SELYA, Circuit Judge.
The outcome of this appeal is controlled by important questions of Rhode Island law and public policy as to which we have found no dispositive precedent. Because the Rhode Island Supreme Court is the ultimate arbiter of matters of Rhode Island law, we certify these unsettled questions to that court for guidance.
See
R.I. Sup.Ct. R. 6.
I. BACKGROUND
Joseph Caramadre believed that he had found the Holy Grail of investment strategies: a way to speculate in high-risk securities while shielding himself from the adverse effects of losses. To implement this scheme, he exploited a perceived loophole in certain annuities issued by, inter alia, plaintiff-appellant Western Reserve Life Assurance Company of Ohio. Because certain features of Western Reserve’s annuities are integral both to Caramadre’s contrivance and to the issues on appeal, we start by outlining those features.
The classic annuity offers a person a stream of periodic payments during his life that are actuarially calculated and fixed in amount. In exchange, the person makes an up-front, lump-sum premium payment to the issuing insurance company. For example, an investor might pay an insurance company a one-time $1,000,000 premium in exchange for a promise to pay him $5,000 per month for the rest of his life.
Over the years, annuity products have evolved, and the WRL Freedom Premier III annuity issued by the appellant is a far cry from the classic model. We sketch some of the salient differences.
To begin, the appellant’s product is a variable annuity, not a fixed annuity. Instead of ceding control of his premium dollars to the insurance company, the investor retains the right to direct that those dollars be invested in certain pre-selected securities. Moreover, the annuity does not necessarily entail fixed periodic payments to the beneficiary; rather, it presents a diverse menu of payment options, including payouts that are determined by the value, from time to time, of the acquired securities. Consequently, the amounts paid to the beneficiary may ebb and flow
with the performance of the investment portfolio.
In addition, the investor in a WRL Freedom Premier III annuity can use the lifetime of someone other than himself (the annuitant) as a measuring device to determine how long the annuity payments will last. The investor (the owner) provides the premium dollars, directs the investment strategy, and selects the recipient of the periodic payouts called for by the annuity (the beneficiary). The beneficiary, who may or may not be the owner, will receive those payouts as long as the annuitant remains alive. Significantly, a WRL Freedom Premier III annuity contains no requirement that the owner and the annuitant be one and the same person; in fact, the annuity contract does not require any extrinsic tie between the two.
Unlike a classic annuity, the WRL Freedom Premier III annuity allows the owner to infuse more than a single premium payment into the annuity. Up until a specified date (not relevant here), the appellant will accept premium payments, as long as the total investment does not exceed $1,000,000.
Last — but far from least — the owner can elect a “Double Enhanced Death Benefit’
by agreeing to pay an additional daily charge. An owner who elects this benefit designates a beneficiary, who, upon the annuitant’s death, will be entitled to receive the greater of: (1) the total premiums invested in the policy, plus interest accrued at
5%
per annum,- or (2) the highest value of the policy (that is, the highest value of its investment portfolio, adjusted for certain deposits and withdrawals) on any annual anniversary of the policy. The owner and the beneficiary may be one and the same person.
Caramadre figured out that if an individual named himself (or an entity he controlled) as both the owner and the beneficiary of a WRL Freedom Premier III annuity and elected the death benefit, that individual could engage in high-risk market speculation without any downside exposure. Caramadre decided that this scheme could best be perpetrated by applying for an annuity with a relatively low initial premium, invested conservatively so as to avoid red flags. The owner/beneficiary (whether Caramadre himself or his nominee) would subsequently deposit a substantially more munificent incremental premium and steer the investment of the aggregate premium dollars into speculative securities. The upshot was a “heads. I win, tails you lose” scenario: if the investment gamble paid off, he would reap the fruits of his speculation when the annuitant died; and if the speculation backfired, the death benefit guaranteed that he would fare no worse than a full return of premiums paid (plus interest). In the latter event, the insurance company would be left holding a collection of nearly worthless securities.
Despite the cleverness of Caramadre’s scheme, there was a rub: one had to be sure that the death benefit would be triggered within a relatively short time after the risky investments were made.
That timing would ensure that the owner/beneficiary of the annuity (Caramadre or his nominee) would receive either the benefit of a strike-it-rich investment gamble or, at worst, the return of his bet. Thus, the linchpin of the scheme was locating and recruiting potential annuitants whose lifespans were predictably short: the terminally ill-
Caramadre rose to this challenge. Among other recruitment .tools, he circulated flyers promising up-front cash payments to terminally-ill patients for agreeing to let their names be used.
Charles Buckman, a Rhode Island resident suffering from chronic obstructive pulmonary disease, received one of Cara-madre’s flyers from a visiting nurse. His interest piqued, Buckman followed up on the flyer and contacted Estate Planning Resources, a Caramadre-controlled company.
To make a tawdry tale tolerably terse, Buckman accepted a cash payment to identify himself as the annuitant oh an application for a WRL Freedom Premier III variable annuity. The application designated a Caramadre nominee, defendant-appellee ADM Associates, LLC (ADM), as the prospective owner and beneficiary of the annuity. The application specifically requested inclusion in the annuity contract of the Double Enhanced Death Benefit. Buckman and ADM were wholly unrelated parties; indeed, up to that point Buckman had never heard of ADM. •
The appellant received the application on or about September 11, 2008. Four days later, it approved the application and issued a WRL Freedom Premier III annuity (the Policy). Pertinently, the Policy provided that it would be “incontestable from the Policy Date.”
The application had been accompanied by an initial $250,000 - premium payment. Roughly four months after the Policy went into effect, ADM made an additional premium payment of $750,000.
At some point, the appellant apparently learned of Caramadre’s scheme and came to believe that the Policy was an iteration of it.
It developed an acute case of seller’s remorse and, approximately a year after the Policy’s inception date, notified both Buckman and ADM that it intended to rescind the Policy.
Acting on this stated intention, the appellant sued ADM in the United States District Court for the District of Rhode Island, seeking rescission of the Policy and a declaration that the Policy was either void ab initio or had been properly rescinded. It also asserted claims for fraud, civil liability for crimes, and civil conspiracy.
ADM moved to dismiss the complaint. The district court heard this motion along with motions in similar cases. After careful consideration, the court dismissed the claims brought against ADM.
See W. Reserve Life Assurance Co. v. Conreal LLC,
715 F.Supp.2d 270, 276-81 (D.R.I.2010). With respect to the appellant’s prayers for rescission and declaratory relief, it held that the presence of the death benefit did not transform the Policy into a life insurance contract under Rhode Island law.
See id.
at 276-79. Hence, the absence of an insurable interest neither rendered the Policy void nor justified its rescission.
See id.
.Although the court acknowledged that “the whole point of the [scheme] was to capitalize on the death benefits,” it con-
eluded that the “[defendants [had] figured out how to game a flaw in the product.”
Id.
at 278.
The district court based its dismissal of the tort claims on the incontestability clause.
See id.
at 280-81. In holding that this provision did not offend public policy, the court reasoned that Rhode Island’s default two-year incontestability period can be supplanted by any shorter period more favorable to the insured.
See id.
at 280 (citing R.I. Gen. Laws § 27-4-6.2(a)).
The appellant filed an amended complaint. The district court dismissed this complaint on substantially the same grounds.
See W. Reserve Life Assurance Co. v. Caramadre,
847 F.Supp.2d 329, 349-50 (D.R.I.2012). With no claims remaining against ADM, the court certified the judgment as final as between ADM and the appellant.
See
Fed.R.Civ.P. 54(b);
Nystedt v. Nigro,
700 F.3d 25, 29-30 (1st Cir. 2012). This timely appeal followed.
II. DISCUSSION
This case implicates two important issues of state law, either or both of which may determine the outcome of this litigation. We describe these issues and, for what assistance it may provide, limn the considerations that have prompted us to certify them to Rhode Island’s highest court.
A.
The Insurable Interest Question.
We begin with the question of whether ADM’s status as a stranger to its annuitant invalidates the Policy. Over a century ago, the Rhode Island Supreme Court declared that “a purely speculative contract on the life of another” procured by one without an insurable interest is contrary to public policy and “may properly be held to be void.”
Cronin v. Vt. Life Ins. Co.,
20 R.I. 570, 40 A. 497, 497 (1898). The Rhode Island General Assembly subsequently endorsed the insurable interest concept, requiring relatives named as beneficiaries to have “a substantial interest” in an insured’s life “engendered by love and affection,” and other beneficiaries to have “a lawful and substantial economic interest in having the life, health, or bodily safety of the individual insured continue.” R.I. Gen. Laws § 27-4-27(e). ADM concedes that it did not have an insurable interest in the life of its annuitant. The question, then, reduces to whether the Policy is the kind of contract that Rhode Island’s insurable interest requirement renders infirm.
The answer may depend on taxonomy: that is, on whether the Policy should be classified either as an “insurance contract” or as an “annuity.” If the former rubric applies, the statutory iteration of the insurable interest requirement, which prohibits the procurement of “any insurance contract upon the life or body of another individual” if “the benefits under the contract are payable to” someone other than the insured, his personal representatives, or “a person having ... an insurable interest in the individual insured,”
id.
§ 27-4-27(a), may resolve the question. In other words, if the death benefit makes the Policy an insurance contract upon the life of the annuitant, the want of an insurable interest would defeat the Policy.
Were an'inquiring court to look solely at the label that the appellant affixed to the Policy, the premise of that proposition would be easily refuted: the document designates itself as a variable annuity, not an insurance contract. BuLlabels can be misleading, and the Rhode- Island courts have sometimes looked beyond the title of a document to deem its substance to be insurance. For example, in
Sisson ex rel. Nardolillo v. Prata Undertaking Co.,
49 R.I. 132, 141 A.. 76 (1928), the Rhode Island Supreme -Court went beyond the “burial contradi ]” designation attached to
certain agreements and held that those agreements actually comprised burial insurance.
Id.
at 76-77.
There is some reason to believe that such label-piercing might be appropriate here. The common understanding of life insurance encompasses “[a]n agreement between an insurance company and the policyholder to pay a specified amount to a designated beneficiary on the insured’s death.” Black’s Law Dictionary 1010 (9th ed.2009). The General Assembly seems to have embraced that understanding, declaring that life insurance is simply “every insurance upon the lives of human beings and every insurance appertaining to that life.” R.I. Gen. Laws § 27-4-0.1(c).
In this instance, an insurance company (the appellant) issued the Policy, which obligated it to pay at least the aggregate premiums invested in the annuity (plus interest) upon the annuitant’s death. To this extent, the Policy is, at least in part, the functional equivalent of a life insurance policy.
See
Kendall J. Burr et al.,
Stranger-Initiated Annuity Transactions and the Case for Insurable Interest,
19 Conn. Ins. L.J. 113, 126 (2012).
Here, however, there is more to the story. Unlike a traditional life insurance policy, the death benefit did not promise a defined sum upon death but, rather, promised either the market value of the Policy or the aggregate premiums paid into it (whichever was greater). Moreover, the appellant did not treat the Policy as if it were life insurance: though cognizant of the insurable interest requirement for life insurance, it made no effort to verify that the owner of the Policy had any relationship with the annuitant. By the same token, it did not engage in the sort of medical underwriting that might have enabled it to calculate the annuitant’s mortality risk.
Last, the appellant chose to structure and market the transaction as an annuity; and venerable precedent indicates that annuities (unlike life insurance contracts) are not normally deemed violative of public policy for want of an insurable interest.
See Cronin,
40 A. at 497. This distinction between life insurance and annuities is firmly entrenched in positive law. The General Assembly’s definition of annuities explicitly excludes “payments made in connection with a life insurance policy.” R.I. Gen. Laws § 27-4-0.1(a).
As the appellant reminds us, though, the Policy is not a nineteenth-century annuity. In that era, the “annuities” to which courts referred had “traditionally and customarily ... been fixed annuities, offering the annuitant specified and definite amounts beginning with a certain year of his or her life.”
SEC v. Var. Annuity Life Ins. Co. of Am.,
359 U.S. 65, 69, 79 S.Ct, 618, 3 L.Ed.2d 640 (1959). Variable annuities (particularly those carrying death benefit riders) are a fairly recent innovation.
See id.
(noting that the first variable annuity appeared in or about 1952). Thus, the Policy is twice removed from its nineteenth-century ancestors — once by reason of its variable nature and once again by reason of its death benefit. It is not clear to us whether these innovations are sufficient to bring the Policy outside the boundaries of Rhode Island’s historic exclusion of annuities from the insurable interest requirement.
We are equally uncertain as to how the relatively recent Life Settlements Act,
see
R.I. Gen. Laws §§ 27-72-1 to 27-72-18, affects this inquiry. Through that legislation, the General Assembly made pellucid that “[sjtranger-originated life insurance” — that is, “a practice or plan to initiate a life insurance policy for the benefit of a third-party investor who, at the time of policy origination, has no insurable interest
in the insured” — is forbidden under Rhode Island law.
Id.
§ 27-72-2(26);
see id.
§§ 27-72-2(9)(i)(A)(X), 27-72-14(a)(1). This proscription expresses hostility to certain classes of transactions that are different than, but obviously similar to, the transaction in this case.
There is one further complication. The appellant’s attack on the Policy focuses on the unseemly nature of the scheme that Caramadre devised. Nevertheless, an insurable interest requirement for contracts like the Policy would not only affect vulture-like arrangements in which the payoff is geared toward the annuitant’s quick demise but also contracts in which the payoff is geared to the annuitant’s longevity. For instance, an owner who purchases a variable annuity measured by the life of a 22-year-old healthy man might do so with a long-lasting income stream in mind and without any thought of capitalizing on the opportunity for risk-free speculation associated with the annuity’s death benefit. This type of hypothetical owner would come much closer than an owner in the Caramadre model to being a counterpart of the owner of a conventional fixed annuity. And that would be so even though both the hypothetical owner and the Cara-madre-style owner had purchased essentially the same variable annuity.
The short of it is that the Policy does not seem to fit neatly into either the category of life insurance contracts or the conception of annuities. And to muddy the waters further, we. think that it might be plausible to treat the Policy as a hybrid. If so, the analysis would likely turn to whether requiring an insurable interest would comport with the broader purpose undergirding the insurable interest requirement.
This possibility prompts us to take a step back in time. Historically, the insurable interest requirement prevented a contract from becoming a vehicle for gambling. -In
Cronin,
for example, the court traced the insurable interest requirement for life insurance to the English Parliament’s Life Assurance Act of 1774, 14 Geo. 3, c. 48, § 1 (Eng.), which “prohibited insurance on the life of a person in which the beneficiary shall have no interest, or by way of gaming or wagering.” 40 A. at 497. That rule was followed in the United States “as declaratory of the common law.”
Id.
By requiring owners of life insurance policies to have “an interest of some sort in the insured life,” courts could ensure that these contracts did not become “mere wager policies.”
Conn. Mut, Life Ins. Co. v. Schaefer,
94 U.S. 457, 460/24 L.Ed. 251 (1876);
see Winward v. Lincoln,
23 R.I. 476, 51 A. 106, 112 (1902) (“[P]ublic policy forbids the enforcement of all wagers by our courts, and it has been so held by this court.”). Without such an interest, a life insurance policy could afford a perverse incentive for the “early termination” of the insured’s life.
Wamock v. Davis,
104 U.S. 775, 779, 26 L.Ed. 924 (1881).
Whether an insurable interest was required for the Policy may therefore depend on whether it can be fairly characterized as a contract wagering on life. The outcome of such an attempt at characterization seems problematic. On the one hand, a straightforward argument can be made that, without an insurable interest, the Policy is a wager on life. From that vantage, the owner/beneficiary of the Policy, a stranger to the annuitant, will receive money when the annuitant dies and, thus, cash in on what amounts to a wager.
See
Winward,
51 A. at 112 (defining a wager as “an agreement between two or more that a sum of money or some valuable thing, in contributing which all agree to take part, shall become the property of one or some of them on the happening in the future of an event at the present uncertain”).
On the other hand, there is a strong counterargument: a rule requiring an insurable interest for the Policy could also sweep up any contract with a death benefit. That result would propel the insurable interest requirement far beyond the traditional province of life insurance.
Furthermore, the connection between ADM’s wagering gain and the life of the annuitant is indirect. In a typical life insurance wager, there is a direct correlation between the timing of the insured’s death and the policy owner’s reward. Here, by contrast, the primary source of anticipated gains is apt to come by way of market performance; the predictably short life of the annuitant simply hedges against the possibility of investment losses. This may be a relevant distinction: one might be hard-pressed to say that an owner who pays $1,000,000 in premiums plus death benefit fees for an annuity and is assured of receiving no less than $1,000,000 (plus interest) upon the death of his annuitant is much of a gambler.
In sum, we find three aspects of the insurable interest question puzzling. First, we are uncertain as to how best to classify the Policy (as an annuity, a life insurance contract, or a hybrid). Second, we are uncertain as to whether the absence of an insurable interest renders the Policy infirm. Third, we are uncertain as to whether the Policy constitutes an unenforceable wagering contract.
B.
The Incontestability Question.
The second issue with which we are concerned revolves around the Policy’s incontestability clause. On its face, that clause appears to bar any challenge, including one based on the lack of an insurable interest, to the validity of the Policy. But this case presents two potential obstacles to such an unconditional reading.
The threshold obstacle is the possibility that Rhode Island public policy would refuse to countenance an incontestability clause — like this one — that wholly eliminates any contestability period. The Rhode Island Supreme Court has not spoken definitively to this question, and courts elsewhere are divided.
See
16 Richard A. Lord,
Williston on Contracts
§ 49:95 (4th ed.2013).
The Rhode Island cases dealing with incontestability clauses have sent mixed messages. For example, the Rhode Island Supreme Court has observed that a two-year incontestability clause is “not an absolute stipulation to waive all defenses and to condone fraud.”
Murray v. State Mut. Life Ins. Co.,
22 R.I. 524, 48 A. 800, 800 (1901) (dictum). This dictum is suggestive, especially since the fear of condoning fraud has led some courts to refuse to enforce incontestability clauses analogous to the clause at issue here.
See, e.g., Reagan v. Union Mut. Life Ins. Co.,
189 Mass. 555, 76 N.E. 217, 218 (1905).
But
Murray
offers some solace to proponents of an incontestability clause that does away entirely with any opportunity to contest the bona fides of a policy (what we shall call an “immediate” incontestability clause). There, the court emphasized the importance of holding the underwriter of a policy to its duty “to fulfill its plain and deliberately assumed obligation.” 48 A. at 801. This emphasis might be seen as favoring the literal enforcement of incontestability clauses, particularly those that operate in favor of the insured.
Cf.
R.I. Gen.
Laws § 27-4-6.2(a) (condoning in the life insurance context contracts that are “more favorable to policyholders” than that required by statute).
In addition, the special risk of fraud associated with immediate incontestability clauses may be offset by a built-in protection. Thus, some courts have upheld immediate incontestability clauses, noting that the insurer had an unlimited time, prior to issuing the policy, to investigate the incidence of fraud.
See, e.g., Pac. Mut. Life Ins. Co. v. Strange,
223 Ala. 226, 135 So. 477, 478 (1931).
Even if the potential obstacle presented by the immediacy of an incontestability clause can be overcome, an issue remains as to the force of such a clause with regard to a defense premised on the lack of an insurable interest. This, too, is an issue that is largely unaddressed in Rhode Island case law — and there is no consensus across other jurisdictions.
A majority of courts hold “that a life insurance policy lacking an insurable interest is void as against public policy, and thus never comes into force, making the incontestability provision inapplicable.”
PHL Var. Ins. Co. v. Price Dawe 2006 Ins. Trust ex rel. Christiana Bank and Trust Co.,
28 A.3d 1059, 1065, 1067 n. 18 (Del. 2011) (collecting cases);
see also
7 Lord,
supra,
§ 17:5.
Cronin
— a century-old decision that did not specifically address the insurable interest requirement in the context of an incontestability clause — hints that Rhode Island might align itself with this view.
See Cronin,
40 A. at 497 (commenting, in dictum, that “a purely speculative contract on the life of another ... may properly be held to be void”).
Some courts, however, have held that an insurable interest defense cannot trump an incontestability clause.
See, e.g., Bogacki v. Great-West Life Assurance Co.,
253 Mich. 253, 234 N.W. 865, 865-67 (1931). These courts hold that the lack of an insurable interest renders an insurance policy merely voidable, not void ab initio, so that an insurable interest defense can be defeated by an incontestability clause.
See, e.g., New Eng. Mut. Life Ins. Co. v. Caruso,
73 N.Y.2d 74, 538 N.Y.S.2d 217, 535 N.E.2d 270, 272-73 (1989);
cf. Monast v. Manhattan Life Ins. Co.,
32 R.I. 557, 79 A. 932, 936 (1911) (deeming it “clear both upon principle and authority that a life insurance policy is not void because the premiums have been paid ... by one having no insurable interest in the life of the assured”). This minority position has evolved subsequent to the
Cronin
dictum.
For all of these reasons, we conclude that it is fairly debatable whether Rhode Island would permit an immediate incontestability provision to thwart the assertion of a defense predicated on the absence of an insurable interest.
III. CERTIFICATION
Against this chiaroscuro backdrop, we certify to the Rhode Island Supreme Court the following questions:
1. If the owner and beneficiary of an annuity with a death benefit is a stranger to the annuitant, is the annuity infirm for want of an insurable interest?
2. Does a clause in an annuity that purports to make the annuity incontestable from the date of its issuance preclude the maintenance of an action based on the lack of an insurable interest?
We wish to make it clear that these questions may be considered in whatever sequence the Rhode Island Supreme Court prefers and that a particular answer to one question may obviate the need for answering the other. We also wish to make it clear that we would welcome the advice of the Rhode Island Supreme Court on any other aspect of Rhode Island law that the
Justices believe should be clarified in order either to aid in the proper resolution of the certified questions or to give context to their response.
The Clerk of this court is directed to transmit to the Rhode Island Supreme Court, under the official seal of this court, a copy of the certified questions and this accompanying memorandum, along with a copy of the district court record and copies of all briefs and appendices heretofore filed in this court. We stay proceedings before us, while retaining jurisdiction, pending the Rhode Island Supreme Court’s determination of the certified questions.
So Ordered.