Justice KAGAN delivered the opinion of the Court.

A Minnesota law provides that "the dissolution or annulment of a marriage revokes any revocable[] beneficiary designation[] made by an individual to the individual's former spouse." Minn. Stat. § 524.2-804, subd. 1 (2016). That statute establishes a default rule for use when Minnesotans divorce. If one spouse has made the other the beneficiary of a life insurance policy or similar asset, their divorce automatically revokes that designation — on the theory that the policyholder would want that result. But if he does not, the policyholder may rename the ex-spouse as beneficiary.

We consider here whether applying Minnesota's automatic-revocation rule to a beneficiary designation made before the statute's enactment violates the Contracts Clause of the Constitution. We hold it does not.

I

All good trust-and-estate lawyers know that "[d]eath is not the end; there remains the litigation over the estate." 8 The Collected Works of Ambrose Bierce 365 (1911). That epigram, beyond presaging this case, helps explain the statute at its center.

The legal system has long used default rules to resolve estate litigation in a way that conforms to decedents' presumed intent. At common law, for example, marriage automatically revoked a woman's prior will, while marriage and the birth of a child revoked a man's. See 4 J. Kent, Commentaries on American Law 507, 512 (1830). The testator could then revive the old will or execute a new one. But if he (or she) did neither, the laws of intestate succession (generally prioritizing children and current spouses) would control the estate's distribution. See 95 C.J.S., Wills § 448, pp. 409-410 (2011); R. Sitkoff & J. Dukeminier, Wills, Trusts, and Estates 63 (10th ed. 2017). Courts reasoned that the average person would prefer that allocation to the one in the old will, given the intervening life events. See T. Atkinson, Handbook of the Law of Wills 423 (2d ed. 1953). If he'd only had the time, the thought went, he would have replaced that will himself.

Changes in society brought about changes in the laws governing revocation of wills. In addition to removing gender distinctions, most States abandoned the common-law rule canceling whole wills executed before a marriage or birth. In its place, they enacted statutes giving a new spouse or child a specified share of the decedent's estate while leaving the rest of his will intact. See Sitkoff & Dukeminier, Wills, Trusts, and Estates, at 240. But more important for our purposes, climbing divorce rates led almost all States by the 1980s to adopt another kind of automatic-revocation law. So-called revocation-ondivorce statutes treat an individual's divorce as voiding a testamentary bequest to a former spouse. Like the old common-law rule, those laws rest on a "judgment about the typical testator's probable intent." Id., at 239. They presume, in other words, that the average Joe does not want his ex inheriting what he leaves behind.

Over time, many States extended their revocation-on-divorce statutes from wills to "will substitutes," such as revocable trusts, pension accounts, and life insurance policies. See Langbein, The Nonprobate Revolution and the Future of the Law of Succession, 97 Harv. L. Rev. 1108, 1109 (1984) (describing nonprobate assets). In doing so, States followed the lead of the Uniform Probate Code, a model statute amended in 1990 to include a provision revoking on divorce not just testamentary bequests but also beneficiary designations to a former spouse. See §§ 2-804(a)(1), (b)(1), 8 U.L.A. 330, 330-331 (2013). The new section, the drafters wrote, aimed to "unify the law of probate and nonprobate transfers." § 2-804, Comment, id., at 333. The underlying idea was that the typical decedent would no more want his former spouse to benefit from his pension plan or life insurance than to inherit under his will. A wealth transfer was a wealth transfer — and a former spouse (as compared with, say, a current spouse or child) was not likely to be its desired recipient. So a decedent's failure to change his beneficiary probably resulted from "inattention," not "intention." Statement of the Joint Editorial Bd. for Uniform Probate Code, 17 Am. College Trust & Est. Counsel 184 (1991). Agreeing with that assumption, 26 States have by now adopted revocation-on-divorce laws substantially similar to the Code's.1 Minnesota is one.

Under prior Minnesota law, a divorce alone did not affect a beneficiary designation — but a particular divorce decree could do so. Take first the simple case: Joe names his wife Ann as beneficiary of his insurance policy, later gets divorced, but never changes the designation. Upon his death, Ann would receive the insurance proceeds — even if Joe had just forgotten to redirect the money. In other words, the insurance contract's beneficiary provision would govern after the divorce, exactly as it would have before. See Larsen v. Northwestern Nat. Life Ins. Co., 463 N.W.2d 777, 779 (Minn.App.1990). But now introduce a complication, in the form of a court addressing a spousal designation in a divorce decree. In Minnesota, as across the nation, divorce courts have always had "broad discretion in dividing property upon dissolution of a marriage." Maurer v. Maurer, 623 N.W.2d 604, 606 (Minn.2001); see 24 Am.Jur.2d, Divorce and Separation § 456 (2008). In exercising that power, a court could revoke a beneficiary designation to a soon-to-be ex-spouse; or conversely, a court could mandate that the old designation remain. See, e.g., Paul v. Paul, 410 N.W.2d 329, 330 (Minn.App.1987); O'Brien v. O'Brien, 343 N.W.2d 850, 853 (Minn.1984). Either way, the court, rather than the insured, would decide whether the ex-spouse would stay the beneficiary.

In contrast to the old law, Minnesota's new revocation-on-divorce statute starts from another baseline: the cancellation, rather than continuation, of a beneficiary designation. Enacted in 2002 to track the Code, the law provides that "the dissolution or annulment of a marriage revokes any revocable[] disposition, beneficiary designation, or appointment of property made by an individual to the individual's former spouse in a governing instrument." Minn. Stat. § 524.2-804, subd. 1. The term "governing instrument" is defined to include an "insurance or annuity policy," along with a will and other will substitutes. § 524.1-201. So now when Joe and Ann divorce, the clause naming Ann as Joe's insurance beneficiary is automatically revoked. If nothing else occurs before Joe's death, his insurance proceeds go to any contingent beneficiary named in the policy (perhaps his daughter Emma) or, failing that, to his estate. See § 524.2-804, subd. 2.

Something else, however, may well happen. As under Minnesota's former law, a divorce decree may alter the natural state of things. So in our example, the court could direct that Ann remain as Joe's insurance beneficiary, despite the normal revocation rule. See § 524.2-804, subd. 1 (providing that a "court order" trumps the rule). And just as important, the policyholder himself may step in to override the revocation. Joe, for example, could agree to a marital settlement ensuring Ann's continued status as his beneficiary. See ibid. (providing that such an agreement controls). Or else, and more simply, he could notify his insurance company at any time that he wishes to restore Ann to that position.

But enough of our hypothetical divorcees: It is time they give way to Mark Sveen and Kaye Melin, whose marriage and divorce led to this case. In 1997, Sveen and Melin wed. The next year, Sveen purchased a life insurance policy. He named Melin as the primary beneficiary, while designating his two children from a prior marriage, Ashley and Antone Sveen, as the contingent beneficiaries. The Sveen-Melin marriage ended in 2007. The divorce decree made no mention of the insurance policy. And Sveen took no action, then or later, to revise his beneficiary designations. In 2011, he passed away.

In this action, petitioners the Sveen children and respondent Melin make competing claims to the insurance proceeds. The Sveens contend that under Minnesota's revocation-on-divorce law, their father's divorce canceled Melin's beneficiary designation and left the two of them as the rightful recipients. Melin notes in reply that the Minnesota law did not yet exist when her former husband bought his insurance policy and named her as the primary beneficiary. And she argues that applying the later-enacted law to the policy would violate the Constitution's Contracts Clause, which prohibits any state "Law impairing the Obligation of Contracts." Art. I, § 10, cl. 1.

The District Court rejected Melin's argument and awarded the insurance money to the Sveens. See Civ. No. 14-5015, 2016 WL 9000457 (D.Minn., Jan. 7, 2016), App. to Pet. for Cert. 9a-16a. But the Court of Appeals for the Eighth Circuit reversed. It held that a "revocation-upon-divorce statute like [Minnesota's] violates the Contract Clause when applied retroactively." 853 F.3d 410, 412 (2017).

We granted certiorari, 583 U.S. ___, 138 S.Ct. 542, 199 L.Ed.2d 422 (2017), to resolve a split of authority over whether the Contracts Clause prevents a revocation-on-divorce law from applying to a preexisting agreement's beneficiary designation.2 We now reverse the decision below.

II

The Contracts Clause restricts the power of States to disrupt contractual arrangements. It provides that "[n]o state shall ... pass any ... Law impairing the Obligation of Contracts." U.S. Const., Art. I, § 10, cl. 1. The origins of the Clause lie in legislation enacted after the Revolutionary War to relieve debtors of their obligations to creditors. See Keystone Bituminous Coal Assn. v. DeBenedictis, 480 U.S. 470, 502-503, 107 S.Ct. 1232, 94 L.Ed.2d 472 (1987). But the Clause applies to any kind of contract. See Allied Structural Steel Co. v. Spannaus, 438 U.S. 234, 244-245, n. 16, 98 S.Ct. 2716, 57 L.Ed.2d 727 (1978). That includes, as here, an insurance policy.

At the same time, not all laws affecting pre-existing contracts violate the Clause. See El Paso v. Simmons, 379 U.S. 497, 506-507, 85 S.Ct. 577, 13 L.Ed.2d 446 (1965). To determine when such a law crosses the constitutional line, this Court has long applied a two-step test. The threshold issue is whether the state law has "operated as a substantial impairment of a contractual relationship." Allied Structural Steel Co., 438 U.S., at 244, 98 S.Ct. 2716. In answering that question, the Court has considered the extent to which the law undermines the contractual bargain, interferes with a party's reasonable expectations, and prevents the party from safeguarding or reinstating his rights. See id., at 246, 98 S.Ct. 2716; El Paso, 379 U.S., at 514-515, 85 S.Ct. 577; Texaco, Inc. v. Short, 454 U.S. 516, 531, 102 S.Ct. 781, 70 L.Ed.2d 738 (1982). If such factors show a substantial impairment, the inquiry turns to the means and ends of the legislation. In particular, the Court has asked whether the state law is drawn in an "appropriate" and "reasonable" way to advance "a significant and legitimate public purpose." Energy Reserves Group, Inc. v. Kansas Power & Light Co., 459 U.S. 400, 411-412, 103 S.Ct. 697, 74 L.Ed.2d 569 (1983).

Here, we may stop after step one because Minnesota's revocation-on-divorce statute does not substantially impair pre-existing contractual arrangements. True enough that in revoking a beneficiary designation, the law makes a significant change. As Melin says, the "whole point" of buying life insurance is to provide the proceeds to the named beneficiary. Brief for Respondent 16. But three aspects of Minnesota's law, taken together, defeat Melin's argument that the change it effected "severely impaired" her ex-husband's contract. Ibid. First, the statute is designed to reflect a policyholder's intent — and so to support, rather than impair, the contractual scheme. Second, the law is unlikely to disturb any policyholder's expectations because it does no more than a divorce court could always have done. And third, the statute supplies a mere default rule, which the policyholder can undo in a moment. Indeed, Minnesota's revocation statute stacks up well against laws that this Court upheld against Contracts Clause challenges as far back as the early 1800s.3 We now consider in detail each of the features that make this so.

To begin, the Minnesota statute furthers the policyholder's intent in many cases — indeed, the drafters reasonably thought in the typical one. As earlier described, legislatures have long made judgments about a decedent's likely testamentary intent after large life changes — a marriage, a birth, or a divorce. See supra, at 1816-1817. And on that basis, they have long enacted statutes revoking earlier-made wills by operation of law. Legislative presumptions about divorce are now especially prevalent — probably because they accurately reflect the intent of most divorcing parties. Although there are exceptions, most divorcees do not aspire to enrich their former partners. (And that is true even when an ex-spouse has custody of shared children, given the many ways to provide them with independent support.) The Minnesota statute (like the model code it tracked) applies that understanding to beneficiary designations in life insurance policies and other will substitutes. See supra, at 1819-1821. Melin rightly notes that this extension raises a brand-new constitutional question because "an insurance policy is a contract under the Contracts Clause, and a will is not." Brief for Respondent 44 (internal quotation marks omitted). But in answering that question, it matters that the old legislative presumption equally fits the new context: A person would as little want his ex-spouse to benefit from his insurance as to collect under his will. Or said otherwise, the insured's failure to change the beneficiary after a divorce is more likely the result of neglect than choice. And that means the Minnesota statute often honors, not undermines, the intent of the only contracting party to care about the beneficiary term. The law no doubt changes how the insurance contract operates. But does it impair the contract? Quite the opposite for lots of policyholders.

And even when presumed and actual intent diverge, the Minnesota law is unlikely to upset a policyholder's expectations at the time of contracting. That is because an insured cannot reasonably rely on a beneficiary designation remaining in place after a divorce. As noted above, divorce courts have wide discretion to divide property between spouses when a marriage ends. See supra, at 1820-1821. The house, the cars, the sporting equipment are all up for grabs. See Judgment and Decree in 14-cv-5015 (D Minn.), p. 51 (awarding Melin, among other things, a snowmobile and all-terrain vehicle). And (what matters here) so too are the spouses' life insurance policies, with their beneficiary provisions. Although not part of the Sveen-Melin divorce decree, they could have been; as Melin acknowledges, they sometimes are. See supra, at 1820-1821; Brief for Respondent 38. Melin counters that the Contracts Clause applies only to legislation, not to judicial decisions. See id., at 38-39; see also post, at 1830-1831 (GORSUCH, J., dissenting). That is true, but of no moment. The power of divorce courts over insurance policies is relevant here because it affects whether a party can reasonably expect a beneficiary designation to survive a marital breakdown. We venture to guess that few people, when purchasing life insurance, give a thought to what will happen in the event of divorce. But even if someone out there does, he can conclude only that ... he cannot possibly know. So his reliance interests are next to nil. And as this Court has held before, that fact cuts against providing protection under the Contracts Clause. See, e.g., El Paso, 379 U.S., at 514-515, 85 S.Ct. 577.

Finally, a policyholder can reverse the effect of the Minnesota statute with the stroke of a pen. The law puts in place a presumption about what an insured wants after divorcing. But if the presumption is wrong, the insured may overthrow it. And he may do so by the simple act of sending a change-of-beneficiary form to his insurer. (Or if he wants to commit himself forever, like Ulysses binding himself to the mast, he may agree to a divorce settlement continuing his ex-spouse's beneficiary status. See supra, at 1820-1821.) That action restores his former spouse to the position she held before the divorce — and in so doing, cancels the state law's operation. The statute thus reduces to a paperwork requirement (and a fairly painless one, at that): File a form and the statutory default rule gives way to the original beneficiary designation.

In cases going back to the 1800s, this Court has held that laws imposing such minimal paperwork burdens do not violate the Contracts Clause. One set of decisions addresses so-called recording statutes, which extinguish contractual interests unless timely recorded at government offices. In Jackson v. Lamphire, 3 Pet. 280, 7 S.Ct. 679 (1830), for example, the Court rejected a Contracts Clause challenge to a New York law granting title in property to a later rather than earlier purchaser whenever the earlier had failed to record his deed. It made no difference, the Court held, whether the unrecorded deed was "dated before or after the passage" of the statute; in neither event did the law's modest recording condition "impair[] the obligation of contracts." Id., at 290. Likewise, in Vance v. Vance, 108 U.S. 514, 2 S.Ct. 854, 27 S.Ct. 808 (1883), the Court upheld a statute rendering unrecorded mortgages unenforceable against third parties — even when the mortgages predated the law. We reasoned that the law gave "due regard to existing contracts" because it demanded only that the mortgagee make a "public registration," and gave him several months to do so. Id., at 517, 518, 2 S.Ct. 854. And more recently, in Texaco, Inc. v. Short, 454 U.S. 516, 102 S.Ct. 781, 70 L.Ed.2d 738 (1982), the Court held that a statute terminating pre-existing mineral interests unless the owner filed a "statement of claim" in a county office did not "unconstitutionally impair" a contract. Id., at 531, 102 S.Ct. 781. The filing requirement was "minimal," we explained, and compliance with it would effectively "safeguard any contractual obligations or rights." Ibid.

So too, the Court has long upheld against Contracts Clause attack laws mandating other kinds of notifications or filings. In Curtis v. Whitney, 13 Wall. 68, 20 S.Ct. 513 (1872), for example, the Court approved a statute retroactively affecting buyers of "certificates" for land offered at tax sales. The law required the buyer to notify the tax-delinquent property owner, who could then put up the funds necessary to prevent the land's final sale. If the buyer failed to give the notice, he could not take the land — and if he provided the notice, his chance of gaining the land declined. Still, the Court made short work of the Contracts Clause claim. Not "every statute which affects the value of a contract," the Court stated, "impair[s] its obligation." Id., at 70. Because the law's notice rule was "easy [to] compl[y] with," it did not raise a constitutional problem. Id., at 71. Similarly, in Gilfillan v. Union Canal Co. of Pa., 109 U.S. 401, 3 S.Ct. 304, 27 S.Ct. 977 (1883), the Court sustained a state law providing that an existing bondholder's failure to reject a settlement proposal in writing would count as consent to the deal. The law operated to reduce the interest received by an investor who did not respond. Yet the Court rebuffed the ensuing Contracts Clause suit. "If [the bondholder did] not wish to abandon his old rights and accept the new," the Court explained, "all he ha[d] to do [was] to say so in writing." Id., at 406, 3 S.Ct. 304. And one last: In Conley v. Barton, 260 U.S. 677, 43 S.Ct. 238, 67 S.Ct. 456 (1923), the Court held that the Contracts Clause did not bar a State from compelling existing mortgagees to complete affidavits before finally foreclosing on properties. The law effectively added a paperwork requirement to the mortgage contracts' foreclosure terms. But the Court said it was "only [a] condition, easily complied with, which the law, for its purposes, requires." Id., at 681, 43 S.Ct. 238.

The Minnesota statute places no greater obligation on a contracting party — while imposing a lesser penalty for noncompliance. Even supposing an insured wants his life insurance to benefit his ex-spouse, filing a change-of-beneficiary form with an insurance company is as "easy" as, say, providing a landowner with notice or recording a deed. Curtis, 13 Wall., at 71. Here too, with only "minimal" effort, a person can "safeguard" his contractual preferences. Texaco, 454 U.S., at 531, 102 S.Ct. 781. And here too, if he does not "wish to abandon his old rights and accept the new," he need only "say so in writing." Gilfillan, 109 U.S., at 406, 3 S.Ct. 304. What's more, if the worst happens — if he wants his ex-spouse to stay as beneficiary but does not send in his form — the consequence pales in comparison with the losses incurred in our earlier cases. When a person ignored a recording obligation, for example, he could forfeit the sum total of his contractual rights — just ask the plaintiffs in Jackson and Vance. But when a policyholder in Minnesota does not redesignate his ex-spouse as beneficiary, his right to insurance does not lapse; the upshot is just that his contingent beneficiaries (here, his children) receive the money. See supra, at 1820-1821. That redirection of proceeds is not nothing; but under our precedents, it gives the policyholder — who, again, could have "easily" and entirely escaped the law's effect — no right to complain of a Contracts Clause violation. Conley, 260 U.S., at 681, 43 S.Ct. 238.

In addressing those precedents, Melin mainly urges us to distinguish between two ways a law can affect a contract. The Minnesota law, Melin claims, "operate[s] on the contract itself" by "directly chang[ing] an express term" (the insured's beneficiary designation). Brief for Respondent 51; Tr. of Oral Arg. 57. In contrast, Melin continues, the recording statutes "impose[] a consequence" for failing to abide by a "procedural" obligation extraneous to the agreement (the State's recording or notification rule). Brief for Respondent 51; Tr. of Oral Arg. 58. The difference, in her view, parallels the line between rights and remedies: The Minnesota law explicitly alters a person's entitlement under the contract, while the recording laws interfere with his ability to enforce that entitlement against others. See Tr. of Oral Arg. 57-59; see also post, at 1830-1831 (GORSUCH, J., dissenting).

But we see no meaningful distinction among all these laws. The old statutes also "act[ed] on the contract" in a significant way. Tr. of Oral Arg. 59. They added a paperwork obligation nowhere found in the original agreement — "record the deed," say, or "notify the landowner." And they informed a contracting party that unless he complied, he could not gain the benefits of his bargain. Or viewed conversely, the Minnesota statute also "impose[s] a consequence" for not satisfying a burden outside the contract. Brief for Respondent 51. For as we have shown, that law overrides a beneficiary designation only when the insured fails to send in a form to his insurer. See supra, at 1823-1824. Of course, the statutes (both old and new) vary in their specific mechanisms. But they all make contract benefits contingent on some simple filing — or more positively spun, enable a party to safeguard those benefits by taking an action. And that feature is what the Court, again and again, has found dispositive.

Nor does Melin's attempt to distinguish the cases gain force when framed in terms of rights and remedies. First, not all the old statutes, as a formal matter, confined the consequence of noncompliance to the remedial sphere. In Gilfillan, for example, the result of failing to give written consent to a settlement was to diminish the interest rate a bondholder got, not to prevent him from enforcing a claim against others. And second, even when the consequence formally related to enforcement — for example, precluding an earlier purchaser from contesting a later one's title — the laws in fact wiped out substantive rights. Failure to record or notify, as noted earlier, would mean that the contracting party lost what (according to his agreement) was his land or mortgage or mineral interest. See supra, at 1824-1825. In Texaco, we replied to an argument like Melin's by saying that when the results of "eliminating a remedy" and "extinguishing a right" are "identical," the Contracts Clause "analysis is the same." 454 U.S., at 528, 102 S.Ct. 781; see El Paso, 379 U.S., at 506-507, 85 S.Ct. 577. That statement rebuts Melin's claim too. Once again: Just like Minnesota's statute, the laws discussed above hinged core contractual benefits on compliance with noncontractual paperwork burdens. When all is said and done, that likeness controls.

For those reasons, we reverse the judgment of the Court of Appeals and remand the case for further proceedings consistent with this opinion.

It is so ordered.

Justice GORSUCH, dissenting.

The Court's argument proceeds this way. Because people are inattentive to their life insurance beneficiary designations when they divorce, the legislature needs to change those designations retroactively to ensure they aren't misdirected. But because those same people are simultaneously attentive to beneficiary designations (not to mention the legislature's activity), they will surely undo the change if they don't like it. And even if that weren't true, it would hardly matter. People know existing divorce laws sometimes allow courts to reform insurance contracts. So people should know a legislature might enact new laws upending insurance contracts at divorce. For these reasons, a statute rewriting the most important term of a life insurance policy — who gets paid — somehow doesn't "substantially impair" the contract. It just "makes a significant change." Ante, at 1822.

Respectfully, I cannot agree. Minnesota's statute automatically alters life insurance policies upon divorce to remove a former spouse as beneficiary. Everyone agrees that the law is valid when applied prospectively to policies purchased after the statute's enactment. But Minnesota wants to apply its law retroactively to policies purchased before the statute's adoption. The Court of Appeals held that this violated the Contracts Clause, which guarantees people the "right to `rely on the law... as it existed when the[ir] contracts were made.'" Metropolitan Life Ins. Co. v. Melin, 853 F.3d 410, 413 (C.A.8 2017) (quoting Whirlpool Corp. v. Ritter, 929 F.2d 1318, 1323 (C.A.8 1991)). That judgment seems to me exactly right.

I

Because legislation often disrupts existing social arrangements, it usually applies only prospectively. This longstanding and "sacred" principle ensures that people have fair warning of the law's demands. Reynolds v. McArthur, 2 Pet. 417, 434, 7 S.Ct. 470 (1829); 3 H. Bracton, De Legibus et Consuetudinibus Angliae 530-531 (1257) (T. Twiss ed. 1880). It also prevents majoritarian legislatures from condemning disfavored minorities for past conduct they are powerless to change. See, e.g., Landgraf v. USI Film Products, 511 U.S. 244, 266, 114 S.Ct. 1483, 128 L.Ed.2d 229 (1994); Vermeule, Veil of Ignorance Rules in Constitutional Law, 111 Yale L.J. 399, 408 (2001).

When it comes to legislation affecting contracts, the Constitution hardens the presumption of prospectivity into a mandate. The Contracts Clause categorically prohibits states from passing "any ... Law impairing the Obligation of Contracts." Art. I, § 10, cl. 1 (emphasis added). Of course, the framers knew how to impose more nuanced limits on state power. The very section of the Constitution where the Contracts Clause is found permits states to take otherwise unconstitutional action when "absolutely necessary," if "actually invaded," or "wit[h] the Consent of Congress." Cls. 2 and 3. But in the Contracts Clause the framers were absolute. They took the view that treating existing contracts as "inviolable" would benefit society by ensuring that all persons could count on the ability to enforce promises lawfully made to them — even if they or their agreements later prove unpopular with some passing majority. Sturges v. Crowninshield, 4 Wheat. 122, 206, 4 S.Ct. 529 (1819).

The categorical nature of the Contracts Clause was not lost on anyone, either. When some delegates at the Constitutional Convention sought softer language, James Madison acknowledged the "`inconvenience'" a categorical rule could sometimes entail "`but thought on the whole it would be overbalanced by the utility of it.'" Kmiec & McGinnis, The Contract Clause: A Return to the Original Understanding, 14 Hastings Const. L.Q. 525, 529-530 (1987). During the ratification debates, these competing positions were again amply aired. Antifederalists argued that the proposed Clause would prevent states from passing valuable legislation. Id., at 532-533. Federalists like Madison countered that the rule of law permitted "property rights and liberty interests [to] be dissolved only by prospective laws of general applicability." Id., at 532. And, of course, the people chose to ratify the Constitution — categorical Clause and all.

For much of its history, this Court construed the Contracts Clause in this light. The Court explained that any legislative deviation from a contract's obligations, "however minute, or apparently immaterial," violates the Constitution. Green v. Biddle, 8 Wheat. 1, 84, 5 S.Ct. 547 (1823). "All the commentators, and all the adjudicated cases upon Constitutional Law agree[d] in th[is] fundamental propositio[n]." Winter v. Jones, 10 Ga. 190, 195 (1851). But while absolute in its field, the Clause also left significant room for legislatures to address changing social conditions. States could regulate contractual rights prospectively. Ogden v. Saunders, 12 Wheat. 213, 262, 6 S.Ct. 606 (1827). They could retroactively alter contractual remedies, so long as they did so reasonably. Sturges, supra, at 200. And perhaps they could even alter contracts without "impairing" their obligations if they made the parties whole by paying just compensation. See West River Bridge Co. v. Dix, 6 How. 507, 532-533, 12 S.Ct. 535 (1848); El Paso v. Simmons, 379 U.S. 497, 525, 85 S.Ct. 577, 13 L.Ed.2d 446 (1965) (Black, J., dissenting). But what they could not do is destroy substantive contract rights — the "Obligation of Contracts" that the Clause protects.

More recently, though, the Court has charted a different course. Our modern cases permit a state to "substantial[ly] impai[r]" a contractual obligation in pursuit of "a significant and legitimate public purpose" so long as the impairment is "`reasonable.'" Energy Reserves Group, Inc. v. Kansas Power & Light Co., 459 U.S. 400, 411-412, 103 S.Ct. 697, 74 L.Ed.2d 569 (1983). That test seems hard to square with the Constitution's original public meaning. After all, the Constitution does not speak of "substantial" impairments — it bars "any" impairment. Under a balancing approach, too, how are the people to know today whether their lawful contracts will be enforced tomorrow, or instead undone by a legislative majority with different sympathies? Should we worry that a balancing test risks investing judges with discretion to choose which contracts to enforce — a discretion that might be exercised with an eye to the identity (and popularity) of the parties or contracts at hand? How are judges supposed to balance the often radically incommensurate goods found in contracts and legislation? And does this test risk reducing the "Contract Clause's protection" to the "Court's judgment" about the "`reasonableness'" of the legislation at hand? Simmons, 379 U.S., at 529, 85 S.Ct. 577 (Black, J., dissenting). Many critics have raised serious objections along these and other lines. See, e.g., ibid.; Kmiec & McGinnis, supra, at 552; Rappaport, Note, A Procedural Approach to the Contract Clause, 93 Yale L.J. 918, 918 (1984); Epstein, Toward a Revitalization of the Contract Clause, 51 U. Chi. L. Rev 703, 705-717 (1984); J. Ely, The Contract Clause: A Constitutional History 7-29 (2016). They deserve a thoughtful reply, if not in this case then in another.

II

Even under our modern precedents, though, I still do not see how the statute before us might survive unscathed. Recall that our recent precedents indicate a state law "substantially impairing" contracts violates the Contracts Clause unless it is "reasonable" in light of a "significant and legitimate public purpose."

Start with the substantial impairment question. No one pays life insurance premiums for the joy of it. Or even for the pleasure of knowing that the insurance company will eventually have to cough up money to someone. As the Court concedes, the choice of beneficiary is the "`whole point.'" Ante, at 1821-1822. So when a state alters life insurance contracts by undoing their beneficiary designations it surely "substantially impairs" them. This Court has already recognized as much, holding that a law "displac[ing] the beneficiary selected by the insured ... and plac[ing] someone else in her stead... frustrates" a scheme designed to deliver proceeds to the named beneficiary. Hillman v. Maretta, 569 U.S. 483, 494, 133 S.Ct. 1943, 186 L.Ed.2d 43 (2013) (quoting Wissner v. Wissner, 338 U.S. 655, 659, 70 S.Ct. 398, 94 S.Ct. 424 (1950) (internal quotation marks omitted)). As Justice Washington explained long ago, legislation "changing the objects of [the donor's] bounty ... changes so materially the terms of a contract" that the law can only be said to "impair its obligation." Trustees of Dartmouth College v. Woodward, 4 Wheat. 518, 662, 4 S.Ct. 629 (1819) (concurring opinion). Just so.

Cases like ours illustrate the point. Kaye Melin testified that, despite their divorce, she and the decedent, Mark Sveen, agreed (repeatedly) to keep each other as the primary beneficiaries in their respective life insurance policies. Affidavit of Kaye Melin in No. 14-cv-05015, Dkt. No. 46, ¶¶ 3, 4, 10-14. Ms. Melin noted that they adopted this arrangement not only because they remained friends but because they paid the policy premiums from their joint checking account. Deposition of Kaye Melin in No. 14-cv-0515, Dkt. No. 45-4, pp. 26-27, 64-65. Of course, we don't know for sure whether removing Ms. Melin as beneficiary undid Mr. Sveen's true wishes. The case comes to us after no one was able to meet Minnesota's clear and convincing evidence standard to prove Mr. Sveen's intent. But what we do know is the retroactive removal of Ms. Melin undid the central term of the contract Mr. Sveen signed and left in place for years, even after his divorce, until the day he died.

Nor are arrangements like the ones Ms. Melin described so unusual. As the federal government has recognized, revocation on divorce statutes cannot be assumed to "effectuat[e] the insured's `true' intent" because a policyholder "might want his ex-spouse to receive insurance proceeds for a number of reasons — out of a sense of obligation, remorse, or continuing affection, or to help care for children of the marriage that remain in the ex-spouse's custody." Brief for United States as Amicus Curiae in Hillman v. Maretta, O.T. 2012, No. 11-1221, p. 28. After all, leaving your ex-spouse life insurance proceeds can be a cheaper, quicker, and more private way to provide for minor or disabled children than leaving the matter to a trustee or other fiduciary. See, e.g., Feder & Sitkoff, Revocable Trusts and Incapacity Planning: More Than Just a Will Substitute, 24 Elder L.J. 1, 15-18 (2016). For these reasons, the federal government and nearly half the states today do not treat divorce as automatically revoking insurance beneficiary designations. Brief for Petitioners 8-9, and nn. 1-2; Hillman, supra, at 494-495.

Consider next the question of the impairment's reasonableness. Our cases suggest that a substantial impairment is unreasonable when "an evident and more moderate course would serve [the state's] purposes equally well." United States Trust Co. of N.Y. v. New Jersey, 431 U.S. 1, 31, 97 S.Ct. 1505, 52 L.Ed.2d 92 (1977); see also Allied Structural Steel Co. v. Spannaus, 438 U.S. 234, 247, 98 S.Ct. 2716, 57 L.Ed.2d 727 (1978) (analyzing whether an impairment of private contracts "was necessary to meet an important general social problem"). Here, Minnesota's stated purpose is to ensure proceeds aren't misdirected to a former spouse because a policyholder forgets to update his beneficiary designation after divorce. But the state could have easily achieved that goal without impairing contracts at all. It could have required courts to confirm that divorcing couples have reviewed their life insurance designations. See Va. Code Ann. § 20-111.1(E) (2017); Utah Code § 30-3-5(1)(e)(i) (2018). It could have instructed insurance companies to notify policyholders of their right to change beneficiary designations. It could have disseminated information on its own. Or it could have required attorneys in divorce proceedings to address the question with affected parties. A host of women's rights organizations have advocated for these and other alternatives in various states. See, e.g., Brief for Women's Law Project et al. as Amici Curiae 34-35. Yet there's no evidence Minnesota investigated any of them, let alone found them wanting.

III

What's the Court's reply? It says that we don't have to decide whether the statute reasonably impairs contracts because it doesn't substantially impair them in the first place. It's easy enough to see why the Court might take this tack given the many obvious and less burdensome alternatives Minnesota never considered. To save the law, the Court must place all its chips on a "no substantial impairment" argument. The gamble, though, proves a tricky one.

The Court first stresses that individuals sometimes neglect their beneficiary designations after divorce. Because of this, it says, Minnesota's law affords "many" persons what they would want if only they had thought about it. Ante, at 1822 1823. But as we've seen the law depends on a stereotype about divorcing couples that not everyone fits. A sizeable (and maybe growing) number of people do want to keep their former spouses as beneficiaries. Brief for Women's Law Project 25-26. Even the Court admits the law's presumption will sometimes prove "wrong." Ante, at 1823-1824. And that tells us all we need to know. That the law is only sometimes wrong in predicting what divorcing policyholders want may go some way to establishing its reasonableness at the second step of our inquiry. But at the first step, where we ask only whether the law substantially impairs contracts, the answer is unavoidable. The statute substantially impairs contracts by displacing the term that is the "`whole point'" of the contract. Ante, at 1821-1822. This Court would never say a law doesn't substantially burden a minority's religious practice because it reflects most people's preferences. See Church of Lukumi Babalu Aye, Inc. v. Hialeah, 508 U.S. 520, 113 S.Ct. 2217, 124 L.Ed.2d 472 (1993). Equally, I do not see how a statute doesn't substantially impair contracts just because it reflects "many" people's preferences. Ante, at 1822-1823. The Contracts Clause does not seek to maximize the bottom line but to protect minority rights "from improvident majoritarian impairment." L. Tribe, American Constitutional Law § 9-8, p. 613 (2d ed. 1988).

The Court's answer to this problem introduces an apparent paradox. If the statute substantially impairs contracts, it says, the impairment can be easily undone. Anyone unhappy with the statute's beneficiary re-designation can just re-re-designate the beneficiary later. Ante, at 1823-1824. Yet the Court just finished telling us the statute is justified because most policyholders neglect their beneficiary designations after divorce. Both claims cannot be true. The statute cannot simultaneously be necessary because people are inattentive to the details of their insurance policies and constitutional because they are hyperaware of those same details.

Perhaps seeking a way out of this problem, the Court offers an entirely different line of argument. Here the Court suggests the statute doesn't substantially impair contracts because it does no more than a divorce court might. Ante, at ___-___. But this argument doesn't work either. Courts may apply pre-existing law to alter a beneficiary designation to ensure an equitable distribution of marital property in specific cases. That hardly means legislatures may retroactively change the law to rearrange beneficiary designations for everyone. A court can fine you for violating an existing law against jaywalking. That doesn't mean a legislature could hold you retroactively liable for violating a new law against jaywalking that didn't exist when you crossed the street. No one would take that idea seriously when it comes to crime, and the Contracts Clause ensures we don't when it comes to contracts, either. After all, the Clause applies only to the "law[s]" legislatures "pass," not to the rulings of courts. Tidal Oil Co. v. Flanagan, 263 U.S. 444, 451, 44 S.Ct. 197, 68 S.Ct. 382 (1924) (emphasis deleted). 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.

The Court finally claims that its course finds support in cases where we've approved retroactive legislation. Ante, at 1823-1825. Those cases, though, involved statutes altering contractual remedies. Home Building & Loan Assn. v. Blaisdell, 290 U.S. 398, 434, and n. 13, 54 S.Ct. 231, 78 S.Ct. 413 (1934) (noting that each of the 19th century cases relied on by the Court today affected only "remedial processes"). And Minnesota's law changes the key contractual obligation — who gets the insurance proceeds — not the method by which the contract's existing obligation is satisfied. Although the Constitution allows legislatures some flexibility to address changing social conditions through retroactive remedial legislation, it does not permit upsetting settled expectations in contractual obligations. See, e.g., Fletcher v. Peck, 6 Cranch 87, 137-138, 3 S.Ct. 162 (1810); Simmons, 379 U.S., at 526 (Black, J., dissenting). We must respect that line found in the text of the Constitution, not elide it. Indeed, our precedent teaches that if remedial changes are just disguised efforts at impairing obligations they will violate the Constitution too. Blaisdell, 290 U.S., at 434, n. 13, 54 S.Ct. 231 (collecting cases).

Consider just how different our case is from the classic remedial change the Contracts Clause permits. In Jackson v. Lamphire, 3 Pet. 280, 7 S.Ct. 679 (1830), a shady landowner sold the same tract to two people. Id., at 287-288. The Court held that the second buyer was entitled to keep the land because he recorded the deed as a retroactive law required. Id., at 289-290. At the same time, nothing in Jackson or the new statute stopped the first buyer (who failed to record his deed) from obtaining damages from the seller for breach of contract. See id., at 287-291. The statute altered the first buyer's remedy, but he remained free to enforce the obligation found in his contract. By contrast, the statute here changes the "`whole point'" of the contract's obligation, substituting a new beneficiary in place of the one found in the contract's terms. Ante, at 1821-1822.

Even the remedial case on which the Court leans most heavily does little to help its cause. In Gilfillan v. Union Canal Co. of Pa., 109 U.S. 401, 3 S.Ct. 304, 27 S.Ct. 977 (1883), the Court upheld a statute requiring bondholders to enforce their contract rights within a shortened timeframe (that is, altering the remedy) or else accept a reorganization plan that threatened a poorer rate of interest. Id., at 402-403, 406, 3 S.Ct. 304. The Court gave three primary reasons for upholding this change. It emphasized that the bonds at issue were "of a peculiar character" because "each bondholder under them enter[ed] by fair implication into certain contract relations with" the other bondholders who approved the reorganization. Id., at 403, 3 S.Ct. 304. It observed that "`a calamity common to all'" had occurred, as the company that issued the bonds "was bankrupt" and payment of "its debts in the ordinary way was impossible." Id., at 405, 3 S.Ct. 304. Finally, it added that the plaintiff challenging the statute had "actual notice" of the law and so faced no difficulty in asserting his contract rights in a timely manner. Id., at 406, 3 S.Ct. 304. These considerations, the Court concluded, justified shortening the limitations period for obtaining full relief even though it might reduce a late-moving party's interest rate a few points. No comparable considerations are present here. And this statute doesn't just reduce Ms. Melin's remedy; it denies her one altogether.

*

The judicial power to declare a law unconstitutional should never be lightly invoked. But the law before us cannot survive an encounter with even the breeziest of Contracts Clause tests. It substantially impairs life insurance contracts by retroactively revising their key term. No one can offer any reasonable justification for this impairment in light of readily available alternatives. Acknowledging this much doesn't even require us to hold the statute invalid in all applications, only that it cannot be applied to contracts formed before its enactment. I respectfully dissent.