Jesse ARONSTEIN, individually and on behalf of all others similarly situated, Plaintiff, Appellant/Cross-Appellee, v. MASSACHUSETTS MUTUAL LIFE INSURANCE COMPANY, Defendant, Appellee/Cross-Appellant, C.M. Life Insurance Company, Defendant.
When a life insurance company cuts the interest rate of an annuity in half, it must make that change clear to its consumers. In choosing to change an interest rate by an endorsement its own staff warned would sow consumer confusion, defendant Massachusetts Mutual Life Insurance Company (“MassMutual”) introduced ambiguity into its annuity certificate. Because of that lack of clarity, plaintiff Jesse Aronstein believed that he had bought an annuity that guaranteed him 3.0% annual interest. MassMutual has taken the position that it clearly promised only 1.5%. After a bench trial, the district court ruled against MassMutual; it also ruled against Aronstein's class action claims.1
Finding no error, we affirm the district court's denial of class certification, entry of judgment for Aronstein, and award of prejudgment interest.
In 2003, MassMutual decided to cut the minimum guaranteed interest rates -- the lowest rates annuities can earn -- paid to purchasers of some of its annuities. For the New York version of its “Odyssey” annuity, MassMutual reduced the rate from 3.0% to 1.5%. To effect that change, MassMutual considered two options: updating the entire certificate or attaching an amendment to the certificate, called an endorsement or rider, to override the certificate's terms. MassMutual chose the latter option, despite internal documents stating that creating such a conflict between the endorsement and the certificate would be “[c]onfusing to the client.” MassMutual entitled this rider the “GUARANTEED INTEREST RATE ENDORSEMENT.” It also changed references in its internal documents from “guaranteed interest rate” to “guaranteed minimum interest rate.” (Emphasis added.) But MassMutual did not make a similar clarification in the endorsement to educate all of its consumers.
We briefly describe the Odyssey annuity certificate, as purportedly amended by the endorsement. The body of the certificate explains that interest will accrue at a minimum rate of 3.0% per year. On the first substantive page of the certificate, the minimum guaranteed interest rate is listed as 3.0%, and the policy promises that “[t]he interest rate credited to this Certificate shall never be less than the Minimum Guaranteed Interest Rate shown above” (i.e., 3.0%). The certificate repeats that promise several times. The certificate also contains payout schedules based on the 3.0% rate. No payout figures based on a 1.5% rate are included. Additionally, the certificate promises 4.0% interest for the first year of the term. The certificate also provides that “[t]he entire Certificate consists of this Certificate, the application, if any, and any riders or endorsements attached to this Certificate.”
The endorsement on which MassMutual relies appears on a separate sheet following the certificate. The endorsement is entitled “GUARANTEED INTEREST RATE ENDORSEMENT,” but the certificate refers to the relevant rate as the “Minimum Guaranteed Interest Rate.” Below the title, the endorsement reads:
This endorsement modifies the Contract to which it is attached. In case of a conflict with any provision in the Contract, the provisions of this endorsement will control. The effective date of this endorsement is the date the endorsement is attached to the Contract. Where appropriate, the word “Certificate” shall be substituted for the word “Contract”. The Contract is modified as follows:
The Minimum Guaranteed Interest Rate has been changed to 1.5%.
The single endorsement page was only one of thirty-four pages sent to consumers. At the start of the certificate, the table of contents provides no reference to the endorsement. The second page of the certificate schedule lists a single “rider” to the policy: a “GUARANTEED INTEREST RATE ENDORSEMENT.”
New York regulators approved the interest rate change in July 2003, and MassMutual put the changed interest rate into effect for annuities sold after December 31, 2003. In the lead-up to that change, MassMutual communicated about the new, reduced interest rate with both its own salesforce and affiliated salespeople. It provided them with new marketing materials for the Odyssey, which did not reference the 3.0% rate (or the 1.5% rate).
Before purchasing the annuity, in December 2003, Aronstein met with a third-party salesperson authorized to sell MassMutual annuities at his local bank. Aronstein told the salesperson that he was looking for a secure investment with better interest rates than the bank was advertising for certificates of deposit. The salesperson steered Aronstein towards the Odyssey, which he described as offering a 3.0% minimum interest rate. The salesperson also gave Aronstein one of MassMutual's marketing brochures reflecting the 3.0% rate. The salesperson did not tell Aronstein that the minimum guaranteed interest rate would soon be halved. Had Aronstein bought the annuity that day, he would have received a certificate guaranteeing him a 3.0% return. But by the time Aronstein had reviewed the materials with his wife and decided to purchase the annuity, MassMutual had lowered the interest rate. And when Aronstein returned to purchase the annuity, the salesperson did not tell him about the change. Aronstein executed the papers to purchase the annuity believing that he would always receive at least 3.0% annual interest.
When Aronstein received in the mail from MassMutual a package of the key documents, he reviewed it. But he did not notice the single-page endorsement tucked between the certificate and a copy of his application. He read the nineteen pages of the certificate, and he specifically noted that the certificate listed the minimum guaranteed interest rate as 3.0%. He then “thumbed through” the four pages of the copy of his application at the end of the package. He did not catch the endorsement sandwiched between the certificate and the application materials. In fact, Aronstein did not notice the endorsement until years later when, upon receiving an annual statement, he realized that MassMutual had credited him less than 3.0% interest for several years. Even then, he only discovered it attached to his certificate after a MassMutual employee sent him a copy and Aronstein matched the copy against the one in his materials. After discovering the endorsement, Aronstein obtained counsel who sued on behalf of Aronstein and a purported class of similarly situated customers. He alleged causes of action for breach of contract and unfair and deceptive trade practices.
Following discovery, the district court allowed in part and denied in part MassMutual's motion for summary judgment. It dismissed the unfair and deceptive trade practices claim, holding that it was time barred. It allowed the breach of contract claim to continue to trial, holding that the meaning of the certificate was ambiguous. The district court then denied Aronstein's motion for class certification, permitting him to pursue his individual claim. Relying on its earlier determination that the contract was ambiguous, it concluded that extrinsic evidence would be needed to determine what interest rate each class member believed the annuity guaranteed. Thus, the district court held that common issues did not predominate over individual ones, precluding class certification. See Fed. R. Civ. P. 23(b)(3). The district court then tried Aronstein's individual claim without a jury and entered judgment for Aronstein. It also awarded Aronstein prejudgment interest.
Aronstein appealed from the denial of class certification. MassMutual cross-appealed from the judgment against it and conditionally cross-appealed from the grant of prejudgment interest.
After a bench trial, we review the district court's findings of fact for clear error and its legal conclusions de novo. Calandro v. Sedgwick Claims Mgmt. Servs., Inc., 919 F.3d 26, 33 (1st Cir. 2019). In this diversity case, the parties agree that we apply New York law. See Erie R.R. Co. v. Tompkins, 304 U.S. 64, 78, 58 S.Ct. 817, 82 L.Ed. 1188 (1938); see also Borden v. Paul Revere Life Ins. Co., 935 F.2d 370, 375 (1st Cir. 1991) (“Where ․ the parties have agreed about what law governs, a federal court sitting in diversity is free, if it chooses, to forgo independent analysis and accept the parties’ agreement.”).
We begin with MassMutual's challenge to the district court's judgment for Aronstein. MassMutual argues that the annuity unambiguously sets the minimum guaranteed interest rate at 1.5%, entitling it to judgment as a matter of law. That argument fails.
The annuity contract, composed of several documents, was ambiguous under New York law as to the Odyssey minimum guaranteed interest rate at the time Aronstein purchased it.
New York courts read contracts as a “harmonious and integrated whole.” See Westmoreland Coal Co. v. Entech, Inc., 100 N.Y.2d 352, 763 N.Y.S.2d 525, 794 N.E.2d 667, 670 (2003). Several documents can comprise a single contract. See Cnty. of Columbia v. Cont'l Ins. Co., 83 N.Y.2d 618, 628, 612 N.Y.S.2d 345, 634 N.E.2d 946 (1994) (“[I]t is settled that in construing an endorsement to an insurance policy, the endorsement and the policy must be read together, and the words of the policy remain in full force and effect except as altered by the words of the endorsement.”). Whether a contract is ambiguous is a question of law. See Banos v. Rhea, 25 N.Y.3d 266, 11 N.Y.S.3d 515, 33 N.E.3d 471, 476 (2015). “ ‘Ambiguity in a contract arises when the contract, read as a whole, fails to disclose its purpose and the parties’ intent,’ or where its terms are subject to more than one reasonable interpretation.” Universal Am. Corp. v. Nat'l Union Fire Ins. Co. of Pittsburgh, Pa., 25 N.Y.3d 675, 16 N.Y.S.3d 21, 37 N.E.3d 78, 80 (2015) (quoting Ellington v. EMI Music, Inc., 24 N.Y.3d 239, 997 N.Y.S.2d 339, 21 N.E.3d 1000, 1003 (2014)) (citations omitted). When the language of a header conflicts with the language of text below it, that difference may create ambiguity. Kowalczyk v. Flintkote Co., 63 A.D.2d 846, 405 N.Y.S.2d 852, 853 (1978).
We identify three elements of the certificate and the endorsement that create ambiguity.
First, the certificate promises that “[t]he interest rate credited to this Certificate shall never be less than the Minimum Guaranteed Interest Rate shown above.” (Emphasis added). The minimum guaranteed interest rate shown above that paragraph is 3.0%. The definitions section of the certificate similarly defines the minimum guaranteed interest rate by reference to the rate “shown on the Certificate Schedule.” (Emphasis added). A consumer could construe those representations to mean that the rate actually printed on the certificate schedule, not the rate in the endorsement, controls. This would be reasonable because the endorsement does not actually say that the rate in the certificate and the rate in the endorsement conflict.
Second, the heading of the endorsement is misleading. The endorsement is entitled “GUARANTEED INTEREST RATE ENDORSEMENT,” but the policy guarantees multiple interest rates. Along with the minimum guaranteed interest rate, the policy promises to credit interest to the annuity at a “current interest rate” for the first year of the term. Thus, one could reasonably look to the endorsement's heading and understand that the endorsement modified the other guaranteed interest rate and not the minimum guaranteed interest rate. See id. Indeed, as the district court permissibly inferred, MassMutual internally corrected how it referred to the interest rate in to “avoid any confusion among MassMutual employees regarding the particular interest rate to be changed.” (Emphasis added.)
Third, a consumer could understand the endorsement as not modifying the guaranteed interest rate. The lone reference to a “rider” in the certificate informs the consumer that there is a “GUARANTEED INTEREST RATE ENDORSEMENT,” but as we have concluded, that language is itself misleading. The certificate's table of contents omits the endorsement entirely. Both the table of contents and the pagination of the nineteen numbered pages 2 that comprise the certificate would lead a consumer to believe reasonably that there was no modification to the minimum guaranteed interest rate. After reviewing eighteen pages premised on a 3.0% interest rate -- including payout tables based on that figure -- a consumer must find the endorsement sandwiched between the certificate and his application. Taking the certificate and the endorsement as a whole, as we must under New York law, we cannot conclude that a consumer would read the endorsement and understand that it fundamentally altered the economic terms of the deal. Finally, although it is not necessary to our analysis, it seems unlikely that a consumer would even see that there was an endorsement given these same factors.
The three elements render the certificate's minimum guaranteed interest rate susceptible to two reasonable interpretations. A consumer could reasonably read the certificate as MassMutual does and conclude that the minimum guaranteed interest rate was 1.5%. But he could also reasonably read it as Aronstein did and conclude that he was entitled to a 3.0% interest rate. Therefore, the provisions setting the minimum guaranteed interest rate are ambiguous.
“Where, as here, the language of a contract is ambiguous, its construction presents a question of fact” under New York law. Pepco Constr. of N.Y., Inc. v. CNA Ins. Co., 15 A.D.3d 464, 790 N.Y.S.2d 490, 491 (2005). So we must determine whether the district court's construction of the contract was clearly erroneous.
When a contract is ambiguous, New York courts look to extrinsic evidence of the parties’ intent. Carlson v. Am. Int'l Grp., Inc., 30 N.Y.3d 288, 67 N.Y.S.3d 100, 89 N.E.3d 490, 498 (2017). If that evidence does not resolve the ambiguity, then New York courts resolve the ambiguity against the insurer. See Fairchild v. Genesee Patrons Coop. Ins. Co., 238 A.D.2d 841, 656 N.Y.S.2d 544, 545 (1997).
Based on the extrinsic evidence, the district court found that each party believed that the certificate provided for a different minimum guaranteed interest rate. Because the extrinsic evidence did not resolve the ambiguity, the district court construed the ambiguity against MassMutual as drafter of the contract.
We find no error. Indeed, MassMutual has waived any argument to the contrary by failing to raise it in its briefs. See United States v. Zannino, 895 F.2d 1, 17 (1st Cir. 1990).
We turn next to the district court's denial of Aronstein's motion for class certification. We review legal questions de novo, factual determinations for clear error, and the district court's overall class certification decision for abuse of discretion. In re Nexium Antitrust Litig., 777 F.3d 9, 17 (1st Cir. 2015).
Aronstein argues that New York courts would, but have not yet, recognize two doctrines of contract interpretation, each of which would render extrinsic evidence irrelevant.3 If extrinsic evidence were irrelevant, then the district court's class certification determination would be weakened, as it held that individualized inquiry into extrinsic evidence would destroy predominance over common questions of fact and law. The district court ruled against the argument, and we find no error.
To apply the substantive law of New York to this dispute, we look foremost to the decisions of the New York Court of Appeals. See Philibotte v. Nisource Corp. Servs. Co., 793 F.3d 159, 165 (1st Cir. 2015). In making an informed “Erie prediction” when state courts have not spoken directly, federal courts are restrained. See generally 19 A. Miller, Federal Practice and Procedure § 4507 (3d ed. Apr. 2021 update). A “plaintiff, who made a deliberate choice to sue in federal court rather than in ․ state court, is not in a position to ask us to blaze a new trail that the [state] courts have not invited.” Jones v. Secord, 684 F.3d 1, 11 (1st Cir. 2012).
Aronstein first argues that New York courts would construe an “intentionally ambiguous” contract against its drafter. That is not the present state of New York law.
Under New York law, courts must first examine extrinsic evidence of the parties’ intent before turning to doctrinal presumptions. See Perella Weinberg Partners LLC v. Kramer, 153 A.D.3d 443, 58 N.Y.S.3d 384, 389 (2017). Doctrines like contra proferentem “may only be applied as a last resort, if the extrinsic evidence is inconclusive.” Id. (emphasis added); see also Nationstar Mortg. LLC v. Goeke, 151 A.D.3d 1237, 57 N.Y.S.3d 223, 227 (2017) (collecting cases). That rule is particularly important because it preserves parties’ right to a jury trial when extrinsic evidence conflicts. See Hartford Acc. & Indem. Co. v. Wesolowski, 33 N.Y.2d 169, 350 N.Y.S.2d 895, 305 N.E.2d 907, 909 (1973); Nationstar Mortg., 57 N.Y.S.3d at 227.
The most important factor in predicting whether New York courts would adopt a rule is what New York courts have to say on the matter. Here they are silent.4 And no other court has adopted Aronstein's proposed rule either. The closest case he points to is McMullin v. McMullin, 338 S.W.3d 315 (Ky. Ct. App. 2011). In McMullin, the Kentucky Court of Appeals interpreted a property division agreement between divorcing spouses. Id. at 317. The husband intentionally introduced an ambiguous provision into the agreement. Id. at 319, 323. Because of that chicanery, the court refused to enforce a purported waiver of contra proferentem and instead ruled for the wife. See id. 322-23. But far from refusing to consider extrinsic evidence, the court explicitly considered the wife's testimony about her intent in contracting. Id. at 321 (“Her testimony may be considered on this matter because courts may look to extrinsic evidence where an ambiguity exists in a contract.”). Thus, McMullin does not support the rule Aronstein offers.
Aronstein next argues that New York courts would construe standard-form contracts in the same manner for all parties. He derives that rule from § 211(2) of the Restatement (Second) of Contracts (Am. L. Inst. 1981), which provides that a standardized contract is “interpreted wherever reasonable as treating alike all those similarly situated, without regard to their knowledge or understanding of the standard terms of the writing.” We need not address whether New York courts would adopt that provision of the Restatement; even if they would, the provision provides no aid to Aronstein.
A reporter's note to § 211 explains that “[w]hen an employee of the dominant party explains a term in a standardized agreement to the other party, parol evidence may be admitted to show the explanation.” Restatement (Second) of Conts. § 211 reporter's note to cmt. b; see, e.g., Darner Motor Sales, Inc. v. Universal Underwriters Ins. Co., 140 Ariz. 383, 682 P.2d 388, 395–96 (1984); Sutton v. Banner Life Ins. Co., 686 A.2d 1045, 1050-51 (D.C. 1996). It is true that in some cases, parol evidence is irrelevant because the parties’ expectations are subordinated to the language of the contract. See Kolbe v. BAC Home Loans Servicing, LP, 738 F.3d 432, 440–41 (1st Cir. 2013) (lead opinion of equally divided en banc court) (collecting cases). But that is not so when the drafting party orally clarifies an ambiguity to most of its customers. After all, the purpose of § 211 is to give consumers the bargain they “reasonably expected.” See, e.g., Darner, 682 P.2d at 391. The purpose is not to allow consumers to obtain benefits they never expected by creative legerdemain. While the provision may provide an “advantage” to “some sophisticated customers who contracted with knowledge of an ambiguity or dispute,” it cannot reasonably be construed to allow a host of customers to exploit a provision they fully understood. See Restatement (Second) of Conts. § 211 cmt. e.
What is more, Aronstein did not establish that he is similarly situated to most other Odyssey purchasers as application of the Restatement provision would require. As the district court explained, although Aronstein was never told that MassMutual reduced the interest rate to 1.5%, MassMutual produced evidence that it “engaged in an extensive marketing campaign to inform sales agents of the minimum guaranteed interest rate change, its marketing materials were modified to reflect this change, and sales agents generally explained this key interest rate to potential purchasers orally.” And Aronstein has adduced no evidence to show that many Odyssey purchasers were confused by the interest rate they obtained. Indeed, the record contains evidence of only one Odyssey purchaser who was not informed that the rate had been lowered: Aronstein himself. Thus, he cannot show that application of § 211(2) would allow for common interpretation of the endorsement across the proposed class.
Finally, we address MassMutual's appeal of the district court's award of prejudgment interest. MassMutual conditionally cross-appealed from the award. In its opening brief, MassMutual stated that it would drop the issue if we either affirmed the denial of class certification or reversed the judgment for Aronstein. Because we affirm the denial of class certification, MassMutual has waived its challenge to prejudgment interest.
The decisions below are affirmed. Costs awarded to Aronstein. See Fed. R. App. P. 39(a).
1. No reporter has published and no electronic database contains the relevant district court decisions. They can be found, however, on the district court's docket. Aronstein v. Mass. Mut. Life Ins. Co., No. 3:15-cv-12864-MGM (D. Mass. judgment entered Nov. 12, 2020), ECF Nos. 169, 212, 223, 229, 230.
2. Each page of the certificate has a footer with the page number listed out of eighteen total pages (e.g., Page 1 of 18). Because the certificate contains both a page 3A and a page 3B, it has nineteen pages in total.
3. Aronstein also argues that the district court committed reversible error by failing to predict whether the New York courts would adopt these doctrines. Not so. The district court held that New York courts would not buy Aronstein's theories. In any case, because our review of these legal questions is de novo, the argument is irrelevant.
4. Aronstein also tries to argue that New York courts do not look to extrinsic evidence in interpreting ambiguous insurance contracts. Of course, New York law says nothing of the kind. See, e.g., Fairchild, 656 N.Y.S.2d at 545 (“[W]here an insurance policy is found to be ambiguous, the parties may submit extrinsic evidence to aid in construction. It is only where such evidence does not resolve the equivocality that the ambiguity must be resolved against the insurer.” (citation omitted)).
LYNCH, Circuit Judge.