unquestioningly.  Accusing the majority of “pointless formalism”, he hypothesized a lengthy plea colloquy in which a defendant admitted to burglarizing a house at “10 Main St.” in the face of lengthy questioning from the judge regarding whether this address might represent a yacht, house boat, trailer, or tent.  (The hypothetical defendant’s response to this last query was said to be, “No, it’s made of brick.  I scraped my knee on the brick.”)  Even in this case, he lamented, “[a]s the Court sees things, none of this would be enough.”

Before addressing some of the implications of this decision for immigration purposes, I will pause to note that Justice Alito’s lengthy hypothetical colloquy, humorous though it may be, strikes me as not really supporting his argument, and perhaps even as weakening it.  In the real world, a judge would almost never go through such a lengthy discussion of the nature of the premises burgled if that did not affect the crime of which the defendant was to be convicted or the punishment for which the defendant was eligible.  If the defendant had actually broken into a car parked right outside the garage of the house at 10 Main Street to steal money and jewelry, rather than breaking into a car parked inside the garage and the garage itself to steal the same money and jewelry, one would not expect him to start an argument with the judge, in the context of a law which made the penalty for the two versions of the crime exactly the same.  Rather, if asked whether he broke into a “structure” at 10 Main Street, there is a good chance that such a hypothetical defendant would simply say “yes”, and that would be the end of it.  To quote from Justice Kagan’s majority opinion:

“At trial, and still more at plea hearings, a defendant may have no incentive to contest what does not matter under the law; to the contrary, he “may have good reason not to”—or even be precluded from doing so by the court. . . .  When that is true, a prosecutor’s or judge’s mistake as to means, reflected in the record, is likely to go uncorrected. See ibid. Such inaccuracies should not come back to haunt the defendant many years down the road by triggering a lengthy mandatory sentence.”

Justice Alito’s hypothetical colloquy, which I admit was quite funny, draws its humor partly from its unrealistic nature.

And just as the sorts of uncorrected inaccuracies to which Justice Kagan refers “should not come back to haunt the defendant many years down the road by triggering a lengthy mandatory sentence”, as Mathis makes clear, they also should not come back to haunt the defendant by triggering deportation.  In footnote 3 of Mathis, Justice Kagan specifically notes a scenario in which the Mathis rule will apply to immigration cases:

To see the point most clearly, consider an example arising in the immigration context: A defendant charged under a statute that criminalizes “intentionally, knowingly, or recklessly” assaulting another—as exists in many States, see, e.g., Tex. Penal Code Ann. §22.01(a)(1) (West Cum. Supp. 2015)—has no apparent reason to dispute a prosecutor’s statement that he committed the crime intentionally (as opposed to recklessly) if those mental states are interchangeable means of satisfying a single mens rea element. But such a statement, if treated as reliable, could make a huge difference in a deportation proceeding years in the future, because an intentional assault (unlike a reckless one) qualifies as a “crime involving moral turpitude,” and so requires removal from the country.

Under Mathis, if recklessness and intentional assault are indeed interchangeable means of satisfying the same mens rea requirement under a particular statute, then for immigration purposes the statute cannot be divided between them.

That is, even someone who seemingly pled guilty to intentional conduct, under such an indivisible statute, should be considered as if he or she had only pled guilty to reckless conduct, because, as Moncrieffe explained, “we must presume that the conviction “rested upon [nothing] more than the least of th[e] acts” criminalized,” 133 S. Ct. at 1684.  As Mathis’s footnote 3 explains, this makes more sense than might at first glance appear, because someone convicted under such a statute would have had no reason to dispute the allegation that he or she had acted intentionally rather than recklessly—unless perhaps he or she had consulted immigration counsel prior to entering a plea.  Allowing the proverbial hairs of a state statute’s text to be split, adversely to the noncitizen, beyond the point where the distinction makes any difference under state law, would penalize those who did not have immigration consequences in mind at the time of their plea or trial.  Certainly, for a variety of reasons, all noncitizens charged with a crime should consult with a competent immigration attorney before pleading to any charge or otherwise proceeding with their criminal case, but the law should not unnecessarily and unfairly penalize those who fail to heed this advice.

The distinction between recklessness and intentional conduct is not the only context in which this means/elements distinction may have relevance for immigration law.  For example, attorneys whose clients have been convicted of possession of a controlled substance, under state laws covering at least some substances not federally controlled, should explore whether the identity of the substance is an element or a means under the relevant state law—whether, in order to obtain a conviction, the state is required to prove which controlled substance a defendant possessed.  If the identity of the substance is a means and not an element, then the conviction may, under Mathis, fall within the protection of the Supreme Court’s decision last year in Mellouli v. Lynch, which required a controlled substance conviction to relate to a federally controlled substance in order to cause adverse immigration consequences.  (The Third Circuit’s 2013 en banc decision in Rojas v. Attorney General rejected what it called the “formal categorical approach” in this controlled-substance context, but it is not clear that this aspect of Rojas can survive the combination of Mellouli and Mathis in states where the identity of the controlled substance is a means and not an element under state law, although we will have to wait and see how the case law develops to be sure.)  There will be other areas, as well, where a statute which lists multiple ways of committing a crime is actually indivisible under state law, and so a noncitizen is entitled under Moncrieffe and Mathis to the assumption that he or she committed the crime in whichever way is least harmful for immigration purposes.

Mathis is therefore good news for a significant number of immigrants, and their attorneys, even though this small piece of good news may pale in comparison to the disappointment of U.S. v. Texas and the continued injunction against DAPA and DACA+, which are bad news for substantially more immigrants.  With respect to the latter, we can hope for, and fight for, the possibility that the November elections may bring more good news.