BACKGROUND

Pep Boys sells automotive products at its stores nationwide, with over 145 stores in California. Hundreds of people, including more than 500 in California, filed claims against Pep Boys alleging harm from exposure to asbestos in products Pep Boys sold. In 2004, Pep Boys sought coverage for hundreds of these claims from insurers who sold a “tower” of commercial general liability policies providing coverage between February 1, 1981, and July 1, 1982. At the base of the tower, Protective National Insurance Company of Omaha (Protective), provided primary coverage. The policy originally covered from February 1, 1981, to February 1, 1982, up to a limit of $500,000 per occurrence and $500,000 in the aggregate “during each annual period while this policy is in force commencing from its effective date.” Protective later extended the policy period by endorsement to June 30, 1982, in exchange for an additional prorated premium. New England Reinsurance Corporation (New England) provided the first layer of umbrella coverage. New England’s policy initially covered from February 1, 1981, to June 30, 1982, up to a limit of $10,000 per occurrence and $10 million “in the aggregate for each annual period.” The policy described its aggregate limit as an “annual aggregate limit . . . on account of

2 all occurrences during each policy year . . . .” (Capitalization omitted.) In exchange for an additional prorated premium, New England later extended its policy to July 1, 1982. Old Republic and American Excess shared the second layer of excess coverage. As originally written, the Old Republic policy covered the period from February 1, 1981, to June 30, 1982. It provided excess coverage up to $10 million per occurrence and $10 million “in the aggregate for each annual period during the currency of this policy.” A letter from Pep Boys’ broker submitting the application for the insurance said that Pep Boys wanted the policy to cover 17 months to “get [its] insurance program concurrent with [its] fiscal year end accounting.” The broker calculated the premium due for the 17 months by prorating the 12-month premium. The premium Old Republic charged was consistent with the broker’s calculation. Old Republic later extended the policy period by one day to July 1, 1982, for an additional prorated premium. American Excess issued the other policy in the second layer, and it originally covered from February 1, 1981, to February 1, 1982. The policy states that American Excess’s liability would be “limited, where and as applicable, to the amount stated” in the declarations as the insurer’s “ ‘aggregate’ with respect to loss excess of the Underlying Insurance which occurs during the term of this Certificate,” which was $5 million. An endorsement later extended the policy to July 1, 1982, for $11, which was one day’s prorated premium.1 This endorsement also

1 The obvious implication of this one-day endorsement to July 1, 1981, is that there was a previous endorsement that extended the term from February 1 to June 30, 1982. The record does

3 states that “the total premium is amended to $5,171.00 and the total annual premiums [remain] $3,665.00.” Fireman’s Fund participated in the third and final layer of excess coverage. Its policy covered from April 3, 1981, to July 1, 1982. It covered up to $15 million per occurrence and $15 million aggregate “for all damages sustained during each annual period of this policy” as part of a total third layer of $25 million. Another insurance company that is now insolvent provided the balance of coverage in the third layer. After Pep Boys demanded coverage, Protective became insolvent and went into receivership. But it agreed to provide two $500,000 aggregate limits payments, one for the period from February 1, 1981, to February 1, 1982, and a second for the period from February 1, 1982, to June 30, 1982. Although it had taken the position that Protective owed two aggregate limits payments, New England nonetheless asserted that its own policy provided only one aggregate annual limit. Accordingly, after New England paid $10 million, it notified Pep Boys that its policy was exhausted. Old Republic, American Excess, and Fireman’s Fund took the same position regarding their respective policies. Pep Boys therefore filed a declaratory judgment action against all four insurers, seeking a ruling that each policy provided two aggregate annual limits, one for the first 12 months of the policies and one for the remaining period.

not contain a copy of this endorsement, but nothing turns on its language so the omission is not significant.

4 Pep Boys moved for summary adjudication of these coverage issues as to each insurer. Fireman’s Fund cross-moved for summary judgment, arguing that Pennsylvania law governed the interpretation of the policy and under Pennsylvania law all of the claims against Pep Boys arose from one occurrence. According to Fireman’s Fund, the per occurrence limit in its policy therefore capped Fireman’s Fund’s liability at $15 million, regardless of how the court construed the aggregate limits provision. Pep Boys reached a settlement with New England and dismissed it before the hearing on the motions. The trial court denied Pep Boys’ motion, ruling that the policies of Old Republic, American Excess, and Fireman’s Fund each provided only a single aggregate limit. The court denied Fireman’s Fund’s motion as moot. The parties agreed that the court’s ruling fully disposed of Pep Boys’ claims, so they stipulated to entry of judgment.

DISCUSSION

A party is entitled to summary adjudication if there is no triable issue as to any material fact and the matter can be adjudicated as a matter of law. (London Market Insurers v. Superior Court (2007) 146 Cal.App.4th 648, 655.) “On appeal, we independently review the trial court’s ruling and apply the same legal standard that governs the trial court.” (Ibid.) The trial court’s reasons for its ruling “are not binding on us because we review its ruling, not its rationale.” (Ram’s Gate Winery, LLC v. Roche (2015) 235 Cal.App.4th 1071, 1079.) We “must affirm on

5 any ground supported by the record.” (Jimenez v. County of Los Angeles (2005) 130 Cal.App.4th 133, 140.) “ ‘ “While insurance contracts have special features, they are still contracts to which the ordinary rules of contractual interpretation apply.” [Citations.] “The fundamental goal of contractual interpretation is to give effect to the mutual intention of the parties.” [Citation.] “Such intent is to be inferred, if possible, solely from the written provisions of the contract.” [Citation.] “If contractual language is clear and explicit, it governs.” [Citation.]’ [Citation.] “ ‘ “A policy provision will be considered ambiguous when it is capable of two or more constructions, both of which are reasonable.” [Citations.] . . . “ ‘[L]anguage in a contract must be construed in the context of that instrument as a whole, and in the circumstances of that case, and cannot be found to be ambiguous in the abstract.’ ” [Citation.] “If an asserted ambiguity is not eliminated by the language and context of the policy, courts then invoke the principle that ambiguities are generally construed against the party who caused the uncertainty to exist (i.e., the insurer) in order to protect the insured’s reasonable expectation of coverage.” [Citation.]’ [Citation.] [¶] . . . [S]tandard form policy provisions are interpreted under the same rules of construction. ‘ “[W]hen they are examined solely on a form, i.e., apart from any actual agreement between a given insurer and a given insured, the rules stated above apply mutatis mutandis. That is to say, where it is clear, the language must be read accordingly, and where it is not, in the sense that satisfies the hypothetical

6 insured’s objectively reasonable expectations.” ’ ” (Powerine Oil Co., Inc. v. Superior Court (2005) 37 Cal.4th 377, 390–391 (Powerine Oil).) Because the language of each of the policies at issue is different, we examine the policies each individually. a. Old Republic I.Policy language Old Republic’s policy establishes the benefits limit as $10 million “ultimate net loss in all in respect of each occurrence subject to a limit of” $10 million “in the aggregate for each annual period during the currency of this policy.” The term of the policy is 17 months, from February 1, 1981, to July 1, 1982. There is no question that Old Republic must pay up to $10 million for the first 12 months of the term. The parties disagree over how to apply the reference to “each annual period” to the remaining five months in the policy term. Old Republic treats “annual period” as applying to the entire 17-month term of the policy, such that the policy provides a total of $10 million of benefits, while Pep Boys reads the policy term as consisting of two annual periods, one for the first 12 months and the second for the remaining 5 months, so that the policy provides a total of $20 million of benefits. The policy language cannot be applied literally as written. “Annual” as a modifier of “period” means “covering the period of a year.” (Merriam-Webster Dict. Online (2023) [as of December 28, 2023].) Both parties ask us to apply “annual

7 period” to terms more than or less than a year: 17 months, in Old Republic’s view, or 5 months, under Pep Boys’ approach. As a textual matter, neither reading accords with the literal meaning of “annual,” and neither is more reasonable than the other. We turn next to the limited extrinsic evidence that the parties have provided, to determine whether the circumstances of this case clarify the matter. (See Powerine Oil, supra, 37 Cal.4th at p. 391.) According to the letter from its broker to Old Republic’s representative, Pep Boys wanted a 17-month policy because it wanted to align the expiration of its insurance policies with its fiscal year. Nowhere is there any suggestion in the letter that Pep Boys wanted to reduce the costs for its insurance or make any changes to the level of its coverage. Pep Boys’ desire merely to extend its insurance, rather than reduce its scope or expense, together with the fact that Pep Boys paid a prorated premium, suggests it intended to receive the same level of coverage as it had been, rather than diluting it. This evidence persuades us that Pep Boys intended to receive $10 million in protection during the last 5 months of the term just as it had for the first 12 months. Old Republic’s approach, by contrast, dilutes Pep Boys’ coverage by spreading the same aggregate limit over 17 months. This might be reasonable if the extrinsic evidence showed that Pep Boys wanted to reduce its coverage or reduce costs. Of course, in such a scenario one would also expect the amount of the premium for the extension to be something less than a simple proration, given that the increased time on the risk would not

8 increase the insurer’s maximum exposure. (Board of Trustees of University of Illinois v. Insurance Corp. of Ireland, Ltd. (N.D.Ill. 1990) 750 F.Supp. 1375, 1383–1384 [discussing premiums insurers would be expected to charge for different lengths and limits of policies], affd. (7th Cir. 1992) 969 F.2d 329.) But since Pep Boys merely intended to extend the date of a policy for administrative convenience and paid a straight prorated premium to do so, it is not reasonable to reduce the amount of Pep Boys’ benefits. This extrinsic evidence may not be definitive, but any remaining ambiguity must be construed against Old Republic as the party who caused the uncertainty and in favor of Pep Boys’ reasonable coverage expectations. (Powerine Oil, supra, 37 Cal.4th at p. 391 [“ ‘ “ambiguities are generally construed against the party who caused the uncertainty to exist (i.e., the insurer) in order to protect the insured’s reasonable expectation of coverage” ’ ”].) Reading the policy as containing two aggregate limits periods is consistent with Pep Boys’ reasonable coverage expectations. Old Republic’s interpretation, by contrast, would be surprising, both to Pep Boys and to any other insured who bought such an insurance policy for a period between one and two years. In Old Republic’s view, the “annual period” clause was intended to make sure that no new aggregate limit would apply to a period of less than one year. In effect, this means that an insured buying or extending a policy for 1 year and 364 days in exchange for a prorated annual premium — which would be almost exactly

9 double the premium for a single year — would have one year’s aggregate limit diluted over 1 year and 364 days, thereby receiving, in effect, half the benefits than if it had purchased two full-year policies for almost the same amount of money. No one would expect this result. If Old Republic intended its policy to operate in this fashion, either for Pep Boys or any other insured, it should have written its policy or fashioned an endorsement that made it clear. It failed to do so and must now abide by the policy it wrote.2 We recognize that our consideration of Pep Boys’ expectations for its insurance coverage is artificially constrained because Pep Boys only seeks a ruling on its insurance policies covering some or all of the period between February 1981 and July 1, 1982. Pep Boys has not provided any indication of whether it had insurance for the remainder of 1982, or the coverage or limits of such insurance. If Pep Boys did obtain insurance for a period following July 1, 1982, and that insurance covered asbestos claims, our interpretation of Old Republic’s policy as providing a full aggregate limits benefit during the period from March 1 to July 1, 1982, could lead to Pep Boys receiving more coverage than it expected for the calendar year of 1982. Conversely, however, adopting Old Republic’s interpretation could lead to Pep Boys effectively having a gap in coverage. “Neither result seems wholly fair, but it is not clear that there is a wholly fair result that is possible under the

2 We note that Protective, Pep Boys’ primary insurer, had policy language similar to Old Republic’s and made two aggregate limit payments for its policy that was extended from 12 to 17 months.

10 policy’s language.” (Union Carbide Corp. v. Affiliated FM Ins. Co. (N.Y. 2011) 947 N.E.2d 111, 114; see id. p. 115 [finding that factual disputes precluded summary judgment regarding amount of aggregate limits applicable to two-month extension of policy]; but see Stonewall Ins. Co. v. National Gypsum Co. (S.D.N.Y., July 29, 1992, No. 86 CIV. 9671 (JSM)) 1992 WL 188433, *1 [giving insured full limits benefit of two separate policies each in effect for six months of a year arguably gives insured a windfall, but “it cannot be disputed that each of the insurers is simply being held to its contract”], affd. in part & revd. in part on other grounds sub nom. Stonewall Ins. Co. v. Asbestos Claims Management Corp. (2d Cir. 1995) 73 F.3d 1178 (Stonewall).) The most conceptually satisfying resolution of this case would allow us to avoid incongruous results by considering Pep Boys’ insurance policies as a whole throughout all periods at issue in the asbestos claims against it, in the hopes of finding a resolution that gives Pep Boys the benefits of all of its insurance policies without gaps or double coverage. Such a theoretically fair resolution might involve pro rata calculation of limits, to bridge coverage between different policies. However satisfying in theory, such a resolution is impossible in practice, in this case or any other. Even if Pep Boys intended to create a cohesive framework of multiple layers of insurance coverage for all periods of its business operations, we interpret insurance policies, not multi-year, multi-layer insurance policy frameworks, and we must apply each policy’s language as written. Even the few policies at issue here do not

11 line up cleanly, since Old Republic’s and American Excess’s policies started on February 1, 1981, while Fireman’s Fund’s policy did not begin until April 3, 1981. And as we determine post, some policies’ language sets aggregate limits for each annual period in a policy term or fraction thereof, while other policies impose one aggregate limit for the entire term of the policy, even if it stretches beyond a year. Meanwhile, no policy in this case or any other we have discovered explicitly allows for proration of policy limits. As a practical matter, then, the vagaries of the different language and terms of different policies effectively foreclose any possibility of uniform layers and years of coverage, and proration cannot solve the issue. II.Decisions from other jurisdictions While this is, somewhat remarkably, the first case involving this issue to arise in California, many courts across the country have previously confronted similar issues. The decisions on point appear to be roughly evenly split. (See Seaman & Schulze, Allocation of Losses in Complex Insurance Coverage Claims (2023) § 8:3 [collecting cases].) The cases congruent with our analysis of the policy language and circumstances here are better reasoned and more on point. Stonewall, supra, 73 F.3d at pages 1216–1218 considered three sets of policies. The first two policies established limits “in the aggregate for each annual period during the currency of this policy” and did not define “annual period.” (Id. at pp. 1216– 1217.) One of these was an approximately 14-month policy canceled after about 8 months for a prorated refund of the

12 premium, and the 12-month term of the second policy was extended for almost 4 more months for an approximately prorated premium.3 (Ibid.) The appellate court affirmed the district court’s ruling that full aggregate limits benefits applied to the shortened term and extension period, respectively. (Ibid.) It noted that the policies did not address the effect of a fractional period on the aggregate limits and construed the resulting ambiguity against the insurers. (Ibid.) The court also noted that the prorated premiums for the shorter periods were in exchange for a reduction for the time the insurer was on the risk. (Id. at p. 1217.) The insured’s third set of policies were excess policies that the insured canceled approximately a year and nine months into their three-year terms and replaced with policies from the same insurers with higher limits. (Stonewall, supra, 73 F.3d at p. 1217.) The policies stated the aggregate limits applied “for each annual period where applicable,” with “annual period” defined as “each consecutive period of one year commencing from the inception date” of the policies.4 (Ibid.) Stonewall affirmed the district court’s ruling that the insured intended only to increase its coverage by replacing the policies, not to obtain separate coverages, so it did not give the insured the benefit of a separate aggregate limit benefit during the initial nine-month period before the cancellations. (Id. at pp. 1217–1218.)

3 The premium for the almost four-month extension was about 41% of the full year’s premium, so the premium charged was actually slightly higher than a strictly prorated amount. (See Stonewall, supra, 73 F.3d at p. 1217.) 4 The first excess policy included this language, and the second incorporated it by reference. (Stonewall, supra, 73 F.3d at p. 1217.)

13 Stonewall supports Pep Boys’ argument here, especially since the operative policy language in the first two policies at issue there was identical to that of Old Republic’s policy. If Pep Boys had chosen to align its insurance policies with its fiscal year by buying two policies for successive one-year terms and canceling one of them after only five months, or by purchasing a one-year policy and then paying extra to extend the term by five months, Stonewall would require Old Republic to provide a full aggregate limits payment for the fractional period. (Stonewall, supra, 73 F.3d at pp. 1216–1217; accord, USM v. American Ins. Co. (N.J.Super.Ct.App.Div. 2002) 792 A.2d 500, 507–508, 518– 525 [following Stonewall; two-week extension of policy created additional aggregate limit of liability, in part because premium for extension was prorated]; IMO Industries v. Transamerica (N.J.Super.Ct.App.Div. 2014) 101 A.3d 1085, 1110 [following USM v. American Ins. Co.; pro-rating policy limits would reduce insurer’s risk twice, “once by its time on the risk and a second time by the pro-rating of the policy limit”]; Board of Trustees of University of Illinois v. Insurance Corp. of Ireland, Ltd., supra, 750 F.Supp. at p. 1384 [prorated premium for extension of term supports application of an additional aggregate limit period].) There is no evident reason to apply “annual period” differently merely because almost all of the fractional period was included in the policy term when Pep Boys first bought its policy and paid the additional prorated premium. Old Republic offers no reason to treat fractional periods differently depending on whether they

14 arise by cancellation, extension, or creation at the outset, when the premium is prorated in all three scenarios. Stonewall’s affirmance of the lower court’s ruling regarding the third set of excess policies that the insured replaced with higher limit policies is not inconsistent with the rest of its analysis or our analysis here. The touchstone for the interpretation of ambiguous policies remains the insured’s reasonable expectations, so providing only one aggregate limit in such circumstances was reasonable. (Stonewall, supra, 73 F.3d at pp. 1217–1218.) Because there is no indication that Pep Boys bought extended policies in order to change the amount of its coverage, as opposed to obtaining the same amount of coverage for a new period, it is reasonable to give it the benefit of two full periods’ worth of aggregate limits. Old Republic dismisses Stonewall and IMO Industries because prorating aggregate limits would have made the per occurrence limits of the policies larger than the aggregate limits, which is “logically incoherent.” Another case that Pep Boys cites, Unigard Sec. Ins. Co. v. North River Ins. Co. (S.D.N.Y. 1991) 762 F.Supp. 566, 595–596, affd. in part & revd. in part on other grounds (2d Cir. 1993) 4 F.3d 1049, did refuse to prorate aggregate limits in part because of this interplay between per occurrence and aggregate limits. But Unigard also noted that the proration of premiums “reflects the shortened length of time for which the insurer is exposed to the risk of loss, not a reduced quantum of protection available if the risk materializes in the stub period, however short it may be.” (Unigard, at p. 596.) In

15 any event, neither Stonewall nor IMO Industries cited Unigard or discussed the relationship between per occurrence and aggregate limits. Old Republic also argues Stonewall has no application here because it did not consider whether a single aggregate limit should apply to a period of more than 12 but less than 24 months. This is not correct, since Stonewall applied a full aggregate limit to the second policy at issue there that was extended to 16 months. (Stonewall, supra, 73 F.3d at p. 1217.) The insurer may only have advocated for prorating and not for eliminating entirely the aggregate limit benefit during the extension period, but this distinction is immaterial. (Ibid.) Stonewall’s denial of an attempt to prorate the aggregate limit, which would still have given the insured some additional coverage during the extension, indicates that it would have rejected an attempt to provide no additional coverage at all. Old Republic cites several decisions from other jurisdictions that reached different results from those cases cited ante, but they do not persuade us.5 UNR Industries, Inc. v. Continental Ins. Co. (N.D.Ill. 1988) 682 F.Supp. 1434, 1457–1459, considered whether policies issued for a period shorter than one year or canceled less than one year into a three-year term should have prorated aggregate limits. The policies declared that their

5 Old Republic also cites an unpublished superior court ruling. Rule 8.1115(a) of the California Rules of Court prohibits citation of unpublished Court of Appeal or superior court appellate division opinions. Old Republic asserts this does not prohibit citation of superior court rulings for their persuasive value. This hyper-technical reading of the rule is nonsensical. It would be absurd to prohibit citation of appellate court rulings, even for their persuasive value, but permit citation to decisions from trial courts for persuasive value. (See Bolanos v. Superior Court (2008) 169 Cal.App.4th 744, 761 [citation to trial court order as legal authority was improper].)

16 aggregate limits applied “separately to each annual period.” (Id. at p. 1458.) The district court concluded the policies provided full aggregate limits benefits and refused to prorate, which is consistent with Stonewall and our approach here. (Id. at p. 1459.) But the district court also ruled that another policy whose aggregate limits were defined “for each annual period during the currency of the policy” and whose term was three years and five weeks provided three aggregate limits periods, not four. (Id. at pp. 1459–1460.) The court interpreted what it considered to be the policy’s plain meaning and did not consider the insured’s reasonable expectations. (Ibid.) The court relied on the fact that the policy required premiums to be paid in installments on the anniversary of the policy and defined the “first anniversary” of the policy as a date one year and five weeks after the beginning of the policy. (Id. at p. 1459; accord, UNR Industries, Inc. v. Continental Ins. Co. (N.D.Ill., Nov. 9, 1988, Nos. 85 C 3532, 83 A 2523) 1988 WL 121574, *2–*3 [same court reaching identical conclusion as to policy extended by endorsement for two months where endorsement amended anniversary date of the policy to 14-month period].) This approach is doubtful because, as we conclude post regarding one of the policies here, annual calculation of premiums does not necessarily control the determination of aggregate limits. In any event, this aspect of the decision is not applicable here because Old Republic’s policy does not have any language defining the 17-month term of the policy as one annual period, either for premium calculation or any other purpose.

17 William Powell Co. v. OneBeacon Ins. Co. (Ohio Ct.App. 2016) 75 N.E.3d 909, 918–919, considered whether one policy issued for three years but canceled after 14 months and another issued for one year but extended by endorsement for an additional 32 days had one aggregate limit each or two. One policy declared that if it were “ ‘issued for a period of three years, the limits of the company liability shall apply separately to each consecutive annual period thereof.’ ” (Id. at p. 918, italics omitted.) The other policy stated that “ ‘[i]f this policy is issued for a period of three years any limit of the company’s liability stated in this policy as “aggregate” shall apply separately to each consecutive annual period thereof.’ ” (Id. at p. 919.) The court nonetheless concluded that there were no consecutive annual periods because the policies were only for a one-year period followed by a one-month or two-month period, not consecutive annual periods. (Ibid.) This analysis simply assumes the conclusion that an annual period must be a full 12-month period. The court also noted that the second policy was not originally issued for three years. (Ibid.) This focus on the duration of the policies’ original terms would suggest that if a one-year policy were extended to three years by endorsement there would still be only a single annual period. We find none of this reasoning persuasive. Finally, the policy in Gen. Refractories v. Ins. Co. of N. America (Pa.Super.Ct. 2006) 906 A.2d 610, 611 (General Refractories) said that if it were issued for three years, then the limits of liability would “ ‘apply separately to each consecutive

18 policy year thereof.’ ” The insured’s policy was written for three years and extended by endorsement by one month for a prorated premium because the cost of a renewal policy from the insurer had risen significantly and the insured wanted to find a new policy from a different insurer. (Ibid.) The appellate court affirmed the trial court’s ruling that the policy was not ambiguous, extrinsic evidence was not necessary to interpret it, and the extension served only to expand the term of the policy, not to provide a fourth limits period. (Id. at pp. 612–613.) We disagree with this reasoning as explained ante. However, we note that if General Refractories had consulted the extrinsic evidence, the fact that the insurer provided the extension as an accommodation to the insured in lieu of charging the increased rate for the renewal policy could have supported the outcome the court reached. (Id. at p. 611.) In such circumstances, the insured likely could not have reasonably expected a full aggregate limit benefit during the extension period, since the prorated premium the insured paid was significantly lower than the going rate the insurer would have charged for a new policy. b. American Excess We turn next to American Excess, whose policy provided the balance of Pep Boys’ second layer of excess insurance coverage. American Excess’s policy demonstrates the effect that different policy language will have. The policy defines the $5 million limit set forth in the declarations as applying “with respect to loss excess of the Underlying Insurance which occurs during the term of this Certificate.” This language is

19 unambiguous. The limits of the policy are set for the entire duration of the policy, not based on annual periods within the policy term. Accordingly, when American Excess extended the term of the policy from one year to 17 months, the same limit set in the declaration page applied to the new, extended term. Pep Boys’ reasonable expectations of coverage play no role in the interpretation of American Excess’s policy because its language is unambiguous. Amazingly, Pep Boys does not even quote this limits language in its briefing. Instead, it directs us to the language of the endorsement that extended the policy term. After establishing the new policy expiration date, the endorsement states, “It is further agreed and understood that the total premium is amended to $5,171.00 and the total annual premiums [remain] $3,655.” Pep Boys construes this as meaning the premiums were calculated on an annual basis. Pep Boys then lumps in American Excess with the other insurers and says that the references to “annual” in all three policies indicate an intent to provide separate aggregate limits for each annual period in the policies. Pep Boys’ attempt to treat the premium calculation in the endorsement as governing the limits provision in the policy is unavailing. American Excess disputes whether the premiums were calculated separately for the annual periods within the policy. But even if they were, it would be insufficient to overcome the policy’s plain definition of its limits as applying to the entire policy period. The endorsement dispels any uncertainty on the

20 matter by stating (in more language that Pep Boys ignores) that it does not “alter, vary, or extend any provision or condition” of the policy “other than as above stated.” Since the endorsement amended only the expiration date of the policy, it did not affect the definition of the limits of the policy as applying throughout the policy term. c. Fireman’s Fund A. Policy language Fireman’s Fund’s policy defines its aggregate limits using the phrase “annual period,” so our interpretation of Fireman’s Fund’s policy tracks that of Old Republic’s. The policy sets a $15 million aggregate limit “for all damages sustained during each annual period of this policy.” Fireman’s Fund argues that there is only a single aggregate limits period because the policy was “written for only one annual period—a single 15-month policy period” and the premium was set as a flat charge. This argument subtly tries to conflate “annual period” with “policy period,” but the policy uses the former term and not the latter. And as with Old Republic’s policy, Fireman’s Fund’s approach of treating the 15-month term of the policy as a single annual period is no more consistent with the literal meaning of “annual period” than Pep Boys’ view that the three months remaining after the first year of the policy are a separate annual period. This policy is ambiguous as well. The same extrinsic evidence we considered ante comes into play again here. The letter from Pep Boys’ broker to the Old Republic representative states that Pep Boys wanted to get its

21 “insurance program” concurrent with its fiscal year. The reference to Pep Boys’ “insurance program,” together with the fact that both Old Republic’s and Fireman’s Fund’s policies expired on the same date of July 1, 1982, indicates that Pep Boys intended its Fireman’s Fund policy to last more than one year for the same reasons as its Old Republic policy. Because there is no indication that Pep Boys intended to reduce coverage or save money, its reasonable expectation was that it was obtaining a new full aggregate limits benefit for the additional three-month period. But even if we disregard this extrinsic evidence as pertaining only to Old Republic, the result would be the same under the rule that ambiguities in an insurance policy are construed against the insurer. Fireman’s Fund cites only two cases that Old Republic did not. Uniroyal Inc. v. American Re-Insurance Co. (N.J.Super.Ct.App.Div., Sept. 13, 2005, No. A-6718-02T1) 2005 WL 4934215, *13, *19–*22, held that a one-month extension of a three-year policy with aggregate limits “for each annual period during the currency of” the policy did not create an additional aggregate limit benefit. Uniroyal found the language of the relevant policy unambiguous, which we disagree with as explained ante. (Id. at *21.) The court also misconstrued Board of Trustees of University of Illinois v. Insurance Corp. of Ireland, Ltd., supra, 969 F.2d at page 334, stating it reformed a policy so that the extension period was treated as a new policy. (Uniroyal, at *22.) University of Illinois did affirm the district court’s reformation of the policy at issue, but the reformation was to

22 make the single policy conform to the parties’ expectation that the initial policy period and its extension would each have separate aggregate limits periods, not to create two separate policies. (University of Illinois, at p. 334 [“the parties to this contract have clearly expressed their intent that separate $5,000,000 limits apply to both the stub and one-year periods, and that the facts demonstrate that the policy as written does not reflect that intent,” italics added].) The district court was also clear that it would have reached the same result because of the prorated premium, even in the absence of facts to support reformation. (Board of Trustees of University of Illinois v. Insurance Corp. of Ireland, Ltd., supra, 750 F.Supp. at p. 1384 [“it makes no difference that the parties did not talk about the coverage issue — the result flows from the one item of totally objective evidence, the premium itself”].) Moreover, there was significant extrinsic evidence in Uniroyal that is not present here. Providing a full aggregate limits benefit for the extension would have given the insured far more coverage than it bargained for, given its documented aims for its general insurance program. (Uniroyal, supra, 2005 WL 4934215 at *13, *20 [Uniroyal intended to secure $100 million in insurance annually, so construing the thirty-day extension period to provide an additional $30 million in coverage “would have created a far larger overall insurance coverage for that time period than Uniroyal bargained for”].) The insurers would not have agreed to the extension had they known the insured wanted an additional aggregate limit benefit. (Id. at *20.) And, perhaps

23 most importantly, as in General Refractories, the insured obtained the extension for a prorated extension of the prior premium because the insurer would have charged the insured a greatly increased premium to renew the policy. (Ibid.) Under these circumstances, we might also have concluded that an additional limits benefit would have been inconsistent with the insured’s reasonable expectations. But there is no similar evidence here. Diamond Shamrock Chemicals v. Aetna (N.J.Super.Ct.App.Div. 1992) 609 A.2d 440, 468–469, held that a one-month extension of a three-year policy did not provide an additional per occurrence limit, where the policy did not define the per occurrence limit in annual terms. Because the aggregate limit in Fireman’s Fund’s policy, like Old Republic’s, is unambiguously calculated on an annual basis, this opinion is inapposite. B. Choice of law As an alternative basis for affirming the trial court’s denial of Pep Boys’ motion for summary adjudication, Fireman’s Fund argues that the trial court should have granted Fireman’s Fund’s motion for summary judgment. Fireman’s Fund argued in that motion that Pennsylvania law governs and considers all the asbestos claims against Pep Boys to arise from a single occurrence, so that the policy’s per occurrence limit cuts off Fireman’s Fund’s liability regardless of the number of aggregate limits at issue. The trial court denied Fireman’s Fund’s motion

24 as moot when it ruled that Fireman’s Fund’s policy had only one aggregate limit. There is some uncertainty regarding the proper choice of law analysis for interpreting contracts like insurance policies. “In California, ‘general choice-of-law rules have been formulated by courts through judicial decisions rendered under the common law, rather than by the legislature through statutory enactments.’ [Citation.] As the forum state, California will apply its own law ‘unless a party litigant timely invokes the law of a foreign state.’ [Citation.] “To determine which jurisdiction’s law will govern, a trial court applies the governmental interest test, which sets out a three-step inquiry: ‘First, the court determines whether the relevant law of each of the potentially affected jurisdictions with regard to the particular issue in question is the same or different. Second, if there is a difference, the court examines each jurisdiction's interest in the application of its own law under the circumstances of the particular case to determine whether a true conflict exists. Third, if the court finds that there is a true conflict, it carefully evaluates and compares the nature and strength of the interest of each jurisdiction in the application of its own law “to determine which state’s interest would be more impaired if its policy were subordinated to the policy of the other state” [citation], and then ultimately applies “the law of the state whose interest would be the more impaired if its law were not applied.” ’ ” (Chen v. Los Angeles Truck Centers, LLC (2019) 7 Cal.5th 862, 867–868.)

25 Despite the statement in Chen that general choice of law rules are established in judicial decisions and not by statute, Fireman’s Fund directs us to Civil Code section 1646,6 which states, “A contract is to be interpreted according to the law and usage of the place where it is to be performed; or, if it does not indicate a place of performance, according to the law and usage of the place where it is made.” Our Supreme Court has not cited this statute after it adopted the governmental interest approach summarized in Chen. (See Reich v. Purcell (1967) 67 Cal.2d 551, 555 [adopting governmental interest approach in tort case]; Bernkrant v. Fowler (1961) 55 Cal.2d 588, 594–596 [examining governmental interests in contract case]; Beneficial Fire & Casualty Ins. Co. v. Kurt Hitke & Co. (1956) 46 Cal.2d 517, 526 [last California Supreme Court case to cite § 1646].) Frontier Oil Corp. v. RLI Ins. Co. (2007) 153 Cal.App.4th 1436, 1447–1461, engaged in a lengthy analysis of the histories of section 1646 and the governmental interest test for choice of law problems. Frontier Oil concluded that section 1646 is a choice of law rule (although the California Supreme Court has never cited it as such) and the California Supreme Court has not judicially abrogated it or limited it by interpretation. (Frontier Oil, at pp. 1449, fn. 5, 1459–1460.) Frontier Oil sought to harmonize the statute and case law by concluding that “the choice-of-law rule in Civil Code section 1646 determines the law governing the interpretation of a contract, notwithstanding the application of the governmental interest analysis to other choice-of-law issues.”

6 Subsequent undesignated statutory citations are to the Civil Code.

26 (Id. at p. 1459; but see Textron Inc. v. Travelers Casualty & Surety Co. (2020) 45 Cal.App.5th 733, 743, 747–748 [using governmental interest test for choice of law analysis of insurance policy, despite earlier citing Frontier Oil]; Code Civ. Proc., § 1857 [“The language of a writing is to be interpreted according to the meaning it bears in the place of its execution, unless the parties have reference to a different place”]; Frontier Oil, at p. 1448, fn. 4 [declining to decide whether Code Civ. Proc., § 1857 is a choice of law rule or merely interpretive rule].) However, Frontier Oil never examined the other state’s law that arguably applied, so it is not clear whether the two tests would have led to different outcomes. (Frontier Oil, at p. 1462.) Given that Frontier Oil later concluded that California and the other state’s law did not materially differ when it came time to apply the policy provision at issue, Frontier Oil’s entire discussion, however well-reasoned or well-researched, may well be dicta. (Id. at pp. 1464–1466.) In any event, we need not wade into these deep waters and determine which choice of law analysis governs. We will assume for the sake of argument that Pennsylvania and California law materially differ in how they construe “occurrence” in insurance policies, as Fireman’s Fund argues. We will likewise assume Frontier Oil’s analysis is correct and that we must start with section 1646 to determine which state’s law to use to interpret the Fireman’s Fund policy. Even granting Fireman’s Fund these premises, Fireman’s Fund has not shown that Pennsylvania law should apply.

27 Under section 1646, Fireman’s Fund’s policy must be interpreted “according to the law and usage of the place where it is to be performed,” if it indicates a place of performance. Contrary to Fireman’s Fund’s view, “indicat[ing] a place of performance” under section 1646 is not limited to situations in which the contract expressly states a place of performance. Frontier Oil explained that “the use of the word ‘indicate’ in section 1646 in lieu of a more restrictive word such as ‘specify’ or ‘state’ suggests that the Legislature intended a less restrictive meaning.” (Frontier Oil, supra, 153 Cal.App.4th at p. 1450.) Instead, the court held, “A contract ‘indicate[s] a place of performance’ within the meaning of section 1646 if the contract expressly specifies a place of performance or if the intended place of performance can be gleaned from the nature of the contract and its surrounding circumstances.” (Id. at p. 1443, italics added; see also id. at pp. 1450–1451.) Frontier Oil further held that the place of performance of an insurance policy is the place of the insured risk. (Id. at pp. 1461–1462.) Accordingly, under section 1646 the policy must be interpreted according to the law of the different states where Pep Boys had stores selling asbestos- containing products that would be covered by Fireman’s Fund’s policy. Our choice of law inquiry runs into a dead end at this point. The record does not identify any states besides California in which Pep Boys sold the products at issue in the underlying tort claims for which it seeks coverage. The only relevant evidence is a declaration by an employee of Pep Boys’ parent company

28 stating that Pep Boys operates stores nationwide, including 145 stores in California, and that Pep Boys faces hundreds of claims based on exposure to asbestos in those products, including claims of over 500 individuals in California. There is not even an indication that Pep Boys faces any claims at all for products sold in Pennsylvania, the state whose law it would have us apply. This does not provide us sufficient information to determine which states’ laws should be applied under section 1646.7 Because the policy does not indicate any single place of performance, Fireman’s Fund would have us look instead to the place where the policy was made, which it asserts is Pennsylvania. But Fireman’s Fund cites no authority requiring the policy to have only one place of performance, and we see no reason why it must. Our research demonstrates that it is accepted in insurance law that a single policy can be interpreted and applied differently in different states. “A liability insurance policy issued on a nationwide basis may be construed in accordance with the law of the jurisdiction in which a particular claim arises. (See Stonewall Surplus Lines Ins. Co. v. Johnson Controls, Inc. (1993) 14 Cal.App.4th 637, 646–647.) Thus, the same policy language may receive different construction and application in different jurisdictions. Parties to an insurance contract understand this.” (Downey Venture v. LMI Ins. Co. (1998) 66 Cal.App.4th 478, 514.) “ ‘Where a multiple risk policy

7 Pep Boys’ primary insurance policy lists various stores and service facilities that the company owned at the time, some of which were in Pennsylvania, some in California, and some elsewhere. However, this does not establish that the claims against Pep Boys for which it seeks coverage in this case relate to any of the stores in Pennsylvania. We also note that there are considerably more California locations than Pennsylvania locations.

29 insures against risks located in several states, it is likely that the courts will view the transaction as if it involved separate policies, each insuring an individual risk, and apply the law of the state of principal location of the particular risk involved.’ ” (Johnson Controls, at pp. 646–647.) Thus, according to section 1646, the policy should be interpreted according to the law of every state where the insured risks were located. The analysis that section 1646 calls for would likely prove unworkable in practice, since Pep Boys operated stores nationwide. It seems likely, then, that finding a practical solution to the choice of law question would likely require us to apply the governmental interest test after all. But since Fireman’s Fund has not analyzed the competing interests of California and Pennsylvania or any other states under that test, either, the end result would be the same: Fireman’s Fund has failed to demonstrate that any foreign states’ laws should apply. Accordingly, we fall back on the default choice of law principle that a California court will apply California law. (Chen v. Los Angeles Truck Centers, LLC, supra, 7 Cal.5th at p. 867.) Fireman’s Fund’s motion for summary judgment based on choice of law rules therefore does not provide an alternative basis on which to affirm the trial court’s judgment.

DISPOSITION

The trial court’s judgment is reversed.

BROWN, P. J.

WE CONCUR:

30

STREETER, J.

GOLDMAN, J.

The Pep Boys et al. v. Old Republic (A166574)

31 Trial Court: San Francisco City & County Superior Court

Trial Judge: Hon. Murlene J. Randle

Counsel: Roxborough, Pomerance, Nye & Adreani LLP, Drew E. Pomerance; Yoka & Smith LLP, Walter Michael Yoka, Michelle J. An for Plaintiff and Appellant.

Nicolaides Fink Thorpe Michaelides Sullivan LLP, Matthew C. Lovell, Alison V. Lippa, Kimberly A. Hartman for Defendant and Respondent Fireman’s Fund Insurance Company.

Clausen Miller, PC, Ian R. Feldman, Amy R. Paulus, Don R. Sampen for Defendant and Respondent Old Republic Insurance Company

Craig & Winkelman LLP, Bruce H. Winkelman for Defendant and Respondent New England Reinsurance Corporation and Executive Risk Indemnity, Inc fka American Excess Insurance Company.