Of Legislatures, Courts, and the Contracts Clause

By Michael Dorf

In my latest Verdict column, I recap my (side of the) debate on originalism with Prof. Randy Barnett Monday night at the Soho Forum. (No video available yet; I'll update when there is.) Regular readers of this blog and my other work will not be especially surprised by the core of my argument.

So-called public-meaning originalism unjustifiably claims the supposed virtue of constraint that intentions-and-expectations originalism claimed (justifiably or not), while supposedly avoiding the pitfalls of intentions-and-expectations originalism, but in fact public-meaning originalism is functionally equivalent to living Constitutionalism; meanwhile, the respectability that scholars like Prof. Barnett bring to the term "originalism" is systematically misused and abused by judges and justices who talk the public-meaning-originalism talk but walk the intentions-and-expectations-originalism walk.

Thus, I conclude with a quotation from Justice Scalia's dissent in Obergefell v. Hodges. Despite having championed public-meaning originalism since the 1980s, the pith of Justice Scalia's argument was the bad old intentions-and-expectations originalism. He wrote: "When the Fourteenth Amendment was ratified in 1868, every State limited marriage to one man and one woman, and no one doubted the constitutionality of doing so. That resolves these cases."

It certainly does not, but this sort of move is all too familiar in the work of the justices who nominally profess public-meaning originalism. I illustrate the point in my column by discussing Justice Gorsuch's dissent in Monday's decision in Sveen v. Melin. Seemingly unaware that he is engaged in methodological inconsistency, Justice Gorsuch invokes original public meaning just paragraphs after employing intentions-and-expectations originalism.

In the balance of this essay, I want to pivot from my methodological critique to the merits of Justice Gorsuch's dissent. Although I am not ultimately persuaded by his dissent, I will suggest that it points to tensions in various lines of constitutional doctrine and perhaps to a flaw in the Constitution itself.

As I explain in the Verdict column, Sveen is a Contracts Clause case posing the question whether Minnesota can apply its default rule overriding the designation of a subsequently divorced spouse in a life insurance policy to a policy purchased before the law changing the default rule went into effect. The column explores one of Justice Gorsuch's objections. The majority opinion (authored by Justice Kagan and joined by all of the other justices) finds no Contracts Clause violation because, in the majority's view, the change in the default rule does not "substantially impair" the life insurance policy.

Justice Gorsuch disagrees on two main grounds. First, he thinks that changing the beneficiary changes the essence of the policy and thus necessarily amounts to a substantial impairment. Second, he thinks that the line of cases requiring substantial impairment--as opposed to any impairment--cannot be reconciled with the text and history of the Contracts Clause. Interested readers should consult my column and the case itself for more details.

Let's turn now to one of the reasons Justice Kagan gives for her conclusion that the Minnesota law did not substantially impair existing life insurance contracts. She says that even under prior law, a judge issuing a divorce could change the beneficiary in an insurance policy. Justice Gorsuch cries foul. Citing the 1924 case of Tidal Oil Co. v. Flanagan, which in turn cited still older cases, Justice Gorsuch observes that the Contracts Clause simply does not apply to state judicial rulings that seem to undercut contractual obligations, only to legislative ones.

Justice Kagan responds that . . . well, duh . . . she knows that. Justice Kagan says she was invoking the substantial possibility of judicial revision during a divorce proceeding only for the proposition that no one could have had a very strong reliance interest in having been named a beneficiary to a life insurance policy by a now-divorcing spouse. Put differently, both the majority and the dissent accept that the Contracts Clause limits state legislatures but not courts. Which leads me to wonder why.

Here is Justice Gorsuch's rationalization of the distinction:
[T]he [Contracts] Clause applies only to the “law[s]” legislatures “pass,” not to the rulings of courts. Tidal Oil Co. v. Flanagan. That’s because legislatures exist to pass new laws of general applicability responsive to majoritarian will, often upsetting settled expectations along the way. The same does not hold true for courts that are supposed to apply existing laws to discrete cases and controversies independently and without consulting shifting political winds. 
Wait, what? Sure, in most cases courts "apply existing laws to discrete cases and controversies," but sometimes they change the common law. And one would think that it's in exactly those cases--when a court discards an old common-law rule or adopts a new one--that the party ending up on the short end of a contract case might want to protest that what the state court has done impaired the obligations of a contract in violation of the Contracts Clause. Suppose that instead of the Minnesota legislature changing the default rule governing ex-spouses named as beneficiaries in life insurance policies via a statute, the Minnesota Supreme Court had changed the default rule in the exercise of its common law power. As I read Justice Gorsuch, that would not implicate the Contracts Clause, even though the effect on ex-spouses would be identical.

Indeed, it's notable that the distinction between state courts and state legislatures arose long before the Court, in 1938 in Erie Railroad v. Tompkins, acknowledged that state courts can be just as much lawmakers as state legislatures. Here is how Justice Brandeis put the modern view in Erie: "whether the law of the State shall be declared by its Legislature in a statute or by its highest court in a decision is not a matter of federal concern."

The Erie Court was construing the Rules of Decision Act, not the Contracts Clause. But it is by no means clear why that should make a difference. Both were adopted in 1789. In Erie, the Court said there is no good reason to construe the phrase "laws of the several States" to exclude judge-made common law. And from a practical perspective, neither is there a good reason to construe a "law impairing the obligation of contracts" to exclude judge-made common law rules that impair contracts.

Pushing in the other direction are two factors. One is a textual difference I have not yet mentioned. The full Contracts Clause says that "No state shall . . . pass any . . . law impairing the obligation of contracts." By contrast, the Rules of Decision Act (which was originally Section 34 of the Judiciary Act of 1789) does not include the word "pass." It is fair to say that only a legislature can pass a law, whereas both legislatures and courts can make law. On this view, the framers should have made the Contracts Clause applicable to changes in judge-made common law, but they goofed in failing to do so, and we're just stuck with what they wrote.

Also pushing this way is the force of longstanding precedent. Contracts and contracting have done just fine these many years, despite the courts' failure to treat the Contracts Clause as limiting judge-made changes in the common law of contracts. Maybe that's because courts introduce wholesale changes to the law less frequently than legislatures. Whatever the reason, the sorts of factors ordinarily needed to overcome stare decisis do not appear to be present.

Note, however, that Justice Gorsuch does not expressly invoke either of these arguments for retaining the distinction between legislative and judicial changes in the law governing contracts. He does not appear to be aware of the problem. He says simply that state courts are supposed to "apply existing law," thereby ignoring the small but significant body of cases in which state courts change the law.

Before concluding, I should say a word about a different problem that would result from applying the Contracts Clause to judge-made changes in state law: State court judges would have to change the common law, if at all, only prospectively; yet norms of judicial decision making reject pure prospectivity--the idea that a court would announce a new rule but not apply it in the case before it.

What's the solution to that problem? Three possibilities come to mind.

(1) In the federal system, Article III's case-or-controversy limit generally counsels against pure prospectivity, but state courts are not bound by Article III. Just as some states have looser standing rules and permit advisory opinions, so state courts could, consistent with the existing Constitution if the Contracts Clause were applied to state judicial rulings, change the common law of contracts only prospectively.

(2) What about states that follow the federal practice of disallowing pure prospectivity? For them, one solution would be to simply disallow judge-made substantial changes in the common law if a similar change by the legislature would violate the Contracts Clause. It appears that Justice Gorsuch might favor this approach, given his odd claim that state courts only ever apply existing law. I hasten to add that I do not favor this approach, which seems rooted in a pre-legal-realist view of common law adjudication.

(3) My preferred solution, if I could wave a magic wand, would be to amend the Constitution to repeal the Contracts Clause. That is, I would "level down" rather than "level up," by having the Contracts Clause apply to neither state courts nor state legislatures. Why? Because the Contracts Clause is peculiar. Retroactive application of new rules of law (whether judge-made or legislated) can upset reliance interests, but there is no reason to think that problem is special to contracts.

Consider how, in the early-to-mid-20th-century states abolished the rule of privity in tort--under which a manufacturer was only liable for a defective product to the purchaser of the product, not to third parties who were injured by it. That change was accomplished in some states by common law rulings, in others by legislation. In either event, the change, when applied retroactively, may have imposed a burden on the makers of products who, in designing those products, took account of the likely cost in lawsuits by purchasers of the products but not of lawsuits by potentially injured third parties. (If you don't think product manufacturers made such calculations, I have a Ford Pinto I'd like to sell you.) That burden can be just as great as or greater than the burden imposed by a retroactive change in contract law. And yet the Constitution contains no "Torts Clause."

To my mind, the only limits on retroactive application of civil duties ought to be the very loose limit imposed by the Due Process Clause and those imposed by the Takings Clause. So long as I'm waving my magic wand, I would make explicit that the Ex Post Facto Clauses of Article I, Sections 9 and 10 apply only to criminal liability.

Of course, I don't possess a Constitution-amending magic wand. Given our existing Constitution, I think the least bad course is to do what the Court has done: to accept on stare decisis and perhaps textual grounds that the Contracts Clause doesn't apply to retroactive judicial changes in the law of contracts, consoling myself that the inconsistency at the theoretical level may also be justified on the ground that the Contracts Clause states a principle that doesn't belong in the Constitution in the first place. Accordingly, there is no good reason to extend it beyond its terms and traditional scope, even if drawing the line between legislatures and courts is hard to justify based on first principles.