The Volokh Conspiracy

Mostly law professors | Sometimes contrarian | Often libertarian | Always independent


Severability Doctrine in the Fifth Circuit

At least two judges favor Justice Thomas's approach to severability


Last month the en banc Fifth Circuit decided Collins v. Mnuchin. The majority opinion, by Judge Willett, found that the Federal Housing Finance Agency was unconstitutionally structured. This agency was headed by a single director, who could only be removed "for cause." However, the court divided on the remedy. Judge Haynes, joined by (my count) nine other justices, struck down the "for cause" provision, and severed the remainder of the statute:

When addressing the partial unconstitutionality of a statute such as this one, we seek to honor Congress's intent while fixing the problematic aspects of the statute. Thus, in this case, the appropriate—and most judicially conservative—remedy is to sever the "for cause" restriction on removal of the FHFA director from the statute.

Judge Oldham, joined by Judge Ho, disagreed with this analysis.

First, they explained that courts do not "strike" down laws. Really, they don't. I cringe whenever anyone uses this phrase. It is simply incorrect, and maintains an inaccurate description of what courts do. Rather, Oldham and Ho favor the framework advanced by Justice Thomas in Murphy v. NCAA.

The second problem we have with the remedy endorsed by a majority of our Court is that we do not believe Article III of the Constitution permits us to "strike" the FHFA Director's for-cause protection from the statute. See Murphy v. NCAA, 138 S. Ct. 1461, 1485 (2018) (Thomas, J., concurring) (explaining that "[e]arly American courts did not have a severability doctrine" because "[t]hey recognized that the judicial power is, fundamentally, the power to render judgments in individual cases").

Second, Oldham and Ho recount that the Framers expressly rejected a "Council of Revision" during the Constitutional Convention. Instead, our courts can only decide "cases" and "controversies" between parties. They cannot, to use Jonathan Mitchell's phraseology, employ a "writ of erasure."

In the final Constitution, the judiciary was given only the power to decide cases and controversies—to resolve legal disputes between parties and order remedies to redress injuries. Thus, when a court concludes that a statute is unconstitutional, it is not "striking down" or "voiding" or "invalidating" the law. It is merely holding that the law may not be applied to the parties in the dispute. The Constitution does not empower courts to delete sections of state and federal codes. The Founders expressly considered the possibility of a judicial veto, and they rejected it multiple times during the Constitutional Convention.

Who then do we blame for the myth that courts can "strike down" laws? Who else? John Marshall.

This history has been obscured by rhetoric that Chief Justice Marshall used in Marbury v. Madison, 5 U.S. (1 Cranch) 137 (1803), to explain judicial review. In that case he famously declared that a statute found unconstitutional by a court becomes "entirely void," "invalid," and "not law." Id. at 177–78. Subsequent cases have compounded the confusion. See, e.g., The Civil Rights Cases, 109 U.S. 3, 26 (1883) (holding "void" sections 1 and 2 of the Civil Rights Act of 1875). Nevertheless, it is indisputable that courts do not have the power to erase duly enacted statutes. Instead, they may decline to enforce them or enjoin their future enforcement to resolve cases and controversies.

The second citation to the Civil Rights Cases is especially apt. That infamous decision held that Congress could not use its Section 5 powers to prohibit privately enforced of segregation in hotels and theaters. But what about publicly enforced segregation, like the Jim Crow law at issue in Plessy. This law was certainly preempted by the Civil Rights Act of 1875. But the Plessy Court bought into the fiction of judicial supremacy.  It held that the Civil Rights Act of 1875 was "unconstitutional and void." Therefore, the state law was not preempted. Even those unsympathetic to Oldham and Ho's general world view should recognize the role that the "striking down" fiction played in Plessy.

Third, Oldham and Ho suggested that the Fifth Circuit should not perpetuate this confusion:

Our Court should not add to the confusion about the judiciary's limited powers by claiming to "sever" a statute based on open-ended speculation about how Congress would have solved the separation-of-powers problem. And we certainly should not rewrite the statute while pretending such legislative activity is the most modest judicial remedy. We would instead remand to the district court with instructions to fashion a remedy that actually redresses Plaintiffs' harms.

There was some discussion on Twitter that Oldham and Ho proposed declaring the entire statute unconstitutional. To the contrary. Oldham and Ho preferred a remand to determine which portions of the law actually injure the Plaintiffs. Only those provisions can be enjoined.

The Department of Justice took this same position in the Affordable Care Act case. (Many people, present company company included, erroneously contended that the government favored setting aside the entire law.) The final sentence of the government's brief explained the position: "Accordingly, the court's judgment should be affirmed on the merits, except insofar as it purports to extend relief to ACA provisions that are unnecessary to remedy plaintiffs' injuries."

And the government reiterated this position during oral arguments. Attorney August Flientje explained:

Flientje: We think it is an Article III issue, so yes we did raise it in our brief for the first time, we do think, given that, it would be appropriate to remand to consider the scope of the judgment on that point. We think that's more of a technical point, because the severability analysis requires looking at the statute altogether. Obviously, there is precedential impact of this court's decision or a higher court's decision that could make a lot of sorting out those details unnecessary down the road.

If DOJ is correct, then the correct remedy after a declaration of inseverability is a district court proceeding to determine what provisions injure the individual Plaintiffs.

Judge Elrod explored this point in a colloquy with Douglas Letter, the lawyer for the House (at 1:41:30):

Judge Elrod: If we held, hypothetically, that it was severable, we would say the district court, do your best severability in the first instance, take out your blue pencil.

Letter: No, you [that is the 5th Circuit] would do that.

Judge Elrod: Why would we do that? In any other normal case, you would send it back to the district in the first instance to make its best stab at trying to implement the ruling that we made. That would be the normal proceeding in hundred cases that we have this month

Judge Elrod had a similar colloquy with Kyle Hawkins, the Texas Solicitor General:

Judge Elrod: If the court ruled on the partial summary judgment, and then you have to go back for the relief, the remedy has not been spoken of yet

Hawkins: That's right. We will go back to district court.

Judge Elrod: You're not to that process yet. You have a partial summary judgment.

Judge Elrod appears to be on the same page as Judges Oldham and Ho. They recognize that under Justice Thomas's approach to severability, the district court can only enjoin those provisions that in fact injure the Plaintiffs.

NEXT: Can a federal agency’s policy pre-empt state law?

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  1. They had a severability doctrine in the French Revolution, too, hi-yo!

    1. Tip the veal, don't forget to try your waitress!

  2. "Funny" that John Marshall, who served in the American Revolution and was at Valley Forge with Washington, who served with James Madison on the Virginia Ratifying Convention to win that state's support for the new Constitution, who, in short might have been considered a "Founder", should have had so little understanding of what the Founders intended. One can also wonder what would have happened if Marshall had died early in Jefferson's first term, allowing Tom to replace him with someone more "Republican" in his thinking. It's almost as if the "Founders' intent" is a mere mask for contemporary conservative political operators!

    1. Well, to be fair to Marshall, the concept of judicial review is implicit in the set up of the Constitution, provided the elected branches don't quash the courts (which they did think would happen more than it actually does), so it is sorta Founders' intent.

      1. My "point", pretty much, is that Justice Thomas always predicates his decision on the "Founders' intent", of which he always claims infallible knowledge, when the Founders often disagreed among themselves as to what the Constitution "meant". And "implicit concepts" are easy to find. Justice Thomas did so himself recently, in FTB v. Hiatt.

        1. My “point”, pretty much, is that Justice Thomas always predicates his decision on the “Founders’ intent”

          That is not, in fact, the basis of Justice Thomas' jurisprudential theories, nor those of any other originalist I am familiar with.

        2. The "Founders' intent" (more specifically the ratifiers' intention) is the very thing we're trying to figure out when we interpret the Constitution. Otherwise, why would we even begin the inquiry with our Constitution? We could just look to Pride and Prejudice, or the Soviet Constitution, or our own consciences.

          It's true that we can't often discern with certainty what the ratifiers actually intended, which is why Justice Thomas and others give primacy to the text rather than legislative history. But the inquiry is still the same. Fundamentally we're trying to figure out what some group of human beings intended when the thing we call a Constitution became a governing legal document.

          1. The “Founders’ intent” (more specifically the ratifiers’ intention)

            You are equating here two things that can be very different - what the drafters thought they were writing, and what the ratifiers thought they were voting for.

            1. I was not attempting to equate the two. I was redirecting. We care about the "Founders' intent" only insofar as it possibly sheds light on the ratifiers' intent. If it is a true fact that the Founders intended X in drafting Y, but the ratifiers intended Z in ratifying Y, Z controls. Fundamentally Z is the only thing we care about when we are engaged in interpretation.

              1. Even if you look to the ratifiers intent rather than the drafters, They still divided into two factions, the anti-federalists and the federalists.

                The anti-federalists wanted a central government that was stronger than the articles of confederation provided, but only marginally so, with the balance of power still favoring the several states.

                The federalists on the other hand wanted to reduce the state governments to mere corporations, fully subjects of the federal government.

                What we got was/is generally a muddled compromise, that at first pass on the text seems to favor the anti-federalist position, but contains a number of federalist supported provisions (N&P clause) that would over time allow the federal government to increase it's authority at the expense of the several states.

                1. Sure, for the same reasons we might not defer to the self-serving statements of Founders we might not defer to the self-serving statements of ratifiers. But that's because we doubt that those statements evidence real intent.

                  The bigger problem is whether something like a collective can have an intent. So textualism walks in. But textualism itself is an original intent concept, since we are still looking to some thing written and adopted by other humans. Textualism resolves doubt and ambiguity, but it doesn't change the overall experiment. Interpretation is about figuring out intent. There is no other theory of interpretation, just different methods.

                  1. Hmm. In my opinion, you start with the text written by the Founders and ratified by the States. Any other source is mere guesswork, strongly influenced by the desires of the litigant.

                    1. Right, that's how interpretation works, everywhere, throughout history. If I stumble across hieroglyphics I begin my analysis with reference to the symbols.

                      We care about the ratifiers because the act of ratifying is the communicative process. We look to Founders' intent because it sheds some (small) light on what people were thinking when they ratified. The text is the best evidence, which is why textualism is the overwhelming method for interpreting written statutes in the US.

        3. Sure, you can see whatever you want in the text, but from the Founding Era, such "implicit concepts" about the Supreme Court and judicial review were *explicitly* debated in The Federalist Papers and the writings of Anti-Federalists, such as Brutus.

    2. One could as well ask how John Adams could be so eager to sign and enforce the Alien and Sedition Acts, or how Thomas Jefferson could think up an absolute trade embargo.

  3. If it looks like a duck, swims like a duck, and quacks like a duck, it's a duck. If Supreme Court decisions cause the practice of the law and legislation to change, then laws have been, in fact, 'struck down.' If you believe that things should be otherwise, then say so. Just don't say that up is down because you've decided it's so.

    1. I guess that, when we are feeling pedantic, we can insist on the narrowest distinctions. And maybe we lawyers can find some value in distinguishing between "struck down" and "struck down for all intents and purposes."
      To the extent that we want laypeople to be able to understand a judicial decision; I have no problem saying "Law X is stuck down." instead of saying, "Law X may not be enforced, now or in the future, against any carbon-based life forms."

      (There might be instances where making such a distinction is absolutely important. I can't think of any. But I'll follow this thread, to see if others can give some examples.)

      1. The Affordable Care Act (among the many things litigated about) also included a provision repealing the authority of commercial banks to lend money to students which the federal government would underwrite if the student defaulted - essentially a scam that allowed banks to issue guaranteed repayment loans, albeit at lower than market rates.

        That repeal would have left no one with standing, because an authorization that only affected corporate entities was ended. For context, this was one of the tricks used to lower the CBO budget score, and an independently good idea.

  4. "Thus, when a court concludes that a statute is unconstitutional, it is not "striking down" or "voiding" or "invalidating" the law. It is merely holding that the law may not be applied to the parties in the dispute."

    So 1,000,000 other people all have to re-litigate the same issue from scratch and the government gets to try and hoodwink a different judge into a more favorable result?

    1. "So 1,000,000 other people all have to re-litigate the same issue from scratch..."

      It depends on the holding. If SCOTUS says it won't enforce the statute to the parties in dispute because it is facially unconstitutional, it's not going to enforce it to any other parties, either.

      1. So under this reading of the Court's Article III powers the "Overbreadth Doctrine" is illegitimate.

        Statutes are deemed unconstitutional when the Court suspects they would reach a substantial zone of protected speech, even when the law would be constitutional as applied to the actual party.

        I always thought this violated the "case or controversy" clause anyway

        1. It doesn't, because the result of the finding of overbreadth still affects the litigants of the present controversy.

          1. But when the court uses the doctrine it rests its holding on the law being unconstitutional as applied to hypothetical cases not before the court, instead of waiting until each situation arises to make determinations of constitutionality.

            Under Blackman's reading--when a case is before the court involving law x and situation y and law x--courts only have the power to enjoin enforcement of law x to situation y. With overbreadth, Courts deem law x constitutional because of its application to unripe hypothetical controversy z

            1. That's the common law system!

              There's a tugboat accident. Suit for negligence. Goes up on appeal. The appellate court announces a test that can be used for lots of negligence cases, not just the one before the Court.

              That doesn't violate Article III. That IS Article III. Use cases to announce rules that can then be applied to other cases.

              Now, overbreadth is a little different, but basically you have a statute that courts should not enforce. Now, overbreadth IS an exception to normal third party standing rules, but third party standing rules are not constitutionally required. There are many examples of third party standing, such as next friends and qui tam suits.

              Don't listen to Blackman. 🙂

            2. I think where your analysis breaks down is the "unripe hypothetical controversy z". The Court isn't using the hypothetical to resolve a non-existent, unripe conflict. It is using the hypothetical to resolve an existing conflict between parties in situation y. If it helps simplify, just treat the overbreadth doctrine as a slippery slope argument.

        2. I'm not sure I agree with the Court's overbreadth doctrine, but to answer your question, it wouldn't be "illegitimate" since the Court presumably believes its overly protective view (of hypothetical speech restricted by the statute) is itself constitutionally mandated by the First Amendment. That is, statutes that arguably prohibit certain types of speech are for that reason unconstitutional (facially even? I can't remember).

      2. If the judge the article is about expects the holding to be binding on the federal government in the next case with a different party on the other side, his complaint is pure semantics, there is no difference in the end effect of the decision.

        However, that's not how what he wrote sounds to me.

  5. This seems to be correct, but is only a technical issue.

    If the Court enjoins an offending passage but otherwise leaves a statute intact, it in effect has severed that passage from the rest of the law. It may not be called severability, but that is in reality what it is.

    1. Huh, so does the court enjoin the offending passage in general or only applied to petitioner and others similarly situated?

  6. So under this reading of the Court’s Article III powers the “Overbreadth Doctrine” is illegitimate.

    Statutes are deemed unconstitutional when the Court suspects they would reach a substantial zone of protected speech, even when the law would be constitutional as applied to the actual party.

    I always thought this violated the “case or controversy” clause anyway

  7. This is outrageous.

    Justice Thomas' fever dreams are not law. They aren't anything. He sits around delivering dissents that will never get 5 votes on the Court, taking positions far outside the judicial mainstream, and lower court judges don't have any right to pick and choose SCOTUS dissents and concurrences to follow.

    There are rules with respect to severability. Refusing to follow Supreme Court precedent is a violation of a lower court judge's oath of office.

    1. And yet, here we are, where judges violate Supreme Court precedent all the time, and Supreme Court dissents are often used by lower courts instead of the main decision. For example, Justice Breyer's Heller dissent, creating an "interest-balancing approach" that is but lipstick on intermediate scrutiny to justify high levels of gun control, rather than using the majority opinion's originalist approach.

      In general, do lower courts follow higher courts, sure, when it doesn't matter that much to their preferred policy outcome.

      1. I think Heller is a deliberately vague opinion.

        I have a personal spin on it (that rather than levels of scrutiny, it calls for certain limitations on weapons with long pedigrees to be upheld and others to be struck down), but I don't have a lot of confidence as to what it means.

        But you are right that no court should use Breyer's dissent as the test.

  8. So the state of Texas passes a law that says people named "Josh" may not post on the internet.

    Blackman sues, claiming a violation of his First Amendment rights, and the Supreme Court rules in his favor.

    What has the court done, if not strike down the Texas statute?

    1. It's said that Texas can't enforce the law against Prof. Blackman.

      I don't know that this sort of analysis is right but I don't think it's conceptually all that difficult to understand.

      1. So it can enforce the law against Josh Smith until he sues?

        1. That certainly seems to be the position that Judges Oldham and Ho take. I'm doubtful that they would even allow Josh Smith to point to a judgement for Prof. Blackman in a similar case for an easy win.

  9. So historical analysis about how 18th and early 19th century actors interpreted the brand new constitution can show how the foreign emoluments clause doesn't apply to the President, but we have to dismiss as just wrong Chief Justice Marshall's reasoning in Marbury. Gotta love originalism.

  10. How does this impact statute with express severability clauses?

    Let me just consider a thought experiment, suppose that the law here had said something like

    If a provision of this act or its application to any person or circumstance is held invalid, the invalidity does not affect other provisions or applications of the act that can be given effect without the invalid provision or application, and to this end the provisions of this act are severable.

    Now, the application of the act to the specific plaintiff is being held invalid due to the "for cause" requirements. That invalidity does not affect the other provisions, which can be given effect without the invalid "for cause" provision.

    This seems straightforwards.

    The final requirement then would be first, to strongly encourage Congress to make severability an explicit part of each statute and, second, to adopt a default policy when the law does not have one.

    1. The final requirement then would be first, to strongly encourage Congress to make severability an explicit part of each statute and, second, to adopt a default policy when the law does not have one.

      Bright line tests make too much sense. They require the courts to abide by such bright line tests. The beauty of fuzzy legal theories is you can decide the same issue in two (or more) completely different ways.

      It is much easier to make new laws when you're unencumbered by past precedent and don't have to worry about politically inconvenient future cases coming out the wrong way.

  11. I'm sympathetic to the conceptual argument about the distinction between providing relief to the parties before the Court and "striking down laws." But, the government's position regarding relief and Josh's interpretation of it is at complete odds with the majority of the government's brief and the practical effect of a victory.

    Here's what the government says in their brief about the effect of the ACA if the mandate is struck down:

    "Once the individual mandate and the guaranteed-issue and community-rating provisions are invalidated, the remaining provisions of the ACA should not be allowed to remain in effect—again, even if the government might support some
    individual provisions as a policy matter." (Br. at 43.)

    "In this unique context, comparatively “minor,” ancillary provisions that were tacked on to the bill should be invalidated once the core provisions have been struck down because “[t]here is no reason to believe that Congress would have enacted them
    independently.” (Br. at 48)(citing NFIB, 567 U.S. at 705 (joint dissent))

    "California’s emphasis on the “far-reaching consequences that would result from making major changes to the ACA,” Cal. Br. 36, merely underscores the dangers of crafting a judicial remedy that picks and chooses among the provisions that Congress enacted to work together in this highly complex area. Rather than having a court select which provisions should remain in force—and thus, select winners and losers in the insurance markets and other spheres—that task should be left to Congress." (Br. in 48-49.)

    And here's the sentence of the conclusion that Josh cited:

    "The district court correctly held that the individual mandate is unconstitutional in light of the elimination of its penalty, that the guaranteed-issue and community-rating provisions are inseverable from the mandate, and that the remainder of the ACA is inseverable in turn." (Br. at 49.)

    Claiming that they only want to restrict relief to ACA provisions that actually harm the plaintiffs argument is completely disingenuous. If you conclude that the entire act is "invalid" as the government expressly argues, but then you request declaratory judgment to only portions of it, that doesn't make any sense, unless you're asking the Court to "please put some dicta in the opinion we can use later so that we can justify not enforcing the act in its entirety."

  12. On the use of the word "void," there is of course Hamilton's discussion of judicial review: "By a limited Constitution, I understand one which contains certain specified exceptions to the legislative authority; such, for instance, as that it shall pass no bills of attainder, no ex-post-facto laws, and the like. Limitations of this kind can be preserved in practice no other way than through the medium of courts of justice, whose duty it must be to declare all acts contrary to the manifest tenor of the Constitution void."

  13. This is a good example of the harm done when the Court blatantly ignores the text and extraneous evidence in order to reach a desired result. The mandate was expressly argued not to be a tax, by the very government now urging it to one. I recall Mr Justice Douglas’ test: Do you want a program or not? Sounds legislative more than judicial in approach.

  14. The Supreme Court has a well-established severability doctrine. Where there are established Supreme Court precedents, The duty of an inferior federal court is to apply those precedents, not to go back to the Federalist papers and argue the question as an original matter.

    It appears there are judges in the fifth circuit who are just plain going rogue, taking a position that everything is up for grabs and staking our positions as if previous Supreme Court decisions simply never happened.

  15. The Supreme Court’s doctrines of facial and as applied challenge cover these questions. Only a a facial challenge results in striking down a law in its entirety.

    It could be argued there should be no such thing as a facial challenge. But the Supreme Court says there is, and an inferior federal court has to accept that even if the judges think there shouldn’t be.

Please to post comments