LOKEN, Circuit Judge.

This case turns on the validity of a $5 million life insurance policy issued in 2007 by PHL Variable Life Insurance Company (PHL) insuring the life of William Close. When Close died in 2011, the policy was a stranger-owned-life-insurance policy ("STOLI policy"), owned after interim transfers by appellant Bank of Utah as custodian for life insurance policy investors. Bank of Utah as beneficiary demanded payment of the death benefit. PHL commenced this action, seeking a declaratory judgment that the policy was "void ab initio" for lack of an insurable interest. The district court granted PHL summary judgment on this issue of Minnesota law, a ruling that became an appealable final judgment after other issues were resolved. PHL Var. Ins. Co. v. Bank of Utah, No. 12-1256, 2013 WL 6190345, at *13 (D.Minn. Nov. 27, 2013). The issue turns on proper application of an ancient common law life insurance principle to recent, controversial developments in the marketing of life insurance policies as investment opportunities. Reviewing the grant of summary judgment de novo, we reverse. See Johnson v. Securitas Sec. Servs. USA, Inc., 769 F.3d 605, 611 (8th Cir.2014) (standard of review).

I.

A "viatical settlement" permits a dying insured to obtain continued medical care and other provisions by selling his life insurance policy at a discount to a purchaser who will pay more than the cash surrender value the insurer would pay. "The viatical settlements industry was born in the 1980s in response to the AIDS crisis." Life Partners, Inc. v. Morrison, 484 F.3d 284, 287 (4th Cir.2007). By the mid-1990s, the market had expanded to include other terminal illnesses, and there were some sixty companies in the viatical settlement business. See Martin, Betting on the Lives of Strangers: Life Settlements, STOLI, & Securitization, 13 U. Pa. J. Bus. L. 173, 185-86 (2010). Supporting the practice, Congress amended the Internal Revenue Code in 1996 to exclude from an insured's taxable income qualifying proceeds received from a licensed viatical settlement provider. Many States responded to the new industry with viatical settlement statutes regulating the impact of the practice on insureds and insurers. See Life Partners, 484 F.3d at 294-300.

Investor demand for life insurance policies insuring the terminally ill exceeded supply as the treatment of AIDS became more effective. To meet this lucrative demand, life insurance agents and life settlement brokers changed the name of the practice from "viatical settlements" to "life settlements" and undertook on a massive scale to persuade senior citizens to purchase life insurance policies in high-value amounts "not for the purpose of protecting his or her family, but for a current financial benefit." Martin, supra at 187. The practice poses risks and rewards for insurers, insureds, and investors that are well illustrated by the facts of this case.

In 2006, William Close, a 74-year old retiree, was persuaded by a referring broker to meet with Brad Friedman, an agent of Lextor Financial, an agency licensed to sell insurance for PHL. Close completed an application for a $5 million life insurance policy, far more than he could afford. As submitted to PHL, the policy application falsely stated that Close had a net worth ten times greater than actual and an annual income of $350,000, and failed to disclose his prior felony conviction for receiving illegal kickbacks as a union pension fund trustee. With Friedman's guidance, Close submitted a loan application falsely stating his net worth and obtained a two-year, $300,225 premium financing loan from CFC of Delaware. Funding for the loan came from New Stream Insurance, LLC (New Stream), a now-bankrupt hedge fund that invested in life settlements and premium finance loans. The policy was pledged as collateral for the non-recourse loan; Close personally guaranteed twenty-five percent of the loan in the event of default. CFC and New Stream determined that the policy would be worth $1.3 million in two-years, when it became "incontestable" under Minnesota law. See Minn.Stat. § 61A.03, subd. 1(c). Close was told he would likely be able to sell the policy in the secondary market for ten percent of its face value ($500,000) at the end of the two-year period.

PHL had previously approved CFC as a funding source for the purchase of PHL policies. PHL approved Close's application with minimal investigation and issued the $5,000,000 policy in September 2007. From the loan proceeds, PHL received insurance premiums of $272,025; CFC received $14,200 in origination and closing fees; and Friedman and a CFC employee split substantial commissions for procuring the policy.

As part of the Financing Arrangement with CFC, Close formed an irrevocable trust to own the insurance policy, naming Mrs. Close as trust beneficiary. The trustee was BNC National Bank. A Minnesota lawyer was named Trust Protector, a position intended to "give the insured and his family some input over the ongoing trust administration." In March 2009, six months before the loan was due, BNC sent Close a letter explaining his options for repaying it — refinance with the lender or a third party, sell the policy and use the sale proceeds to repay the loan, or relinquish his interest in the policy to the lender. Close sought Friedman's help in selling the policy, but the secondary market had crashed by the fall of 2009, and Friedman's efforts were unsuccessful. Unable to sell the policy, Close surrendered the policy to New Stream in full satisfaction of the loan.1 New Stream filed for bankruptcy in June 2011. Its portfolio of life insurance policies, including the Close policy, was sold to Limited Life Assets Services Limited (LLAS). When Close died in November 2011 from lung cancer, Bank of Utah held the policy as securities intermediary for LLAS. Bank of Utah filed a claim for the death benefit in January 2012.

PHL's claim investigation revealed the fraudulent misrepresentations on Close's policy application. But any claim to rescind the policy for fraud in its procurement was foreclosed by the two-year incontestability provision in Minn.Stat. § 61A.03, subd. 1(c). See Sellwood v. Equitable Life Ins. Co. of Iowa, 230 Minn. 529, 42 N.W.2d 346, 351 (1950) (an incontestability provision "limit[s] the time within which the policy may be contested for fraudulent answers in its procurement"). Therefore, PHL asserted in this declaratory judgment action that the policy was void ab initio as contrary to public policy for lack of an insurable interest. The district court agreed and, relying on decisions from other jurisdictions, ruled "that a policy may be challenged for lack of insurable interest beyond the contestability period."

II.

This diversity action is governed by Minnesota law. The securitization of life settlements for purchase by investors, and the dramatic increase in suspect marketing practices to sell STOLI policies, raise legitimate public policy and legislative concerns that have led to legislation and regulation in nearly every State, and have prompted a raft of litigation around the country, illustrated by this case. See generally Martin, supra at 197-216. The fact patterns in many cases were similar to this case, but each decision necessarily turned on the governing statutes and judicial precedents of a particular State. Thus, while decisions from other jurisdictions may be highly instructive, as a federal court exercising diversity jurisdiction we must remain grounded in Minnesota law and determine how the Supreme Court of Minnesota would apply that law in this case. See, e.g., Larson v. Nationwide Agribusiness Ins. Co., 739 F.3d 1143, 1146 (8th Cir.2014).

The Anglo-American principle that an insurable interest is required to purchase a life insurance policy dates from the Life Assurance Act of 1774, enacted by the British Parliament to regulate a popular English gambling activity-using insurance to bet on strangers' lives. Martin, supra at 176. The principle was universally adopted in this country and became a part of federal common law (pre-Erie) and the common law of the States, including Minnesota. Many States enacted statutes defining the elusive term "insurable interest." Minnesota did not, until the Legislature, responding in 2009 to problems created by STOLI marketing practices, enacted the Minnesota Insurable Interest Act. Minn.Stat. §§ 60A.078 et seq. This Act is prospective only and thus does not apply to the policy here at issue. See 2009 Minn. Sess. Law Serv. Ch 52, § 11. Thus, this case is governed solely by Minnesota common law. Decisions from other jurisdictions having relevant statutes or differing judicial precedents must be applied with great care.

The core of the common law insurable interest rule is "that a policy, issued to one who has no interest in the continuation of the life of the person insured, is both a gambling contract, and a contract which creates a motive for desiring the termination of such life." Christenson v. Madson, 127 Minn. 225, 149 N.W. 288, 289 (1914). In this country, the dominant public policy underlying the rule is to eliminate a form of "moral hazard." As the hazard was graphically described by one commentator, "Insurance policies that compensate beneficiaries upon the death of a person or destruction of property that the beneficiary does not have an interest in preserving give beneficiaries an incentive to murder the insured person or destroy the insured property."2 Justice Holmes succinctly stated this public policy in Grigsby v. Russell, 222 U.S. 149, 155, 32 S.Ct. 58, 56 L.Ed. 133 (1911): "The very meaning of an insurable interest is an interest in having the life continue, and so one that is opposed to crime." Note that the policy is based upon the relationships between the insured, the person purchasing a life insurance policy, and the death benefit beneficiary. It does not address whether the insurer should be permitted to renege on its contractual obligations.

The insurable interest rule is satisfied when a person purchases insurance on his or her own life. See Conn. Mut. Life Ins. Co. v. Schaefer, 94 U.S. 457, 460, 24 L.Ed. 251 (1876). In Grigsby, the Supreme Court noted that, "[s]o far as reasonable safety permits, it is desirable to give to life policies the ordinary characteristics of property." 222 U.S. at 156, 32 S.Ct. 58. Applying this somewhat countervailing public policy, the Court overruled prior dicta and held that a man who purchased insurance on his own life could validly assign or sell the policy to a person lacking an insurable interest in the insured's life. But the Court noted a caveat to its ruling: "cases in which a person having an interest lends himself to one without any, as a cloak to what is, in its inception, a wager, have no similarity to those where an honest contract is sold in good faith" to a stranger. Id. (emphasis added). That caveat frames the issues to be decided in this case.

III.

PHL's claim is based on two contentions: (A) that the Close policy was "void ab initio" for lack of an insurable interest, and (B) that any policy void ab initio is never "in force" and therefore PHL's defense to paying the death benefit is not barred by the incontestability provision in Minn.Stat. § 61A.03, subd. 1(c).3 In PHL Var. Ins. Co. v. Lucille E. Morello 2007 Irrev. Trust, 645 F.3d 965, 968-69 (8th Cir.2011), PHL sued to rescind a policy before it became incontestable, and the parties agreed the policy was void ab initio. Here, by contrast, both issues are contested. Unlike the district court, we conclude that neither PHL contention is supported by Supreme Court of Minnesota decisions applying the applicable Minnesota common law.

A. The district court did not discuss the first question, simply accepting PHL's assertion that life insurance policies lacking an insurable interest violate public policy and are void ab initio, an assertion PHL supported entirely by citing cases from other jurisdictions. In an earlier District of Minnesota diversity case in which an insurer sought to invalidate a life policy on this ground, the district court properly looked to Supreme Court of Minnesota common law decisions and concluded:

Sun Life Assur. Co. of Canada v. Paulson, No. 07-3877, 2008 WL 451054, at *2 (D.Minn. Feb. 15, 2008). But that categorical statement does not reflect the Supreme Court of Minnesota's far more nuanced discussion in Christenson, a case in which the insured's children claimed a life insurance policy death benefit because the woman named as beneficiary had no insurable interest:

Nowhere in this discussion is there even a hint that a policy purchased by an insured on his own life would ever be "void ab initio" at the instance of the insurer.

Our research has uncovered four Supreme Court of Minnesota cases discussing whether a life insurance policy was supported by an insurable interest. See Hogue v. Minn. Packing & Provision Co., 59 Minn. 39, 60 N.W. 812, 813 (1894); Rahders, Merritt & Hagler v. People's Bank of Mpls., 113 Minn. 496, 130 N.W. 16, 17 (1911); Christenson, 149 N.W. at 289-90; Peel v. Reibel, 205 Minn. 474, 286 N.W. 345, 346 (1939). Each of these cases involved competing claims to the death benefit; none included a claim by the insurer that the policy would be "void ab initio" if a beneficiary or assignee was found to lack an insurable interest.4 The Supreme Court of Minnesota has never discussed the issue, and its decisions awarding death benefits to the prevailing claimants in these cases is strong, indeed in our view compelling evidence that the "void ab initio" principle urged by PHL and accepted without discussion by the district court is not consistent with Minnesota common law.

As the Court explained in Christenson, when a person other than the insured purchases life insurance on a stranger, naming himself as beneficiary, the insurance policy is "against public policy and void." But when a person purchases insurance on his own life and later assigns it to a stranger, the contract between the insured and insurer is valid unless voidable for fraud or other defenses that are subject to the incontestability bar. This court had earlier applied the same principle in Gordon v. Ware Nat'l Bank, 132 F. 444, 448 (8th Cir.1904), explaining that "an insurable interest in the assignee of a policy of life insurance is not essential to the validity of the assignment if the party to whom it was originally issued had such an interest, and the assignment is not made as a cover for the issue of a wager policy."

We acknowledge the caveat stated in Grigsby and restated more than once by the Supreme Court of Minnesota and by this court — an assignment of a life insurance policy is valid if "made in good faith and not as a mere cover for taking out insurance in the beginning in favor of one without insurable interest." Peel, 286 N.W. at 346; see Rahders, 130 N.W. at 17; Bankers' Reserve Life Co. v. Matthews, 39 F.2d 528, 529 (8th Cir.1930) ("Any person has a right to procure an insurance on his own life and assign it to another provided it be not done by way of cover for a wager policy.)." As articulated by the Supreme Court of Minnesota, this exception to the free transferability of life insurance policies is narrow.

In Sun Life v. Paulson, the district court applied this oft-repeated caveat and concluded that a scheme or agreement between the insured and a third person at the time the policy is procured to transfer or assign the policy to an identified person who lacked an insurable interest would render the policy void ab initio under Minnesota common law. 2008 WL 5120953, at *5. This is a plausible application of the dicta in Peel and Rahders and Grigsby. But we conclude it would not be adopted by the Supreme Court of Minnesota because it ignores an overriding principle of Minnesota law: "The court's power `to declare a contract void for being in contravention of sound public policy is a very delicate and undefined power, and... should be exercised only in cases free from doubt.'" Katun Corp. v. Clarke, 484 F.3d 972, 976 (8th Cir.2007) (Supreme Court of Minnesota quotation omitted).

In this context, the Supreme Court of Minnesota has recognized "the theory that a policy, issued to one who has no interest in the continuation of the life of the person insured, is ... against public policy and void." Christenson, 149 N.W. at 289. But when the policy has been purchased by the person insured, we believe the Supreme Court of Minnesota would conclude the public policy issue is not free from doubt. The question is whether Minnesota public policy requires that we permit an insurer who collected over $500,000 in premiums — a windfall it will keep if we affirm — to renege on its contractual obligation because a third party "schemed" with the insured before the policy issued to help him achieve his intent to purchase the policy for resale, an intent which, if unilateral, was consistent with the public policy recognizing that life insurance policies are legitimate investments, as well as insurance. Accord First Penn-Pacific Life Ins. Co. v. Evans, 313 Fed.Appx. 633, 636 (4th Cir.2009) (a person's unilateral intent to purchase a policy on his own life to exploit the secondary market for life policies does not make the policy void ab initio under the common law as declared in Grigsby). Recalling that the Supreme Court of Minnesota has never even considered an insurer's claim that its policy should be invalidated on this common law ground, we conclude that the Court would not declare the Close policy void ab initio, permitting PHL to walk away from its bargain.

If our disagreement with the decision in Sun Life v. Paulson is debatable, we further note the district court went far beyond that limited ruling when it concluded that PHL was entitled to summary judgment despite the absence of proof of an agreement to resell the policy to an identified person. Relying on cases applying Delaware and New York law that were governed by quite different statutes and judicial precedents, the district court declared the Close policy void as against public policy because it "was procured by a scheme to assign it to a party lacking an insurable interest and with the mutual intent of circumventing the law against wagering policies." This reasoning bears little if any relationship to the "moral hazard" on which both federal and Minnesota common law are grounded. Moreover, it would permit life insurers to resist paying a death benefit any time there is some evidence that an insured used premium financing to obtain a policy he or she planned to sell. Subjecting insureds and their beneficiaries to this inquiry cannot be squared with the public policy declaration in Grigsby that "it is desirable to give to life policies the ordinary characteristics of property."5 We conclude this is not a result the Supreme Court of Minnesota would find acceptable in exercising its "very delicate and undefined power" to declare a contract void as contrary to sound public policy.

B. We likewise disagree with the district court's answer to the second issue on which PHL must prevail — that its claim to avoid paying the death benefit for lack of an insurable interest is not foreclosed by Minnesota's incontestability statute. Adopting what it described as the majority view from other jurisdictions, the district court concluded, "A policy that is void ab initio never comes into force, and so the incontestability provision of such policy has no effect." 2013 WL 6190345, at *9. The district court in Sun Life v. Paulson did not address this issue, dismissing Sun Life's complaint for lack of evidence of a scheme, existing when the policy was procured, to sell or assign it to an identified third party.

We conclude the Supreme Court of Minnesota would not agree. The purpose behind Minn.Stat. § 61A.03, subd. 1(c), is "to protect an insured (and designated beneficiaries) from a dilatory challenge to the insurance policy while also encouraging the insurer to be diligent in performing its duty to investigate within a specified period, and to penalize it if it does not." PHL Var. Life Ins. Co. v. U.S. Bank Nat'l Ass'n, No. 10-1197, 2010 WL 3926310, at *5 (D.Minn. Oct. 4, 2010).6 Whether the insured has an agreement with an insurance agent or broker or a premium financing company at the time the policy is issued that it will be sold, either to an identified person who lacks an insurable interest or, more typically, into a secondary market of insurance policy investors, is a risk the insurer can promptly investigate (assuming it is relevant to the decision to insure). Therefore, absent a supervening statute, the defense is subject to the incontestability provision of § 61A.03, subd. 1(c).7 To declare that a facially valid policy on which PHL collected substantial premiums for over four years was never "in force" is simply a fiction.

IV.

Without question, an aggressive secondary market for life insurance policies raises serious public policy issues. While passive investors of securitized policies are unlikely to murder the insureds, the life settlement market functions in part on the truism that a policy is worth more to an investor if the insured is elderly or in poor health. See Martin, 13 U. Pa. J. Bus. L. at 185-86. Thus, it is entirely reasonable for legislators and insurance regulators to conclude that many STOLI premium financing programs and marketing practices should be curtailed or banned because these practices induce elderly insureds to purchase high-value life insurance policies that are not needed for insurance purposes under terms ensuring that life settlement promoters and premium financing companies will ultimately collect the substantial death benefits. But these are issues that go far beyond Minnesota common law decisions, none of which even suggest that an existing life insurance policy will be declared void ab initio for lack of an insurable interest at the behest of an insurer that wishes to avoid paying the death benefit.

The judgment of the district court is reversed and the case is remanded with directions to dismiss PHL's complaint for declaratory judgment relief, and for further proceedings not inconsistent with this opinion.

COLLOTON, Circuit Judge, concurring in the judgment.

The dispositive question on this appeal is whether the life insurance policy issued by PHL Variable Insurance Company on the life of William Close is void ab initio as against public policy under Minnesota common law. The district court ruled that PHL's challenge to the policy was not time-barred under the Minnesota incontestability statute, Minn.Stat. § 61A.03, subd. 1(c), on the view that a policy issued to one lacking an insurable interest is void from inception, so if the claim has merit, the policy's incontestability provision never came into force and has no effect. See Lincoln Nat'l Life Ins. Co. v. Joseph Schlanger 2006 Ins. Trust, 28 A.3d 436, 440 (Del.2011). Bank of Utah does not appeal this ruling, and any challenge to the timeliness of PHL's challenge is therefore waived. Jenkins v. Winter, 540 F.3d 742, 751 (8th Cir.2008).

This appeal is governed by Minnesota law. In my view, the district court in Sun Life Assurance Co. of Canada v. Paulson, Civil No. 07-3877, 2008 WL 451054 (D.Minn. Feb. 15, 2008), was correct in its prediction that "the Minnesota Supreme Court would consider a life insurance policy void as against public policy if the policy was procured under a scheme, purpose, or agreement to transfer or assign the policy to a person without an insurable interest in order to evade the law against wagering contracts." Id. at *2 (internal quotation omitted). The stated rule is the better reading of the Minnesota Supreme Court's considered dicta on the subject in Peel v. Reibel, 205 Minn. 474, 286 N.W. 345 (1939) and Rahders, Merritt & Hagler v. People's Bank of Minneapolis, 113 Minn. 496, 130 N.W. 16 (1911). In Rahders, the court emphasized that "good faith in the transaction is required, and the courts do not hesitate to condemn a policy issued for the purpose of having it assigned." 130 N.W. at 17. In Peel, the court said that an assignment of life insurance is valid if "made in good faith and not as a mere cover [for taking out insurance in the beginning in favor of one without insurable interest]." 286 N.W. at 346 (emphasis added). The Minnesota court thus has stated its view that the case for declaring an insurance contract void in these circumstances is free from doubt. The Paulson rule, moreover, was expressly accepted by appellant Bank of Utah in this case. Appellant's Br. 20-21 & n.9. The Bank never urged the contrary position adopted by the court, and PHL had no occasion to address it.

In applying the Minnesota rule to the record in this case, however, PHL is not entitled to summary judgment. To establish that an insurance contract is void ab initio as a cover for a wagering contract, the Minnesota court likely would require an insurer to show that the scheming parties agreed that the insured would resell the policy to an identified person without an insurable interest. A broader rule — that the insurer need only show that the parties intended that the insured would resell the policy to someone without an insurable interest — would interfere with the ability of insureds to use premium financing, and could not meaningfully be distinguished from the generally accepted rule that permits an insured to purchase a life insurance policy with his own funds while harboring a unilateral intent to resell the policy to a stranger. See Grigsby v. Russell, 222 U.S. 149, 155-56, 32 S.Ct. 58, 56 L.Ed. 133 (1911). The district court thought it undisputed that Close intended to transfer the policy to "a third party lacking an insurable interest on his life," R. Doc. 174 at 23, but that conclusion is insufficient to render judgment for PHL.

On the contrary, my assessment of the record leads to the conclusion that Bank of Utah is entitled to judgment, because there is no evidence that Close intended from the outset to transfer the policy to CFC of Delaware, the premium financing company, or to New Stream, the entity that funded the loans to Close and ultimately acquired the policy. The evidence shows at best that Close intended to sell the policy to an unknown third party on the secondary market. Close's wife testified that the agent who marketed the policy explained that he would look for an investor to buy the policy within two years of issuance. Close himself eventually asked the agent to sell the policy while it was in force. PHL does not urge on appeal that Close intended all along to transfer the policy to CFC or New Stream. It relies instead on a different rule of law, one that would declare the policy void even if the parties had different intentions about how the policy would be transferred to an investor. Appellee's Br. 35. But that is not the rule that the Minnesota courts likely would adopt. Because there is insufficient evidence to establish that the parties agreed to resell the life insurance policy to an identified person, Bank of Utah is entitled to judgment.

For these reasons, I concur in the judgment.