MEMORANDUM DECISION AND ORDER GRANTING IN PART AND DENYING IN PART DEFENDANT’S MOTION FOR PARTIAL SUMMARY JUDGMENT AND DENYING PLAINTIFFS’ MOTION FOR PARTIAL SUMMARY JUDGMENT
McMAHON, District Judge:Plaintiffs Martin Fleisher (“Fleisher”), as Trustee of the Michael Moss Irrevocable Life Insurance Trust II, and Jonathan Berck (“Berck,” and, together with Fleisher, “Plaintiffs”), as Trustee of the John L. Loeb, Jr. Insurance Trust, initiated this class action against Defendant Phoenix Life Insurance Company (“Phoenix”). The only remaining claim (Count One) alleges that Phoenix breached the terms of certain insurance policies owned by the Trusts.
Both the Plaintiffs and the Defendant move for partial summary judgment on liability for Fleisher’s claim pursuant to Rule 56 of the Federal Rules of Civil Procedure. For the reasons discussed below, the Defendant’s motion in granted in part *460and denied in part, and the Plaintiffs’ motion is denied in its entirety.
BACKGROUND1
A. The Parties
Plaintiff Fleisher is the trustee of the Michael Moss Irrevocable Life Insurance Trust II. Plaintiff Berck is the trustee of the John L. Loeb, Jr. Insurance Trust. These two trusts own life insurance policies issued by Defendant Phoenix. See PI. 56.1 Statement at ¶ 2.
Plaintiff Fleisher brings his breach of contract claim on behalf of a class. The Court previously certified a class consisting of:
All owners of flexible-premium “universal life” insurance policies issued by Phoenix Life Insurance Company that were subjected to the Cost of Insurance rate increase announced by Phoenix on or about November 1, 2011 (excluding defendant Phoenix, its officers and directors, members of their immediate families, and the heirs, successors or assigns of any of the foregoing).
Compl. ¶¶ 37; see also Docket No. 135. We have come to refer to this class as the “2011 Class” to distinguish it from a second class, formerly represented by Plaintiff Berck and now decertified, who brought similar claims relating to a cost of insurance rate increase imposed in 2010.
This motion deals only with the claim for breach of contract asserted by Fleisher and the 2011 Class.
B. The PAUL Policies
Plaintiff Fleisher and the members of the 2011 Class own (or owned)2 “substantively identical” Phoenix Accumulator Universal Life (“PAUL”) insurance policies. Def. 56.1 Statement Resp. at ¶ 3.
Generally speaking, there are two categories of life insurance: whole life insurance and term life insurance. Term life insurance protects the policyholder for a specified period of time. Whole life policies, by contrast, remain in existence throughout the life of an insured. In general, premiums on term insurance policies pay only for the cost of providing the insurance, while at least some whole life policies have some type of participatory investment or savings feature. The PAUL policies at issue here are a type of whole life insurance called universal life insurance.
Traditional whole life insurance policies require payment of a fixed monthly premium. The cost of life insurance for any insured increases over time as the insured ages and becomes more likely to die. In order to spread out this insurance cost into fixed monthly premium payments, the premium charged earlier in the life of the insured must be greater than the actual insurance cost, and the premium charged later in life must be less than the actual insurance cost. The amount of early year premiums paid in excess of actual insurance costs goes into a cash reserve that accumulates in value (the “policy value”). This policy value functions like a savings account. When the insurance cost exceeds the fixed monthly premium later in life, the policy value is used to supplement the fixed premium in order to cover the total actual insurance cost. See New York State Department of Financial Services, *461“Basic Types Of Policies,” http://www.dfs. ny.gov/consumer/cli — basic.htm (last visited April 29, 2014).
During the life of the insured, the policyholder may choose to cash out his accumulated policy value by “surrendering” the policy. Though surrender causes the policyholder to lose life insurance protection, he is able to withdraw the balance of the policy value in cash, subject to any surrender charges specified in the contract. See id.
Universal life insurance is similar to traditional whole life insurance but with a central distinguishing feature — the policyholder is not required to pay a fixed monthly premium.
Under the terms of these “flexible premium” PAUL policies, a policyholder must pay a Minimum Initial Premium, which is specified in his contract. See Compl. Ex. B at 1, 3. This amount covers his up-front costs. Any amount in excess of the Minimum Initial Premium that the insured chooses to pay is deposited into his savings account — his “Policy Value,” on which Phoenix pays interest.
Thereafter, the policyholder has options. The only requirement is that he must pay enough each month to cover the monthly insurance expenses (referred in the policy as the “Monthly Deduction”). If he fails to do that, the policy will lapse.
Otherwise, the policyholder has flexibility to determine the amount and timing of his premium payments. He may choose to make monthly payments equal to the Monthly Deduction and nothing more— rather like buying a term life insurance policy. See Def. 56.1 Statement Resp. at ¶ 5. If the policyholder chooses to pay this minimum amount every month, his Policy Value will never increase above zero. Such a policyholder would not be utilizing the savings component of his policy.
Alternatively, a policyholder can pay an amount in excess of the minimum Monthly Deduction. The excess payment will be added to his Policy Value.
Once his Policy Value is high enough, the policyholder may elect not to make premium payments for a while, and instead allow Phoenix to draw down his Monthly Deduction from his accumulated Policy Value. The policyholder can do this until the Policy Value is depleted — at which point the Policy Value account must be replenished, or the policy will lapse. This strategy allows the policyholder flexibility in the timing of his payments; he can pay excess premiums in times when he has better cash flow and then use his accumulated Policy Value to cover the periods when his cash flow deteriorates. Meanwhile, the money in the policyholder’s “savings account” — his Policy Value — accrues interest for as long as it sits in the “savings account.”
Because of these characteristics, Phoenix marketed PAUL policies as “offering [policyholders] flexibility” to “[a]djust the amount and timing of premium payments to fit [a policyholder’s] cash flow needs.” Lewis Deck Ex. 7 at 965.
Phoenix asks each policyholder to estimate his “Planned Premium” during the insurance application process. See Compl. Ex. B at 3. The policy defines the “Planned Premium” as “the premium that is selected in the application or as later changed by you for this policy that you intend to be pay [sic] on a regular modal basis.” Id. However, the policyholder is not required to adhere to his estimate of his Planned Premium. The term is not mentioned elsewhere in the policy as a required payment. In fact, Section 10 of the policy, titled “Premiums,” makes no mention of the policyholder’s Planned Premium. See id. Ex. B at 13. Instead, the policy repeatedly refers to how Phoenix will handle *462“any premium payment” it receives, bolstering the understanding that a policyholder is not required to pay premiums in any particular amount. Id. This lack of any required premium payment— “Planned” or otherwise — is why these policies are referred to as “flexible premium.”
A policyholder can also choose to pay excess premiums every month in order to accumulate a high Policy Value. So if a policyholder finds Phoenix’s current interest rate attractive, he may choose to invest more money with Phoenix; if he does not, his Policy Value will increase each month only by the amount of interest earned on that Value. See id.; PI. 56.1 Statement at ¶ 6.
Phoenix marketed this “Cash Accumulation” aspect of the policies, stating: “A competitive, interest — sensitive return helps your policy’s cash value increase over time.” Lewis Decl. Ex. 7 at 964. Phoenix is able to pay interest to policyholders because it invests the funds it receives from them and earns returns on those investments.
In sum, it is entirely up to the policyholder whether to utilize the savings (or “Cash Accumulation”) feature of his policy. He may choose to pay the minimum Monthly Deduction each month, or he may take advantage of the option to pay excess premiums, to build up his Policy Value, and to use the accumulated Policy Value to cover his Monthly Deductions. He may maintain a high Policy Value, a low Policy Value, or no Policy Value at all. As one would expect, Phoenix uses experience studies to predict how many policyholders will use their policies as investment vehicles and how many will not; but Phoenix can have no expectation that any particular policyholder will use the savings feature or keep money a high Policy Value, because that is at the policyholder’s option. These are truly “flexible premium” policies.
C. Computation of the Monthly COI Charge
One of the insurance charges included in the Monthly Deduction is the “Cost of Insurance” (“COI”) charge. The COI charge represents Phoenix’s risk of paying the death benefit (the policy’s “Face Amount”) to the policy’s beneficiary upon the death of the insured. The policyholder’s payment of COI charges to cover Phoenix’s risk is the policy’s insurance component. See Compl. Ex. B at 12.
The policy sets forth a formula for calculating the monthly COI charge. The formula is this: Monthly COI Charge = COI Rate x Net Amount at Risk. See id.
The policy defines the “Net Amount at Risk” (“NAR”) as the Total Face Amount of the policy (the death benefit) divided by 1.0032737 less the Policy Value in any given month.3 See id. Ex. B at 7, 12. Because the denominator used in calculating Net Amount at Risk is approximately 1, for shorthand purposes I will describe the NAR as the Total Face Amount minus the Policy Value — although I recognize that the formula differs slightly from that rounded calculation.
The “Total Face Amount” is equal to the “Basic Face Amount” plus the “Supplemental Face Amount;” these two components of the death benefit are selected by the policyholder and shown on the face of the policy. For example, Fleisher’s Basic Face Amount is $6 million, and his Supplemental Face Amount is zero. Thus, his Total Face Amount — the total to be paid to the beneficiaries upon the death of the insured — is $6 million. See id. Ex. B at 3.
The “Policy Value” (a defined term) is “determined by accumulating with interest *463the Policy Value for the prior day increased by Net Premiums credited and decreased by withdrawals and, on the Monthly Calculation Day, the Monthly Deductions from Policy Value ...” Id. Ex. B at 8. The “Net Premium” is equal to any premium payment submitted by a policyholder less a percentage “Sales Charge,” which is a fee deducted from each premium payment. See id. Ex. B at 12. The “Monthly Calculation Day” is the “date on which monthly deductions are assessed from the Policy Value,” which occurs once a month. Id. Ex. B at 8. In other words, the Policy Value on any given day is equal to the prior day’s Policy Value plus any premiums or interest added to the account, less any expenses or withdrawals deducted from the account.
The policy does not contain a simple, straightforward definition of the “COI rate,” as it does for the other terms used. Instead, Section 9 of the policy — entitled “Policy Value” — explains how to calculate something called the “Cost of Insurance Charge.” The COI rate is used to compute the “Cost of Insurance Charge,” and, as Section 9 explains, there are two different ways to calculate the COI rate, depending on when it is being calculated.
The first paragraph (“Paragraph A”) explains how to calculate the “rates for the Cost of Insurance Charge” on the date the policy comes into force:
The rates for the Cost of Insurance Charge as of the Policy Date are based on the sex, if applicable, Age, Risk Classification, Basic Face Amount, Supplemental Face Amount, Net Amount at Risk, and duration that the coverage has been in force for the insured.
Id. Ex. B at 12. Paragraph A explains that the COI rate “as of the Policy Date,” (a date identified on the face of the policy) will be calculated based on certain factors. The Policy Date is not necessarily the same as the “Issue Date” — which is defined separately — but in Fleisher’s policy the Policy Date and the Issue Date are the same. The insured’s “sex,” “Age,” and “Risk Classification” are identified on the face of the policy in accordance with the insured’s personal and health characteristics. For example, Fleisher’s policy states that the insured, as of the issue date, was a 72-year-old male whose Risk Classification was “Non-Smoker, Percentage Substandard 150.0%.” Id. Ex. B at 3. For any policy in which the Policy Date and the Issue Date (the date when coverage comes into force) are the same date, the “duration that the coverage has been in force” drops out as a factor in calculating the COI rate as of the Policy Date, because that value is zero.
The policy then states: “The first Monthly Calculation Date is the Policy Date.” Id. Ex. B at 8. Thus, a Monthly Deduction is due on the Policy Date, and the first monthly COI charge is equal to the NAR multiplied by the initial COI rate that Phoenix determines “based on” the factors listed in Paragraph A: sex, age, risk classification, basic and supplemental Face Amount, NAR, and duration of coverage.
The next paragraph (“Paragraph B”) explains how to calculate the COI Charge (which the policy defines as “the charge for the Net Amount at Risk”) for “a specific Policy Month.” Id. Ex. B at 12. A “Policy Month” is defined as “the period from one Monthly Calculation Date up to, but not including, the next Monthly Calculation Date.” Id. Ex. B at 8. In simpler terms, a Policy Month is any month during which the policy is in force. Paragraph B states:
The Cost of Insurance Charge for a specific Policy Month is the charge for the Net Amount at Risk. The charge for the Net Amount at Risk is an amount *464equal to the per dollar cost of insurance rate for that month multiplied by the Net Amount at Risk, and such rates will be based on our expectations of future mortality, persistency, investment earnings, expense experience, capital and reserve requirements, and tax assumptions. The Maximum Monthly Rates at any Age are shown in Section 2....
Id. Ex. B at 12. This lawsuit is about the meaning of this paragraph.
To establish the COI Charge for any particular Policy Month, Phoenix multiplies the “per dollar cost of insurance rate for that month” by the NAR (Total Face Amount less Policy Value). Id. (emphasis added). Paragraph A discussed above explains only how to calculate the COI rate that will be used to determine what Monthly Deduction is due on the Policy Date — what I will call the “initial COI rate.” In any succeeding month, the COI rate may be adjusted in accordance with the terms of Paragraph B before calculating the COI Charge.
The policy says that “such rates” (i.e., the COI rates for any Policy Month) will be based on Phoenix’s “expectations” about six identified factors: future mortality, persistency, investment earnings, expense experience, capital and reserve requirements, and tax assumptions. Notably, this list does not include the factors that are used to calculate the COI rate “as of the Policy Date.” I assume this is because the initial COI rate as of the Policy Date is a baseline rate off which future rate adjustments are calculated, and those factors were taken into account when the initial COI rate was calculated.
Phoenix is not required to calculate a new COI rate every month. The company starts with the initial COI rate in force on the Policy Date and “review[s] [its] COI rates periodically, and may re-determine Cost of Insurance rates at such time ...” Id.
The COI rate is also subject to certain maximum rates set forth in the policy, which vary in accordance with the age of the insured. See id. Ex. B at 7.
One of the breach of contract allegations in this lawsuit is that, in calculating a COI rate adjustment in 2011 (the “2011 COI Rate Adjustment”) for some (but not all) of the PAUL policies, Phoenix violated the contract by using factors in its calculation of the “cost of insurance rate for that month” that are impermissible under the terms of the policy. Whether plaintiffs are correct about that turns on the correct interpretation of Paragraph B.
D. The 2011 COI Rate Adjustment
In November 2011, Phoenix announced that it was imposing the 2011 COI Rate Adjustment; it changed the COI rates for some (but not all) of the PAUL policies, including those held by Fleisher and the members of the 2011 Class. The company sent letters to the affected policyholders to tell them that the 2011 COI Rate Adjustment would take effect on their next policy anniversaries. See PI. 56.1 Statement at ¶ 28. The members of the 2011 Class are those policyholders whose COI rates increased. See Fleisher v. Phoenix Life Ins. Co., No. 11 Civ. 8405(CM)(JCF), 2013 WL 4573530, at *1 (S.D.N.Y. Aug. 26, 2013).
It is undisputed that, in setting a new “cost of insurance rate for that month,” Phoenix took Policy Values into consideration. Phoenix admits that it “analyzed funding ratios ... in the process of defining the class” that would be subject to the 2011 COI Rate Adjustment. Def. 56.1 Statement Resp. at ¶ 30. A policy’s funding ratio is equal to the Policy Value divided by the Total Face Amount. Essentially, Phoenix determined which policies had low funding ratios (and thus, lower Policy *465Values) — ie., it determined which policies were not being used as investment vehicles, which made them less remunerative to Phoenix — and then identified groups of policies that more or less matched the policies with low funding ratios (PAUL series IIIA policies with insureds age 68 or older and face amounts of $1 million or more, and PAUL series IIIB/C policies with insureds age 65 or older and face amounts of $1 million or more) and subjected these groups to the 2011 COI Rate Adjustment. See Def. 56.1 Statement at ¶ 2.
Fleisher argues that Policy Value (or funding ratio, which derives from Policy Value) is not among the factors that can be taken into consideration in adjusting COI rates for months after the Policy Date, because it is not one of the six factors enumerated in Paragraph B.
Phoenix admits that Policy Value is not explicitly listed as one of the factors that can be factored into the calculation of an adjusted COI rate. But Phoenix argues that it is implicitly permitted to incorporate Policy Value into its calculation of a new, adjusted COI rate because Policy Value is a component of one of three specifically enumerated factors: “persisten-cy,” “mortality,” and “investment earnings.”
Fleisher also contends that the 2011 COI Rate Adjustment violated two other aspects of the insurance contracts:
We review our Cost of Insurance rates periodically, and may redetermine Cost of Insurance rates at such time on a basis that does not discriminate unfairly within any class of insureds. Any change in rates will be determined prospectively. We will not distribute past gains or recoup prior losses, if any, by changing the rates.
Compl. Ex. B at 12 (emphasis added). Fleisher argues that the rate adjustment breached the contract because (1) Phoenix “unfairly” discriminated within a class of insureds by applying the rate adjustment to only a subset of PAUL policies, and (2) Phoenix designed the rate adjustment to recoup prior losses, not to restore prospective profitability.
It is clear enough, both from the record and from the paucity of the arguments for and against summary judgment, that the latter two issues would have to be tried because there are disputed issues of fact applicable to both. The only possibility of disposing of this case on motion is if Fle-isher is correct that, as a matter of simple contract interpretation, Phoenix used impermissible factors in setting the 2011 COI Rate Adjustment.
E. The NYSID Settlement
In 2011, the New York State Insurance Department (“NYSID”)4 — the state agency responsible for supervising and regulating insurance carriers — investigated a COI rate adjustment that Phoenix imposed on certain other PAUL policies in 2010 (“the 2010 COI Rate Adjustment”). Plaintiff Berck was one of the policyholders subject to the 2010 COI Rate Adjustment, and his breach of contract claim is derived from this adjustment. However, the 2010 COI Rate Adjustment did not affect the policies of Fleisher or the 2011 Class, and this adjustment is not the subject of the pending summary judgment motions.
During its investigation of the 2010 COI Rate Adjustment, the NYSID made very similar allegations as those put forth by Fleisher with respect to the 2011 COI *466Rate Adjustment; the NYSID alleged that Phoenix used impermissible factors to calculate the 2010 COI Rate Adjustment— specifically, that Phoenix used a policyholder’s funding ratio (Policy Value divided by the Total Face Amount) in adjusting the COI rates. See Sklaver Decl. dated Jan. 11, 2013, Ex. J at 1 (NYSID Order, Sept. 6, 2011). This issue was never formally litigated; indeed, there was never a final regulatory finding. Phoenix, while denying liability, chose to settle the 2010 COI Rate Adjustment dispute with the NYSID. See Fleisher v. Phoenix Life Ins. Co., 997 F.Supp.2d 230, 231-34, No. 11 Civ. 8405(CM), 2014 WL 550391, at *1-3 (S.D.N.Y. Feb. 7, 2014). There is, therefore, no regulatory finding that might play into the discussion that follows.
During Phoenix’s settlement discussions with the NYSID regarding the 2010 COI Rate Adjustment, Phoenix voluntarily sent the agency an “Informational Submission” regarding the planned 2011 COI Rate Adjustment. See O’Connell Decl. Ex. D. An NYSID agent responded to the submission with an informal email, stating: “We have no objection to the company going ahead with these changes.” O’Connell Decl. Ex. E. This email was later described to the Court as a regulatory imprimatur on the 2011 COI Rate Adjustment — a fact to which I referred in earlier opinions in this case — but it is now the Court’s understanding that the 2011 COI Rate Adjustment is still under review at the NYSID. See Docket No. 234.
DISCUSSION
I. Standard of Review
A party is entitled to summary judgment pursuant to Rule 56 when there is no “genuine issue of material fact” and the undisputed facts warrant judgment for the moving party as a matter of law. See Fed.R.Civ.P. 56(c); Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 247-48, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986). In addressing a motion for summary judgment, “the court must view the evidence in the light most favorable to the party against whom summary judgment is sought and must draw all reasonable inferences in [its] favor.” Matsushita Elec. Indus. Co. Ltd. v. Zenith Radio Corp., 475 U.S. 574, 587, 106 S.Ct. 1348, 89 L.Ed.2d 538 (1986). Whether any disputed issue of fact exists is for the court to determine. See Balderman v. United States Veterans Admin., 870 F.2d 57, 60 (2d Cir.1989). The moving party has the initial burden of demonstrating the absence of a disputed issue of material fact. See Celotex v. Catrett, 477 U.S. 317, 323, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986). Once such a showing has been made, the non-moving party must present “specific facts showing that there is a genuine issue for trial.” Fed. R. Civ. P. 56(e). The party opposing summary judgment “may not rely on conclusory allegations or unsubstantiated speculation.” Scotto v. Almenas, 143 F.3d 105, 114 (2d Cir.1998).
Moreover, not every disputed factual issue is material in light of the substantive law that governs the case. “Only disputes over facts that might affect the outcome of the suit under the governing law will properly preclude summary judgment.” Anderson, 477 U.S. at 248, 106 S.Ct. 2505. Finally, the non-moving party “must do more than simply show that there is some metaphysical doubt as to the material facts.” Matsushita, 475 U.S. at 586, 106 S.Ct. 1348. To withstand a summary judgment motion, sufficient evidence must exist upon which a reasonable jury could return a verdict for the non-moving party. See Anderson, 477 U.S. at 248, 106 S.Ct. 2505.
II. The Filed Rate Doctrine Does Not Bar the Plaintiffs Claim.
Because the NYSID did not formally approve the 2011 COI Rate Adjust*467ment, the filed rate doctrine does not bar Fleisher’s breach of contract claim.
Prior to implementing the 2011 COI Rate Adjustment, Phoenix informed the NYSID of its intent to do so. See O’Connell Decl. Ex. D. An NYSID agent informally responded that the agency had “no objection.” O’Connell Decl. Ex. E. The Court had previously understood this email to mean that the 2011 COI Rate Adjustment “had arguably been approved by the regulators before it went into effect.” Fleisher, 2013 WL 4573530, at *1.1 thus requested briefing from the parties on whether the filed rate doctrine barred Fleisher’s breach of contract claim. Under this doctrine, “any ‘filed rate’ — that is, one approved by the governing regulatory agency — is per se reasonable and unassailable in judicial proceedings brought by ratepayers.” Wegoland Ltd. v. NYNEX Corp., 27 F.3d 17, 18 (2d Cir.1994). New York courts apply the filed rate doctrine to bar claims challenging the reasonableness of insurance rates that the NYSID has formally approved. See Roussin v. AARP, Inc., 664 F.Supp.2d 412, 415-19 (S.D.N.Y.2009); Minihane v. Weissman, 226 A.D.2d 152, 640 N.Y.S.2d 102, 102 (1996).
However, the doctrine does not apply on the facts of this case. It has become clear to this Court that the NYSID never formally approved the 2011 COI Rate Adjustment.
With respect to universal life insurance policies, New York insurance law only requires insurance companies to obtain formal NYSID approval of policy forms, not COI rates. See N.Y. Ins. Law § 3201(b)(1); New York State Department of Financial Services, “Guidance for Insurers When Submitting Life and Annuity Product Filings,” http://www.dfs.ny.gov/ insurance/lifeindx.htm (last visited April 29, 2014). There is a formal filing process for fulfilling this statutory obligation called the State Electronic Rate and Forms Filing (“SERFF”) system.
The NYSID formally approved Phoenix’s PAUL IIIA and PAUL IIIB/C policy forms in March 2006. See O’Connell Decl. Ex. D. However, the agency did not comment on the COI rates for these policies, as they vary by insured and had not yet been determined.
While there are a few specific types of life insurance policies whose premium rates must be submitted for formal NY-SID approval, this process does not apply to universal life insurance policy COI rates. See New York State Department of Financial Services, “General SERFF Guidelines for Form Filings,” http://www. dfs.ny.gov/insurance/serflife.htm (last visited April 29, 2014). Accordingly, Phoenix did not formally submit the 2011 COI Rate Adjustment for approval via the SERFF system.
Phoenix merely informally notified the NYSID about the 2011 COI Rate Adjustment as part of a good faith effort to settle the issues surrounding the 2010 COI Rate Adjustment. The NYSID agent’s brief email stating that it had “no objection” to the 2011 COI Rate Adjustment (which the agency was not statutorily required to approve) is not equivalent to the agency engaging in a thorough, public, and legally-required evaluation of a policy form or premium rate. See, e.g., In re Empire Blue Cross and Blue Shield Customer Litig., 164 Misc.2d 350, 622 N.Y.S.2d 843, 845 (N.Y.Sup.Ct.1994). Thus, the filed rate doctrine does not bar Fleisher’s breach of contract claim.
The long and the short of the matter is that no regulator has ever issued any sort of final determination about whether, under the terms of the policies, Phoenix may consider Policy Value when adjusting the COI rate in Policy Months after the initial *468COI rate is set on the Policy Date. It falls to this Court to decide the matter — or to decide whether a jury must determine it.
III. Phoenix Did Not Rely on Impermissible Factors in Adjusting COI Rates.
On behalf of the 2011 Class, Plaintiff Fleisher moves for summary judgment on liability for the breach of contract claim; he contends that there is no genuine issue of material fact as to whether Phoenix breached the class members’ insurance contracts by imposing the 2011 COI Rate Adjustment. Defendant Phoenix cross moves for summary judgment, arguing that it did nothing improper.
To establish a breach of contract under New York law, a plaintiff must show: (1) the existence of an agreement, (2) adequate performance of the contract by the plaintiff, (3) breach of contract by the defendant, and (4) damages. See Stadt v. Fox News Network LLC, 719 F.Supp.2d 312, 318 (S.D.N.Y.2010).
Under New York law, “The construction of an insurance contract is ordinarily a matter of law to be determined by the court.” U.S. Underwriters Ins. Co. v. Affordable Hous. Found., Inc., 256 F.Supp.2d 176, 181 (S.D.N.Y.2003). A written contract must be interpreted according to the parties’ intent, which is “derived from the plain meaning of the language employed” in a contract, In re Lehman Bros. Inc., 478 B.R. 570, 585-86 (S.D.N.Y.2012), when it is “read as a whole.” WWW Assocs., Inc. v. Giancontieri, 77 N.Y.2d 157, 162, 565 N.Y.S.2d 440, 566 N.E.2d 639 (N.Y.1990). Divining the parties’ intent requires a court to “give full meaning and effect to all of [the contract’s] provisions.” Katel Ltd. Liab. Co. v. AT & T Corp., 607 F.3d 60, 64 (2d Cir.2010) (quotation marks omitted).
In a dispute over the meaning of a contract, the threshold question is whether the contract terms are ambiguous. See Krumme v. WestPoint Stevens Inc., 238 F.3d 133, 138 (2d Cir.2000). “Whether or not a writing is ambiguous is a question of law to be resolved by the courts.” WWW Assocs., 77 N.Y.2d at 162, 565 N.Y.S.2d 440, 566 N.E.2d 639.
“A contract is unambiguous where the contract’s terms have ‘a definite and precise meaning, as to which there is no reasonable basis for a difference of opinion.’ ” In re Lehman Bros. Inc., 478 B.R. 570, 586 (S.D.N.Y.2012) (quoting Lockheed Martin Corp. v. Retail Holdings, N.V., 639 F.3d 63, 69 (2d Cir.2011)). “Conversely, an ambiguity does exist in an insurance policy where a term or terms might be susceptible to two or more reasonable interpretations.” D.C. USA Operating Co., LLC v. Indian Harbor Ins. Co., No. 07 Civ. 116(CM), 2007 WL 945016, at *7 (S.D.N.Y. Mar. 27, 2007). “If the court finds that the contract is unambiguous, the court should assign the plain and ordinary meaning to each term and interpret the contract without the aid of extrinsic evidence.” Id.
Where an insurance contract is ambiguous, New York law recognizes a well-established contra proferentem rule, pursuant to which any ambiguity in an insurance policy must be construed against the insurer. See Breed v. Ins. Co. of N. Am., 46 N.Y.2d 351, 353, 413 N.Y.S.2d 352, 385 N.E.2d 1280 (N.Y.1978); Hartol Prods. Corp. v. Prudential Ins. Co., 290 N.Y. 44, 49, 47 N.E.2d 687 (N.Y.1943). “[A] contract of insurance, drawn by the insurer, must be read through the eyes of the average man on the street or the average housewife who purchases it.” Lacks v. Fidelity & Cas. Co. of New York, 306 N.Y. 357, 363, 118 N.E.2d 555 (N.Y.1954). This rule is “based on the fact that it is the *469insurance company that drafted the policy, and therefore the insurance company that is responsible for any ambiguities therein.” Checkrite Ltd., Inc. v. Illinois Nat. Ins. Co., 95 F.Supp.2d 180, 189 (S.D.N.Y.2000).
As the Second Circuit stated in Haber v. St. Paul Guardian Insurance Co., 137 F.3d 691 (2d Cir.1998), “where a policy of insurance is so framed as to leave room for two constructions, the words used should be interpreted most strongly against the insurer.” In order for its proposed construction to prevail, a defendant insurer “bears the heavy burden of demonstrating that it would be unreasonable for the average man reading the policy to construe it as the insured does and that [the insurer’s] interpretation of the insurance policy provisions is the only construction that fairly could be placed on the policy.” Id. at 698 (emphasis added); see also Vargas v. Ins. Co. of N. Am., 651 F.2d 838, 840 (2d Cir.1981); Filor, Bullard & Smyth v. Ins. Co. of N. Am., 605 F.2d 598, 602 (2d Cir.1978); Pan Am. World Airways, Inc. v. Aetna Cas. & Sur. Co., 505 F.2d 989, 1000 (2d Cir.1974); Sincoff v. Liberty Mut. Fire Ins. Co., 11 N.Y.2d 386, 390, 230 N.Y.S.2d 13, 183 N.E.2d 899 (N.Y.1962); Lacks, 306 NY at 365, 118 N.E.2d 555; Hartol, 290 N.Y. at 49, 47 N.E.2d 687.
Thus, a New York court interpreting an insurance contract whose meaning is disputed must answer a simple question: “is the insurer’s interpretation of the contract the only reasonable and fair construction as a matter of law?” Vargas, 651 F.2d at 840. If not, the contract must be construed against the insurer. See id. at 842.
' For our purposes, the question is whether Phoenix has established that it is unreasonable for the Plaintiff policyholders to construe the PAUL policy as precluding Phoenix from taking Policy Values into account when calculating the COI rate, and instead shown that Phoenix’s construction (which permits the use of Policy Values) is the “only construction” that could fairly be placed on the policy. Haber, 137 F.3d at 698.
A. Paragraph A Defines Only the Initial COI Rate as of the Policy Date.
As explained above, the policy sets forth how Phoenix must calculate the initial COI rate used on the “Policy Date” — the date on which the first Monthly Deduction must be paid:
The rates for the Cost of Insurance Charge as of the Policy Date are based on the sex, if applicable, Age, Risk Classification, Basic Face Amount, Supplemental Face Amount, Net Amount at Risk, and duration that the coverage has been in force for the insured.
Compl. Ex. B at 12. This paragraph deals solely with the factors used to compute the initial COI rate, off which adjustments may be made in the future.
Paragraph A specifically mentions NAR as a factor that will be used to calculate the initial COI rate. Since the NAR is a function of, inter alia, the Policy Value (NAR equals Total Face Amount less Policy Value), the Policy Value is obviously used to set the initial COI rate as of the Policy Date. The fact that the words “Policy Value” do not appear in the first paragraph is of no moment, because the phrase “Net Amount at Risk” does appear, and the NAR is defined (on the very same page) as the Total Face Amount less the Policy Value.
B. Paragraph B Circumscribes Which Factors Can Be Used to Calculate an Adjustment to the COI Rate for any Policy Month After the Policy Date.
The next paragraph in the policy (Paragraph B) explains how to calculate the *470Cost of Insurance Charge in any specific “Policy Month” (the period between the Monthly Calculation Dates on which COI charges are assessed):
The Cost of Insurance Charge for a specific Policy Month is the charge for the Net Amount at Risk. The charge for the Net Amount at Risk is an amount equal to the per dollar cost of insurance rate for that month multiplied by the Net Amount at Risk, and such rates will be based on our expectations of future mortality, persistency, investment earnings, expense experience, capital and reserve requirements, and tax assumptions.
Id. In interpreting the meaning of this paragraph, the Court is guided by the Latin maxim expressio unius est exclusio alterius — specific inclusion of one thing is the exclusion of another. See Cornell University v. UAW Local 2300, 942 F.2d 138, 139 (2d Cir.1991). This principle dictates that the formula set out in Paragraph B applies in months after the first month that begins with the Policy Date.
Paragraph A specifically states that its factors are to be used in setting the COI rate on the “Policy Date,” and Paragraph B enumerates the factors to be used in setting the COI rate for a “Policy Month.” Given these two separate COI rate definitions, the contract must be interpreted to mean that only the initial COI rate is calculated in accordance with Paragraph A. Any subsequent COI rate adjustments — COI rates for “Policy Months”— must be calculated in accordance with Paragraph B.
C. The Only Factors That Phoenix May Consider in Adjusting the COI Rates Are the Ones Specifically Enumerated in Paragraph B.
The COI Charge for any given Policy Month is the NAR (Total Face Amount less Policy Value) multiplied by the per dollar COI rate in effect for that month. The per dollar cost of insurance rate for any given month “will be based on [Phoenix’s] expectations of six specifically enumerated factors: “future mortality, per-sistency, investment earnings, expense experience, capital and reserve requirements, and tax assumptions.” Compl. Ex. B at 12.
Fleisher contends that Paragraph B limits Phoenix to considering only the enumerated pricing factors when it adjusts COI rates. He argues that the phrase “based on,” which precedes the list of factors, should be construed as a limiting phrase, not as a “such as” or “including but not limited to;” COI rate adjustments must be “based on” the enumerated factors and only those factors.
Phoenix contends that the enumerated pricing factors are merely illustrative, and that the phrase “based on” should not be construed as limiting. Under Phoenix’s proposed construction of the contract, the company would be allowed to consider both the enumerated factors and other relevant factors in adjusting COI rates— including Policy Values.
Though I personally find Fleisher’s proposed definition to be the more natural reading of “based on,” the policy is ambiguous in that both readings are plausible. In fact, in other cases involving PAUL— like policies, some courts have adopted each reading. Compare Norem v. Lincoln Ben. Life Co., 737 F.3d 1145, 1155 (7th Cir.2013) (holding that “based on” is illustrative), with Yue v. Conseco Life Ins. Co., No. 08 Civ. 1506(AHM), 2011 WL 210943, at *8-10 (C.D.Cal. Jan. 19, 2011) (holding that “based on” is exhaustive), and Jeanes v. Allied Life Ins. Co., 168 F.Supp.2d 958, 974-76 (S.D.Iowa 2001) (same), rev’d in *471part on other grounds 300 F.3d 938 (8th Cir.2002).
Ambiguity alone compels me to reject Phoenix’s interpretation of the phrase “based on.” Applying New York’s doctrine of contra proferentem in the insurance context, if there are two or more reasonable interpretations of a phrase in an insurance contract, the Court must prefer the one advanced by the insured to the one advanced by the insurer. See Lachs, 306 N.Y. at 363, 118 N.E.2d 555; Haber, 137 F.3d at 698. Since there is no conceivable argument that Fleisher’s reading is “unreasonable,” or that Phoenix’s is the only fair interpretation of the contract language, Phoenix loses; the phrase “based on” must be deemed exhaustive.
First, it would be entirely reasonable for an insured to interpret the phrase “based on” as exhaustive. The Macmillan Online Dictionary lists as synonyms for “based on” phrases like “founded upon” and “consist of,” while the Farlex Free Dictionary, another online dictionary, includes “relying on,” “built on,” and “contingent on” as synonyms. See Macmillan Publishers Limited, “Macmillan Dictionary,” http:// www.maemillandictionary.com/us (last visited April 29, 2014); Farlex, Inc., “The Free Dictionary,” http://www.thefree dictionary.com/ (last visited April 29, 2014). At least some, and arguably all, of these synonyms fairly imply that the factors upon which something is “based” are the exclusive factors.
In the insurance policies, the COI rate is “based on” six enumerated factors. In everyday parlance, when you make a calculation “based on” specific factors, you take only those factors into account; the calculation is made “relying on” or “building on” those factors. You calculate a baseball player’s batting average “based on” his number of hits and his number of at bats — nothing more and nothing less. The area of a rectangle is calculated “based on” its width and length, while velocity is calculated “based on” distance and time.
So it would be perfectly plausible — and certainly not unreasonable — for an average insured to conclude, as Fleisher argues, that when Phoenix says it will calculate the COI rate for a particular Policy Month “based on” six specifically enumerated factors, those are the only six factors it will take into account when adjusting the rate.
Other courts have adopted precisely such a construction of “based on,” holding that when a universal life insurance policy states that the policyholder’s COI rate is “based on” certain pricing factors, that list of factors is exhaustive, not illustrative.
In Jeanes v. Allied Life Insurance Co., 168 F.Supp.2d 958 (S.D.Iowa 2001), rev’d in part on other grounds 300 F.3d 938 (8th Cir.2002), the universal life policy at issue stated: “Monthly cost of insurance rates will be determined by us based on our expectations as to future mortality experience.” Like Fleisher, the plaintiff policyholder contended that this provision prohibited the defendant insurance company from considering factors other than mortality experience in increasing COI rates. The district court agreed, stating:
The plain language of the policies appears to provide no basis for increases in the cost of insurance other than increases due to expectations of future mortality. There is no language included in the above portion of the policy indicating that expectations of future mortality are merely one factor upon which a cost of insurance increase may be founded.
Id. at 974. The defendant did not appeal the district court’s holding that the defendant breached this contract. See Jeanes v. *472Allied Life Ins. Co., 300 F.3d 938, 941 (8th Cir.2002).
In Yue v. Conseco Life Insurance Co., No. 08 Civ. 1506(AHM), 2011 WL 210943 (C.D.Cal. Jan. 19, 2011), the universal life policies at issue stated: “Current monthly-cost of insurance rates will be determined by the Company based on its expectation as to future mortality experience.” In setting COI rates, the defendant insurance company used a calculation method that took into account both lapse rates and interest rates in addition to mortality rates. The plaintiff policyholder contended that this method breached the insurance contract, because the contract limited the company to considering only mortality experience in adjusting COI rates. The court agreed and held that the insurance company’s approach was “impermissible because it takes into account factors other than ‘mortality.’ ” Id. at *9. Accordingly, the court granted summary judgment in favor of the plaintiff. In doing so, the court implicitly held that the phrase “based on” was exclusive, and that the enumerated factor was the sole permissible basis for a rate adjustment.
Expressio unius bolsters the Court’s conclusion that the list of enumerated factors is exhaustive. Phoenix’s decision to list six specific factors fairly implies that all other factors are excluded from COI rate adjustment calculations. See Cornell, 942 F.2d at 139. It is not likely that a reasonable insured would read the list of enumerated factors and interpret it as merely informational; he would believe that these are the only six factors that Phoenix will take into account in adjusting COI rates.
One court has apparently adopted Phoenix’s point of view: Phoenix cites Norem v. Lincoln Benefit Life Co., 737 F.3d 1145 (7th Cir.2013), in support of its argument that the phrase “based on” should be construed as merely illustrative rather than limiting.
In Norem, the insurance contract at issue stated: “The cost of insurance rate is based on the insured’s sex, issue age, policy year, and payment class.” Id. at 1147. Like Fleisher, the plaintiff in Norem argued that the defendant insurance company breached the contract when it took into account other factors — including such things as lapse rates, agent commissions, and anticipated death benefit costs — when calculating COI rates.
The Seventh Circuit concluded that “based on” plainly meant that the listed factors were merely illustrative rather than exhaustive. It cited several dictionary definitions of “base,” including: (1) “a main ingredient,” (2) “a supporting or carrying ingredient,” and (3) “the fundamental part of something.” Id. at 1149 (quoting Merriam-Webster’s Collegiate Dictionary 101 (11th ed.2007)).5 The court reasoned that these definitions did not imply exclusivity. It also determined that the common understanding of “based on” was not limiting, stating: “[N]o one would suppose that a cake recipe ‘based on’ flour, sugar, and eggs must be limited only to those ingredients.” Id. at 1150. Accordingly, the court held that “based on” did not prohibit the insurance company from considering factors beyond those listed in setting COI rates. It stated that “the provision is most reasonably read as a description of those components of the COI rate relevant to an individual insured.” Id. at 1152 (emphasis in original).
It seems to me that the Seventh Circuit has unnecessarily complicated a simple issue. Though there are literally dozens of *473dictionary definitions of “base” for use in different literal and figurative contexts, the phrase “based on” is commonly understood to mean that something is created in reliance on identified factors.
Moreover, the Norem analogy to a cake recipe is inaccurate. A recipe is a list of ingredients that are combined to create something. In the cookbooks I read, recipes are exhaustive lists of all the ingredients needed to bake a cake, or fricassee a chicken, or roast a saddle of mutton. See, e.g., Irma S. Rombauer & Marion Rombauer Becker, Joy of Cooking (Bobbs-Merrill Co. 1975) (1931). Highly experienced chefs might be able to play with recipes, but the average home cook (the person analogous to the average insured under New York law) follows them slavishly, without adding other, undisclosed ingredients. The cakes they bake are “based on” the ingredients listed in the recipe — they include those ingredients and none other.
The insurance policies at issue in both Norem and in this case are similar. Both identify the COI rate as one of the two factors that, taken together, fix the amount that will be charged for universal life coverage, and then go on to explain that changes in the COI rate will be calculated in terms of specific factors on which the rate is “based.” There are four such factors in the Norem policy, see 737 F.3d at 1147; in this case, the policy identifies six. There is nothing in either policy to suggest that the listed factors are merely a starting point for the rate calculation, and that the insurance company is free to add a dollop of Undisclosed Factor A and a dash of Undisclosed Factor B in order to “season” the COI rate to its liking. If we must resort to cake analogies, then I prefer this one: the listed factors are the ingredients that, taken together, make up the recipe for calculating an adjusted COI rate.
Additionally, Norem can — indeed, must — be distinguished because it involved insurance policies construed under Illinois law. I know nothing about Illinois insurance law, but I know a great deal about New York’s views on the interpretation of insurance contracts. I cannot fathom a scenario in which a New York court would construe the phrase “based on” as precluding an average insured individual from understanding the phrase to be exhaustive. See Lachs, 306 N.Y. at 363, 118 N.E.2d 555; Haber, 137 F.3d at 698. Interestingly, the Seventh Circuit thought that an insurer was free to list some but not all the factors it would take into account in calculating COI rates because the four that the insurer chose to list would be “relevant to an individual insured.” Norem, 737 F.3d at 1152. Perhaps they would be the most relevant factors, or the most easily understood; but that does not mean that the use of other factors would be “irrelevant” or that an average insured would understand that it was fair game for the insurer to use undisclosed factors. A New York policy would have to say “including but not limited to” or “such as” in order to alert an average insured that factors in addition to those specifically listed could be used to adjust the COI rate.
So I, like the court in deanes, would conclude that “based on” unambiguously precludes Phoenix from considering factors beyond those six factors enumerated in Paragraph B when adjusting COI rates. But as distinguished courts elsewhere have reached two different conclusions on that score, there is at the very least ambiguity. That said, there is no way that one could find that “illustrative” is the “only construction that fairly could be placed on” the phrase “based on.” Haber, 137 F.3d at 698 (emphasis added). That ambiguity must be construed against the insurance *474company that drafted the policy. See id.; Breed, 46 N.Y.2d at 353, 413 N.Y.S.2d 352, 385 N.E.2d 1280. Since the Norem court’s construction — which would give the company carte blanche to adjust COI rates using factors of which the insured has no notice — is also Phoenix’s construction, it must be rejected as a matter of New York law.
So when adjusting the COI rate — that is, in departing from the initial COI rate— Phoenix may take into account its expectations about the factors enumerated in Paragraph B, and none other. Those factors are: “future mortality, persistency, investment earnings, expense experience, capital and reserve requirements, and tax assumptions.” Compl. Ex. B at 12. That list is exhaustive.
In so concluding, I do not suggest that including other “ingredients” in the “recipe” for new COI rates would not make economic. Rather, I conclude that, under New York law, the PAUL policies as drafted limit Phoenix to taking the six enumerated factors into account when adjusting COI rates. This case does not raise issues of sound economics, but of contract construction. Phoenix, the drafter of the policies, had the ability to alert consumers to the fact that its list of “ingredients” for adjusting COI rates was not exhaustive, but merely illustrative; in this state, at least, its failure to do so will not be corrected simply because some other construction of the contract makes more economic sense.
D. Because the Enumerated Factors “Mortality” and “Persistency” Cannot Reasonably Be Interpreted to Incorporate Policy Values, Phoenix Cannot Rely On Policy Value as a Component of Either Factor.
Phoenix argues that Policy Values influence its “expectations” about three of the six enumerated factors that the company may use to adjust COI rates, and so can be used to calculate a COI rate adjustment. Those three factors are: “mortality,” “persistency,” and “investment earnings.”
None of these three factors is defined in the PAUL policy. But there was really no need to define them. All three terms are so ubiquitous in the insurance industry as to have common and settled meanings. And as to two of those three factors— mortality and persistency — Phoenix’s argument that Policy Values have something to do with them is easily rejected.
According to the Dictionary of Insurance Terms, “mortality” simply means “frequency of death.” Rubin, Dictionary of Insurance Terms 318 (4th ed.2000). The “mortality rate” is the “relationship of the frequency of deaths of individual members of a group to the entire group membership over a particular period of time,” and a “mortality table” is a “chart showing rate of death at each age in terms of the number of deaths per thousand.” Id. at 318-19. Mortality tables are used by actuaries to calculate insurance rates. It will come as no surprise to learn that the data in mortality tables is not static; “mortality adjustments” are changes made to a mortality table “to reflect changing levels of mortality due to advancement in medicine, geriatrics, and sanitation.” Id. at 318.
In short, “mortality,” as used in the insurance business, has to do with predicting, aetuarially, how likely it is that a person will die. Mortality is not a function of Policy Values; it has nothing whatever to do with Policy Values, and no average insured would conclude that Phoenix’s “expectations of future mortality” would be affected by whether his Policy Value was high (because he uses the savings feature *475of the policy and pays more than the minimum monthly premium) or low (because he pays only enough to cover the cost of his insurance, and provides Phoenix with no additional funds that it can invest).
“Persistency” is also a common term in the insurance industry; the same insurance dictionary defines it as the “percentage of life insurance or other insurance policies remaining in force; percentage of policies that have not lapsed.” Id. at 377 (emphasis in original). The industry even recognizes a term “persistency bonuses,” as “financial incentives credited to the policy to encourage the policy-owner to keep the policy in force.” Id. Again, this definition of this term does not subsume any aspect of Policy Value — which is calculated, according to the PAUL policy, by taking the Policy Value (with any interest) for the day prior to the day on which the calculation is made, adding any Net Premiums paid and subtracting any withdrawals, including (on the appropriate day) the Monthly Deduction. As so defined, Policy Value has nothing whatever to do with lapse rates.
Furthermore, were some other understanding of the term “persistency” plausible — for example, if Phoenix were correct that the term “persistency” could mean either “policy persistency” or “premium persistency” — Phoenix’s construction would not prevail, under New York’s contra proferentem rule for insurance contracts. The ambiguity would be construed against Phoenix. See Breed, 46 N.Y.2d at 353, 413 N.Y.S.2d 352, 385 N.E.2d 1280; Haber, 137 F.3d at 698.
There is no reason to believe — and Phoenix certainly has offered no evidence tending to suggest — that as used in the PAUL policies, these common insurance terms do not bear the meanings that are ordinarily ascribed to them in the insurance industry. Indeed, the very fact that “mortality” and “persistency” are not otherwise defined in the policy strongly suggests that they are intended to be understood as insurance people commonly use them — and in the way that a policyholder can look up, read, and understand.
E. Policy Values are a Logical Component of Phoenix’s “Expectations of ... Investment Earnings.”
Finally, Phoenix argues that Policy Values influence its “expectations of ... investment earnings” because Policy Values affect the amount of money that Phoenix has available to invest.
The term “investment earnings” is anything but ambiguous. It is defined in the insurance dictionary as “investment income.” Rubin, Dictionary of Insurance Terms 258 (4th ed.2000). The dictionary goes on to explain:
Insurance companies invest part of their premiums that are not immediately needed for claims and administrative expenses. These earnings are critical to an insurance company. A property and casualty company depends on investment earnings to balance underwriting losses. A life company depends on the investment earnings to help build policy cash values.
Id. This is a perfectly straightforward explanation of an easily understood term. “Earnings” equates with “income” — it is the amount of money that the insurance company receives from the investments it makes. “Expectations of ... investment earnings” means the amount of money that Phoenix expects (believes, estimates, predicts) it will make on its investments in the future (since expectancy necessarily refers to the future). In other words, the term bears its plain meaning.
*476Phoenix does not disagree with the Court’s analysis. On the contrary, so obvious is the plain meaning of the phrase “expectations of ... investment earnings” that Phoenix does not even bother to offer a definition of that policy language. It simply makes arguments predicated on the Court’s acceptance of the plain meaning of the words used in the policy. See Def. Opp. at 14-15.
Attempting to create ambiguity where there is none, Fleisher contends that “investment earnings” as used in the insurance industry refers to the “spread” Phoenix was “earning on [its] investments, i.e., the difference between the percent earned from the percent paid to policyholders (the ‘credited interest rate’).” Lewis Decl. Ex. 1 at 45 (Report of Larry N. Stern dated Sept. 16, 2013). This convoluted definition finds no support in the industry dictionary, and there is no evidence that it would occur to an average insured reading the policy (it certainly did not occur to the Court when I read the policy). Indeed, by using “percent earned” instead of “amount earned” in his definition, Fleisher’s expert effectively defines “investment earnings” as “net rate of return.” This makes no sense at all; while “earnings” and “rate of return” are functions of one another, they are not the same thing.
I thus cannot conclude that there is any ambiguity about the meaning of the phrase “expectations of ... investment earnings,” or that an average insured would understand the phrase to mean anything other than what it plainly says. There is no legal basis to construe the phrase against Phoenix, for it would indeed be unreasonable for an average insured to construe the phrase in any manner other than in aceor-dance with the plain meaning of the words used in the policy.
So where does that take us?
Both sides’ experts testified that insurance companies estimate their expected investment earnings by making predictions about two factors — oiie of which is how much money the company is going to have available to invest.6 See French Decl. dated Sept. 16, 2013 at ¶¶ 35, 43, 49; French Decl. dated Oct. 23, 2013 at ¶ 13; Stern Dep. dated Nov. 22, 2013 at 67:5-69:19; Lewis Decl. Ex. 2 at ¶ 5 (Rebuttal Report of Larry N. Stern dated Oct. 23, 2013). Phoenix argues that it can take Policy Values7 into account when calculating its “expectation of ... investment earnings,” because Policy Values impact the amount of money the company has available for investment. The argument goes like this: the more money a PAUL policyholder places in his savings account, and the less he elects to deduct from that account to cover his minimum monthly premiums — which is to say, the higher his Policy Value — the more money Phoenix will have available to invest. And the more Phoenix has available to invest, the higher its expected “investment earnings” for any given rate of return. See French Decl. dated Sept. 16, 2013, supra.
Phoenix’s argument is not only entirely logical; it is the sort of statement that a teenager of a certain era would have greeted with a “well, duh!” Insurance companies extrapolate from prior investment experience when they estimate (predict) what they will earn on investments in the future — -just as they extrapolate from prior experience all actuarial expectancies (including mortality and persistency). It would be utterly unreasonable for an aver*477age insured to expect that Phoenix would ignore any segment of its assets available for investment when calculating its investment earnings expectations. Policy Values — which are nothing more than the excess of premiums paid over the actual cost of providing insurance coverage, plus any past earnings on those excess premiums that remain in the policyholder’s savings account — represent the segment of Phoenix’s income from PAUL policies that the company can invest. As a matter of simple logic, therefore, one would expect Phoenix to take Policy Values into account when predicting future investment earnings.
Admittedly, the PAUL policy does not explain how Phoenix will calculate its expected investment earnings — what steps it will go through to calculate expected earnings, what factors it will use to make that calculation, and how it will weigh those factors. Nor does the policy explicitly state that a policyholder’s choice to maintain a lower Policy Value might negatively affect the company’s “expectations” for future investment earnings, or that making such a choice (a choice that lies entirely within the policyholder’s discretion) could lead to a COI rate increase if Phoenix’s then — operative “expectations” about how much money policyholders will invest turns out to be incorrect (as was apparently the case with at least some PAUL policies).
But the fact that Paragraph B of the PAUL policies does not set out in detail the formula Phoenix will use to predict its future investment earnings neither creates any sort of contractual ambiguity nor militates against the conclusion that it would be unreasonable for an average insured to expect Phoenix to ignore Policy Values when estimating future investment results.
It is the policy language that must be analyzed for ambiguity and construed against Phoenix if there is any such; and as discussed above, the phrase used in the policy (“expectations of ... investment earnings”) is simply not ambiguous — which is to say, it is not susceptible of more than one reasonable interpretation. There is only one plausible meaning of the phrase that appears in the contract: Phoenix’s prediction about how much income it will receive from investments in the future. One might, I suppose, say, “Investment earnings from what?” but, significantly, Fleisher does not; Plaintiff does not argue that the policy language is ambiguous because it does not specify “expectations of ... investment earnings” from any particular product or business segment (PAUL policies only, universal life only, all products).8 He simply' argues that the policy language should be defined in a way that makes absolutely no sense — especially when considered from the perspective of *478an unsophisticated purchaser of insurance — and so creates no ambiguity.
As for not putting the formula for calculating investment earnings expectancy in the policy, I am constrained to note that the policy also does not set out the formulas for calculating Phoenix’s expectations of mortality or persistency, or for making assumptions about taxes — all of which Phoenix can take into account when adjusting the COI rate. The absence of such formulas does not render the words “mortality” or “persistency” or “tax assumptions” ambiguous. I would be hard-pressed to conclude that the PAUL policies were ambiguous for failing to include such formulas in their text; indeed, an average insured would probably have a great deal of difficulty making sense of complicated actuarial formulas. What matters is that the Paragraph B places the average insured on notice that these factors — however Phoenix calculates them— will be used to adjust COI rates, once the initial rate has been set with reference to such factors as age, face amount, and NAR. Of course, an average insured would, and should, expect Phoenix to make such calculations in a manner consistent with what other insurance companies do; but as to that there is no disputed issue of fact — both sides’ experts agree that all insurance companies estimate future investment earnings by extrapolating the amount that will be available for investment from past experience.
And that is what Phoenix did. Phoenix looked at its investment earnings from all insurance products, company-wide, and concluded that those earnings were declining. It then went back to analyze where the problem lay-less money available for investment, or lower rates of return, or both. The answer was that investment earnings from the universal life insurance sector had declined because of low funding ratios (which are a function of Policy Values). And the decrease was attributable to two types of policies: PAUL series IIIA policies with insureds aged 68 or older and face amounts of $ 1 million or more, and PAUL series IIIB/C policies with insureds age 65 or older and face amounts of $1 million or more. See O’Connell Dep. dated June 18, 2013 at 181:7-183:11; French Decl. dated Sept. 16, 2013 at ¶¶43, 49; French Decl. dated Oct. 23, 2013 at ¶ 13.
Phoenix thus determined that its “expectations of ... investment income” for the company overall was reduced because of the low funding ratios for policies within these two groups. Policy Values (or, more to the point, lack of Policy Values) most definitely factored into that analysis; but nothing in the text of the PAUL policies suggests that Policy Values should be overlooked in calculating “expectations of ... investment earnings,” and logic compels the conclusion that they ought not be. Giving the words of the policy their plainest possible meaning, Paragraph B places the average insured on notice that Phoenix can adjust COI rates based on its “expectation” (prediction, best guess) about how much income the company is going to earn on its investments. And since (1) both experts agree that any rational insurer will estimate future investment earnings based on, inter alia, past amounts available for investment, and (2) Policy Values approximate how much money Phoenix will have available for investment from policies that generate Policy Values (including not just PAUL policies but other whole life products as well), there would seem to be nothing left to try — Phoenix can take Policy Values (or funding ratios, which derive from Policy Values) into account when adjusting COI rates.
The real question is whether anything in the policy prohibits Phoenix from doing what it did once its “expectations of ... *479investment earnings” took a nose-dive as a result of less-than-expected earnings from certain PAUL policies — adjusting the COI rate only for those types (“classes”) of policies. But that is not the question of contract interpretation with which we are presently wrestling; that issue is addressed below. See infra at § IV.
In sum, by taking Policy Values into account in its calculation of its “expectations of ... investment earnings,” Phoenix did not rely on impermissible factors; Policy Values are a logical thing to consider when predicting expected investment earnings. Thus, Phoenix did not breach the policy language in Paragraph B, and it is entitled to summary judgment dismissing Plaintiffs’ breach of contract claim on this theory.
IV. An Issue of Fact Remains as to Whether Phoenix Unfairly Discriminated Within a Class of Insureds.
In the alternative, Fleisher contends that the 2011 COI Rate Adjustment “discriminate[d] unfairly within any class of insureds,” in violation of the policy terms. Compl. Ex. B at 12. This allegation cannot be resolved on a motion for summary judgment.
Fleisher contends that Phoenix violated this provision of the contract by imposing the 2011 COI Rate Adjustment only on two groups of PAUL policyholders: PAUL series IIIA policies with insureds age 68 or older and face amounts of $1 million or more, and PAUL series IIIB/C policies with insureds age 65 or older and face amounts of $1 million or more. The question is whether Phoenix “discriminate[d] unfairly within any class of insureds” by subdividing the PAUL policies based on age and face amount, and then imposing the 2011 COI Rate Adjustment on only the two subsets.9
Though the policy does nothing to elucidate the meaning of “discriminate unfairly within any class of insureds,” the Court is guided by a nearly identical provision in New York Insurance Law, which states:
No life insurance company doing business in this state ... shall ... make or permit any unfair discrimination between individuals of the same class and of equal expectation of life, in the amount or payment or return of premiums, or rates charged for policies of life insurance or annuity contracts, or in the dividends or other benefits payable thereon, or in any of the terms and conditions thereof ...
N.Y. Ins. Law. § 4224(a)(1) (emphasis added). Thus, the interpretation of this policy provision should track the statutory definition of what constitutes “unfair discrimination:” it is discrimination that does not have a proper underwriting basis. New York deems discrimination to be “fair” as long as the insurance company’s differential treatment of insureds is appropriate under generally accepted actuarial standards. See Polan v. State of New York Ins. Dept., 3 A.D.3d 30, 768 N.Y.S.2d 441, 444 (2003); Health Ins. Ass’n v. Corcoran, 154 A.D.2d 61, 551 N.Y.S.2d 615, 619 (1990);10 N.Y. Gen. Counsel Op. 2-12-2002 *480(# 4), 2002 WL 33011225, at *2 (Feb. 12, 2002); N.Y. Gen. Counsel Op. No. 12-13-2000 (# 2), 2000 WL 34630175, at *3 (Dec. 13,2000).
The leading New York General Counsel Opinion interpreting § 4224(a)(1) states:
The language of N.Y. Ins. Law § 4224(a)(1) is clear that only UNFAIR discrimination is prohibited. All types of discrimination are not prohibited. All risk classification necessarily involves discrimination, with each person paying according to the risk he or she represents. Classifying and pricing risks is the very essence of insurance ...
An insurer is free to impose any appropriate rules for classifying, selecting, and pricing risks that it believes are required based on sound underwriting practices and in accordance with accepted insurance and actuarial principles, provided such rules are not contrary to law.
Id. (citations omitted). Thus, an insurance company does not unfairly discriminate within a class of insureds “when differential treatment has a proper underwriting basis.” Corcoran, 551 N.Y.S.2d at 619 (emphasis added).
So imposing the 2011 COI Rate Adjustment on the classes of insureds identified will not be deemed “unfair” as long as it can be justified under accepted actuarial principles.
There is a genuine issue of material fact concerning whether what Phoenix did can be justified under accepted actuarial principles. Phoenix’s expert claims that the company’s classifications of policies based on age and face amount met the requirements of the official guidance of the Actuarial Standards Board — the Actuarial Standards of Practice (“ASOP”). See French Decl. dated Sept. 16, 2013 at ¶¶ 14, 92-95. Fleisher’s expert disagrees. See Lewis Decl. Ex. 2 at ¶ 6 (Rebuttal Report of Larry N. Stern dated Oct. 23, 2013).
Intuitively, it seems obvious that age would be an appropriate way to classify insureds in the life insurance context — the older a person is, the higher his risk of death. However, it is not so apparent that face value is an appropriate way to classify insureds; this factor is not clearly tied to life expectancy, and it seems more closely associated with profitability. The parties’ experts disagree about whether face amount and profitability are appropriate classification considerations under accepted actuarial standards. See id. at ¶¶ 3-7; French Decl. dated Sept. 16, 2013 at ¶¶ 14, 94-95.
The text of ASOP 2 (the actuarial principle cited by Phoenix) does not clearly resolve the factual dispute. With respect to “Policy Classes,” the rule states:
Policies will usually be grouped into classes for purposes of determining non-guaranteed charges or benefits. The determination policy may include a definition of the policy classes to be used. If the policy classes have not been defined in the determination policy, the actuary should establish policy classes considering criteria such as the following:
a. the similarity of the policy types;
*481b. the structure of policy factors and nonguaranteed charges or benefits;
c. the similarity of anticipated experience factors;
d. the time period over which the policies were issued; and
e. the underwriting and marketing characteristics of the policies.
Actuarial Standard of Practice No. 2, Non-guaranteed Charges or Benefits for Life Insurance Policies and Annuity Contracts, at § 3.4 (Actuarial Standards Board, March 2013). This language offers only general guidance and does not settle the issue of whether it is appropriate under actuarial standards to classify insureds based on their age and face amounts. Given the experts’ differing opinions regarding the relevant actuarial standards, an issue of fact remains.
Fleisher also argues that Phoenix was required to define any “class of insureds” at the time of policy issuance. Thus, Fle-isher reasons, the groupings Phoenix identified at issuance — the PAUL IIIA and PAUL IIIB/C policy series — constituted the relevant “classes.” Fleisher contends that Phoenix was not permitted to further subdivide the existing policy series classes by age and face amount and to treat each subdivision differently.
The New York General Counsel has recognized that an insurance company “may” define all policies within a certain policy series as a class, although no authority has stated that insurance companies must do so. See N.Y. Gen. Counsel Op. No. 12-13-2000 (#2), 2000 WL 34630175, at *4. I have no idea whether this is something that may only occur upon issuance or whether definitions can be modified after issuance (as Phoenix did by adding age and face amount components to the classes that were identified at issuance). Nothing in the record helps me to determine this issue; it would be very useful to have the views of the Superintendent of the New York Department of Financial Services.
The Court thus denies the parties’ cross motions for summary judgment on this breach of contract claim.
V. An Issue of Fact Remains as to Whether Phoenix Recouped Prior Losses.
Fleisher’s final breach of contract allegation is that the 2011 COI Rate Adjustment “reeoup[ed] prior losses,” in violation of the policy terms. See Compl. Ex. B at 12. There is a disputed issue of fact on this alleged breach as well. The parties’ experts disagree over whether Phoenix’s method for calculating the 2011 COI Rate Adjustment merely restored profitability prospectively using the company’s original profitability goals, or restored profitability retrospectively by compensating for prior losses through the increased COI rates. See French Decl. dated Sept. 16, 2013 at ¶¶ 73-75; Lewis Decl. Ex. 2 at ¶8 (Rebuttal Report of Larry N. Stern dated Oct. 23, 2013). Accordingly, the Court must deny the cross motions for summary judgment on this breach of contract claim as well.
VI. An Issue of Fact Remains as to the Amount of Damages.
There is no basis for summary judgment motions addressed to damages issues when liability issues remain to be decided. All issues relating to damages will be taken up at the trial.
CONCLUSION
For the foregoing reasons, Phoenix’s motion for partial summary judgment in granted in part and denied in part; Fleisher’s motion for partial summary judgment is denied in its entirety. The Clerk of the Court is directed to close out the motions *482at Docket Nos. 184 and 190 and to remove same from the Court’s list of pending motions. The parties should confer and propose a trial schedule.
. These facts are taken from the parties’ Rule 56.1 Statements, the accompanying exhibits, and the insurance policies themselves. They are not in dispute.
. All members of the 2011 Class owned their policies at the time of the alleged breach of contract in 2011. Some class members have since allowed their policies to lapse. See Def. 56.1 Statement Resp. at ¶ 2.
. Net Amount at Risk = (Total Face Amount / 1.0032737) — Policy Value
. The New York State Insurance Department has since been consolidated with another agency and renamed the New York State Department of Financial Services.
. The Seventh Circuit apparently did not look for definitions of or synonyms for the phrase "based on,” which are the words actually used in the policy.
. The other relevant factor is what the rate of return on investment is likely to be in view of anticipated market conditions.
. As a component of funding ratios.
. Not all of Phoenix’s "investment earnings” come from the excess premiums paid on PAUL policies. Phoenix offers many other life insurance products. See The Phoenix Companies, Inc., "Product Prospectuses,” https://phoenixwm.phl.com/public/products/ regulatory/index.jsp (last visited April 29, 2014). The asset base from which Phoenix derives its "investment earnings” includes not only excess premiums paid on PAUL policies (which can logically be approximated by aggregating PAUL Policy Values), but also excess premiums paid on other types of whole life insurance products, and even "unused” premiums (those not needed to fund current claims and expenses) paid on term policies. The language of Paragraph B does not restrict "expectations of ... investment earnings” to earnings from PAUL policies, so an average insured might well understand that predicted earnings from all of those insurance products would be factored into Phoenix’s “expectations of ... investment earnings.” I suspect that Fleisher does not argue that the lack of specificity about earnings "from what” creates an ambiguity because if he were to argue that the phrase referred only to investment earnings from PAUL policies, his suggestion that Policy Values ought not be factored into that calculus would be ridiculous on its face.
. These are, of course, the subsets that "approximate” policies with low funding ratios. However, not every policy in those classes was underfunded; in some cases, the policyholder had paid in substantially more than the minimum Monthly Deduction. Nonetheless, all policies in those classes were subjected to the 2011 COI Rate Adjustment.
. Health Ins. Ass’n v. Corcoran interpreted the parallel statute that applies to accident and health insurance, N.Y. Ins. Law. § 4224(b)(1). See 551 N.Y.S.2d at 619. However, this statute contains language that is nearly identical to § 4224(a)(1):
No insurer doing in this state the business of accident and health insurance ... shall *480... make or permit any unfair discrimination between individuals of the same class in the amount of premiums, policy fees, or rates charged for any policy of accident and health insurance, or in the benefits payable thereon, or in any of the terms or conditions of such policies, or in any other manner whatsoever ...
N.Y. Ins. Law. § 4224(b)(1) (emphasis added). Thus, Corcoran's, analysis regarding what constitutes unfair discrimination is equally applicable in the life insurance context.