dissenting.

As a matter of state law, the Maine Supreme Judicial Court imposed a constructive trust on the proceeds of Sergeant Ridgway’s life insurance. The trust effectuates a settlement agreement and an express judicial decree that commanded Ridgway to maintain the policy in effect for the benefit of his minor children.1 The propriety of the imposi*72tion of a constructive trust under Maine law is, of course, not a matter for us to review.2 Unless the application of this well-established equitable doctrine does “major damage” to “clear and substantial federal interests,”3 we must respect it.

Notwithstanding the absence of any such major damage, the Court today decides that the Maine court’s decision conflicts with two provisions of the Servicemen’s Group Life Insurance Act (SGLIA), 38 U. S. C. §§765-776.4 The Court finds a conflict with § 770(a) of the statute, which gives the serviceman the right to designate his beneficiary, and with § 770(g), which exempts the insurance proceeds from taxation and from seizure by legal or equitable process. Because the Court in Wissner v. Wissner, 338 U. S. 655, relied on similar provisions of the National Service Life Insurance Act of 1940, 54 Stat. 1008, in rejecting a claim to insurance proceeds paid under that statute, the Court today concludes that Wissner is controlling and that it must reach a similar result.

Unquestionably, there is a strong federal interest in protecting federally supported benefits from claims of the recipient’s commercial creditors.6 There is also a federal interest, much less clearly defined, in permitting a federal serviceman *73to designate the beneficiary of his insurance policy. Both of these federal interests supported the rejection of the estranged wife’s claim in Wissner. A careful examination of this case, however, demonstrates that neither of these interests is compromised by the decision of the Maine Supreme Judicial Court.

I

Since the alleged conflict with the exemption provision is more obvious in this case, and concerns a more substantial federal interest, I address it first. The statute provides:

“Payments of benefits due or to become due under Servicemen’s Group Life Insurance or Veterans’ Group Life Insurance made to, or on account of, a beneficiary shall be exempt from taxation, shall be exempt from the claims of creditors, and shall not be liable to attachment, levy, or seizure by or under any legal or equitable process whatever, either before or after receipt by the beneficiary.” 38 U. S. C. § 770(g).

This provision prohibits a commercial creditor from securing insurance proceeds in the hands of the beneficiary, regardless *74of any contrary agreement made by the insured or any terms of state law. Although the majority concludes that this provision also prohibits the state court from recognizing respondents’ claim in this case, ante, at 60, it is most unlikely that Congress intended § 770(g) to operate as a bar to claims advanced by an insured’s dependents for support.

The language used in the “anti-attachment” provision of the SGLIA is comparable to that found in so-called “spendthrift clauses” that have protected trust beneficiaries from the claims of commercial creditors for centuries.6 As stated by Dean Griswold, “[i]t is widely held, however, that even where such trusts are generally valid, the interest of the beneficiary may be reached for the support of his wife or children, or for the payment of alimony to his wife.” E. Griswold, Spendthrift Trusts 389 (2d ed. 1947).7 Prior to the decision of this Court in Wissner, a number of courts had held that statutory “spendthrift” provisions did not bar a claim for alimony or support.8 Many of these cases in fact *75concerned exemption provisions applicable to veterans’ benefits programs. As summarized in one treatise:

“And claims for the support and care of minor children of an incompetent veteran have been held not to be subject to the exemption, as the obligation of a father to support his minor children is not a debt within the meaning of the statute, but is an obligation growing out of the parental status and public policy.” R. Kimbrough & J. Glen, American Law of Veterans 32 (2d ed. 1954).9

A thoughtful and expansive opinion of Justice Rutledge, then a member of the United States Court of Appeals for the District of Columbia, best explains the rationale of these decisions. In Schlaefer v. Schlaefer, 71 App. D. C. 350, 112 F. 2d 177 (1940), the court considered a claim for arrears in alimony payments. Plaintiff sought sequestration of her former husband’s property, including $100 per month that he received as disability benefit payments under the Life Insurance Act for the District of Columbia. Defendant responded that these payments were exempted specifically from process under the express language of § 16(a) of that federal statute.10

*76The court in Schlaefer stated that “[t]he basic issue boils down to whether Congress intended to relieve the disabled insured to the extent of his disability payments from legally enforceable obligation to support his family and those legally dependent upon him.” Id., at 358, 112 F. 2d, at 185. The court recognized:

“So far as general creditors are concerned the purpose is clear, with the exceptions stated, to make the disposition of these funds a matter solely for his judgment. Congress regarded it as better for the creditors to go unpaid than to deprive the debtor and his dependents of this means of support when earning capacity would be cut off. Hence it used broad language prohibiting recourse to the fund by legal process.” Ibid.

The court determined, however, that the insured’s legal dependents were not to be classified, for purposes of the statute, “with strangers holding claims hostile to his interest.” Ibid. The court noted that “the usual purpose of exemptions is to relieve the person exempted from the pressure of claims hostile to his dependents’ essential needs as well as his own personal ones, not to relieve him of familial obligations and destroy what may be the family’s last and only security, short of public relief.” Ibid.

The court concluded that this construction was “not inconsistent with giving full effect to the statute.” Id., at 359, 112 F. 2d, at 186. As explained by the court:

“The protection remaining is broad, applying both to ‘debts’ and to ‘liabilities.’ Furthermore, it renders the *77statute consistent with others which provide methods for enforcement of the husband’s and the father’s duty of support. Any other would nullify them in circumstances where the disability payments constitute the sole source of livelihood, though they might be adequate to support the insured and all his dependents in luxury. We cannot believe that Congress intended to create an exemption so broad and so inconsistent with the policy which it has declared in other acts.” Ibid, (footnote omitted).

The court further noted that its construction of the exemption statute was consistent with other authorities, which had held that a claim for support was not a “debt” or a “liability” in the ordinary usages of those terms.11

In Wissner, the Court did not repudiate this distinction between family and business obligations. Rather, in ruling that the exemption statute was applicable in that case, the Court expressly recognized this distinction and placed the estranged wife’s community property claim in the business category. As stated by the Court, “we must note that the community property principle rests upon something more than the moral obligation of supporting spouse and children: the business relationship of man and wife for their mutual monetary profit.” 338 U. S., at 660.12 As a result, it simply can*78not be said that Wissner commands that an exemption statute such as that present in this case stands as a bar to claims based on familial obligations.

Although Wissner left open the question presented in this case, there is nothing in the language of the SGLIA or its legislative history that evidences an intent by Congress to repudiate this distinction between commercial and family obligations.13 The federal interest incorporated within exemption statutes is an interest in preventing federally supported benefits from satisfying claims of commercial credi*79tors. Although such claims are certainly valid, they arise solely from a personal obligation of the debtor, and should not be borne by the public through payment from general revenues. Claims based on familial obligation, however, are of a different character, and indeed may be precisely the type of claim for which the federal benefit was intended.14 Absent some indication that Congress intended the standard exemption provision contained in the SGLIA to bar a minor child’s claim for support, I am unwilling to conclude that this provision of the statute pre-empts the application of state law in this case.

II

When the exemption provision is put to one side, the only support for the Court’s pre-emption holding is the statutory provision giving the serviceman the right to designate the beneficiary of his insurance policy.15 In order to determine whether the decision of the Maine court has done “major damage” to the federal interests underlying this statutory provision, it is first appropriate to identify those federal interests precisely.

The right to designate the beneficiary of an insurance policy is a common feature in insurance contracts. It surely is not a right that can be characterized as uniquely federal in any sense. Moreover, the mere fact that the right has its *80source in a federal statute does not require that it be given a construction different from that given a comparable right created by state law or by private contract. As stated by this Court in Hisquierdo v. Hisquierdo, 439 U. S. 572, 583, “[t]he federal nature of the benefits does not by itself proscribe the entire field of state control.”

To be sure, the Court in Wissner speculated that “[possession of government insurance, payable to the relative of his choice, might well directly enhance the morale of the serviceman.” 338 U. S., at 660. This interest in permitting a serviceman to designate the beneficiary of his insurance policy is not compromised in this case, however. It cannot be said that state law forces a distribution of the insurance proceeds that is inconsistent with the federal policy of permitting Sergeant Ridgway to choose his beneficiary. In a freely negotiated child custody and support settlement, Ridgway agreed to maintain his former wife as the beneficiary of the policy for the benefit of his minor children. Ridgway himself made that choice; the question presented in this case, therefore, is whether any provision of the statute espouses a federal interest in permitting him to change his beneficiary in derogation of an accepted obligation to provide support for his children. I can find no section of the statute that expresses such an interest. The result reached by the Court today surely cannot be justified by the need to maintain the “morale” of our Armed Forces.

The history of the statutory provision defining the serviceman’s right to designate his beneficiary supports the conclusion that § 770(a) does not pre-empt state law in this case. Originally, servicemen could name as beneficiaries only those persons who fell within a limited, defined class.16 At the *81time Wissner was decided, servicemen could designate only a spouse, child, grandchild, parent, or sibling as a beneficiary of a National Service Life Insurance policy. The designation provision at issue in Wissner thus added support for the proposition that insurance proceeds were intended to benefit only immediate family members and dependents of the serviceman, and not any other party.

When Congress enacted the SGLIA in 1965, however, it removed all limitations on eligible beneficiaries. 79 Stat. 883. Any person may be named as beneficiary of the policy, including a commercial creditor. Today, the Court gives priority to the claim of any such designated beneficiary. Thus, as a result of its decision, a loan shark, a camp follower, or a total stranger designated as beneficiary would have priority over claims of dependent family members, even though those claims were incorporated in a voluntary settlement agreement and an express judicial decree. This result simply was not possible at the time Wissner was decided. No federal interest justifies such an absolute and unqualified priority for the designated beneficiary.17

*82It is ironic that today’s decision may harm federal interests in a more tangible way than that ascribed to the decision of the Maine Supreme Judicial Court. As a result of the holding today, a commitment to keep military insurance in effect for one’s children is not legally binding. In the future, a serviceman in divorce negotiations may be forced to purchase new insurance from a private insurer in order to provide fair assurance that his support obligation will remain satisfied in the event of his death. For many servicemen, such private insurance may not be easy to obtain. Surely there is no federal interest in depreciating the value of this insurance.

I respectfully dissent.

delivered the opinion of the Court.

This case presents the issue whether an insured serviceman’s beneficiary designation under a life policy issued pursuant to the Servicemen’s Group Life Insurance Act of 1965 (SGLIA), Pub. L. 89-214, 79 Stat. 880, prevails over a constructive trust imposed upon the policy proceeds by a state-court decree.

*48I — I

The Facts

Richard H. Ridgway was a career sergeant in the United States Army. April D. Ridgway was his wife. Richard and April were the parents of three children, Hayley, Laurie, and Brady, all minors. The Ridgways’ marriage, however, ended with a divorce granted by a Maine court on December 7, 1977. The state divorce judgment, entered on April’s complaint and apparently following property settlement negotiations, ordered Richard, among other things, to pay specified amounts monthly for the support of the three children. App. 13. It also ordered him

“to keep in force the life insurance policies on his life now outstanding for the benefit of the parties’ three children. If any of such insurance policies should subsequently be terminated for any reason, defendant shall immediately replace it with other life insurance of equal amount for the benefit of the children.” Id., at 14.

Sergeant Ridgway’s life was then insured under a $20,000 policy issued by Prudential Insurance Company of America pursuant to a group contract with the Administrator of Veterans’ Affairs. At the time of the Ridgways’ divorce, April was the designated beneficiary of that policy.

On March 28, 1978, less than four months after the divorce, Ridgway married his second wife, Donna, the individual petitioner here. Six days later, the sergeant, as insured, changed the policy’s beneficiary designation to one directing that its proceeds be paid as specified “by law.” This referred to the statutory order of beneficiary precedence set forth in 38 U. S. C. § 770(a). See also 38 CFR §9.16(i) (1980). Under that statutory prescription, the policy proceeds, in the event of Ridgway’s death, would be paid to his *49“widow,” that is, his “lawful spouse ... at the time of his death.” 38 U. S. C. § 765(7).

Sergeant Ridgway died on January 5, 1979. Donna survived him and was his lawful wife at the time of his death. Both April and Donna filed claims for the proceeds of the policy. April based her claim, which was on behalf of the children, on the divorce decree. Donna’s claim rested on the beneficiary designation and her status as Ridgway’s widow.

April thereafter instituted the present suit in the Superior Court for Androscoggin County, Me. As legal representative of the three minor children, she sued Prudential, seeking both to enjoin the payment of the policy proceeds to Donna, and to obtain a declaratory judgment that those proceeds were payable to the children. Donna joined the litigation and was aligned as a plaintiff asserting a claim to the proceeds. April then filed a cross-claim against Donna, praying for the imposition of a constructive trust, for the benefit of the children, on any policy proceeds paid to Donna. Prudential supported Donna’s position.

The Superior Court rejected April Ridgway’s claims. It acknowledged that the terms of the judgment of divorce and the beneficiary designation were inconsistent.1 But it felt that the imposition of a constructive trust would interfere with the operation of the federal SGLIA, and that such a disposition would therefore run afoul of the Supremacy Clause, U. S. Const., Art. VI, cl. 2. App. 38-43.

On the ensuing appeal to the Supreme Judicial Court of Maine, the parties stipulated, inasmuch as the policy proceeds by that time had been deposited in court, that the sole *50issue was “[wjhether or not the presiding justice erred in ruling that, on the basis of the facts found, he could not impose a constructive trust on the proceeds of Sergeant Ridgway’s insurance.” Id., at 48. That court, sympathetic to April, vacated the Superior Court’s dismissal of her cross-claim, and remanded the case with directions to enter an order naming Donna as constructive trustee of the policy proceeds. The Court Clerk, who held the proceeds, was directed to pay them to April for and on behalf of the three children. Ridgway v. Prudential Ins. Co. of America, 419 A. 2d 1030, 1035 (1980).

We granted certiorari, 450 U. S. 979 (1981), to review the important issue presented by the case.

II

The Statutory Background

In order to make life insurance coverage available to members of the uniformed services on active duty, particularly in combat zones, Congress in 1965 enacted the SGLIA. See H. R. Rep. No. 1003, 89th Cong., 1st Sess., 7 (1965). The impetus for the legislation was the escalating level of hostilities and casualties in the then ongoing Vietnam conflict; this had prompted private commercial insurers to restrict coverage for service members.2 See 111 Cong. Rec. 24339 (1965) (remarks of Rep. Teague, Chairman of the House Committee on Veterans’ Affairs); see also S. Rep. No. 619, 89th Cong., 1st Sess., 3 (1965). The earlier program of federally sponsored life insurance for service members, see National Service Life Insurance Act of 1940, 54 Stat. 1008, and National Service Life Insurance Act of 1958, as amended, 38 U. S. C. § 701 et seq. (NSLIA), placed in effect shortly before the involvement of this country in World War II, had been allowed *51to lapse after the end of the Korean hostilities when commercial insurance generally became available to service members.3 Accordingly, NSLIA coverage could not be obtained by many service members on active duty in 1965. See 111 Cong. Rec. 24339 (1965) (remarks of Rep. Teague).

Although its purposes and provisions resemble those of the NSLIA in many respects, the SGLIA differs from the predecessor program in that it directs the Administrator of Veterans’ Affairs to purchase coverage from one or more qualified commercial insurers instead of offering coverage by the United States itself. See 38 U. S. C. §766. Thus, under the SGLIA, the Government is the policyholder, rather than the insurer. The Administrator has contracted with petitioner Prudential Insurance Company of America, which now serves as the primary insurer under the SGLIA and which operates, under Veterans’ Administration supervision and pursuant to 38 U. S. C. § 766(b), the Office of Servicemen’s Group Life Insurance in Newark, N. J.

The SGLIA initially provided insurance only for members serving in specified services. 79 Stat. 880. The maximum coverage allowed was then $10,000. Id., at 881. Since 1965, however, statutory changes have expanded both eligibility for coverage and the amount of insurance available.4 The program is operated on a presumptive enrollment basis; coverage is provided automatically and premiums are withheld from the service member’s pay, unless the insurance is expressly declined or is terminated by written election. 38 U. S. C. §§ 767(a) and 769.5

*52In order to make the insurance available through a commercial carrier at a reasonable rate, notwithstanding the special mortality risks that service members often must assume, Congress undertook to subsidize the program. See S. Rep. No. 91-398, p. 2 (1969). A sum representing the extra premium for special mortality risks is periodically deposited by the United States into a revolving fund that is used to pay premiums on the master policy. See 38 U. S. C. §§ 769(b) and (d)(1). The fund otherwise is derived primarily from deductions withheld from service members’ pay. §§ 769(a)(1) and (d)(1). Accordingly, depending upon the conditions faced by service members at any given time, the program may be financed in part with federal funds. See S. Rep. No. 91-398, at 2.

The SGLIA establishes a specified “order of precedence,” 38 U. S. C. § 770(a), for policy beneficiaries. By this statutory provision, the proceeds of a policy are paid first to such “beneficiary or beneficiaries as the member . . . may have designated by [an appropriately filed] writing received prior to death.” If there be no such designated beneficiary, the proceeds go to the widow or widower of the service member or, if there also be no widow or widower, “to the child or children of such member . . . and descendants of deceased children by representation.” Parents, and then the representative of the insured’s estate (an obvious bow at this point in the direction of state law), are next in order. Ibid. See also 38 CFR §9.16(i) (1980).

In 1970, by Pub. L. 91-291, §5, 84 Stat. 330, Congress added an anti-attachment provision. With certain exceptions not applicable here, this provision shields payments made under § 770(a) “from taxation” and from “claims of creditors,” and states that the payments “shall not be liable to attachment, levy, or seizure by or under any legal or equitable *53process whatever, either before or after receipt by the beneficiary.” § 770(g).

Pursuant to his general rulemaking authority over veterans’ programs, § 210(c)(1), the Administrator has promulgated regulations implementing the SGLIA. These provide that the insured “may designate any person, firm, corporation or legal entity” as a policy beneficiary, and any such “designation or change of beneficiary . . . will take effect only if it is in writing, signed by the insured and received [by the appropriate office] prior to the death of the insured.” 38 CFR §§ 9.16(a) and (d) (1980). A change of beneficiary “may be made at any time and without the knowledge or consent of the previous beneficiary.” § 9.16(e). And “[n]o change or cancellation of beneficiary ... in a last will or testament, or in any other document shall have any force or effect unless such change is received by the appropriate office.” § 9.16(f).

l — l l — l l — l

The foregoing description of the statutory plan adopted by Congress, and implemented by the Administrator’s regulations, demonstrates the pervasive and detailed characteristics of the congressional specifications. The obvious and stated concern of Congress was to provide coverage for the member, no matter how hazardous the duty, and thus protection for the member’s designated beneficiaries. The legislation itself says nothing about contrary dictates of state law or state judgments.

The Supreme Judicial Court of Maine, however, concluded that the order of beneficiary precedence set forth in 38 U. S. C. § 770(a) “does not reflect any federal interest in permitting a serviceman to evade the responsibility to provide for his minor children imposed both by virtue of his voluntary agreement and by the express provision of a valid state court decree.” 419 A. 2d, at 1033. That court further concluded that the anti-attachment provision, § 770(g), “has no application to the instant case since its purpose is to protect the pro*54ceeds of the insurance from the claims of creditors.” It pointed out that it was concerned “not with the claim of a creditor but with the claims of minor children who assert an equitable interest in the proceeds arising from their deceased father’s voluntary agreement and a valid judicial decree.” Thus, it said, the accomplishment of the objectives of the federal statute “is neither obstructed nor interfered with by imposing a constructive trust on the insurance proceeds.” Ibid.

We forthwith acknowledge, of course, that this Court’s “only power over state judgments is to correct them to the extent that they incorrectly adjudge federal rights.” Herb v. Pitcairn, 324 U. S. 117, 125-126 (1945). It follows that the decision of the Supreme Judicial Court of Maine is subject to disturbance here only to the extent that it fails to honor federal rights and duties.

Notwithstanding the limited application of federal law in the field of domestic relations generally, see McCarty v. McCarty, 453 U. S. 210, 220 (1981); Hisquierdo v. Hisquierdo, 439 U. S. 572, 581 (1979); In re Burrus, 136 U. S. 586, 593-594 (1890), this Court, even in that area, has not hesitated to protect, under the Supremacy Clause, rights and expectancies established by federal law against the operation of state law, or to prevent the frustration and erosion of the congressional policy embodied in the federal rights. See McCarty v. McCarty, supra; Hisquierdo v. Hisquierdo, supra; Free v. Bland, 369 U. S. 663 (1962); Wissner v. Wissner, 338 U. S. 655 (1950); McCune v. Essig, 199 U. S. 382 (1905). Cf. Yiatchos v. Yiatchos, 376 U. S. 306, 309 (1964). While “[sjtate family and family-property law must do ‘major damage’ to ‘clear and substantial’ federal interests before the Supremacy Clause will demand that state law be overridden,” Hisquierdo, 439 U. S., at 581, with references to United States v. Yazell, 382 U. S. 341, 352 (1966), “[t]he relative importance to the State of its own law is not material when there is a conflict with a valid federal law, for the *55Framers of our Constitution provided that the federal law must prevail.” Free v. Bland, 369 U. S., at 666. See also Gibbons v. Ogden, 9 Wheat. 1, 210-211 (1824). And, specifically, a state divorce decree, like other law governing the economic aspects of domestic relations, must give way to clearly conflicting federal enactments. McCarty v. McCarty, supra; Hisquierdo v. Hisquierdo, supra. That principle is but the necessary consequence of the Supremacy Clause of our National Constitution.

In Wissner v. Wissner, supra, an insured under an NSLIA policy named his parents as beneficiaries. Upon his death, the serviceman’s widow claimed community property rights in the policy proceeds. The NSLIA specifically provided that the insured had the right to designate and to change the beneficiary. It also had an anti-attachment clause. Despite these provisions, a California court held that the policy proceeds were community property, and it ordered half the proceeds paid to the widow. This Court reversed, noting that “Congress has spoken with force and clarity in directing that the proceeds belong to the named beneficiary and no other.” 338 U. S., at 658. Further, “the judgment below nullifies the soldier’s choice and frustrates the deliberate purpose of Congress. It cannot stand.” Id., at 659. And the diversion, as directed by the state court, of future payments to be received by the beneficiary would be a “seizure” prohibited by the anti-attachment provision. Ibid. These are strong words and a positive ruling.

The same approach has been followed in later cases: Free v. Bland, supra, concerning the right of survivorship in United States Savings Bonds issued in co-ownership form; Hisquierdo v. Hisquierdo, supra, involving the Railroad Retirement Act of 1974, 45 U. S. C. §231 et seq.; and McCarty v. McCarty, supra, concerning military retired pay.

The present case, we feel, is controlled by Wissner. Under §§ 717(a) and 770(a) of the SGLIA, just as under § 602(g) of the predecessor NSLIA, 54 Stat. 1010, at issue in *56Wissner, the insured service member possesses the right freely to designate the beneficiary and to alter that choice at any time by communicating the decision in writing to the proper office. 338 U. S., at 658. Here, as there, it appropriately may be said: “Congress has spoken with force and clarity in directing that the proceeds belong to the named beneficiary and no other.”

There can be no doubt that Congress was aware of the breadth of the freedom of choice accorded the service member under the SGLIA. The pertinent House Report stated flatly: “The serviceman may designate any person as a beneficiary,” H. R. Rep. No. 1003, 89th Cong., 1st Sess., 7 (1965), and the point was emphasized on the floor of the House by Representative Everett: “This bill permits you to leave your insurance to your church, to your college, to your best friend. The beneficiary provision is wide open under this option.” Ill Cong. Rec. 24341 (1965). Thus, the Maine court’s analysis is inconsistent both with the language of the Act and with its legislative history.6

Neither respondents nor the Supreme Judicial Court of Maine has questioned the authority of Congress to control payment of the proceeds of SGLIA policies. Indeed, this Court observed in Wissner:

“Possession of government insurance, payable to the relative of his choice, might well directly enhance the morale of the serviceman. The exemption provision is his *57guarantee of the complete and full performance of the contract to the exclusion of conflicting claims. The end is a legitimate one within the congressional powers over national defense, and the means are adapted to the chosen end.” 388 U. S., at 660-661.

The federal interest is especially strong because a substantial share of the proceeds of an SGLIA policy may be attributable to general tax revenues.

There are, to be sure, some small differences between the SGLIA and the predecessor NSLIA. In the provision granting the service member the right to designate the beneficiary, the words “at all times” appear in the earlier Act, 38 U. S. C. § 717(a), but not in the later one, 38 U. S. C. § 770(a), and the right to change the beneficiary “without the consent” of the one presently named is spelled out in § 717(a) but not in § 770(a). But the later Act’s unqualified directive to pay the proceeds to the properly designated beneficiary clearly suggests that no different result was intended by Congress. And any possible ambiguity was eliminated by the Administrator’s regulations that provide that a “change of beneficiary may be made at any time and without the knowledge or consent of the previous beneficiary.” 38 CFR §9.16(e) (1980). There has been no suggestion that these regulations are unreasonable, unauthorized, or inconsistent with the SGLIA, and such a suggestion would not be supportable.7 See Whirlpool Corp. v. Marshall, 445 U. S. 1, 11-13 (1980); Udall v. Tallman, 380 U. S. 1, 16 (1965).

*58Yiatchos v. Yiatchos, 376 U. S. 306 (1964), relied on by the respondents, but not cited by the Maine court, does not stand to the contrary. In Yiatchos, the Court considered a question left open in Free v. Bland, 369 U. S., at 670-671, namely, the “scope and application” of the doctrine of fraud as an exception “to the regulatory imperative.” 376 U. S., at 307. There, the decedent Yiatchos, a resident of a community property State, purchased United States Savings Bonds with community funds and had them issued in the name of the decedent but payable on his death to his brother. The state court held that this purchase “was in fraud of the rights” of the surviving wife, as “a void endeavor to divest the wife of any interest in her own property.” In re Yiatchos’ Estate, 60 Wash. 2d 179, 181-182, 373 P. 2d 125, 127 (1962). This Court agreed that the bonds could “not be used as a device to deprive the widow of property rights which she enjoys under Washington law.” 376 U. S., at 309. But because the named beneficiary was entitled to the bonds “unless his deceased brother committed fraud or breach of trust tantamount to fraud” by wrongfully disposing of the wife’s property, ibid., the case was remanded to give the widow an opportunity to demonstrate that she had not consented to or ratified the purchase and registration of the bonds. The remand was also for the determination, under state law, whether the widow had an interest in the community’s specific assets, or only a half interest in the estate generally.

Here, in contrast, Sergeant Ridgway’s conduct did not amount to breach of trust or conversion of another’s prop*59erty. A careful reading of the complaint and the amended complaint, App. 11 and 24, in this case reveals no allegation of fraud or breach of trust. And we are not inclined to provide or infer such an allegation when a case comes to us, as this one does, with the record indicating nothing more than a breach of contract on the part of the deceased service member. Indeed, to say that this type of conduct constitutes constructive fraud would be to open the policy proceeds to a suit by any commercial creditor, a result that would render § 770(g) nugatory. As the trial court intimated, respondents may have a claim against the insured’s estate for that breach; the record does not disclose whether a claim of that kind would be collectible.8

There is, finally, a fundamental distinction between respondents’ asserted interests in the SGLIA policy proceeds and the community property concepts at issue in Yiatchos. Federal law and federal regulations bestow upon the service member an absolute right to designate the policy beneficiary. *60That right is personal to the member alone. It is not a shared asset subject to the interests of another, as is community property. Yiatchos had imposed his will upon property in which his wife had a distinct vested community interest. In contrast, only Sergeant Ridgway had the power to create and change a beneficiary interest in his SGLIA insurance. By exercising that power, he hardly can be said to have committed fraud.

We conclude, therefore, that the controlling provisions of the SGLIA prevail over and displace inconsistent state law.9

IV

The imposition of a constructive trust upon the insurance proceeds is also inconsistent with the anti-attachment provision, 38 U. S. C. § 770(g), of the SGLIA. In Wissner, 338 U. S., at 659, this Court invoked the identical anti-attachment provision of the NSLIA as an independent ground for the result reached in that case. The Court rejected, as it did so, id., at 663-664, the dissent’s argument that “Congress was interested in protecting [the fund], not the beneficiary,” which parallels respondents’ argument here in favor of creating a constructive trust after the proceeds have been received by the beneficiary. Any diversion of the proceeds of Sergeant Ridgway’s SGLIA policy by means of a court-imposed constructive trust would therefore operate as a forbidden “seizure” of those proceeds.

The Maine court attempted to limit the reach of § 770(g), as has been noted above, on the theory that the purpose of the anti-attachment provision was to protect the policy proceeds from the claims of creditors, and that the provision has no *61application to minor children asserting equitable interests. 419 A. 2d, at 1033. This contention, however, fails to give effect to the unqualified sweep of the federal statute. Section 770(g), in addition to exempting the policy proceeds “from the claims of creditors,” prohibits, in the broadest of terms, any “attachment, levy, or seizure by or under any legal or equitable process whatever,” whether accomplished “either before or after receipt by the beneficiary.” The reading adopted by the Maine court renders the bulk of the quoted statutory text extraneous. What was said of the statute under consideration in Hisquierdo, supra, is applicable without qualification here:

“Like anti-attachment provisions generally [citing Wissuer], it ensures that the benefits actually reach the beneficiary. It pre-empts all state law that stands in its way. It protects the benefits from legal process ‘[notwithstanding any other law... of any State’. ... It prevents the vagaries of state law from disrupting the national scheme, and guarantees a national uniformity that enhances the effectiveness of congressional policy.” 439 U. S., at 584.10

We find nothing to indicate that Congress intended to exempt claims based on property settlement agreements from the strong language of the anti-attachment provision.11

*62V

We recognize that this unpalatable case suggests certain “equities” in favor of the respondent minor children and their mother. Sergeant Ridgway did have specific obligations to *63the children that were imposed by the 1977 divorce judgment of the Maine court. Those obligations not only concerned life insurance “now outstanding” for the benefit of the children, but also extended to their support, to clothing, to “medical, dental, and optical expense,” and to certain loans and other indebtedness. App. 13-15. Ridgway, instead, chose to name his then new wife as beneficiary of his SGLIA policy.12

A result of this kind, of course, may be avoided if Congress chooses to avoid it. It is within Congress’ power. Thus far, however, Congress has insulated the proceeds of SGLIA insurance from attack or seizure by any claimant other than the beneficiary designated by the insured or the one first in line under the statutory order of precedence. That is Congress’ choice. It remains effective until legislation providing otherwise is enacted.

The judgment of the Supreme Judicial Court of Maine is

Reversed.

dissenting.

The Court holds that the Servicemen’s Group Life Insurance Act of 1965 (SGLIA or Act) broadly pre-empts state law. The Court also finds, as it must in light of previous decisions, that the pre-emptive power of this Act does not extend to eases of fraud or breach of trust. Ante, at 58, citing Yiatchos v. Yiatchos, 376 U. S. 306, 309 (1964).1 See also Free v. Bland, 369 U. S. 663, 670 (1962) (pre-emption may not be used to create a “sanctuary for a wrongdoer’s gains”); Hisquierdo v. Hisquierdo, 439 U. S. 572, 582 (1979) (the “survivorship rules in federal savings bond and military life insurance programs override community property law, absent fraud or breach of trust by the decedent”) (emphasis added).

The Court concludes, however, that there is not even an “allegation of fraud or breach of trust” in this case:

“Sergeant Ridgway’s conduct did not amount to breach of trust or conversion of another’s property. A careful reading of the complaint and the amended complaint, App. 11 and 24, in this case reveals no allegation of fraud *65or breach of trust. And we are not inclined to provide or infer such an allegation when a case comes to us, as this one does, with the record indicating nothing more than a breach of contract on the part of the deceased service member.” Ante, at 58-59.

I

In reaching the conclusion that this case presents “nothing more than a breach of contract,” the Court’s opinion does not linger over the facts.2 The decree divorcing Richard from *66his first wife April and incorporating the agreement of the parties was entered on December 7, 1977. The agreement required several months to negotiate, and it was negotiated in detail. April received custody of the couple’s three children. Richard was entitled to claim one child as a tax exemption in 1977 and two in 1978. He was to make specified monthly support payments beginning in 1978 and increasing in 1979. Although Richard was to pay for his children’s medical and dental expenses, April agreed to incur them, to the extent possible, “so that they will be payable under [Richard’s] serviceman’s insurance.” App. 14. In addition to other particular exchanges and responsibilities, the settlement specified and the decree ordered:

“Defendant [Richard] is ordered to keep in force the life insurance policies on his life now outstanding/or the benefit of the parties’ three children. If any of such insurance policies should subsequently be terminated for any reason, defendant shall immediately replace it with other life insurance of equal amount for the benefit of the children.” Ibid, (emphasis added).3

Less than four months later, Richard remarried and promptly changed the beneficiary clause of his serviceman’s *67policy so that the entire insurance proceeds would go to his new wife, Donna.

I return to the Court’s view that the complaint makes “no allegation of fraud or breach of trust,” and that this is a simple case involving “nothing more than a breach of contract” by Richard. Ante, at 59. Perhaps the complaint, as amended, is inartful. Yet it specifically averred that a constructive trust existed under which Donna — the recipient of the insurance proceeds — was the “constructive trustee . . . for the benefit” of the children. App. 26. The Supreme Judicial Court of Maine explicitly held that a constructive trust existed and that Donna was the constructive trustee of the corpus of this trust.4 In a technical sense, perhaps it can be argued there was “no allegation of fraud or breach of trust” as these precise terms were not used. But the complaint averred, id., at 24, 25, and the substance of this case is, that a constructive trust was created by Richard’s agreement and conduct.

In my view, the Court is plainly wrong in concluding that Richard’s conduct was “nothing more than a breach of contract” and that his obligation was like that of “any commercial” debtor who defaults on a judgment.5 Ante, at 59. *68Familial obligations are not merely commercial. Few legal duties are more universally acknowledged than the duty of a father to support his children. This duty existed in this case by law before the divorce. As a result of the divorce, it was recognized explicitly by Richard’s contract with his family and by the divorce decree ordering him to discharge that duty by maintaining the insurance at issue for the benefit of his children. Yet, the Court today analogizes a father’s support duty to that of a commercial debtor! This holding ignores the difference not only in character of the duties but also in their consequences. A defaulting debtor may be subjected to a judgment and an attachment lien. He may not be sent to jail. But a father who defaults on his duty to support his children or who violates a court decree enforcing that duty may be imprisoned for contempt.

The Court responds to this dissent in its footnotes 8 and 11. Yiatchos and Free are said to involve “a particular type of fraudulent behavior: attempts ‘to divest the wife of any interest in her own property.’” Ante, at 59, n. 8 (emphasis deleted). The Court distinguishes, for the purpose of determining the pre-emption issue, between fraud or breach of trust that affects a wife’s interest in community property and fraud or breach of trust that affects minor children’s interest in a fund set aside for their support. I see no basis for such a distinction. Yiatchos and Free recognize, without qualifica*69tion, that “federal pre-emption [may not be allowed] to shield fraud or breach of trust.” I would not have thought before today’s decision that any court would suggest — much less find — that minor children are less entitled to be “shielded” from this type of conduct than an adult wife. Indeed, because of the difference between the capacity of an adult and that of young children, our law always has reflected a special solicitude for minors. See, e. g., Bellotti v. Baird, 443 U. S. 622, 633-635 (1979).

The Court further argues that “by way of contrast” with a husband’s “divest[ing]” his wife of her interest in community property, “Sergeant Ridgway [merely] misdirected property over which he had exclusive control.” (Emphasis added.) Ante, at 59, n. 8. This is indeed a generous way to describe the Sergeant’s conduct. Moreover, the statement that he had “exclusive control” over the property begs the very question before us: whether Richard retained this control despite his conduct. He had divested himself voluntarily of all interest in the insurance policies by the agreement. The Maine court had approved the agreement, and ordered Richard to comply with it. He had far less “control” over the fund he had thereby created for his children’s support than the husband had in either Yiatchos or Free. He had no interest whatever in the policies to “misdirect” to his new wife unless — contrary to those decisions — the Act is now read to allow fraud or breach of trust.6

*70I would hold that the special nature of the parental legal duty, as expressly manifested in this case by both Richard’s negotiated bargain with his family and by the terms of his divorce decree, imposes a constructive trust upon the proceeds of the insurance for the benefit of Richard’s children as a matter of federal law.7 As the Court acknowledges, ante, at 58, the intention of Congress to supplant state law does not extend to a breach of trust. Here the fund impressed with a trust should be held for its agreed purpose in accordance with the law of Maine.8 I would affirm the judgment of the Supreme Judicial Court of Maine.

II.

Although I think the breach-of-trust issue is dispositive, I would be willing — in the interest of preventing what seems to me a uniquely unjust decision — to join an opinion remanding the case to the Maine Supreme Judicial Court on the issue of fraud.9 There is no specific allegation of fraud, and yet the admitted facts create the strongest inference that Richard intended to evade his support obligation by diverting to his *71new wife the fund he had created for the benefit of his children. The temporal sequence itself is persuasive. Following several months of negotiation, the divorce was granted on December 7, 1977, but only on conditions that included the funding of the support obligation. On March 28, 1978, less than four months thereafter, Richard married Donna. Six days later he stripped his children of this fund by changing the beneficiary clause so that Donna would receive the proceeds. This hardly was an inadvertent act. It is unlikely that even the enchantment of a new wife caused him to forget both his duty and his obligation to his children.

It would be appropriate, however, to afford the state courts an opportunity to address the fraud issue. The Supreme Judicial Court of Maine ruled in April’s favor without considering this alternative theory. This Court should not foreclose this consideration, for whether April will be permitted to advance this argument at this stage of the proceedings is a question of state procedural law.