IN THE SUPREME COURT OF CALIFORNIA

BLAKELY McHUGH et al.,

Plaintiffs and Appellants,

v.

Protective Life Insurance Company,

Defendant and Respondent.

S259215

Fourth Appellate District, Division One

D072863

San Diego County Superior Court

37-2014-00019212-CU-IC-CTL

August 30, 2021

Counsel:

Winters & Associates, Jack B. Winters, Jr., Georg M. Capielo, Sarah D. Ball; Williams Iagmin and Jon R. Williams for Plaintiffs and Appellants.

Law Offices of Daniel D. Murphy and Daniel D. Murphy for California Advocates for Nursing Home Reform, Inc., as Amicus Curiae on behalf of Plaintiffs and Appellants.

Glick Law Group and Noam Glick for California Retired County Employees Association as Amicus Curiae on behalf of Plaintiffs and Appellants.

Neil Granger, in pro. per., as Amicus Curiae on behalf of Plaintiffs and Appellants.

Grignon Law Firm, Margaret M. Grignon; Maynard Cooper & Gale, C. Andrew Kitchen, Alexandra V. Drury, John C. Neiman, Jr.; Noonan Lance Boyer & Banach and David J. Noonan for Defendant and Respondent.

Alston & Bird and Thomas A. Evans for American Council of Life Insurers as Amicus Curiae on behalf of Defendant and Respondent.

Quinn Emanuel Urquhart & Sullivan and Kathleen M. Sullivan for Chamber of Commerce of the United States of America as Amicus Curiae on behalf of Defendant and Respondent.

Matthew Rodriguez, Acting Attorney General, and Lucy F. Wang, Deputy Attorney General, for Ricardo Lara, Insurance Commissioner, as Amicus Curiae, upon the request of the Supreme Court.


Justice Cuéllar authored the opinion of the Court, in which Chief Justice Cantil-Sakauye and Justices Liu, Kruger, and Groban concurred.

Justice Jenkins filed a concurring opinion, in which Justice Corrigan concurred.

Opinion of the Court by Cuéllar, J.

Millions of California consumers manage financial risks for their families by purchasing life insurance. Through these policies, Californians ensure that their families and other designated beneficiaries are protected by a financial safety net — and are able to plan for contingencies — in the event of the policy owners’ untimely death. But there’s a cost: In exchange for continuing coverage, consumers pay regular premiums to their insurers. If consumers fail to do so, insurers have the right to end the policies.

In 2012, the Legislature created certain protections to shield consumers from losing life insurance coverage because of a missed premium payment. Codified in sections 10113.71 and 10113.72 of the Insurance Code,[1] these protections went into effect on January 1, 2013. Soon thereafter, the defendant terminated one of the life insurance policies at issue in this case because the policy owner had failed to make a payment. Plaintiffs claim that the defendant had no right to terminate these policies without complying with the newly codified statutory protections against termination. The Court of Appeal reasoned that sections 10113.71 and 10113.72 did not apply because they appeared to affect only policies issued or delivered after the sections’ January 1, 2013 effective date, and the policy at issue here predated the sections. In reaching this construction of the statutes, the court cited — among other considerations — its deference to Department of Insurance (DOI) staff correspondence and electronic instructions for policy forms.

We conclude that sections 10113.71 and 10113.72 apply to all life insurance policies in force when these two sections went into effect, regardless of when the policies were originally issued. This interpretation fits the provisions’ language, legislative history, and uniform notice scheme, and it protects policy owners — including elderly, hospitalized, or incapacitated ones who may be particularly vulnerable to missing a premium payment — from losing coverage, consistent with the provisions’ purpose. This interpretation does not depend on extending deference to DOI staff correspondence or electronic instructions, neither of which represent the agency’s official interpretation of sections 10113.71 and 10113.72 nor otherwise reflect the agency’s carefully considered, long-standing, and consistent interpretive viewpoint on the sections. Accordingly, we reverse the judgment of the Court of Appeal and remand for proceedings consistent with this opinion.

I.

In March 2005, Chase Life Insurance Company, the predecessor in interest to defendant Protective Life Insurance Company (Protective Life), issued a $1 million term life insurance policy to William McHugh. The policy named McHugh’s daughter, Blakely McHugh, as the designated beneficiary and Trysta Henselmeier, Blakely’s mother and McHugh’s successor in interest, as a contingent beneficiary.

The policy was for a 60-year term, and it set out a schedule of annual premiums to keep the policy in force. For the first 10 years of the policy, the insurance policy set the annual premium at $310; after that, the premium steadily increased each year. The policy included a provision for a 31-day grace period before the policy could be terminated for the failure to pay the premium.

McHugh paid all the yearly premiums through January 2012. That meant his policy was, by its terms, “in force” until February 9, 2013, 31 days after the January 9, 2013 due date for that year’s payment. On December 20, 2012, Protective Life sent McHugh a letter reminding him of the January 9 deadline and that nonpayment by February 9 would cause his policy to lapse or terminate. McHugh failed to pay the premium by the due date. Protective Life sent him a second letter on January 29, which stated that it had not received his premium payment for the year and warned that his policy would lapse if he did not make the payment by February 9, the end of the grace period. McHugh again failed to make the payment, and the policy lapsed. On February 18, Protective Life sent McHugh a letter informing him the grace period had expired, but that he could reinstate the policy if it received his payment by March 12, during his lifetime. McHugh did not pay, and Protective Life formally terminated his policy.

At some point close to when Protective Life sent its last letter, McHugh suffered a serious fall that left him disabled, caused him continuing physical pain, and required surgery. McHugh passed away in June 2013. Henselmeier contacted Protective Life to inquire about the status of McHugh’s policy and whether a claim could be made. Protective Life advised that the policy had been terminated. Thereafter, Henselmeier and Blakely (plaintiffs) sued Protective Life for breach of contract and breach of the implied covenant of good faith and fair dealing. Plaintiffs argued that sections 10113.71 and 10113.72, which came into effect on January 1, 2013, applied to policies issued before this effective date, and that Protective Life failed to comply with the statutes’ requirements before it terminated McHugh’s policy.

In various filings, including its motion for a directed verdict, Protective Life argued the statutes did not apply to policies issued before January 1, 2013. In making this argument, Protective Life relied at times on purported agency interpretations of the statutes. The trial court rejected Protective Life’s argument, concluding that the statutes applied to McHugh’s policy. Ultimately, the jury found for Protective Life. It concluded that: (1) Protective Life and McHugh entered into an insurance contract; (2) McHugh failed to do all, or substantially all, of what the contract required him to do, but he was excused from doing so; (3) all conditions required for Protective Life’s performance occurred and were not excused; (4) Protective Life did something the contract prohibited; but (5) plaintiffs were not harmed by Protective Life’s failure.

Plaintiffs appealed from the special verdict in favor of Protective Life and the denial of plaintiffs’ judgment notwithstanding the verdict motion. What they argued, among other things, is that the trial court erred by declining to decide as a matter of law whether Protective Life had complied with Insurance Code sections 10113.71 and 10113.72, and instead permitting the jury to decide that issue. (McHugh v. Protective Life Ins. (2019) 40 Cal.App.5th 1166, 1171, fn. 4 (McHugh).) Under Code of Civil Procedure section 906, Protective Life requested the Court of Appeal affirm the judgment on the additional ground that Insurance Code sections 10113.71 and 10113.72 do not apply retroactively to McHugh’s policy, and the trial court erred as a matter of law when it ruled otherwise in denying the directed verdict motion. (McHugh, at pp. 1170–1171.) The Court of Appeal affirmed the judgment on this additional ground. (Id. at p. 1171.) In reaching this holding, the court first relied on two sets of DOI documents that it held indicated that sections 10113.71 and 10113.72 applied only to policies issued after January 1, 2013: private correspondence between DOI counsel and insurers, and DOI’s System for Electronic and Form Filing (SERFF) “ ‘Instructions for Complying with [Assembly Bill No.] 1747.’ ” (McHugh, at p. 1172.) It then determined that the statutes’ language supported DOI’s purported interpretation. (Id. at pp. 1175–1177.)

We granted review to resolve whether (1) sections 10113.71 and 10113.72 apply to all life insurance policies in force as of January 1, 2013 — regardless of when those policies had originally been issued — or only to policies that went into effect after this date; and (2) the Court of Appeal properly deferred to DOI guidance in its analysis.

II.

The grace period and notice requirements governing life insurance policies issued before January 1, 2013 depend on whether sections 10113.71 and 10113.72 apply to such policies. To understand the effects of these provisions, we begin by surveying the mechanics of life insurance and the broad legal framework governing such policies.

A.

A life insurance policy “is a contract of indemnity under which, in exchange for the payment of premiums, the insurer promises to pay a sum of money to the designated beneficiary upon the death of the named insured.” (Fairbanks v. Superior Court (2009) 46 Cal.4th 56, 61.) There are two main categories of life insurance. (See Fairbanks v. Farmers New World Life Ins. Co. (2011) 197 Cal.App.4th 544, 547–548 & fn. 3 (Fairbanks); Kaldenbach v. Mutual of Omaha Life Ins. Co. (2009) 178 Cal.App.4th 830, 834.) Cash value life insurance provides “ ‘insurance’ ” in the form of a death benefit for designated beneficiaries upon the policy owner’s death, as well as “ ‘cash value’ ” savings that accumulate and are available during the policy owner’s lifetime. (Kaldenbach, at p. 834.) Term life insurance, which McHugh purchased, provides only a death benefit, and it does so only for a set duration of years. (Fairbanks, at p. 547.) It does not accrue and pay out a cash value. (Estate of Logan (1987) 191 Cal.App.3d 319, 324; see also Logue, The Current Life Insurance Crisis: How the Law Should Respond (2002) 32 Cumb. L.Rev. 1, 17 [“A characteristic of term life insurance is that if the insured fails to renew or cancels his policy, the coverage will cease and any premiums that have been paid (and earned) will not be refunded”].)

Despite the differences between term and cash value life insurance, the importance of one aspect of insurance for the larger public remains relatively constant across policy types: Consumers often find it very difficult and costly to replace a policy, including one that has been cancelled because of a missed premium payment. They may need to spend money so they can purchase a new policy; they may be forced to pay more expensive premiums because they are being insured at an older age and possibly after having developed health conditions; they may need to pay commission charges and similar fees; they may be forced to live with new incontestability or suicide clauses; and they may be barred from accessing the cash value in their new policy for a considerable time. (See, e.g., In re Prudential Ins. Co. of Am. Sales Practice Litig. (3d Cir. 2001) 261 F.3d 355, 359, fn. 2; Mahan v. Charles W. Chan Ins. Agency, Inc. (2017) 14 Cal.App.5th 841, 867, fn. 25; DOI, Life Insurance Guide <http://?www.insurance.ca.?gov/01-?consumers?/105-?type/??95-??guides?/07-life/?life-ins-guide.?cfm>[2] [as of Aug. 30, 2021]; cf. Wilner v. Sunset Life Ins. Co. (2000) 78 Cal.App.4th 952, 965–967.)

In part because of these consequences, the insurance business is a matter of public interest. (Calfarm Ins. Co. v. Deukmejian (1989) 48 Cal.3d 805, 830 (Calfarm); see also 20th Century Ins. Co. v. Superior Court (2001) 90 Cal.App.4th 1247, 1265 & fn. 9.) So insurance contracts are subject to substantial regulation under the state’s police power. (Calfarm, at p. 830.) The state regulates insurance contracts primarily through the Insurance Code. Policies may be required by the code to include certain provisions, and these provisions are deemed to be incorporated into every policy to which they pertain. (California Fair Plan Assn. v. Garnes (2017) 11 Cal.App.5th 1276, 1305, 1309.) The laws in effect at the time of a policy’s issuance generally govern the policy. (See Interinsurance Exchange of the Auto. Club of Southern Calif. v. Ohio Cas. Ins. Co. (1962) 58 Cal.2d 142, 148 (Interinsurance Exchange); 2 Witkin, Summary of Cal. Law (11th ed. 2017) Insurance, § 10, p. 44.) This general rule promotes certainty in the commercial and legal relationship between insurers and insureds. (See Swenson v. File (1970) 3 Cal.3d 389, 394–395.) Subject to certain constitutional guardrails, the state may exercise its police power to enact legislation that affects existing policies. (Calfarm, at pp. 814–815, 829–831.)

At the time Protective Life issued McHugh’s policy, the Insurance Code provided (and continues to provide) that insurers can cancel policies when policy owners fail to pay the premium owed. (§ 484.) The code did not require life insurers to provide notice when cancelling a policy, including for nonpayment of a premium (see Stewart v. Life Ins. Co. of North America (E.D. Cal. 2005) 388 F.Supp.2d 1138, 1142), even though insurers were responsible for providing notice when cancelling other forms of insurance (see, e.g., § 662, subd. (a) [for automobile policies]; 2 Witkin, Summary of Cal. Law, supra, Insurance, § 319, pp. 493–494). But the terms of the policy may create a duty for insurers to provide cancellation notices, or insurers may provide such notices as a general business practice, as Protective Life did. (See 16 Williston on Contracts (4th ed. 2014) § 49:85, pp. 728–731.) Such notice protects policy owners from losing coverage due to their neglect (5 Couch on Insurance (3d ed. 2012) § 76:23, p. 76-52) or enables them to obtain insurance elsewhere before being subject to risk without protection (2 Couch on Insurance (3d ed. 2010) § 32:1, p. 32-7).

Moreover, at the time Protective Life issued McHugh’s policy, the Insurance Code did not require life insurers to provide a grace period before cancelling a life insurance policy for premium nonpayment. But life insurers could, as Protective Life did, provide a grace period as a contractual provision and business practice. (See 5 Couch on Insurance, supra, § 76:47, p. 76-90.)[3] A grace period offers an obvious benefit to policy owners: time to pay a missed premium without an interruption in coverage. (16 Williston on Contracts, supra, § 49:80, p. 699; see Pediatricians, Inc. v. Provident Life & Acc. Ins. Co. (1st Cir. 1992) 965 F.2d 1164, 1168 (Pediatricians).) Grace periods can also provide a business advantage to insurers. Of course, insurers depend on the regular, timely payment of premiums in order to pay death benefits and cover the cost of administering policies. (See New York Life Ins. Co. v. Statham (1876) 93 U.S. 24, 30 [“[I]t must be conceded that promptness of payment is essential in the business of life insurance. All the calculations of the insurance company are based on the hypothesis of prompt payments”]; 16 Williston on Contracts, supra, § 49:75, p. 655.) But grace periods decrease the probability that policy owners will terminate the policy accidentally or because of temporary financial difficulties, and thus increase the likelihood that policy owners will continue the insurance in effect, providing the insurer not just with the premium then due but also future premiums. (16 Williston on Contracts, supra, § 49:80, pp. 699–700; Pickens v. State Farm Mut. Auto. Ins. Co. (S.C. 1965) 144 S.E.2d 68, 71; Pediatricians, at pp. 1168–1169.)

B.

In 2012, the Legislature enacted Assembly Bill No. 1747 (2011–2012 Reg. Sess.), grafting sections 10113.71 and 10113.72 onto the Insurance Code. (Stats. 2012, ch. 315, §§ 1, 2.) These provisions, enacted after Protective Life issued McHugh’s policy but before it cancelled the policy, changed the grace period and notice requirements for life insurance policies in California.

Section 10113.71 established a 60-day grace period after a missed premium. Subdivision (a) states: “Each life insurance policy issued or delivered in this state shall contain a provision for a grace period of not less than 60 days from the premium due date. The 60-day grace period shall not run concurrently with the period of paid coverage. The provision shall provide that the policy shall remain in force during the grace period.”[4] (§ 10113.71, subd. (a).) The section also requires insurers to notify policy owners, as well as persons designated by the policy owners to receive notice (under section 10113.72), at least 30 days before terminating a policy due to a payment lapse. (§ 10113.71, subd. (b)(1).) Subdivision (b)(1) states: “A notice of pending lapse and termination of a life insurance policy shall not be effective unless mailed by the insurer to the named policy owner, a designee named pursuant to Section 10113.72 for an individual life insurance policy, and a known assignee or other person having an interest in the individual life insurance policy, at least 30 days prior to the effective date of termination if termination is for nonpayment of premium.” (Ibid.) And subdivision (b)(3) mandates that the “[n]otice shall be given to the policy owner and to the designee by first-class United States mail within 30 days after a premium is due and unpaid.” (§ 10113.71, subd. (b)(3).)

Section 10113.72 requires life insurance policies to grant policy owners the right to designate at least one other person to receive a notice of an overdue premium and impending lapse or termination of the policy. In subdivision (a), it provides that “[a]n individual life insurance policy shall not be issued or delivered in this state until the applicant has been given the right to designate at least one person, in addition to the applicant, to receive notice of lapse or termination of a policy for nonpayment of premium.” (§ 10113.72, subd. (a).) It also explains that “the insurer shall provide each applicant with a form to make the designation,” and that this form must provide the applicant with the opportunity “to submit the name, address, and telephone number of at least one person, in addition to the applicant, who is to receive notice of lapse or termination of the policy for nonpayment of premium.” (Ibid.) Subdivision (b) provides that insurers “shall notify the policy owner annually of the right to change the written designation or designate one or more persons,” and that policy owners may elect to change the designation more often if they choose to do so. (§ 10113.72, subd. (b).) Finally, subdivision (c) prevents an insurer from ending a policy for an unpaid premium without giving policy owners at least 30 days’ notice. It mandates that no policy “shall lapse or be terminated” for an unpaid premium “unless the insurer, at least 30 days prior to the effective date of the lapse or termination, gives notice to the policy owner and to the person or persons designated pursuant to subdivision (a) . . . .” (§ 10113.72, subd. (c).)

Both of these sections went into effect on January 1, 2013.

III.

Protective Life maintains that the two new statutes have no effect on insurance contracts issued or delivered before January 1, 2013. This case turns on whether the appellate court was right to embrace that conclusion. To resolve this question, we begin by reviewing de novo the Court of Appeal’s interpretation of sections 10113.71 and 10113.72. (Kirby v. Immoos Fire Protection, Inc. (2012) 53 Cal.4th 1244, 1250.) As with any question of statutory construction, our core task here is to determine and give effect to the Legislature’s underlying purpose in enacting the statutes at issue. (California Teachers Assn. v. San Diego Community College Dist. (1981) 28 Cal.3d 692, 698; Calatayud v. State of California (1998) 18 Cal.4th 1057, 1065; Goodman v. Lozano (2010) 47 Cal.4th 1327, 1332.) We first consider the words of the statutes, as statutory language is generally the most reliable indicator of legislation’s intended purpose. (In re H.W. (2019) 6 Cal.5th 1068, 1073 (H.W.).) We consider the ordinary meaning of the relevant terms, related provisions, terms used in other parts of the statute, and the structure of the statutory scheme. (Larkin v. Workers’ Comp. Appeals Bd. (2015) 62 Cal.4th 152, 157.) If the relevant statutory language is ambiguous, we look to appropriate extrinsic sources, including the legislative history, for further insights. (H.W., at p. 1073.) We also extend some deference to DOI’s interpretations of the Insurance Code, to the extent that those interpretations are embodied in quasi-legislative regulations or constitute long-standing, consistent, and contemporaneous interpretations. (See, e.g., Yamaha Corp. of America v. State Bd. of Equalization (1998) 19 Cal.4th 1, 12–13 (Yamaha); Farmers Ins. Exch. v. Superior Court (2006) 137 Cal.App.4th 842, 859.)

A.

The Court of Appeal framed its analysis by relying on a general interpretive principle: the rebuttable presumption that a statute does not operate retroactively, and instead operates on a prospective basis only. (Myers v. Philip Morris Companies, Inc. (2002) 28 Cal.4th 828, 844 (Myers).) The Court of Appeal presupposed that applying sections 10113.71 and 10113.72 to McHugh’s policy was, in fact, retroactive for purposes of applying the presumption against retroactivity. Its only brush with this threshold question came toward the end of its analysis, where — seeking to tether its statutory analysis to constitutional principles — the appellate court explained that well-settled law dictated that “McHugh’s policy is governed by the regulations in effect when it was issued in 2005, and the subsequently enacted sections 10113.71 and 10113.72 are not incorporated into the policy.” (McHugh, supra, 40 Cal.App.5th at p. 1177.) The court’s assumption appears to have been that applying the sections here was retroactive because they imposed new grace period and notice obligations nowhere found in the 2005 regulations, which governed already-existing policies and lacked these grace period and notice requirements.

Applying the presumption against retroactivity, the court emphasized what it took to be the principle’s requirements — that “ ‘a statute may be applied retroactively only if it contains express language of retroactivity or if other sources provide a clear and unavoidable implication that the Legislature intended retroactive application.’ ” (McHugh, supra, 40 Cal.App.5th at p. 1174, quoting Myers, supra, 28 Cal.4th at p. 844, italics added by Myers.)  It then concluded that there was no basis to rebut the presumption. (McHugh, at p. 1174.)

Protective Life reiterates and expands upon the Court of Appeal’s analysis. Regarding whether we face a question of “retroactivity”: Protective Life invokes Interinsurance Exchange, supra, 58 Cal.2d at page 148 to urge that reading the grace period and notice protections into McHugh’s policy would be a retroactive application, since insurance policies are putatively governed by the law in effect when they are issued. Because no statutory law pre-2013 created grace period and notice requirements related to missed life insurance premiums, what Protective Life is arguing, then, is that the absence of regulation should be read into McHugh’s policy, and that holding otherwise would be to rewrite the policy.

Plaintiffs argue, however, that this case involves an entirely prospective statutory application because they seek no more than application of the grace period and notice requirements to missed premium payments occurring after sections 10113.71 and 10113.72 went into effect. In other words, plaintiffs argue that this case merely concerns Protective Life’s postenactment conduct with respect to policies in force as of January 1, 2013. They contend that this statutory application is not “retroactive” at all because it does not materially alter the contractual agreement memorialized in McHugh’s policy, and similarly situated policies, in a way that unfairly undermines the parties’ reliance interests. In the alternative, plaintiffs argue that this statutory application does not implicate principles of retroactivity because any retroactive effect here is minimal and will not substantially impair any vested contractual rights.

We find support for both of plaintiffs’ arguments. Our previous decisions buttress plaintiffs’ understanding of the presumption against retroactivity, and whether it even applies. We conclude that under either alternative theory, the interpretive presumption does not vindicate Protective Life’s position.

1.

The presumption against retroactivity is, at core, a canon to facilitate interpretation rather than an inexorable command. (People v. Superior Ct. (Lara) (2018) 4 Cal.5th 299, 307; see Medical Board v. Superior Court (2001) 88 Cal.App.4th 1001, 1013.) In cases where the presumption is potentially implicated, we must consider both its overall role — helping guide us in our core endeavor of determining and giving full effect to a statute’s underlying purpose — and the specific premise for applying it in that particular case. (People v. Garcia (2016) 62 Cal.4th 1116, 1124; see Tapia v. Superior Ct. (1991) 53 Cal.3d 282, 301 (Tapia); Fox v. Alexis (1985) 38 Cal.3d 621, 629.) We apply the presumption in the absence of explicit legislative indications of retroactivity, doing so based on the fundamental fairness considerations raised by “ ‘imposing new burdens on persons after the fact.’ ” (McClung v. Employment Development Dept. (2004) 34 Cal.4th 467, 475 (McClung); see id. at p. 476 [“ ‘Requiring clear intent assures that [the legislative body] itself has affirmatively considered the potential unfairness of retroactive application and determined that it is an acceptable price to pay for the countervailing benefits’ ”].)

These considerations influence the threshold question courts must answer before even applying the presumption against retroactivity: Is the statutory change in question “ ‘retroactive’ ” or “ ‘prospective’ ”? (Californians for Disability Rights v. Mervyn’s, LLC (2006) 39 Cal.4th 223, 230 (Mervyn’s).) In theory, these concepts are simple; in practice, they often prove more elusive. To distinguish between “retroactive” and “prospective” statutory applications, we explained the need to deploy the following standard: “We must consider ‘ “ ‘the nature and extent of the change in the law and the degree of connection between the operation of the new rule and a relevant past event’ ” ’ ”; “ ‘ “ ‘familiar considerations of fair notice, reasonable reliance, and settled expectations offer sound guidance.’ ” ’ ” (Quarry v. Doe I (2012) 53 Cal.4th 945, 955 (Quarry).) In keeping with these principles, we have generally explained that a new law operates “retroactively” when it changes “ ‘ “the legal consequences of past conduct by imposing new or different liabilities based upon such conduct.” ’ ” (Mervyn’s, at p. 231.) We have asked whether the new law “ ‘ “substantially affect[s] existing rights and obligations.” ’ ” (Ibid., italics added.) Plaintiffs advance this understanding of retroactivity. Yet some of our cases can potentially be read to articulate a broader definition of retroactivity, which Protective Life argues we should apply. In Myers, for example, we stated that a statute operates retroactively when it “ ‘ “affects rights, obligations, acts, transactions and conditions which are performed or exist prior to the adoption of the statute.” ’ ” (Myers, supra, 28 Cal.4th at p. 839, quoting Aetna Cas. & Sur. Co. v. Ind. Acc. Com. (1947) 30 Cal.2d 388, 391 (Aetna).) This broader definition seems to embrace any conceivable statutory impact on the terms of an existing contract — including an insurance contract — as Protective Life urges.

The two differing conceptions of “retroactivity” at play underscore why the term and its antonym “are not always easy to apply to a given statute.” (Quarry, supra, 53 Cal.4th at p. 955; see 2 Sutherland, Statutes and Statutory Construction (7th ed. 2009) § 41:1, p. 385.) Established precedent nonetheless helps us clarify any potential ambiguity and strongly favors plaintiffs’ retroactivity approach.

Consistent with the presumption’s underlying logic, our cases defining “retroactivity” have principally focused on whether the statutory change in question significantly alters settled expectations: by changing the legal consequences of past events, or vitiating substantial rights established by prior law. (See, e.g., Quarry, supra, 53 Cal.4th at p. 956; Strauss v. Horton (2009) 46 Cal.4th 364, 472; Elsner v. Uveges (2004) 34 Cal.4th 915, 937; McClung, supra, 34 Cal.4th at p. 472; Tapia, supra, 53 Cal.3d at p. 290; see also Western Security Bank v. Superior Court (1997) 15 Cal.4th 232, 243 (Western) [a retroactive statute “substantially changes the legal consequences of past events,” and “[a] statute does not operate retrospectively simply because its application depends on facts or conditions existing before its enactment”]; Landgraf v. USI Film Products (1994) 511 U.S. 244, 266, 269–270 (Landgraf) [similar].) Even Myers, supra, 28 Cal.4th at page 839 followed up its potentially more expansive articulation of “retroactivity” by explaining that, “[p]hrased another way, a statute that operates to ‘increase a party’s liability for past conduct’ is retroactive.” Leading cases addressing “retroactive” legislation have confronted such changes in settled expectations.

This was true, for example, of the amendments to title VII of the Civil Rights Act of 1964 (42 U.S.C. § 2000e et seq.) at issue in Landgraf, supra, 511 U.S. 244. The changes matched new remedies to certain statutory violations. (Ibid.) Courts have also turned to drawing the retroactivity-no retroactivity line with respect to the new certificate requirement for Chinese nationals’ reentry in Chew Heong v. United States (1884) 112 U.S. 536 (discussed in Landgraf, at pp. 271–272); the new liability rule added by Proposition 51 (adopted by voters in June 1986) in Evangelatos v. Superior Court (1988) 44 Cal.3d 1188; the increased worker compensation benefits in Aetna, supra, 30 Cal.2d 388; and, finally, the repeal of statutory tort immunity for tobacco companies in Myers, supra, 28 Cal.4th 828. The changes wrought by sections 10113.71 and 10113.72, by contrast, do not disrupt clearly settled expectations in such fashion — so it’s not clear they operate “retroactively” at all.

To wit: The grace period and notice obligations added by sections 10113.71 and 10113.72 do not impact a life insurer’s liability for past, preenactment defaults. Nothing in these sections compels insurers to reinstate any policy cancelled preenactment less than 60 days after a missed premium payment. Nor do the changes otherwise impinge on a contracting party’s substantial rights or unfairly upset the bargain memorialized in the insurance policy, for example, by requiring an insurer to provide substantially expanded coverage without also giving it an opportunity to raise premiums. (Cf. Interinsurance Exchange, supra, 58 Cal.2d at p. 148 [“retroactive” statutory change would have repealed a rule requiring a mandatory provision in automobile insurance policies covering certain permittees; if applied to previously negotiated contracts, this change would have upended the bargain struck].) The grace period and notice rules make relatively cabined, procedural changes to how insurers administer policies routinely subject to public regulation — they require insurers to provide policy owners with limited but critical safeguards to avoid defaulting. These new rules affect contractual relationships in a field pervasively “ ‘affected with a public interest,’ ” and thereby already heavily regulated by the state. (Calfarm, supra, 48 Cal.3d at p. 830.) And these rules do not unfairly “rewrite” existing policies, as Protective Life suggests. They instead merely impose additional rules on insurers as a condition of doing business in California — rules that govern insurers’ conduct postenactment when, in the future, one of their policy owners misses a premium payment.[5]

The context matters. Where a new law makes only moderate, procedural-type adjustments to the rules for conduct that will apply in the event of some future circumstance, in an already highly regulated contractual relationship, the new law’s application to existing contracts could be regarded as prospective rather than retroactive for purposes of the presumption. To say the least, this type of statutory application falls well short of the quintessential understanding of “retroactivity” — the disruption of settled expectations because a statutory change “ ‘imposes a new or additional liability and substantially affects existing rights and obligations’ ” — that can be reasonably gleaned from leading cases such as Tapia, supra, 53 Cal.3d at page 290. The concurrence fails to fully grapple with this established body of law. (Conc. opn., post, at pp. 8–9 [contending only that the majority fails to identify a sufficiently analogous case].)

An analogous case drives home our conclusion. In Mervyns, supra, 39 Cal.4th 223, we addressed whether Proposition 64’s rule, which restricted standing to bring an unfair competition law claim to plaintiffs who have suffered an injury in fact and have lost money or property, applied to pending cases. (Mervyn’s, at pp. 227–228 [law had previously authorized any person acting for the general public to sue, and Prop. 64, approved by voters in November 2, 2004, deleted this language].) We held that applying the proposition to pending cases “is not to apply [it] ‘retroactively,’ as we have defined that term, because the measure does not change the legal consequences of past conduct by imposing new or different liabilities based on such conduct.” (Mervyn’s, at p. 232; see id. at p. 231 [distinguishing Myers, supra, 28 Cal.4th at p. 240, where the new law “subjected tobacco sellers to tort liability for acts performed at a time when they enjoyed the protection of an immunity statute”].)

So too here. The grace period and notice provisions at issue here simply dictate the procedures for terminating policies after January 1, 2013. Applying the provisions to policies already in effect on that date does not appear to impose new or different liabilities based on earlier conduct. (See also Pitts v. Perluss (1962) 58 Cal.2d 824, 835 (Pitts) [similar reasoning with respect to a regulation preventing substantial adverse selection of risks by private insurance companies, as the “regulation looks solely to the future; it provides that the future operation of the plans comply with the standards. No sanction or penalty attaches to any past act of the companies; the companies need only discontinue one or more existing noncomplying plans or establish complying ones”]; see id. at p. 836 [distinguishing, inter alia, Interinsurance Exchange, supra, 58 Cal.2d 142 and Aetna, supra, 30 Cal.2d 388, because those cases involved attempts “to apply the new law of today to the conduct of yesterday”].)

Protective Life contends that some of our cases support a different conclusion: that any impact on the terms of an existing contract represents a retroactive change for purposes of applying the presumption. Our precedent more readily establishes a different proposition. The mere fact that a new law somehow implicates an existing contract does not, by itself, make the law retroactive. (See Western, supra, 15 Cal.4th at p. 243.) The key is the nature of the new law’s impact — whether it works a substantial change in the contracting parties’ rights or obligations. (See, e.g., Tapia, supra, 53 Cal.3d at p. 290.) It’s far from clear that any of the effects identified by Protective Life and industry amici curiae rise to this level.

Protective Life argues simply that applying the grace period and notice requirements to McHugh’s policy and others like it goes against contractual counterparties’ strong interests in avoiding unexpected shifts in their legal relationship. It asserts, as a general matter, that “[t]he agreed-to premium pricing reflected, among other things, the grace period and notice provisions in the policies.” Protective Life’s argument ultimately boils down to an assertion of its expectations that sections 10113.71 and 10113.72 would not apply to previously enacted policies: It expected these policies to be governed by “the old rules” and can’t now change premiums to account for the new rules. We question the prudence of this expectation, since, as we have previously observed, the “highly regulated” nature of the insurance industry means that “further regulation” on policies by the Legislature “can reasonably be anticipated.” (Calfarm, supra, 48 Cal.3d at p. 830.) Even setting this aside, though, Protective Life’s argument fails to persuade. The insurer’s generalized, amorphous allusion to financial impact does not specifically identify any way in which the bargain memorialized in the insurance contract would be substantially upset by applying the grace period and notice provisions to future cancellations of policies issued before January 1, 2013.

Industry amici curiae’s further arguments, though more substantiated, also fail to persuade. The Chamber of Commerce emphasizes that insurers will have to “devote resources” to complying with the notice provisions. But any such resources would seem to be minimal. Notice of the designation right need only be provided annually and can be sent together with a billing statement. As to pretermination notice, it is standard industry practice to provide some notice before terminating an insurance policy for nonpayment of a premium. (Ante, at p. 8.) Indeed, although McHugh’s policy did not require any pretermination notice, Protective Life sent McHugh letters reminding him that his payment was due, informing him that his payment was late, and then informing him that his policy had lapsed but could be reinstated. The American Council of Life Insurers points out that if the insured dies during the extended grace period, the insurer will be required to pay benefits for which it has not received a premium. But this burden is at least somewhat offset, since the insurer would be entitled to deduct the unpaid premium payment from any life insurance benefits it pays out.

Under the circumstances presented, there appears to be only one way an insurer could incur a significant unaccounted-for loss because of sections 10113.71 and 10113.72: If the grace period and notice provisions prevent inadvertent defaults that the insurer had anticipated and baked into its rates — that, after all, is the problem the provisions were designed to address. If, however, insurers did not specifically account for a projected rate of inadvertent default when setting their rates, then any such default would result in a windfall to the insurer. As amicus curiae California Advocates for Nursing Home Reform, Inc., explains: A windfall, or “ ‘Lapse Profit,’ ” arises when a policy owner’s missed premium payment allows the insurers to “ ‘pocket’ years of premium[s],” potentially totaling several thousands of dollars, and “simply walk away from any obligation [to] pay anything to [its] ‘former client[].’ ” One additional consideration for why a windfall may occur: Premiums in a policy’s earlier years exceed the cost of providing coverage. (Fairbanks, supra, 197 Cal.App.4th at pp. 547–548.)[6]

Here, neither Protective Life nor amici curiae argue that insurers accounted for a particular rate of inadvertent default in setting premiums before January 1, 2013. Though they generally reference premium pricing and financial projections, they do not specifically claim to have included in their rate-setting calculations an anticipated percentage of policy owners who, due to illness, incapacity, or other factors, would fail to pay their premiums and lose coverage, thereby relieving the insurers from the obligation to pay out death benefits. In fact, Protective Life argued at trial — and the jury accepted — that the notices and grace period it actually provided to McHugh substantially complied with and largely replicated sections 10113.71 and 10113.72. Taken together, the arguments offered fall short of clearly showing how the sections’ new protections constituted a disruptive contract change of the sort that would qualify as “retroactive” under our precedent.

2.

If it’s far from obvious that applying sections 10113.71 and 10113.72 to existing life insurance policies is retroactive under well-settled case law, it’s also clear that Protective Life’s argument about the scope of these two new sections still fails to persuade us even if one considered the provisions in question to have some retroactive effect. For some observers, the statutory changes in question might be considered to go beyond nominal or trivial alterations to existing contracts. To the extent some of our cases can be read to suggest a broader understanding of “retroactivity” — one potentially embracing any statutory impact on an existing insurance contract (see Myers, supra, 28 Cal.4th at p. 839) — our precedent nonetheless still supports a conclusion in plaintiffs’ favor.

As plaintiffs argue, even if applying sections 10113.71 and 10113.72 in this case is retroactive, it is retroactive only in a relatively narrow sense. True: The sections, if applied to preenactment policies, do create new rules for insurers in administering these policies. Because the grace period and notice provisions expand insurers’ pretermination requirements beyond the bargained-for policy terms, they technically affect “ ‘ “rights, obligations, acts, transactions and conditions” in existence “prior to the adoption of the [sections].” ’ ” (Myers, supra, 28 Cal.4th at p. 839.) But, for the same reasons that one could conclude this case does not present a question of retroactivity, we determine that any nominal retroactive effect arguably at issue here plainly fails to present the type of concern underlying the application of the presumption as we have ordinarily understood it.

The retroactivity presumption is not a “straitjacket.” (In re Estrada (1965) 63 Cal.2d 740, 746.) Nothing in our cases calls on us to apply the presumption for any conceivable type of preenactment impact, however slight. Instead, our case law calls for application of the presumption where applying the new law implicates fundamental fairness concerns, including by “foist[ing] upon past conduct new and onerous legal consequences.” (Pitts, supra, 58 Cal.2d at pp. 835–836.) In Myers, supra, 28 Cal.4th 828, for instance, the retroactive legal change in question subjected tobacco sellers to tort liability for prior acts performed when they enjoyed the protection of an immunity statute. (See also Tapia, supra, 53 Cal.3d at pp. 297–299 [retroactive legal change subjected persons to increased punishment for past criminal conduct, or to punishment for past conduct not formerly defined as criminal].) By contrast, the new grace period and notice requirements do not thrust new legal consequences onto Protective Life’s preenactment policy terminations or otherwise appear to cause the insurer to bear significant and unanticipated costs for its pre-2013 policies. Instead, the new rules simply updated how the regulatory system governing life insurance terminations treats all policies going forward. Therefore, insofar as these new rules operate “retroactively,” that is not the type of retroactivity that warrants our usual level of reluctance to construe statutes retroactively.

In summary, this case may be viewed as not involving “retroactivity” as our cases have generally defined the term, or alternatively as involving retroactivity only in a narrow sense — one different from the type of preenactment impact at the heartland of the presumption’s concerns and at issue in cases applying the presumption.  We need not choose between the two views. Under either, we decline to give the presumption such weight that it determines the outcome of this case.

B.

Having explored the presumption against retroactivity, the broader interpretive question remains: whether sections 10113.71 and 10113.72 apply to McHugh’s policy and similarly situated ones. We conclude that they do.

Before engaging in our interpretive analysis, we make one brief observation: As the concurrence implies, our precedent leaves open the possibility of simply assuming that this case presents an instance of retroactivity and therefore merits application of the presumption against retroactivity. (Conc. opn., post, at pp. 6–7.) And if the presumption applies with its ordinary weight, the indicia of legislative purpose here could rebut it. (See, e.g., id. at pp. 3–4 [emphasizing the breadth of the statutory language]; id. at pp. 5–6 [highlighting the legislative history discussion of problems facing existing policyholders]; cf. post, at pp. 33–37.)

But the most thorough approach, consistent with previous cases, is to address a threshold question: whether sections 10113.71 and 10113.72 even create retroactive changes for purposes of the presumption. (Ante, at p. 15.) That way, we avoid having to apply the canon in a circumstance where it’s not necessary, and where our cases do not definitively indicate that the presumption has been rebutted. (Compare conc. opn., post, at pp. 5–6 with McClung, supra, 34 Cal.4th at p. 475 [presumption not rebutted where legislative history lacked retroactivity discussion].)

1.

We start with the statutory language. (H.W., supra, 6 Cal.5th at p. 1073.)

On the one hand, broadly applicable language found in many of the sections’ subdivisions, and the absence of any temporal qualifiers in this language, supports plaintiffs’ argument that the sections apply to all policies in force as of January 1, 2013. On the other hand, different parts of the provisions plausibly favor the interpretations urged by both parties. One provision, section 10113.72, subdivision (a), unmistakably applies only to new policies; but some provisions likely seem to apply to both new and existing policies, and some could be read either way. Because the parties’ linguistic parsing at times plausibly cut in opposing directions, and because we must interpret these provisions as a package, the net effect is one of some potential statutory ambiguity. To see why, consider each provision in turn.

Section 10113.72, subdivision (a) relates to the right to designate at least one third party recipient to receive a missed premium notice. This provision applies only to new, postenactment policies. (McHugh, supra, 40 Cal.App.5th at pp. 1174–1175.) The language refers repeatedly to the “applicant” making a written designation, rather than the “policy owner,” clearly indicating it applies only to new policies. (§ 10113.72, subd. (a).) Protective Life also persuasively argues that the particular phrasing of subdivision (a)’s command — it instructs that a policy “shall not be issued or delivered . . . until” an applicant has been given the written designation right — further supports that it applies only to new policies as of 2013, since “shall be” commands often signify a statute’s forward-looking nature. (Ibid., italics added; cf. Russell v. Superior Court (1986) 185 Cal.App.3d 810, 818–819.)

But subdivision (b) of section 10113.72 does not refer to “applicant[s],” nor does it use “shall be” language. The subdivision requires insurers to annually notify “policy owner[s]” for their right to update and change their third party notice designation. (§ 10113.72, subd. (b).) It could easily apply to both new policies and those already in force as of January 1, 2013. The subdivision provides an opportunity for the policy owner on an existing policy with no designation to “designate one or more persons” (ibid.) — a right that could apply regardless of when the policy was issued. (See Bentley v. United of Omaha Life Ins. Co. (C.D. Cal., June 22, 2016, No. CV15-7870-DMG (AJWx)) 2016 WL 7443189, p. *4 (Bentley).) Indeed, as plaintiffs persuasively argue, nothing in this language appears to limit this right only to those “policy owners” who purchased insurance postenactment. (See ibid.)

Even so, we can’t determine for sure from the isolated language of section 10113.72, subdivision (b) that the designation right unambiguously applies to all policies. Protective Life counters plaintiffs’ plausible reading of subdivision (b) with its own potentially tenable interpretation: Subdivision (b) does not create a freestanding designation right, but instead repeatedly refers back to and builds off its immediately preceding provision, subdivision (a) — discussing “[t]he insurer,” “the policy owner,” “the right,” “the designation,” and “change[s]” to the designation. (§ 10113.72, subd. (b).) In other words, subdivision (b) can be read to merely clarify the scope of subdivision (a) by requiring that “applicant[s]” who are now “policy owner[s]” be advised that they may exercise the right to designate by changing or initially naming a designee. Plaintiffs likely offer the better interpretation, as it gives full effect to the several instances where subdivision (b) uses meaningfully distinct language from subdivision (a). (See, e.g., Gomez v. Superior Court (2012) 54 Cal.4th 293, 304; Briggs v. Eden Council for Hope & Opportunity (1999) 19 Cal.4th 1106, 1117.)[7] Yet Protective Life also offers a tenable reading.

Far less ambiguity lurks in the notice provisions. Each speaks universally, referring to “policy owner[s]” and describing its requirement without any apparent limitation on the date of issuance. (§§ 10113.71, subd. (b)(1), (3), 10113.72, subd. (c).) Section 10113.71, subdivision (b)(1) states broadly that notice of pending lapse and termination “shall not be effective unless mailed by the insurer” to the policy owner, a designee, and a known assignee “at least 30 days prior to the effective date of termination.” Section 10113.71, subdivision (b)(3) also sweeps broadly, requiring insurers to provide policy owners and designees with notice “within 30 days after a premium is due and unpaid.” Section 10113.72, subdivision (c) similarly provides that “[n]o individual life insurance policy shall lapse or be terminated for nonpayment of premium unless the insurer” gives the requisite 30-day-minimum notice, and that “[n]otice shall be given . . . within 30 days after a premium is due and unpaid.”

As Protective Life observes, however, the notice provisions do include references to the designee for notice under section 10113.72, which suggests the two sets of provisions are meant to work together. That introduces some ambiguity into our reading of the notice provisions insofar as the designee provisions can be considered ambiguous. If all of section 10113.72 were read to apply only to new contracts, that would be some indication that the notice provisions apply only to new contracts as well. But if the more likely reading is that designations can also be made under existing contracts under section 10113.72, subdivision (b), then we have little reason to think that the notice provisions would be restricted to new contracts. (We say little reason, rather than no reason, because section 10113.72, subdivision (c)’s specific cross-reference to subdivision (a) of the same section, and the absence of such a specific cross-reference in section 10113.71, subdivision (b), does slightly muddy the waters.) In any event, if no notice recipient has yet been designated for a policy — which could, of course, happen even in the case of a new policy — then the provisions requiring notice to the designee are simply ineffective.

Finally, section 10113.71, subdivision (a) addresses the grace period provision. Here too, we find support in its language for both parties’ arguments.

Protective Life identifies language in section 10113.71, subdivision (a) that appears future-oriented: The phrases “shall contain” (each policy “shall contain a provision for a grace period”) and “shall provide” (“The provision shall provide that the policy shall remain in force during the grace period”) in subdivision (a) suggest reference to a requirement for policies yet to be issued. (See People v. Allied Architects’ Assn. (1927) 201 Cal. 428, 437.)

Plaintiffs reasonably respond, however, that “shall” in 10113.71, subdivision (a) represents a mandatory directive for all policies (i.e., each policy must be read to contain a grace period), rather than a temporal limitation on the policies to which the grace period applies. (Evangelatos v. Superior Court, supra, 44 Cal.3d at p. 1209, fn. 3; see People v. Ledesma (1997) 16 Cal.4th 90, 95 [“ ‘shall’ ” can be construed as either mandatory or directory as well as denote future operation].) Moreover, they persuasively analyze the past participle “issued or delivered.” As used here in the phrase “[e]ach life insurance policy issued or delivered in this state,” the reference to what’s “issued or delivered” can simultaneously refer to past as well as future events. (See Bernal v. NRA Grp. LLC (7th Cir. 2019) 930 F.3d 891, 895.)[8]

Considering section 10113.71, subdivision (a) in context supports plaintiffs’ interpretation. (See People v. Garcia (2017) 2 Cal.5th 792, 805 (Garcia).) As Protective Life itself observes, the grace period provision has an “intertwined” relationship with the notice provisions. The Legislature evidently designed the grace period to work in conjunction with the notice provisions, which ensure policy owners and their designees receive notification of a pending lapse and termination at least 30 days before the 60-day grace period has expired. (See Assem. Com. on Insurance, Background Information Sheet for Assem. Bill No. 1747 (2011–2012 Reg. Sess.) Feb. 27, 2012, p. 2.) Indeed, it appears intentional that the two 30-day windows provided by the notice provisions operate within and can add up to the 60-day grace period: The insurer has up to 30 days after a missed premium payment to give notice (§§ 10113.71, subd. (b)(3), 10113.72, subd. (c)), and a separate 30 days that must follow before a mailed notice becomes effective to terminate a policy for nonpayment (see §§ 10113.71, subd. (b)(1), 10113.72, subd. (c)). Given this relationship, it certainly seems sensible that the grace period would, as the notice provisions appear to do, apply universally.

Admittedly, the notice provisions can perhaps be capable of operating independently of the grace period provision — the former applying to all policies, the latter to new policies only. The notice provisions themselves effectively establish a grace period of 30 to 60 days. But it seems unlikely that the Legislature meant for the notice provisions to drive the scope of protections conferred for nonpayment in the class of cases involving existing contracts, particularly since it did not draw any express distinctions in any of the provisions on the basis of policy issue date. It seems more reasonable to construe the statutory provisions as a package, as either all applying to existing contracts or not. That the provisions work together — the notice provisions require sending notice to the policy owner and an individual designated under the new designation procedures, within the time frame established by the new grace-period provision — reinforces our conclusion.

In view of how these provisions interact, the broadly applicable language found in most of the relevant statutory passages tends to cut in favor of plaintiffs’ interpretation. (Cf. conc. opn., post, at pp. 3–4.) That said, Protective Life does identify some language cutting in favor of its narrower interpretation — and the statutory sections at issue stop short of conclusively establishing precisely how the provisions work. To resolve any potential ambiguity in the language, we must look to other sources to determine whether, as plaintiffs argue, the provisions’ intended purpose entails applying them to existing contracts. (H.W., supra, 6 Cal.5th at p. 1073.)

2.

Other indicia of purpose, gleaned from context, resolve any latent ambiguity in the language of sections 10113.71 and 10113.72. They indicate that the sections apply to all policies in force as of January 1, 2013.

To begin with, the statutory sections appear to create a single, unified pretermination notice scheme. This scheme appears to include three components: (1) New and existing policy owners must have the opportunity to designate additional people to receive a notice of termination (§ 10113.72, subds. (a), (b)); (2) policy owners and any designees must receive notice within 30 days of a missed premium payment, and any termination for nonpayment will not be effective unless insurers send notice to these parties at least 30 days prior (§§ 10113.71, subd. (b)(1), (3), 10113.72, subd. (c)); and (3) each policy has a 60-day grace period, which lines up with the two 30-day notice windows (§ 10113.71, subd. (a)). In light of these new, detailed statutory notice requirements, it seems doubtful the Legislature contemplated that insurance companies would, going forward, simultaneously implement two vastly different notice schemes: one applying to pre-2013 policies that requires only 31-day notices before termination and no right to designations, and a post-2013 scheme as described. It thus seemed largely assumed that insurance companies would implement the single, new notice scheme, which would have the effect of benefitting both new and existing policy owners.

The legislative history supports this conclusion. Those involved in the legislative process seemed to take for granted a single, standardized notice scheme. (See, e.g., Assem. Com. on Insurance, 3d reading analysis of Assem. Bill No. 1747 (2011–2012 Reg. Sess.) as amended May 9, 2012, p. 2.)

Moreover, the legislative history provides several indications that the Legislature enacted the grace period and notice protections in part to protect existing policy owners from losing the important life insurance coverage they had spent years paying for. The Assembly and Senate materials on Assembly Bill No. 1747 (2011–2012 Reg. Sess.) include purpose and supporting argument statements like the following: “According to the author, the bill provides consumer safeguards from which people who have purchased life insurance coverage, especially seniors, would benefit. Under existing law, individuals can easily lose the critical protection of life insurance if a single premium is accidentally missed (even if they have been paying premiums on time for many years). If an insured individual loses coverage and wants it reinstated, he or she may have to undergo a new physical exam and be underwritten again, risking a significantly more expensive, possibly unaffordable premium if his or her health has changed in the years since purchasing the policy. Therefore, the protections provided by [Assembly Bill No.] 1747 are intended to make sure that policy owners have sufficient warning that their premium may lapse due to nonpayment.” (Assem. Com. on Insurance, Analysis of Assem. Bill No. 1747 (2011–2012 Reg. Sess.) as amended Apr. 26, 2012, pp. 1–2, italics added (hereafter Assem. Com. on Insurance Analysis); see also Sen. Com. on Insurance, Analysis of Assem. Bill No. 1747 (2011–2012 Reg. Sess.) as amended June 7, 2012, p. 3 [longtime policy owners may miss a payment “because they were being hospitalized when the bill came, in others, as a result of a mail mix-up or forgetfulness, etc.”].) Where, as here, the author’s statements are part of committee materials — and are therefore relayed not merely as personal views, but instead as part of the Legislature’s consideration of the bill — they can serve as salient reflections of legislative purpose. (See Carter v. California Dept. of Veterans Affairs (2006) 38 Cal.4th 914, 928.)

Protective Life argues that the legislative history does not clarify the statutory language, but we are not persuaded. True: The materials do not explicitly consider the reach of the broadly worded, but less than crystal clear, grace period and notice provisions; and their references to protecting “seniors” could be to the people whom the Legislature anticipated would benefit from the new law down the line. But the insurer ignores the clear guidance the materials do provide. At the very least, they reflect lawmakers’ (a) awareness that consumers tend to hold life insurance policies for long periods and to pay premiums for many years; and (b) concern that policy owners, “especially seniors” (Assem. Com. on Insurance Analysis, supra, at p. 1), may lose the benefits of these extended payments by failing to pay a single annual premium on time, and thereafter face hardship to regain coverage. It would certainly be consistent with the Legislature’s awareness and concern for it to seek to protect all policy owners from losing coverage. (Cf. Calfarm, supra, 48 Cal.3d at p. 827; conc. opn., post, at pp. 5–6.)

The consequences of Protective Life’s interpretation strongly suggest, given the legislative history, that it wasn’t in the ambit of the Legislature’s purpose for the statute to operate as the insurer describes. (See Copley Press, Inc. v. Superior Court (2006) 39 Cal.4th 1272, 1291 (Copley).) Indeed, the insurer’s interpretation would produce results seemingly incongruous with the legislation’s broader aims of preventing forfeiture and its specific motivating concerns. As the Legislature identified, policy owners may fail to make a payment on time for a host of understandable reasons, including some related to their age or health. But the very consumers the Legislature identified as needing protection the most against this risk — seniors and other longtime, potentially infirm or incapacitated policy owners — would not presently be entitled to the safeguards to help them maintain coverage they and their beneficiaries depend on. They would face a host of adverse financial consequences to resume coverage. (Ante, at p. 6.) Meanwhile, those who arguably need protection the least — younger policy owners, who recently purchased life insurance, are less likely to miss a payment due to infirmity or deteriorating health, and face a lower loss of past premium investment and an easier time regaining coverage — are protected, and with measures that will be likely consequential to them only when they become “seniors” years down the line. If a paradigmatic beneficiary of the new legislation was, say, a 70-year-old life insurance policy owner who had paid premiums for 30 years before missing an annual payment, a new-policy-only construction would mean that a person in this situation wouldn’t garner protection from the new laws before 2043. Even for a forward-thinking Legislature, this seems like a stretch. (Cf. Bentley, supra, 2016 WL 7443189, at p. *4 [declining to give effect to the “absurd result[s]” of Protective Life’s interpretation].)

Assembly Bill No. 1747 (2011–2012 Reg. Sess.) also cuts in favor of reading sections 10113.71 and 10113.72 broadly. The bill not only added these sections to the Insurance Code, but also amended section 10173.2, which concerns when life insurance policies are assigned as security for a debt and the notice that the insurer must give the assignee when the policy owner fails to pay a premium. (§ 10173.2, as amended by Stats. 2012, ch. 315, § 3.) The Legislature amended section 10173.2 by changing some deadlines and revising nonsubstantive language. More notable is what section 10173.2 already said prior to amendment:  “When a policy of life insurance is, after the effective date of this section, assigned in writing as security for an indebtedness . . . .” (§ 10173.2, italics added.) The italicized language by its terms cabins the statute’s application to assignments after section 10173.2’s effective date. (Estate of Coate (1979) 98 Cal.App.3d 982, 986–987; see also Mardirosian v. Lincoln Nat. Life Ins. Co. (9th Cir. 1984) 739 F.2d 474, 477.) In other words, when the Legislature added sections 10113.71 and 10113.72 to the Insurance Code, it knew that another statute — indeed, a statute it amended in the very same bill — used expressly future-oriented language. Despite this, the Legislature did not add similar language to sections 10113.71 and 10113.72. This circumstance provides additional, if modest, support for the conclusion that the grace period and notice provisions apply universally. (Cf. Calfarm, supra, 48 Cal.3d at p. 827 [similar]; conc. opn., post, at pp. 4–5.)

The various contextual arguments raised by Protective Life and industry amici curiae fail to persuade.

First, Protective Life fails to substantiate its argument that construing sections 10113.71 and 10113.72 to apply only to postenactment policies gives effect to a key legislative compromise. It’s well established that “compromises necessary to [a statute’s] enactment may require adopting means other than those that would most effectively pursue the main goal.” (Landgraf, supra, 511 U.S. at p. 286.) But this general proposition doesn’t mean we can strike a bargain the Legislature never struck. (Cf. State Dept. of Public Health v. Superior Court (2015) 60 Cal.4th 940, 956.) Here, Protective Life identifies no indicia of compromise in the legislative history or statutory language; instead, it simply invokes the presumption against retroactivity, which we have determined carries little if any weight in this case. (Ante, at pp. 25–26.)[9]

Protective Life also contends that the Legislature had “good reasons” to restrict the application of sections 10113.71 and 10113.72 to postenactment policies: to avoid unfairly altering the bargained-for grace period and notice rules, which the agreed-to premium pricing had taken into account. Similarly, amicus curiae Chamber of Commerce claims that applying the new grace period to preenactment policies “undermines insurers’ ability to prudently manage their resources” and could leave insurers with “inadequate funds to pay valid claims” statewide. Yet without evidence that Protective Life or other insurers anticipated and accounted for a projected rate of inadvertent defaults when setting their premiums (ante, at p. 23), we have no basis to determine either that (a) the new protections will create a significant financial impact for insurers, or (b) the Legislature would have sought to avoid such a policy outcome.

Moreover, we note that plaintiffs and supporting amici curiae offer their own “good reasons” why the Legislature would apply pretermination procedures to all policies: The procedures (1) promote continuity in the insuring arrangement (cf. Bittinger v. New York Life Ins. Co. (1941) 17 Cal.2d 834, 840 [“forfeitures generally are not favored”]; People v. United Nat. Life Ins. Co. (1967) 66 Cal.2d 577, 600 [“The insurance industry is regulated primarily for the benefit of” insureds]); (2) place the burden on the party who stands to gain financially from an early termination; (3) create standardized rules governing policies; and (4) help prevent payment disputes, which typically arise after policy owners have died. We take into account these considerations insofar as they plausibly counter the policy arguments raised by Protective Life and industry amici curiae, and they help us determine that plaintiffs’ construction “ ‘leads to the more reasonable result.’ ” (Copley, supra, 39 Cal.4th at p. 1291.)

Finally, Protective Life briefly raises a constitutional avoidance argument. (Garcia, supra, 2 Cal.5th at p. 804.) It contends that the Legislature’s decision to restrict sections 10113.71 and 10113.72 to postenactment policies represented an “especially sound” decision in light of contracts clause concerns that would have flowed from altering the terms of existing policies. (U.S. Const., art. I, § 10; see Cal. Const., art. I, § 9.) What we conclude instead — relying on similar logic that applies in our separate analysis of the presumption of retroactivity (see Myers, supra, 28 Cal.4th at p. 841 [explaining the presumption is rooted in constitutional principles]) — is that requiring insurers to observe a 60-day grace period and give 30 days’ notice of impending lapse does not substantially impair Protective Life’s contractual rights under an existing policy. Calfarm, supra, 48 Cal.3d at pages 830–831 supports our conclusion.

C.

The Court of Appeal held that insurance policies already in effect when the Legislature reformed grace period and notice requirements were not affected by sections 10113.71 and 10113.72. In reaching this conclusion, the appellate court cited DOI guidance about these sections and claimed it had an obligation to defer to these agency interpretations. We find otherwise.

According to the Court of Appeal, two sources of DOI guidance established the agency’s position that the sections apply only to policies issued after January 1, 2013. First, the court pointed to SERFF. (McHugh, supra, 40 Cal.App.5th at p. 1172.) SERFF is an internet-based system for insurers to submit rate and form filings to the DOI for review and approval. (Ibid.) According to the Court of Appeal, DOI “mandates the use of SERFF and provides regulatory guidance to insurers through SERFF, including guidance for compliance with the statutes.” (Ibid.) As the court explained, DOI provided its determination that sections 10113.71 and 10113.72 apply only to postenactment policies with its SERFF “ ‘Instructions for Complying with [Assembly Bill No.] 1747,’ ” which stated: “ ‘All life insurance policies issued or delivered in California on or after [January 1, 2013] must contain a grace period of at least 60 days.’ ” (McHugh, at p. 1172.) Second, the Court of Appeal observed that senior DOI personnel consistently communicated in written responses to inquiries from insurance industry representatives that the requirements in Assembly Bill No. 1747 (2011–2012 Reg. Sess.) applied only prospectively. (McHugh, at p. 1172.)

The Insurance Commissioner takes a different position in his amicus curiae letter, arguing that the Court of Appeal erred on both fronts. We agree. Neither the SERFF instruction nor the correspondence represented official guidance on the agency’s construction of sections 10113.71 and 10113.72, and as a result neither merited any measure of presumptive deference (see Yamaha, supra, 19 Cal.4th at pp. 7–8, 11); and Protective Life offers no other good reason why we should defer (see id. at pp. 12–13 [contextual factors such as the agency’s expertise and technical knowledge of the issue, and whether the agency’s interpretation represents its carefully considered, long-standing, and contemporaneous view, determine what level of deference to give]).

In fact, the SERFF instruction does not appear to even constitute an interpretation of sections 10113.71 or 10113.72. The document simply instructs that policies issued on or after January 1, 2013 must contain the 60-day grace period. That instruction enables new, yet-to-be issued policy forms to align with current law, in line with the electronic system’s function. Contrary to the Court of Appeal’s view, the instruction provides no view on whether the grace period or the other requirements in sections 10113.71 and 10113.72 apply to existing policies. As the Insurance Commissioner helpfully explains, SERFF is simply a voluntary system for form and rate submissions, and SERFF instructions like the one here “may include a brief description of the relevant statutes,” but are “not intended to serve as a formal legal opinion of the [DOI].”

Although agency correspondence may express an interpretive view, it does not merit deference here. Although courts should certainly consider the interpretations of an agency advanced in litigation even if these are not associated with formal administrative actions, we agree with the Insurance Commissioner that ordinary agency correspondence provides us with little assistance in our interpretive inquiry. As we explained in Heckart v. A-1 Self Storage, Inc. (2018) 4 Cal.5th 749, 769, footnote 9, these types of private communications offer poor guides because (a) they don’t appear to be the product of “ ‘ “careful consideration” ’ ” of the legal issue, but instead reflect interpretations prepared in ad hoc advice letters by individual staff members; (b) the views expressed in them do not represent “a quasi-legislative rule, promulgated pursuant to delegated lawmaking power” (ibid.); and (c) they were “not disseminated as an annotation by the [DOI] to be considered by anyone other than the recipient, and there is no information regarding how carefully the issue was considered” (ibid., applying Yamaha, supra, 19 Cal.4th at pp. 11–16). For the same reasons, we give no weight to the identical, as well as additional, correspondence that we judicially notice at Protective Life’s request.[10]

IV.

When the Legislature enacted changes to the Insurance Code protecting people who hold life insurance policies from inadvertently losing them, it established limited protections that kept such policies from being revoked when policy owners lapsed in paying premiums. Those provisions clearly establish that life insurance policies must have a 60-day grace period before they can be terminated for a premium lapse, and that insurers cannot terminate policies for a premium lapse until they give at least 30-day mailed notice to the policy owners and to any additional designated individuals. What they don’t explicitly establish is whether these protections apply to people holding life insurance policies issued or delivered before these amendments went into effect.

These sections are nonetheless best read to extend protections to policies issued before these sections went into effect. Key passages in sections 10113.71 and 10113.72 are written in universal terms, best understood to modify policies whether they come into effect after reforms were enacted or were already in effect at the time. Other indicia of purpose resolve any ambiguity that remains from the language. The grace period and notice protections apply to all policies in effect as of the sections’ effective date — and in this case, nothing in the presumption against retroactive application of legislation as ordinarily applied compels another result. The Legislature enacted the sections not only to provide protections to people in the future, but also to ensure that existing policy owners don’t lose the life insurance coverage that they may have spent years paying for and on which their loved ones depend. Accordingly, we reverse the judgment of the Court of Appeal and remand for proceedings consistent with this opinion.

CUÉLLAR, J.

We Concur:

CANTIL-SAKAUYE, C. J.

LIU, J.

KRUGER, J.

GROBAN, J.

Concurring Opinion by Justice Jenkins

I agree with the majority that sections 10113.71 and 10113.72 of the Insurance Code[11] “apply to all life insurance policies in force when these two sections went into effect, regardless of when the policies were originally issued.” (Maj. opn., ante, at p. 2.) I reach this conclusion by a different analytical path. Even if, as defendant Protective Life Insurance Company (Protective Life) argues, this conclusion constitutes retroactive application of the statutes — such that the presumption against retroactivity applies — the relevant statutory language and legislative history are, in my view, “sufficiently clear to compel the inference that the [Legislature] did intend the provisions” to apply retroactively. (Californians for Disability Rights v. Mervyn’s, LLC (2006) 39 Cal.4th 223, 229 (Mervyn’s).) Indeed, the majority acknowledges that “if the presumption applies with its ordinary weight, the indicia of legislative purpose here could rebut it.” (Maj. opn., ante, at p. 26.)

However, I do not endorse the majority’s conclusion that applying the statutes to the policy at issue in this case does not trigger the presumption — or, alternatively, that the presumption applies but with something less than “its ordinary weight” (maj. opn., ante, at p. 26) — because (a) the impact of doing so on Protective Life’s contractual rights and obligations is insufficiently “substantial” to constitute a retroactive legal change (maj. opn., ante, at p. 18), and/or (b) “any nominal retroactive effect . . . plainly fails to present the type of concern underlying the application of the presumption as we have ordinarily understood it” (maj. opn., ante, at pp. 24–25). I therefore concur only in the judgment.

I.

The insurance policy here at issue includes a 31-day “grace period” for payment of the yearly premium, which provides in relevant part: “A grace period of 31 days will be allowed for payment of each premium after the first. This policy will continue in force during the grace period. If the premium remains unpaid at the end of the grace period, coverage will cease.”

As the majority notes, the length of this contractual grace period complied with applicable administrative regulations, which then expressly required at least “a 31-day grace period.” (Maj. opn., ante, at p. 9, fn. 3.) William McHugh, who purchased the policy, failed to pay the premium that was due on January 9, 2013, by the end of the grace period. He died in June 2013. Protective Life advised his named beneficiaries that the policy terminated before his death for nonpayment of the premium. The beneficiaries — plaintiffs Blakely McHugh and Trysta Henselmeier — sued Protective Life for breach of contract and breach of the implied covenant of good faith and fair dealing, arguing that Protective Life improperly terminated the policy without following the requirements of sections 10113.71 and 10113.72. Protective Life asserts that sections 10113.71 and 10113.72 do not apply in this case because they took effect on January 1, 2013, long after McHugh’s policy was issued in 2005. A contrary conclusion, it argues, would constitute a retroactive application of the statutes — by extending the policy’s stated grace period and imposing new notice requirements — and there is insufficient evidence to “overcome” the “presumption ‘that legislation operates prospectively rather than retroactively.’ ”

II.

Even if Protective Life is correct that applying the statutes to the policy at issue here triggers the presumption against retroactivity, Protective Life’s argument ultimately fails because the presumption has been overcome. As the majority notes, “[t]he retroactivity presumption is not a ‘straitjacket.’ ” (Maj. opn., ante, at p. 25.) “Even without an express declaration, a statute may apply retroactively if there is ‘ “a clear and compelling implication” ’ that the Legislature intended such a result.” (People v. Alford (2007) 42 Cal.4th 749, 754.) “We may infer such an intent from the express provisions of the statute as well as from extrinsic sources, including the legislative history.” (Preston v. State Bd. Of Equalization (2001) 25 Cal.4th 197, 222 (Preston); see Alford, at p. 754 [relying on “legislative history” in giving statute retroactive effect].)

In my view, the statutory language and relevant legislative history are “sufficiently clear to compel the inference that the [Legislature] did intend” sections 10113.71 and 10113.72 to apply retroactively. (Mervyn’s, supra, 39 Cal.4th at p. 229.) Section 10113.71, subdivision (a), states that “[e]ach life insurance policy issued or delivered in this state shall contain a provision for a grace period of not less than 60 days from the premium due date.” (Italics added.) As the majority explains, this language “reasonably” may be understood as “a mandatory directive for all policies (i.e., each policy must be read to contain a grace period),” without any “temporal limitation.” (Maj. opn., ante, at p. 31.) Next, section 10113.72, subdivision (c), states that “no individual life insurance policy shall lapse or be terminated for nonpayment of premium unless the insurer, at least 30 days prior to the effective date of the lapse or termination, gives notice to the policy owner and to the person or persons designated pursuant to subdivision (a), at the address provided by the policy owner for purposes of receiving notice of lapse or termination.” (Italics added.) This language states a substantive rule of law — apparently applicable to all policies (“No individual policy” (ibid.)) — that precludes lapse or termination of any policy absent provision of the required notice.

The breadth of this language, and the absence of any language limiting the statutes’ application to policies issued after the statutes’ effective date, are significant given that the Legislature, in the same 2012 measure that added sections 10113.71 and 10113.72, amended section 10173.2. As to life insurance policies “assigned in writing as security for an indebtedness,” section 10173.2 requires “insurer[s]” to mail written notice to the assignees “each time the policy owner has failed or refused to transmit a premium payment to the insurer before the commencement of the policy’s grace period or before the notice is mailed.” As the majority explains, when the Legislature amended section 10173.2 in 2012, the statute contained — and still contains — language giving it prospective-only effect, by providing that the statute’s requirements apply only when a life insurance policy is assigned “ ‘after the effective date of this section.’ ” (Maj. opn., ante, at p. 37; see Stats. 2012, ch. 315, § 3; Stats. 1975, ch. 792, § 1, p. 1816.) Thus, as the majority also explains, “when the Legislature added sections 10113.71 and 10113.72 to the Insurance Code, it knew that another statute — indeed, a statute it amended in the very same bill — used expressly future-oriented language,” but the Legislature “did not add similar [prospective-only] language to sections 10113.71 and 10113.72.” (Maj. opn., ante, at p. 37.) This circumstance indicates the Legislature’s intent to make the grace period and notice provisions applicable to all policies, regardless of issue date. (See Calfarm Ins. Co. v. Deukmejian (1989) 48 Cal.3d 805, 827 (Calfarm) [“necessary inference” from “omission” of language giving statute prospective-only effect is that statute’s application “was not so limited,” given language in simultaneously enacted provision “expressly” giving it prospective-only effect].)

The relevant legislative history reinforces this conclusion. According to one analysis of the proposed legislation, the “[p]urpose of the bill” was “[t]o provide consumer safeguards from which people who have purchased life insurance coverage . . . would benefit.” (Sen. Com. on Insurance, Analysis of Assem. Bill No. 1747 (2011–2012 Reg. Sess.) as amended June 7, 2012, p. 2, italics added, underscoring omitted.) Several other analyses used identical language in describing what, “[a]ccording to the author” of the legislation, the proposed statutes would “provide[].” (Assem. Com. on Insurance, Analysis of Assem. Bill No. 1747 (2011–2012 Reg. Sess.) as amended Apr. 26, 2012, p. 1; see Assem. 3d reading analysis of Assem. Bill No. 1747 (2011–2012 Reg. Sess.) as amended May 9, 2012, p. 2 [same].) In explaining the need for these safeguards, the same analyses explained that under existing law, policyholders — “especially seniors” — could “easily lose” coverage after “many years” of “paying premiums” if they “accidentally missed” making even “a single premium” payment. (Assem. Com. on Insurance, Analysis of Assem. Bill No. 1747 (2011–2012 Reg. Sess.) as amended Apr. 26, 2012, pp. 1–2; Assem. 3d reading analysis of Assem. Bill No. 1747 (2011–2012 Reg. Sess.) as amended May 9, 2012, p. 2; Sen. Com. on Insurance, Analysis of Assem. Bill No. 1747 (2011–2012 Reg. Sess.) as amended June 7, 2012, p. 2.) The Legislature sought to address this problem through the combined effect of the new notice and extended grace period provisions. As the majority observes, to conclude that the Legislature did not intend to extend these new safeguards to existing at-risk policyholders who, according to the legislative history, were the motivation for the legislation, “would produce results seemingly incongruous with” the Legislature’s intent. (Maj. opn., ante, at p. 36.) Given “[t]he evident purpose of” the statutes, “the conclusion is inescapable that” they were “intended to apply to policies in force on the . . . date” they took effect. (Calfarm, supra, 48 Cal.3d at p. 827.) To decline to give the statutes’ retroactive effect would, contrary to our precedent, turn the presumption into a “ ‘straitjacket.’ ”[12] (Maj. opn., ante, at p. 25.)

III.

In light of my conclusion that the statutory language and legislative history are “sufficiently clear to compel the inference that the [Legislature] did intend the provisions” to apply retroactively (Mervyn’s, supra, 39 Cal.4th at p. 229), it is unnecessary for me to decide whether, as Protective Life asserts, applying the requirements of sections 10113.71 and 10113.72 in this case constitutes a retroactive application of the statutes that triggers the presumption. Contrary to what the majority might seem to suggest, this approach is fully “consistent with” precedent (maj. opn., ante, at p. 26) in which we first “assum[ed],” without deciding, that applying a statute would be giving it retroactive effect, and then held, based on “the pertinent legislative materials,” that the Legislature intended the statute to have such effect. (Preston, supra, 25 Cal.4th at pp. 221, 222.) I therefore do not join the majority’s conclusion that the presumption is inapplicable or applies with less than its ordinary force, or with its discussion of those questions.

Although I acknowledge that the majority’s choice to address the question of whether the presumption even applies is “consistent with previous cases,” it is not evident to me that the majority’s approach is “the most thorough” one. (Maj. opn., ante, at p. 26.) I say this because the majority leaves more questions unanswered than it resolves. It does not definitively decide whether applying the statutes here would result in “retroactive changes for purposes of the presumption.” (Ibid.) It also declines to decide whether the presumption is inapplicable — such that it carries no weight — or whether it applies but carries less than “its ordinary weight.” (Ibid.; see id. at p. 38 [the presumption “carries little if any weight” in this case]).[13] Nor does the majority explain, with respect to its alternative holding that the presumption applies but carries less than “its ordinary weight” (maj. opn., ante, at p. 26), what weight the presumption does carry or what is required to overcome it. Indeed, notwithstanding the majority’s alternative holding, its analysis proceeds as if the presumption is completely inapplicable. For example, the majority never mentions the presumption or appears to give it any weight, as evidenced by the majority’s express refusal even “to entertain” Protective Life’s argument “that the Legislature’s failure to enact the sections as part of urgency legislation cuts against rebutting the presumption.” (Id. at p. 38, fn. 9.) In short, by declining to offer answers to what it calls the “threshold question” (id. at p. 26) of retroactivity, the majority’s “approach” ultimately fails to yield an analysis that provides lower courts with clear and adequate guidance for applying the presumption, as the majority puts it, in a way we have not previously “understood it” (id. at p. 25) and with less than “its ordinary weight” (id. at p. 26).

Nor is it clear to me that our precedents “support[]” (maj. opn., ante, at p. 24), much less “strongly favor[]” (id. at p. 17), the majority’s conclusion that the presumption either does not apply at all or applies but carries less than “its ordinary weight” (id. at p. 26). I have found no case — and the majority cites none — in which this court (or a Court of Appeal) has declined to apply the presumption because the impact on contractual rights and obligations of applying a new statute to an existing contract was insufficiently “substantial” to trigger the presumption. (Maj. opn., ante, at p. 18.) Nor have I found a case — and again, the majority cites none — in which this court (or a Court of Appeal) has applied the presumption with less than “its ordinary weight” (id. at p. 26), either for the reason the majority offers here — applying the statutes gives them retroactive effect, but only in a “technical[],” “nominal,” or “trivial” way (id. at p. 24) — or for any other reason.

In support of its view, the majority states that (1) “we have generally explained that a new law operates ‘retroactively’ when it changes ‘ “ ‘the legal consequences of past conduct by imposing new or different liabilities based upon such conduct,’ ” ’ ” and (2) “[w]e have asked whether the new law ‘ “ ‘substantially affect[s] existing rights and obligations.’ ” ’ ” (Maj. opn., ante, at p. 16.) But our precedents also have long declared that a law is “ ‘retroactive’ ” for purposes of the presumption if it “ ‘takes away or impairs vested rights acquired under existing laws . . . or give[s] a right [that] never before existed.’ ” (Davis & McMillan v. Industrial Acc. Com. (1926) 198 Cal. 631, 637–638.) Our precedents also indicate that where a law “ ‘destroy[s] or impair[s] an existing right, or give[s] a right which never before existed,’ ” the law necessarily “ ‘relate[s] to substantial rights’ ” and is therefore retroactive for purposes of the presumption. (Id. at p. 638 [“ ‘Retrospective statutes are usually considered to embrace only those which relate to substantial rights, as those which destroy or impair an existing right, or give a right which never before existed’ ”].) Consistent with this understanding, our modern decisions broadly declare that “ ‘ “ ‘[e]very statute . . . which takes away or impairs vested rights acquired under existing laws . . . , in respect to transactions or considerations already past, must be deemed retrospective.’ ” ’ ” (Strauss v. Horton (2009) 46 Cal.4th 364, 471–472, italics added.) Indeed, the majority acknowledges that at least “some of our cases” articulate a “broad[] definition” of retroactivity that “seems to embrace any conceivable statutory impact on the terms of an existing contract — including an insurance contract.” (Maj. opn., ante, at p. 16.) The majority’s analysis does not convince me that “[o]ur precedent . . . establishes [the] different,” far narrower “proposition” that retroactivity does not exist where application of “a new law” would, in fact, “impact” vested contractual rights by “chang[ing] . . . the contracting parties’ rights or obligations,” but a court decides that the “impact” on those contractual rights is not sufficiently “substantial.” (Maj. opn., ante, at p. 21.) Thus, contrary to the majority’s assertion, my reservations about its analysis and conclusion go far beyond its failure to “identify a sufficiently analogous case.” (Id. at p. 20.) Having failed to find a decision from any court applying our decisions in the way the majority does, and having thoroughly “grapple[d] with this established body of law” (id. at p. 19), it simply is not evident to me that our precedents “support[]” (id. at p. 24) the majority’s novel analysis.

It also is not evident to me that the statutes, as applied to existing policies, merely “make relatively cabined, procedural changes to how insurers administer policies,” by “requir[ing] insurers to provide policy owners with limited but critical safeguards to avoid defaulting.” (Maj. opn., ante, at p. 18.) As the majority acknowledges, “ ‘promptness of payment is essential in the business of life insurance,’ ” and “insurers depend on the regular, timely payment of premiums in order to pay death benefits and cover the cost of administering policies.” (Maj. opn., ante, at p. 9.) As noted above, the policy in this case expressly states that “coverage will cease” if the premium is not paid at the end of “[a] grace period of 31 days.” Under section 10113.71, subdivision (a), coverage must remain in force, notwithstanding the contractual provision and the nonpayment of premium, for at least 60 days. In my view, this mandatory coverage extension arguably constitutes more than a mere “procedural change[].”[14] (Maj. opn., ante, at p. 18.)

In all events, in terms of the presumption’s applicability and operation, our prior decisions eschew reliance on whether a change may be characterized as procedural rather than substantive. (Aetna Cas. & Sur. Co. v. Industrial Acc. Commission (1947) 30 Cal.2d 388, 394.) As we have explained, “ ‘In deciding whether the application of a law is prospective or retroactive, we look to function, not form. [Citations.] We consider the effect of a law on a party’s rights and liabilities, not whether a procedural or substantive label best applies.’ ” (In re Friend (2021) 11 Cal.5th 720, 743.) “In this area of the law, . . . substance and procedure are so interwoven that their attempted segregation into clean-cut categories becomes meaningless; here, as elsewhere, the hoary dichotomy between the substantive and the procedural cannot serve as a talismanic solution to the retroactivity problem.” (People v. Charles (1967) 66 Cal.2d 330, 336.) The majority fails to even acknowledge, let alone “grapple with this established body of law.” (Maj. opn., ante, at p. 19.)

I also am not convinced that under our precedents, “the ‘highly regulated’ nature of the insurance industry” (maj. opn., ante, at p. 21) is a factor that weighs against fully applying the presumption. Indeed, our decision in Interinsurance Exchange of the Auto. Club of Southern Calif. v. Ohio Cas. Ins. Co. (1962) 58 Cal.2d 142, seems to indicate precisely the contrary. That case did not, as the majority indicates, present the question of whether to apply a statutory “change [that] would have upended the bargain struck” in “previously negotiated contracts.” (Maj. opn., ante, at p. 18.) Instead, the issue in the case was whether to apply a statutory change that would have negated a contractual provision that was not expressly contained in the contract — an insurance policy — but that was “written into the policy as a matter of law” and “public policy.” (Interinsurance Exchange, at p. 146.) In answering this question, after stating the “rule” that provisions required by “the statutory and decisional law in force at the time the policy is issued . . . ‘are read into each policy . . . and become a part of the contract with full binding effect upon each party’ ” (id. at p. 148), we explained that “[b]ased upon” the presumption against retroactivity, “this rule is followed even though there has been a subsequent amendment or repeal of the statute incorporated into the policy” (id. at p. 149). Applying the presumption, we then declined to apply the new statute to existing policies, finding “nothing to indicate that the Legislature wished the [statutory] amendment . . . to have such a retroactive effect.” (Ibid.) This analysis and holding seem inconsistent with the majority’s reliance on “the ‘highly regulated’ nature of the insurance industry” (maj. opn., ante, at p. 21) as a factor weighing against application of the presumption. In my view, the “expectations” at issue under the policy here — which arose from an express contractual provision that was mandated by then-applicable administrative regulations and which would be  “disrupt[ed]” by applying the statutes to the policy — are at least as “settled” as — and arguably more settled than — the “expectations” at issue in Interinsurance Exchange, which arose from a provision read into the policy by law and which we found sufficient to trigger application of the presumption. (Maj. opn., ante, p. 19.)

Last, I note that the majority’s analysis of whether and how the presumption applies appears to overlap the inquiry that governs our analysis of whether a retroactive application of a law unconstitutionally impairs contractual rights. As we recently explained, the “threshold question” of the constitutional inquiry is whether the state law “ ‘ “operate[s] as a substantial impairment of a contractual relationship.” ’ ” (Alameda County Deputy Sheriff’s Assn. v. Alameda County Employees’ Retirement Assn. (2020) 9 Cal.5th 1032, 1075.) “In making this determination, we “ ‘consider[] the extent to which the law undermines the contractual bargain, interferes with a party’s reasonable expectations, and prevents the party from safeguarding or reinstating his rights.’ ” (Ibid.) Under the majority’s analysis, whether the presumption against retroactivity fully applies likewise turns on whether the impact of applying the new law to existing contracts is sufficiently “substantial” (maj. opn., ante, at p. 18), and the same factors likewise are considered in deciding these questions (id. at p. 17 [“focus[]” of “ ‘retroactivity’ ” inquiry is “whether the statutory change in question significantly alters settled expectations”], 18 [statutory changes here “do not disrupt clearly settled expectations” in “fashion” that makes application of statutes retroactive and applying them would not “impinge on a contracting party’s substantial rights or unfairly upset the bargain memorialized in the insurance policy”], 22 [Protective Life fails to “specifically identify any way in which the bargain memorialized in the insurance contract would be substantially upset by applying” the new statutory requirements]). Our previous decisions state that the question of whether the presumption applies is separate and different from the question of whether retroactive application of a law unconstitutionally impairs contractual rights. (Hogan v. Ingold (1952) 38 Cal.2d 802, 821 [“the question of the constitutionality of retroactive legislation and the question of the applicability of a rule” against retroactivity “are distinct”]; People ex rel. Thorne v. Hays (1854) 4 Cal. 127, 139, 131, 132 [“there is a broad difference” between the question of whether a statute “not expressly made retrospective in terms, should not be so construed as to affect past transactions” and whether retroactive application of the law unconstitutionally “divest[s] the rights of individuals vested previous to its passage”].) I am not convinced that it is appropriate to make the two inquiries similar in this way or that doing so will not have unforeseen consequences.

In summary, because it is not clear to me the majority’s analysis squares with our jurisprudence, I do not join its conclusion that the presumption against retroactivity is inapplicable or applies with less than its ordinary weight. However, I concur in the judgment because I conclude that the statutory language and relevant legislative history are sufficiently clear to overcome the presumption, assuming it applies.

JENKINS, J.

I Concur:

CORRIGAN, J.

Footnotes

  1. All unspecified section references are to the Insurance Code. ?
  2. All Internet citations in this opinion are archived by year, docket number, and case name at <http://?www.?courts.?ca.?gov/?38324.?htm>. ?
  3. The 31-day grace period Protective Life provided typifies the grace periods found in many policies. (16 Williston on Contracts, supra, § 49:80, p. 699.) California regulations at the time provided for a 31-day grace period. (Cal. Code Regs., tit. 10, § 2534.3.) ?
  4. As originally enacted, section 10113.71, subdivision (a) began with “Every” instead of “Each.” (Stats. 2012, ch. 315, § 1.) The Legislature amended the subdivision as part of a code maintenance bill making “nonsubstantive changes” to various provisions of law. (Legis. Counsel’s Dig., Assem. Bill No. 383 (2013–2014 Reg. Sess.); see also Stats. 2013, ch. 76, § 137.) ?
  5. Sections 10113.71, subdivision (b) and 10113.72, subdivision (c) frame their notice obligations as requirements insurers must observe before terminating policies. Section 10113.71, subdivision (a) arguably operates differently because it mandates that policies include a 60-day grace period provision. That particular way to frame the policies, and the fact that McHugh’s policy already contained a 31-day grace period, may imply to some observers that the subdivision operates here by formally altering an existing contract. But given that the provision does not operate on already-defaulted policies before enactment, it’s not clear that it differs in any substantive way from a provision simply requiring all insurers to, going forward, observe a pretermination 60-day grace period. ?
  6. We recognize that, theoretically, some number of policy owners might deliberately allow a policy to lapse rather than cancelling it to maintain coverage during the grace period. ?
  7. Section 10113.72, subdivision (b) differs from its precedent subdivision in three ways: It (1) applies to a different rightsholder (i.e., a policy owner, versus an applicant); (2) describes the exercise of designation at a different point in time (i.e., after the policy owner’s initial application, a timing which permits both pre- and postenactment policy owners to exercise their designation); and (3) creates a different obligation for insurers (i.e., to notify policy owners annually regarding their designation right, as opposed to providing applicants with a designation form). (§ 10113.72, subds. (a), (b).) ?
  8. The Court of Appeal relied on Ball v. California State Auto. Assn. Inter-Ins. Bureau (1962) 201 Cal.App.2d 85, 87, to read “ ‘issued or delivered’ ” as “customar[il]y” embracing only postenactment policies. (McHugh, supra, 40 Cal.App.5th at p. 1176; see id. at p. 1175.) But nothing in our case law suggests that Ball, which concerned a statute impacting automobile liability policies, provides the type of “definitive judicial construction” of the phrase “issued or delivered” that we can presume the Legislature knew of and sought to adopt here. (Foley v. Interactive Data Corp. (1988) 47 Cal.3d 654, 675.) ?
  9. For this reason, we have no occasion to entertain another of Protective Life’s arguments: that the Legislature’s failure to enact the sections as part of urgency legislation cuts against rebutting the presumption against retroactivity. ?
  10. Plaintiffs argue that applying sections 10113.71 and 10113.72 to the facts here requires us to conclude they are entitled to recover the policy benefits from Protective Life. But this argument would require us to address the correctness of the jury’s verdict. We decline to do so, as plaintiffs did not petition for our review on this issue and it is not squarely before us. (Nationwide Biweekly Administration, Inc. v. Superior Court. (2020) 9 Cal.5th 279, 334, fn. 25.) ?
  11. All unspecified section references are to the Insurance Code. ?
  12. The majority asserts that “our cases do not definitively indicate that the presumption has been rebutted” in this case. (Maj. opn., ante, at p. 26). But my analysis and conclusion are fully in line with, and supported by, our analysis and conclusion in Preston, Calfarm, and Alford, and the majority does not assert otherwise. ?
  13. Given the majority’s alternative holding that the presumption applies and carries some weight, its choice to address the question of whether applying the statutes here involves retroactivity ultimately does not, as the majority asserts, “avoid having to apply” the presumption “in a circumstance where it’s not necessary.” (Maj. opn., ante, at p. 26.) ?
  14. Curiously, despite the majority’s acknowledgement that prompt and timely payment “ ‘is essential’ ” to insurers (maj. opn., ante, at p. 9), the majority later rests its conclusion in part on Protective Life’s failure to “clearly” show how extending the grace period “constituted a disruptive contract change” (maj. opn., ante, at p. 24), its failure to “specifically identify any way in which the bargain memorialized in the insurance contract would be substantially upset by applying the [extended] grace period” (id. at p. 22), and its reliance instead on a “generalized, amorphous allusion to financial impact” (ibid). According to the majority, Protective Life can show a sufficient “financial impact” only by providing “evidence that [it] or other insurers anticipated and accounted for a projected rate of inadvertent defaults when setting their premiums.” (Id. at p. 39.) ?