St. of CA v. Continental Ins. Co. ( 2017 )


Menu:
  • Filed 9/29/17
    CERTIFIED FOR PARTIAL PUBLICATION*
    IN THE COURT OF APPEAL OF THE STATE OF CALIFORNIA
    FOURTH APPELLATE DISTRICT
    DIVISION TWO
    STATE OF CALIFORNIA,
    Plaintiff and Respondent,                     E064518
    v.                                                    (Super.Ct.No. CIV 239784)
    THE CONTINENTAL INSURANCE                             OPINION
    COMPANY et al.,
    Defendants and Appellants.
    APPEAL from the Superior Court of Riverside County. Sharon J. Waters, Judge.
    Affirmed.
    Berkes Crane Robinson & Seal, Steven M. Crane, and Barbara S. Hodous for
    Defendants and Appellants.
    Kamala D. Harris, Attorney General, Kristin G. Hogue, Senior Assistant Attorney
    General, Peter A. Meshot, Supervising Deputy Attorney General, and Darryl L. Doke,
    *     Pursuant to California Rules of Court, rules 8.1105(b) and 8.1110, this
    opinion is certified for publication with the exception of parts III.B.5, IV, V and VII.
    1
    Deputy Attorney General; Law Offices of Roger W. Simpson and Roger W. Simpson for
    Plaintiff and Respondent.
    This is an action by the State of California (State) to recover from various insurers
    the costs of cleaning up the Stringfellow hazardous waste site. It has been pending since
    1993. It has been to this court three times and to the California Supreme Court twice. At
    this point, however, the only remaining insurers are the Continental Insurance Company
    and Continental Casualty Company (collectively Continental), and the only remaining
    issues relate to prejudgment interest.
    In 2015, Continental paid the State its full policy limits of $12 million. The trial
    court ruled that the State was entitled to mandatory prejudgment interest on that amount
    at seven percent, dating back to 1998, and thus totaling $13,914,082.09. In the
    alternative, it also ruled that the State was entitled to discretionary prejudgment interest,
    at seven percent, dating back to 2002, and thus totaling $10,554,082.19.
    Continental appeals. In the published portion of this opinion, we address its
    contentions that the award of mandatory prejudgment interest was erroneous because:
    1. The award was premised on the trial court’s erroneous ruling as to when
    Continental’s policies attached.
    2. The State was not entitled to mandatory prejudgment interest because the
    amount of its damages was uncertain.
    Continental further contends that the award of discretionary prejudgment interest
    was erroneous because the trial court used an inapplicable interest rate.
    2
    Finding no error affecting the award of mandatory prejudgment interest, we will
    affirm it. Accordingly, we need not review the award of discretionary prejudgment
    interest.
    I
    FACTUAL BACKGROUND
    Various insurers issued liability insurance policies to the State, such that the State
    had at least some coverage at all times from 1963 through 1978. Except for the 1963-
    1964 policy period, the State was self-insured for the first $1, $2, or $5 million in
    liability, and it had a series of policies for liability above that, affording total coverage of
    either $50 or $51 million per policy period.
    Continental (and/or its predecessors in interest) issued three of these policies:
    Company                         Period              Limits         Retention
    Continental Casualty            1970-1973           $5 million     $16 million
    Harbor                          1970-1973           $5 million     $16 million
    CNA Casualty                    1973-1976           $2 million     $25 million
    Attachment A is a chart illustrating the amounts and policy periods of all of the
    relevant policies.
    In 1983, the United States and the State, as plaintiffs, filed an action in federal
    court against numerous defendants, alleging that they were responsible for the
    contamination of the Stringfellow site. Some of the defendants filed counterclaims
    3
    against the State. On July 28, 1988, the State gave notice of the counterclaims to its
    insurers, including Continental.
    On September 11, 1998, the federal district court issued a “Judgment Pursuant to
    Rule 54(b).”1 (Capitalization altered.) It defined the “plaintiffs” as the United States and
    the State. It also defined the “counterclaimants” as those defendants who had filed
    counterclaims against the State. These included J.B. Stringfellow, Stringfellow Quarry
    Co., and Stringfellow Quarry Co., Inc. (Stringfellow counterclaimants).
    It declared that the State “is liable to counterclaimants” under both the
    Comprehensive Environmental Response, Compensation, and Liability Act (
    42 U.S.C. § 9601
     et seq.) (CERCLA) and state law.
    For purposes of CERCLA, it found that the United States, the State, and the
    counterclaimants were all “liable persons.” It allocated liability under CERCLA as
    follows:
    a. The State was 65 percent liable “as against all [c]ounterclaimants.”
    b. The Stringfellow counterclaimants were 10 percent liable.
    c. The remaining counterclaimants and the United States were 25 percent liable.
    d. “[A]ny orphan share[] created by any party who is unable to pay its
    apportioned share under CERCLA . . . shall be reallocated among the existing solvent
    1       Rule 54(b) of the Federal Rules of Civil Procedure allows a trial court to
    enter “a final judgment as to one or more, but fewer than all, claims or parties . . . .”
    4
    parties in the same proportionate amounts as the above-described allocated CERCLA
    equitable shares.”
    For purposes of state law, it ruled: “Each counterclaimant is entitled to be paid by
    the State . . . 100% (one hundred percent) of any damages which that counterclaimant has
    incurred or will incur.” It specifically ruled that the counterclaimants’ liability under
    state law was “0% (zero percent).” It added, “The Court finds that the state law claims
    are direct claims for damages.”
    Finally, it provided: “[T]he United States has incurred certain costs of response in
    connection with the site which are consistent with the National Contingency Plan; those
    costs plus interest[] accrued through February 29, 1992 total $80,174,584.22.”
    The State filed an appeal from the Rule 54(b) judgment. It ultimately dismissed
    that appeal in 2002.
    Meanwhile, in December 1998, the State entered into a settlement agreement with
    the counterclaimants. In it, the counterclaimants released the State from all claims for
    past costs incurred at the Stringfellow site. In return, the State agreed to assume all
    liability for the Stringfellow site, including the liability to reimburse the United States for
    past and future costs incurred.
    In April 2001, the State entered into a settlement agreement with the United
    States, in which the State agreed to pay $99.4 million “for Response Costs . . . that the
    United States incurred at the Site . . . .” In July 2001, the federal district court ordered
    5
    this settlement agreement into effect as a consent decree. In August 2001, the State duly
    paid the United States $99.4 million.
    Meanwhile, separate and apart from its liability to the United States and/or the
    counterclaimants, the State also carried out its own remediation work at the Stringfellow
    site. As of the time of trial, remediation work was still going on.
    II
    PROCEDURAL BACKGROUND
    In 1993, the State filed an action against certain insurers, seeking to recover the
    costs of cleaning up the Stringfellow site. In 2002, the State filed a second action against
    additional insurers, including Continental. The two actions were consolidated; the
    defendants in the second action agreed to be bound by previous rulings in the first action.
    One of these previous rulings, in June 1999, was that the policy limits under a policy for
    a multi-year period applied once per policy period, not once per year (no-annualization
    ruling).
    In March 2004, the trial court ruled that each insurer was potentially liable for the
    total amount of the loss (subject to its policy limits); it rejected the contention that each
    insurer could be liable only for the portion of the loss attributable to its own policy
    periods (all-sums ruling). At the same time, however, it also ruled that the State could
    not recover the policy limits in effect for every policy period. Instead, the State had to
    choose one policy period, and it could recover only up to the policy limits of the policies
    in effect during that period (no-stacking ruling).
    6
    In February 2005, the trial court ruled that, for purposes of policy limits, there had
    been only a single covered occurrence (one-occurrence ruling).
    Meanwhile, a number of insurers had settled with the State. In March 2006, the
    trial court ruled that the amounts of these settlements, which totaled, at that point,
    approximately $120 million, had to be used to offset the other defendants’ liability (offset
    ruling).
    At that point, based on the one-occurrence, no-annualization and no-stacking
    rulings, the most the State could recover was $48 million. The $120 million in
    settlements was more than sufficient to offset this. Accordingly, the trial court entered a
    final judgment finding the defendants liable, but awarding the State zero damages.
    The State appealed; the defendants cross-appealed. In 2009, we filed our opinion.
    (State of California v. Continental Ins. Co. (2009) 
    88 Cal.Rptr.3d 288
    , review granted
    Mar. 18, 2009.) We affirmed the no-annualization ruling (id. at pp. 317-319), the all-
    sums ruling (id. at pp. 297-302), and the one-occurrence ruling (id. at pp. 313-317). We
    concluded that the offset ruling was moot. (Id. at pp. 319-320.) However, we reversed
    the no-stacking ruling. (Id. at pp. 302-313.)
    In 2012, the California Supreme Court affirmed our decision. (State of California
    v. Continental Ins. Co. (2012) 
    55 Cal.4th 186
    .)
    On remand, the parties filed cross-motions for summary adjudication on the issue
    of whether Continental’s policies attached immediately upon exhaustion of the specified
    retention for the specified policy period (vertical exhaustion) or only upon exhaustion of
    7
    all retentions across all policy periods (horizontal exhaustion). In October 2014, the trial
    court ruled that vertical exhaustion applied (vertical exhaustion ruling). Thus, it granted
    the State’s motion and denied Continental’s motion.
    In February 2015, the parties stipulated that Continental would pay its policy
    limits of $12 million, without any right of reimbursement, and that the trial court would
    proceed to determine all factual and legal issues regarding prejudgment interest.
    According to Continental, it paid the $12 million on April 6, 2015.
    The parties filed prehearing briefs on the issue of prejudgment interest.
    The State took the position that its damages were certain for purposes of
    prejudgment interest immediately upon the entry of the Rule 54(b) judgment, because the
    Rule 54(b) judgment made Continental liable for its full policy limits.
    Continental argued, among other things, that:
    1. The Rule 54(b) judgment did not make Continental liable at all, because it was
    not a judgment for damages.
    2. The State’s damages had not been certain, and could not have been made
    certain, until specific issues were adjudicated, including:
    a. Whether horizontal exhaustion or vertical exhaustion applied.
    b. How many occurrences there were.
    c. Whether the policy limits applied per occurrence per policy period, or
    per occurrence per year.
    d. Whether the policies could be stacked.
    8
    3. Before prejudgment interest, if any, could be calculated, Continental’s liability
    had to be offset by amounts that the State had recovered in settlements with other
    insurers.
    In July 2015, the trial court held a trial on the issue of prejudgment interest.
    In August 2015, the trial court ruled that the State was entitled to mandatory
    prejudgment interest under Civil Code section 3287, subdivision (a), at a rate of seven
    percent, starting on September 11, 1998, the date of the Rule 54(b) judgment, totaling
    $13,914,082.09. Alternatively, in the event that the award of mandatory prejudgment
    interest was reversed on appeal, it ruled that the State was entitled to discretionary
    prejudgment interest under Civil Code section 3287, subdivision (b), at a rate of seven
    percent, starting on September 11, 2002, the date when this action was filed, in the
    amount of $10,554,082.19. It then entered judgment accordingly.
    III
    THE VERTICAL EXHAUSTION RULING
    Continental contends that the award of prejudgment interest was erroneous
    because it was premised on the trial court’s vertical exhaustion ruling, which was also
    erroneous.
    A.     Additional Factual and Procedural Background.
    In its policies, Continental promised to pay “all sums which the [State] shall
    become obligated to pay by reason of liability imposed by law . . . for damages . . . .”
    9
    The “Limits of Liability” clause provided: “The limit of [Continental’s] liability
    shall be . . . [either $2 or $5 million] excess of [either $16 or $25 million] Ultimate Net
    Loss each occurrence (hereinafter called ‘the insured’s retention’).” (Capitalization
    altered.)
    “Ultimate Net Loss” was defined as “the amount payable in settlement of the
    liability of the [State] arising only from the hazards covered by this policy after making
    deductions for all recoveries and for other valid and collectible insurances excepting
    however policy/ies in respect of the insured’s retention . . . .”
    The “Other Insurance” clause stated:
    “If the Insured has other valid and collectible insurance against a loss covered by
    this policy, the insurance extended by this policy shall be excess only, and not primary or
    contributing.
    “The Insured may be insured when named as an insured or as an additional insured
    under policy/ies for all or any part of the Insured’s retention . . . .”
    Finally, the “Loss Payable” clause provided: “Liability under this policy with
    respect to any occurrence shall not attach unless and until the Insured shall have paid the
    amount of the Insured’s retention on account of such occurrence.”
    In the trial court, the State asserted that “vertical exhaustion” applied, meaning
    that each excess policy attached upon exhaustion of the State’s retention, as specified in
    that policy (either $16 million or $25 million), without regard to the exhaustion of any
    policies or retentions for any other policy period.
    10
    Attachment B is a chart showing how vertical exhaustion would work. Assuming
    vertical exhaustion applies, two of Continental’s policies would attach when the State’s
    covered losses reach $16 million, and all of them would attach when the State’s covered
    losses reach $25 million.
    Continental, on the other hand, asserted that “horizontal exhaustion” applied,
    meaning that no excess policy would attach until, not only the State’s self-insured
    retention, but also the amounts of all lower-lying policies and retentions, over all
    applicable policy periods, had been exhausted. Continental also argued that the State was
    judicially estopped by various previous admissions that horizontal exhaustion applied.
    Attachment C is a chart showing how horizontal exhaustion would work.
    Assuming horizontal exhaustion applies, Continental’s policies would not attach until the
    State’s covered losses reach either $101 million (as to the two policies with a $16 million
    retention) or $155 million (as to the one policy with a $25 million retention). Moreover,
    Continental’s policies would not be fully exhausted until the State’s covered losses reach
    $307 million. Continental has described horizontal exhaustion as the “rising bathtub”
    approach.
    B.     Discussion.
    1.     Continuous loss principles.
    As background, we begin by discussing some general principles governing
    liability insurance coverage for a continuous loss over multiple policy periods. This
    discussion revolves around three California Supreme Court cases.
    11
    First, in Montrose Chemical Corp. v. Admiral Ins. Co. (1995) 
    10 Cal.4th 645
    , the
    Supreme Court adopted the “‘continuous injury’ trigger of coverage.” (Id. at p. 655.)
    “A key inquiry under ‘occurrence’ policies is what fact or event ‘triggers’ the
    insurer’s duties to indemnify and/or defend the insured.” (Croskey et al., Cal. Practice
    Guide: Insurance Litigation (The Rutter Group 2016) ¶ 7:161, p. 7A-83.) “The issue is
    largely one of timing — what must take place within the policy’s effective dates for the
    potential of coverage to be ‘triggered’?” (Montrose Chemical Corp. v. Admiral Ins. Co.,
    
    supra,
     10 Cal.4th at p. 655, fn. 2, italics omitted.) Based on standard policy language,
    Montrose held that “bodily injuries and property damage that are continuous or
    progressively deteriorating throughout successive policy periods are covered by all
    policies in effect during those periods.” (Id. at p. 675; see also id. at p. 685.)
    Next, in Aerojet-General Corp. v. Transport Indem. Co. (1997) 
    17 Cal.4th 38
    , the
    Supreme Court adopted the “all sums” approach.
    In allocating a loss among insurers, some states follow a pro-rata approach, in
    which each insurer is liable only for its equitable share of the loss; usually that is
    determined by dividing the insurer’s time on the risk by the total duration of the loss.
    (See State of California v. Continental Ins. Co., supra, 55 Cal.4th at pp. 198-199.) In
    Aerojet, however, the Supreme Court held that, based on standard policy language, in
    which the insurer promises to pay “all sums” that the insured becomes legally obligated
    to pay as damages, the insurer’s duty to indemnify the insured “extends to all specified
    harm caused by an included occurrence, even if some such harm results beyond the
    12
    policy period. [Citation.]” (Aerojet-General Corp. v. Transport Indem. Co., supra, 17
    Cal.4th at pp. 56-57, fn. omitted.)2
    Finally, in State of California v. Continental Ins. Co., supra, 
    55 Cal.4th 186
    , the
    Supreme Court held that the insured is entitled to “stack” policy limits.
    The question of stacking goes to whether the insured can recover the limits of
    multiple policies across multiple policy periods. Based, yet again, on standard policy
    language, the Supreme Court held that the insured is entitled to stack policy limits. (State
    of California v. Continental Ins. Co., supra, 55 Cal.4th at pp. 201-202.) “The all-sums-
    with-stacking indemnity principle properly incorporates the Montrose continuous injury
    trigger of coverage rule and the Aerojet all sums rule, and ‘effectively stacks the
    insurance coverage from different policy periods to form one giant “uber-policy” with a
    coverage limit equal to the sum of all purchased insurance policies. Instead of treating a
    long-tail injury as though it occurred in one policy period, this approach treats all the
    triggered insurance as though it were purchased in one policy period. The [insured] has
    access to far more insurance than it would ever be entitled to within any one period.’
    [Citation.]” (Id. at p. 201.)
    2      An insurer that pays more than its pro rata share may then be able to obtain
    contribution from the other insurers. (State of California v. Continental Ins. Co., supra,
    55 Cal.4th at p. 200.)
    13
    2.     Primary and excess insurance principles.
    We also need to discuss primary and excess insurance. “Primary insurance . . .
    provides immediate coverage upon the ‘occurrence’ of a ‘loss’ or the ‘happening’ of an
    ‘event’ giving rise to liability: ‘Primary coverage is insurance coverage whereby . . .
    liability attaches immediately upon the happening of the occurrence that gives rise to
    liability.’ [Citations.]” (Croskey et al., 
    supra, ¶ 8:176, p. 8-59
    .) “Excess insurance . . .
    provides coverage after other identified insurance is exhausted.” (Id., ¶ 8:177, p. 8-59.)
    Most excess policies are written as excess to a specified primary policy.
    Alternatively, however, a policy may be written as excess to an insured’s retention.
    “‘The term “retention” (or “retained limit”) refers to a specific sum or percentage of loss
    that is the insured’s initial responsibility and must be satisfied before there is any
    coverage under the policy. . . .’ [Citation.]” (Forecast Homes, Inc. v. Steadfast Ins. Co.
    (2010) 
    181 Cal.App.4th 1466
    , 1474.) “The insured may purchase other insurance to
    cover the [retention] unless the policy clearly requires the insured itself, not other
    insurers, to pay this amount. [Citations.]” (Croskey et al., supra, ¶ 7:385.5, p. 7A-156,
    italics omitted.) If the insured does not purchase insurance to cover the retention and
    instead bears the risk itself, the retention is called a self-insured retention.
    Here, the “Other Insurance” clause permitted (though it did not require) the State
    to buy other insurance covering some or all of its retention. Moreover, in every policy
    period, the State did in fact buy other insurance covering all but the first $1 million, $2
    14
    million, or $5 million of its retention. Only the uncovered portion was a true self-insured
    retention.
    3.     Standard of review.
    The trial court made its vertical exhaustion ruling on a motion for summary
    adjudication. (Code Civ. Proc., § 437c, subd. (f)(1).) “Motions for summary
    adjudication are procedurally identical to motions for summary judgment [citation], and
    our review of rulings on those motions is de novo [citation].” (Dunn v. County of Santa
    Barbara (2006) 
    135 Cal.App.4th 1281
    , 1290.)
    4.     Because the policies were written in excess of a retention, vertical
    exhaustion applies.
    We now turn to Continental’s arguments in favor of horizontal exhaustion. These
    are of two types, arguments based on the policy language, and arguments based on case
    law. We discuss these seriatim.
    “[Exhaustion] cases, like all other insurance cases, look first to the terms of the
    policy.” (Montgomery Ward & Company, Inc. v. Imperial Cas. & Indem. Co. (2000) 
    81 Cal.App.4th 356
    , 368; accord, Community Redevelopment Agency v. Aetna Casualty &
    Surety Co. (1996) 
    50 Cal.App.4th 329
    , 338-340 & fn. 6.)
    “The principles governing the interpretation of insurance policies in California are
    well settled. ‘Our goal in construing insurance contracts, as with contracts generally, is
    to give effect to the parties’ mutual intentions. [Citations.] “If contractual language is
    clear and explicit, it governs.” [Citation.] If the terms are ambiguous [i.e., susceptible of
    15
    more than one reasonable interpretation], we interpret them to protect “‘the objectively
    reasonable expectations of the insured.’” [Citation.] Only if these rules do not resolve a
    claimed ambiguity do we resort to the rule that ambiguities are to be resolved against the
    insurer. [Citation.]’ [Citation.]” (Minkler v. Safeco Ins. Co. of America (2010) 
    49 Cal.4th 315
    , 321.)
    The “Limits of Liability” clause seems plain enough: Continental’s liability
    attaches upon an “Ultimate Net Loss” that is in excess of the specified ($16 or $25
    million) retention. This would seem to be the very definition of vertical exhaustion.
    Continental relies, however, on the definition of “Ultimate Net Loss,” and also on
    the “Other Insurance” clause. These both provided, essentially, that Continental’s policy
    was excess to any other insurance. However, there are two problems with this.
    First, and most glaringly, these clauses are not limited to lower-layer insurance.
    As written, they purport to provide that Continental simply is not liable as long as there is
    any other unexhausted insurance — including policies in the same layer. There is no
    language in Continental’s policies that would make them excess to lower-layer policies
    for other policy periods, but not to same-layer policies for other policy periods. Thus, the
    policy language simply does not support Continental’s proposed rising bathtub approach.
    Second, other-insurance clauses are intended to apply in contribution actions
    between insurers, not in coverage litigation between insurer and insured. In Dart
    Industries, Inc. v. Commercial Union Ins. Co. (2002) 
    28 Cal.4th 1059
    , the Supreme Court
    stated: “‘[A]pportionment among multiple insurers must be distinguished from
    16
    apportionment between an insurer and its insured. When multiple policies are triggered
    on a single claim, the insurers’ liability is apportioned pursuant to the “other insurance”
    clauses of the policies [citation] or under the equitable doctrine of contribution
    [citations]. That apportionment, however, has no bearing upon the insurers’ obligations
    to the policyholder. [Citation.] A pro rata allocation among insurers “does not reduce
    their respective obligations to their insured.” [Citation.] The insurers’ contractual
    obligation to the policyholder is to cover the full extent of the policyholder’s liability (up
    to the policy limits).’ [Citations.] This principle is consistent with ‘the settled rule that
    an insurer on the risk when continuous or progressively deteriorating damage or injury
    first manifests itself remains obligated to indemnify the insured for the entirety of the
    ensuing damage or injury.’ [Citation.]” (Id. at p. 1080; see also Shade Foods, Inc. v.
    Innovative Products Sales & Marketing, Inc. (2000) 
    78 Cal.App.4th 847
    , 899 [Montrose
    “implies that [the insurer] was bound by an obligation to indemnify [the insured] for the
    whole amount of the loss, without taking into account any contribution that [another
    insurer] might be obligated to make.”]; FMC Corp. v. Plaisted & Companies (1998) 
    61 Cal.App.4th 1132
    , 1185 [“[T]he availability of allocation among multiple insurers is
    essentially irrelevant to each individual insurer’s obligation to its insured under the terms
    of the parties’ insurance contract.”], disapproved on other grounds in State of California
    v. Continental Ins. Co., supra, 55 Cal.4th at p. 201.) We see no reason to treat the other
    insurance clause in this case differently just because it was repeated and incorporated into
    the definition of Ultimate Net Loss.
    17
    Continental argues that Dart is inapplicable because it did not involve any issue as
    to excess policies or exhaustion. Nevertheless, the language that we have quoted was not
    dictum. The Supreme Court was explaining why it did not need to know whether a lost
    primary policy had an other-insurance clause to determine whether the issuer had a duty
    to indemnify its insured. (Dart Industries, Inc. v. Commercial Union Ins. Co., supra, 28
    Cal.4th at pp. 1080-1081.) And “‘“[e]ven if properly characterized as dictum, statements
    of the Supreme Court should be considered persuasive. [Citation.]” [Citation.]’
    [Citation.]” (Arteaga v. Superior Court (2015) 
    233 Cal.App.4th 851
    , 865.)
    Third, as the State aptly points out, “Under [Continental]’s approach, a court could
    not determine the amount any insurer owes without first determining what every insurer
    owes . . . .” (Fn. omitted.) For example, if a lower-layer insurer for a different policy
    period happened to claim that some exclusion in its policy applied, a court could not
    determine whether Continental’s policies were triggered without first determining that
    exclusion claim. This would deprive the State of the timely indemnity that it bargained
    for.
    We turn, then, to Continental’s argument based on the case law. Continental relies
    primarily on Community Redevelopment Agency v. Aetna Casualty & Surety Co., supra,
    
    50 Cal.App.4th 329
    , which held that, in the case of a continuous loss across multiple
    policy periods, horizontal exhaustion ordinarily applies to primary insurance. In
    Community, several developers had been sued in related construction defect cases. (Id. at
    pp. 333-334.) Two insurers (United and State Farm) had each issued primary policies,
    18
    covering two different policy periods, to the developers; a third insurer (Scottsdale) had
    issued an umbrella policy to one of the developers. (Id. at p. 334.) The umbrella policy
    provided that it was excess to the State Farm policy — which was specifically listed as
    “underlying insurance” — and also to “the applicable limits of any other underlying
    insurance . . . .” (Id. at p. 335, italics omitted.) United paid for the developers’ defense.
    (Id. at p. 336.) One of the claimants settled with one of the developers, and State Farm
    paid out its policy limits to fund the settlement. (Id. at p. 336.) United then sought to
    compel Scottsdale to contribute toward its defense costs. (Id. at p. 332 & fn. 1.)
    The appellate court held that Scottsdale had no duty to defend under its excess
    policy until both the State Farm and United primary policies were exhausted.
    (Community Redevelopment Agency v. Aetna Casualty & Surety Co., supra, 50
    Cal.App.4th at p. 342.) It stated: “It is settled under California law that an excess or
    secondary policy does not cover a loss, nor does any duty to defend the insured arise,
    until all of the primary insurance has been exhausted. [Citation.]” (Id. at p. 339.) “This
    is a particular problem in continuous loss cases, such as the one before us.” (Id. at
    p. 340.) “[P]rimary policies may have defense and coverage obligations which make
    them underlying insurance to excess policies which were effective in entirely different
    time periods and which may not have expressly described such primary policies as
    underlying insurance. Absent a provision in the excess policy specifically describing and
    limiting the underlying insurance, a horizontal exhaustion rule should be applied in
    continuous loss cases because it is most consistent with the principles enunciated in
    19
    Montrose. In other words, all of the primary policies in force during the period of
    continuous loss will be deemed primary policies to each of the excess policies covering
    that same period. Under the principle of horizontal exhaustion, all of the primary policies
    must exhaust before any excess will have coverage exposure.” (Ibid.; accord, Padilla
    Const. Co., Inc. v. Transportation Ins. Co. (2007) 
    150 Cal.App.4th 984
    , 986-987;
    Olympic Ins. Co. v. Employers Surplus Lines Ins. Co. (1981) 
    126 Cal.App.3d 593
    , 600-
    601; Iolab Corp. v. Seaboard Sur. Co. (9th Cir. 1994) 
    15 F.3d 1500
    , 1504 [applying
    California law].)
    Community is not controlling, because here the applicable policies were not neatly
    divided into a primary level and an excess level. With one negligible exception,3 all of
    the applicable policies were excess to a retention. The lowest-layer policies were excess
    to the State’s true self-insured retention (e.g., in the 1970-1973 policy period, $2 million);
    the higher-layer policies were excess to a retention in a specified dollar amount (e.g.,
    Continental’s policies for the 1970-1973 policy period had a retention of $16 million),
    which the State filled in with other insurance. Thus, no policy was written as excess to
    any other specified policy, although some became excess to another policy or policies as
    the result of the State’s decision to purchase other insurance to cover its retention.
    Community reasoned that a primary policy is qualitatively different from an excess
    policy; the defense and indemnity obligations under a primary policy are immediate,
    3     There was a $1 million primary policy in the 1963-1964 policy period.
    However, it may be disregarded, as it was exhausted under any scenario.
    20
    whereas under an excess policy, they are merely contingent.4 Thus, an excess insurer
    should not be required to defend or to indemnify as long as any primary insurer is still
    sitting on its hands. The same is not true of two insurers who have issued policies that
    are excess to a retention. Their defense and indemnity obligations are both contingent,
    and they have priced their premiums accordingly. We cannot say, from their relationship
    alone, that either one should have to exhaust before the other is liable.
    Community also noted that Scottsdale’s policy provided generally that it was
    excess to “‘the applicable limits of any other underlying insurance collectible by the
    [insured parties].’” (Community Redevelopment Agency v. Aetna Casualty & Surety Co.,
    supra, 50 Cal.App.4th at p. 338.) Here, by contrast, the other insurance provisions
    (including the “Ultimate Net Loss” definition) were not limited to other “underlying”
    insurance; rather, they purported to make Continental’s policies excess to any other
    insurance whatsoever. And as the State points out, all the other policies had similar other
    insurance provisions, purporting to make them excess to any other insurance whatsoever.
    Thus, these provisions did not make any policy excess to any particular primary policy.
    Under these circumstances, Montgomery Ward & Company, Inc. v. Imperial Cas.
    & Indem. Co., supra, 
    81 Cal.App.4th 356
     — which held that in the case of a continuous
    loss across multiple policy periods, vertical exhaustion ordinarily applies to self-insured
    4      As a result (although Community did not specifically mention this), the
    premium per dollar of coverage for a primary policy is generally much higher than for an
    excess policy. (American Safety Indemnity Company v. Admiral Insurance Company
    (2013) 
    220 Cal.App.4th 1
    , 11.)
    21
    retentions — is more apropos. There, Montgomery Ward was sued for environmental
    contamination over a period of years. Four insurers had issued policies to Montgomery
    Ward, each covering part of the relevant time period; each policy was subject to a self-
    insured retention. (Id. at pp. 361-362 & 361, fn. 2.) The trial court ruled that “all [self-
    insured retentions] for all policies in effect over the period of continuing damage . . .
    were the equivalent of primary insurance, and must be exhausted before the excess
    insurers had any obligation to indemnify Montgomery Ward.” (Id. at p. 363.)
    The appellate court reversed. It held that self-insured retentions “are not primary
    insurance and the principle of horizontal exhaustion does not apply.” (Montgomery Ward
    & Company, Inc. v. Imperial Cas. & Indem. Co., supra, 81 Cal.App.4th at p. 364.) It
    noted that, in each policy, the insurer promised to indemnify Montgomery Ward for the
    ultimate net loss in excess of a specified dollar amount (usually $50,000.) (Id. at p. 366.)
    The court acknowledged that the polices also provided that they were excess to other
    valid and collectible insurance, but it reasoned that a self-insured retention is not
    insurance. (Id. at pp. 367-370; accord, Legacy Vulcan Corp. v. Superior Court (2010)
    
    185 Cal.App.4th 677
    , 696-697; California Pacific Homes, Inc. v. Scottsdale Ins. Co.
    (1999) 
    70 Cal.App.4th 1187
    , 1192-1195.)
    The court specifically distinguished Community: “[Community] does not stand for
    the proposition all primary coverage must be exhausted before excess policies may be
    reached without regard to the terms of the excess policies. Indeed, the [Community] court
    said ‘[i]f an excess policy states that it is excess over a specifically described policy and
    22
    will cover a claim when that specific primary policy is exhausted, such language is
    sufficiently clear to overcome the usual presumption that all primary coverage must be
    exhausted.’ [Citation.] Here, the Insurers’ policies state they are excess over a
    specifically described [retention] and will cover a claim when that [retention] is
    exhausted.” (Montgomery Ward & Company, Inc. v. Imperial Cas. & Indem. Co., supra,
    81 Cal.App.4th at pp. 368-369.)
    Admittedly, this is something of a hybrid case. The State argues that
    Continental’s policies were excess to a retention; under Montgomery Ward, self-insured
    retentions exhaust vertically. However, Montgomery Ward involved true self-insured
    retentions. Here, the State covered most of its retentions by buying lower-level policies.
    Accordingly, Continental argues that under Community, primary policies exhaust
    horizontally. However, Community involved true primary policies; here, the State had
    only policies that were excess to various retentions.
    We believe that, under these circumstances, Montgomery Ward, and not
    Community, should be controlling. Certainly if the State had had nothing but true self-
    insured retentions, vertical exhaustion would be the rule; Continental would not be
    entitled to the benefit of any self-insured retentions for any other policy periods. It would
    be paradoxical if the fact that the State prudently decided to protect itself further by
    23
    buying insurance covering most of its retentions actually made it harder for the State to
    obtain indemnity from any one insurer.5
    Under Montgomery Ward, Continental’s rising bathtub model is clearly wrong in
    at least one respect. Continental states that “all underlying policy limits and self-insured
    retentions across the years of continuing property damage must be exhausted . . . before
    any of [Continental’s] excess policies are obligated to pay covered claims.” (Bolding
    omitted, italics added.)   However, the State was not required to exhaust its true self-
    insured retentions for any other policy periods. Continental does not point to any
    particular language in its own policies that even arguably so requires. Each Continental
    policy is excess to the State’s retention for that policy period, but not for any other policy
    periods.
    More generally, however, Continental’s rising bathtub model is also wrong with
    respect to exhaustion of lower-layer policies. Aside from the other-insurance clauses,
    which are not specific to other lower-layer insurance, there was simply no policy
    5        Montrose Chemical Corp. v. Superior Court (2017) ___ Cal.App.5th ___
    [
    2017 Cal. App. LEXIS 759
    ] held that the insured in that case was not entitled to
    summary adjudication that vertical exhaustion applied. It was decided so recently that it
    is not yet final (Cal. Rules of Court, rule 8.264(b)(1)); therefore, we do not discuss it in
    detail. We note, however, that Montrose is distinguishable from our case on multiple
    grounds, including that all of the policies here have (1) mutually conflicting other
    insurance clauses (cf. 
    id.
     at p. *48, fn. 7) , and (2) an at least partly self-insured retention
    (cf. 
    id.
     at pp. *33-*35, *40, fn. 6). If and to the extent that the court’s reasoning is
    irreconcilable with ours, we disagree with it for the reasons we have already stated.
    24
    language requiring exhaustion of any other insurance. Rather, the policies provided that
    Continental’s duty to pay arose as soon as the specified retention was exhausted.
    We therefore conclude that the trial court correctly ruled that only vertical
    exhaustion was required.
    5.     We did not decide this issue in the previous appeal.
    Continental argues that we already held, in the previous appeal, that horizontal
    exhaustion was required. Not so. Continental focuses on that portion of our opinion in
    which we discussed California Pacific Homes, Inc. v. Scottsdale Ins. Co., supra, 
    70 Cal.App.4th 1187
    . (State of California v. Continental Ins. Co., supra, 88 Cal.Rptr.3d at
    pp. 311-312.) California Pacific had held that an insured who sought to recover under
    just one of five policies covering a continuous loss only had to pay the self-insured
    retention under that one policy, not under all five. We distinguished California Pacific
    on the ground that it dealt with an insurer’s attempt to stack retentions, whereas the issue
    then before us was an insured’s attempt to stack policy limits. (State of California v.
    Continental Ins. Co., supra, at pp. 311-312.) Admittedly, we also noted that California
    Pacific had left open the possibility that, if the insured had sought to recover under all
    five policies, it would have had to pay all five retentions (i.e., horizontal exhaustion); we
    added, “That is perfectly consistent with our approach to stacking of limits.” (State of
    California v. Continental Ins. Co., supra, at p. 312.) However, California Pacific did not
    actually hold that horizontal exhaustion of retentions is required. A fortiori, neither did
    we.
    25
    IV
    JUDICIAL ESTOPPEL CONCERNING HORIZONTAL EXHAUSTION
    Continental contends that the State previously asserted that horizontal exhaustion
    applied, and therefore it is judicially estopped from claiming that vertical exhaustion
    applies.
    A.      Additional Factual and Procedural Background.
    In the previous appeal, the State argued that it was entitled to stack policy limits
    across policy periods. In the course of doing so, it argued that the insured’s right to stack
    policy limits across policy periods was analogous to, and thus followed from, an excess
    insurer’s right to stack (i.e., horizontally exhaust) primary policies across policy periods.
    Accordingly, it made the following statements:
    1. “‘[O]ther insurance’ issues arise when a policyholder is protected by multiple
    lines of insurance, each containing one or more levels of coverage. For example, a
    policyholder may be a ‘named insured’ under his own line of coverage and an ‘additional
    insured under another policyholder’s line of coverage, with each line containing a
    primary policy and one or more excess policies. In such cases, California law holds an
    excess policy does not attach until the combined limits of all lower-level policies in all
    lines of coverage have been exhausted.”
    2. “The horizontal exhaustion rule applies to excess as well as primary policies.”
    (Capitalization altered, bolding omitted.)
    26
    3. “In the proceedings below, Respondents agreed that the horizontal exhaustion
    doctrine allows policyholders to combine or ‘stack’ primary policies, but claimed it does
    not allow stacking of excess policies . . . . [¶] Respondents’ argument is disingenuous.
    . . . The fact that the State’s own insurance companies have agreed in other cases that
    horizontal exhaustion also applies to excess coverage again confirms that the State’s
    identical interpretation, at the very least, is a reasonable construction of the policies and
    therefore should be adopted by this Court.”
    4. “[U]nder the horizontal exhaustion doctrine, because the State was held liable
    for contamination occurring during successive policy periods, it again is entitled to
    indemnity up to its combined policy limits.”
    5. “Under the express terms of the [Ultimate Net Loss] clause, the policy’s limits
    are paid after the combined limits of all other insurances, at which point the policy covers
    the policyholder’s remaining liability, if any.”
    6. “[W]here a policyholder is protected by both primary and excess policies, the
    combined limits of all policies in each level of coverage (primary, first-level excess,
    second-level excess, etc.) must be exhausted before the next level of policies is
    triggered . . . .”
    7. “[T]he ‘horizontal exhaustion’ doctrine . . . holds that, where multiple,
    successive policies afford coverage in a continuous damage case, all successive policies
    in a given level of coverage must be exhausted before the next level of coverage applies.”
    27
    B.     Discussion.
    “The doctrine of judicial estoppel precludes a party from taking inconsistent
    positions in judicial or quasi-judicial proceedings. [Citation.]” (Claxton v. Waters
    (2004) 
    34 Cal.4th 367
    , 379, fn. 2.)
    “Judicial estoppel applies when ‘(1) the same party has taken two positions; (2)
    the positions were taken in judicial or quasi-judicial administrative proceedings; (3) the
    party was successful in asserting the first position (i.e., the tribunal adopted the position
    or accepted it as true); (4) the two positions are totally inconsistent; and (5) the first
    position was not taken as a result of ignorance, fraud, or mistake.’ [Citation.]” (RSL
    Funding, LLC v. Alford (2015) 
    239 Cal.App.4th 741
    , 748.)
    Judicial estoppel does not apply here because, in the previous appeal, we did not
    adopt or accept as true the position that Continental attributes to the State. Indeed,
    Continental never explains how or where we supposedly did do so; hence, it has forfeited
    any such argument.
    In the previous appeal, we noted that “stacking” is an ambiguous term; it can refer
    to an insured’s right to stack policy limits across policy periods (State of California v.
    Continental Ins. Co., supra, 88 Cal.Rptr.3d at p. 302), but it can also refer to an excess
    insurer’s right to stack the limits of primary policies (and/or self-insured retentions)
    across policy periods. (Id. at pp. 302, 306, 311-313.) We added: “As a general rule,
    California requires horizontal exhaustion. Vertical exhaustion applies if, and only if, the
    excess policy provides that it is excess to a specified primary policy. [Citations.]” (Id. at
    28
    p. 306.) We concluded that the very existence of a horizontal exhaustion rule in some
    cases “necessarily implies that the insured, too, is entitled to stack the primary
    policies . . . .” (Ibid.) However, we did not need to decide — and we did not decide —
    whether the particular policies in this case actually called for horizontal or vertical
    exhaustion.
    To the extent that Continental relies on the State’s assertions in the trial court,
    rather than in this court, judicial estoppel does not apply because the trial court rejected
    the State’s position.
    To the extent that Continental relies on the State’s assertions in the California
    Supreme Court, judicial estoppel does not apply because the Supreme Court did not
    express any opinion whatsoever regarding horizontal or vertical exhaustion.
    Separately and alternatively, the trial court was not required to apply judicial
    estoppel, even assuming that all of its requirements were satisfied. “We do not invariably
    enforce the judicial estoppel doctrine merely because all of its elements are met.
    ‘[N]umerous decisions have made clear that judicial estoppel . . . is an equitable doctrine,
    and its application . . . is discretionary. [Citations.]’ [Citation.]” (People v. Castillo
    (2010) 
    49 Cal.4th 145
    , 156.) Continental has not even argued that the trial court abused
    its discretion in this regard. We deem it to have forfeited any such argument.
    29
    V
    USING THE RULE 54(B) JUDGMENT AS THE START DATE
    Continental contends that the trial court erred by using September 11, 1998 — the
    date of the Rule 54(b) judgment — as the start date for prejudgment interest.
    A.     Additional Factual and Procedural Background.
    Continental raised this general argument, albeit tersely, in its prehearing briefing
    below.
    The trial court ruled, “The judgment in the federal action found the State’s liability
    as 65% under CERCLA and 100% under the state law claims. It also found that the
    United States had incurred costs of response in connection with the site in the amount of
    $80,174,584.22 . . . . CERCLA . . . provides for liability for all costs of removal or
    remediation incurred by the United States . . . not inconsistent with the national
    contingency plan. Thus, under the CERCLA claims, through this judgment, the State
    was liable for 65% of at least $80,174,584.22. The judgment also declared the State was
    liable to the counterclaimants under the state law claims and essentially was 100%
    responsible for all costs of remediation. Thus, the remaining 35% owed on the
    $80,174,584.22 which was apportioned to the counterclaimants was ultimately required
    to be paid by the State anyway. Reading the judgment in its entirety, it is an adjudication
    that the State must pay the United States at least $80,174,584.22.”
    30
    B.      Discussion.
    As already mentioned (see part III.A., ante), Continental promised to pay “all
    sums which the [State] shall become obligated to pay by reason of liability imposed by
    law . . . for damages . . . .”
    “[T]he term ‘damages’ limits the insurer’s indemnification obligation to ‘money
    ordered by a court,’ i.e., a money judgment entered against the insured in a third party
    suit for damages. [Citation.]” (County of San Diego v. Ace Property & Cas. Ins. Co.
    (2005) 
    37 Cal.4th 406
    , 410.) “It does not cover anticipated liabilities on future claims.”
    (Croskey et al., supra, ¶ 7:148.8, p. 7A-68.) “[S]tatutory and contractual duties . . . are
    not enough to implicate [a] liability policy, even if they constitute separately owed public
    obligations. Those duties do not of their own force create insurance coverage.” (San
    Diego Housing Com. v. Industrial Indemnity Co. (1998) 
    68 Cal.App.4th 526
    , 543.)
    “‘The interpretation of the effect of a judgment is a question of law within the
    ambit of the appellate court.’ [Citation.]” (Mendly v. County of Los Angeles (1994) 
    23 Cal.App.4th 1193
    , 1205.) That is particularly true when, as here, the parties have not
    offered any relevant extrinsic evidence. (Parsons v. Bristol Development Co. (1965) 
    62 Cal.2d 861
    , 865.)
    Under CERCLA, in the event of a release of a hazardous substance, certain classes
    of persons are liable for “all costs of removal or remedial action incurred by the United
    States Government . . . not inconsistent with the national contingency plan . . . .” (
    42 U.S.C. § 9607
    (a)(4)(A); Cooper Industries, Inc. v. Aviall Services, Inc. (2004) 
    543 U.S. 31
    157, 161.) Thus, as the trial court reasoned, the provision of the Rule 54(b) judgment
    finding that the United States had incurred $80,174,584.22 in response costs fixed at least
    a minimum amount of liability, and the provision of the Rule 54(b) judgment finding the
    State 100 percent liable under state law required the State to pay that amount.
    Continental argues that the Rule 54(b) judgment contains no language ordering the
    State to pay the United States. Indeed, as it points out, the State and the United States
    were co-plaintiffs; it was the counterclaimants, not the United States, who sued the State.
    The policies, however, required only that a court order the State to pay money; it did not
    matter whether the State was ordered to pay the United States, the counterclaimants, or
    both.
    In any event, the Rule 54(b) judgment did effectively order the State to pay the
    United States. It found the State and the counterclaimants to be liable persons, and thus
    jointly and severally liable for the $80,174,584.22 in costs incurred by the United States.
    Moreover, while it provisionally allocated the liability under CERCLA (subject to
    reallocation in the event of any “orphan share[]”), it further ruled that the State was 100
    percent liable under state law for the counterclaimants’ damages. By holding the State
    and other parties liable to the United States in the amount of $80,174,584.22, but
    allowing the other parties to pass along their liability for this amount the to the State, it
    effectively held the State liable to the United States for this amount.6
    6    Continental argues that the State was not held liable for any “fixed
    amount.” It does not argue that the State was not held liable for any particular portion of
    the $80,174,584.22. We deem it to have forfeited any such argument.
    32
    Most important, “‘Rule 54(b) permits entry of a partial final judgment only when
    all of one party’s claims or rights have been fully adjudicated, or when a distinct claim
    has been fully resolved with respect to all parties.’ [Citation.]” (Lottie v. West American
    Ins. Co., of Ohio Cas. Group of Ins. Companies (7th Cir. 2005) 
    408 F.3d 935
    , 938.)
    Hence, a Rule 54(b) judgment is not supposed to be used to permit appellate review of a
    decision sustaining a contribution or indemnity claim before the underlying claim has
    been decided. (Factory Mut. Ins. Co. v. Bobst Group USA, Inc. (7th Cir. 2004) 
    392 F.3d 922
    , 924-925.) It is also not supposed to be used to permit appellate review of a decision
    finding liability without determining damages, unless the calculation of damages would
    be “‘mechanical’” or “‘ministerial.’” (Winston Network, Inc. v. Indiana Harbor Belt R.
    Co. (7th Cir. 1991) 
    944 F.2d 1351
    , 1356-1357.) Here, then, simply by entering judgment
    pursuant to Rule 54(b), the federal court demonstrated that it was not just deciding the
    counterclaimants’ claims against the State, but also the United States’ claims against the
    counterclaimants, as well as the amount of those claims.
    Continental asserts that a witness at trial supposedly “conceded . . . that he could
    find nowhere in the judgment where the State was ordered to pay the United States the
    $80 million.” We find no such concession. Witness Donald Robinson had been the
    State’s lead attorney in the federal action; as such, he was involved in drafting the Rule
    54(b) judgment. He testified that it actually did constitute entry of judgment against the
    State in favor of the United States: “[I]t . . . set[s] forth that the State is liable for 100
    percent of all costs at the Stringfellow site, and it also states that the United States has
    33
    expended an certain amount in past costs that have been liquidated and have been
    determined to be not inconsistent with the NCP. I interpret those two provisions together
    as holding the State liable for the U.S. costs.”
    Continental complains that the State did not notify it of the Rule 54(b) judgment
    until November 1998 and even then did not demand indemnity. That is irrelevant. The
    State had already given Continental notice of the federal action. However, Continental
    had denied coverage, because, in its view, horizontal exhaustion was required and had not
    yet occurred.
    Continental also seems to argue that it had no immediate obligation to indemnify
    the State for its liability under the Rule 54(b) judgment because the State filed an appeal
    from that judgment. It cites no authority for this position, which runs counter to the basic
    principle that the duty to indemnify ordinarily arises at entry of judgment. (See 12 Couch
    on Ins. (3d ed. 1998) §§ 170:48, 170.51.)
    Next, Continental argues that the mere entry of the Rule 54(b) judgment, without
    any actual payment by the State, was insufficient to trigger its policies. As mentioned,
    the “Loss Payable” clause provided: “Liability under this policy with respect to any
    occurrence shall not attach unless and until the Insured shall have paid the amount of the
    Insured’s retention on account of such occurrence.” (Italics added.)
    As the State points out, however, as of the entry of the Rule 54(b) judgment, it had
    already paid over $47 million in remediation costs. These were payments “on account of
    [a covered] occurrence.” Moreover, they greatly exceeded the applicable retention
    34
    (either $16 or $25 million). The Loss Payable clause did not require that the amount of
    the retention be paid pursuant to a court order; the State had the right to settle claims in
    an amount up to and including the retention without Continental’s consent. Thus, the
    Loss Payable clause was satisfied. The subsequent entry of the Rule 54(b) judgment,
    requiring the State to pay more than the sum of the amount of its applicable retention,
    plus the amount of the applicable excess layer (i.e., a total of $50 million), was sufficient
    to make Continental liable for its policy limits.
    Continental also argues that full exhaustion required $50 million in liability for the
    1970-1973 policy period and another $50 million in liability for the 1973-1976 policy
    period; the $80,174,584.22 Rule 54(b) judgment was insufficient to exhaust both. As we
    understand this, it merely reiterates the argument that horizontal exhaustion applies — an
    argument we rejected in part III, ante. If it means anything else, it has been forfeited by
    the failure to provide any analysis or authority to support it. (City of Palo Alto v. Public
    Employment Relations Board (2016) 
    5 Cal.App.5th 1271
    , 1302.)
    Finally, Continental argues that, if the Rule 54(b) judgment did make the State
    liable for the $80,174,584.22, then there was no reason for the State to enter into the 1998
    settlement agreement, in which it accepted liability for the $80,174,584.22. As
    Continental itself points out, however, the State had appealed from the Rule 54(b)
    judgment; thus, vis à vis the counterclaimants, its liability was still an open legal question
    and a fit subject for settlement. Moreover, that is not all the 1998 settlement agreement
    35
    did; it also settled liability for remediation costs accrued since February 29, 1992, as well
    as for future remediation costs.
    We therefore conclude that the entry of the Rule 54(b) judgment not only triggered
    but fully exhausted Continental’s policies, making Continental liable for its $12 million
    policy limits. The trial court therefore correctly awarded prejudgment interest on this
    amount dating back to September 11, 1998.7
    VI
    THE AMOUNT OF THE STATE’S DAMAGES WAS CERTAIN
    Continental contends that the trial court erred by awarding prejudgment interest
    because damages could not be ascertained until after the trial court made a series of
    rulings.
    A.     General Background Principles.
    Civil Code section 3287, subdivision (a), as relevant here, provides: “A person
    who is entitled to recover damages certain, or capable of being made certain by
    calculation, and the right to recover which is vested in the person upon a particular day, is
    entitled also to recover interest thereon from that day . . . .”
    7        On the assumption that we will agree that the Rule 54(b) judgment did not
    start interest running, Continental further argues that (1) the December 1998 settlement
    agreement and (2) the July 2001 consent decree (resulting in the August 2001 payment to
    the United States) also were not judgments for damages and therefore also did not start
    interest running.
    Because we hold that the Rule 54(b) judgment did start interest running, we need
    not address these contentions.
    36
    “[T]he primary purpose of [Civil Code] section 3287(a) ‘is to provide just
    compensation to the injured party for loss of use of the [underlying] award during the
    prejudgment period — in other words, to make the plaintiff whole as of the date of the
    injury.’ [Citations.]” (Flethez v. San Bernardino County Employees Retirement Assn.
    (2017) 
    2 Cal.5th 630
    , 643.)
    “‘Under this provision, prejudgment interest is allowable where the amount due
    plaintiff is fixed by the terms of a contract . . . .’ [Citation.] If damages are ‘certain,’
    interest must be awarded as a matter of right. [Citations.]” (National Farm Workers
    Service Center, Inc. v. M. Caratan, Inc. (1983) 
    146 Cal.App.3d 796
    , 809.) “Courts
    generally apply a liberal construction in determining whether a claim is certain . . . .
    [Citation.]” (Howard v. American Nat. Fire Ins. Co. (2010) 
    187 Cal.App.4th 498
    , 535.)
    “‘On appeal, we independently determine whether damages were ascertainable for
    purposes of the statute, absent a factual dispute as to what information was known or
    available to the defendant at the time.’ [Citation.]” (Watson Bowman Acme Corporation
    v. RGW Construction, Inc. (2016) 
    2 Cal.App.5th 279
    , 296.)
    Our Supreme Court has stated that: “[T]he certainty required of Civil Code
    section 3287, subdivision (a), is absent when the amounts due turn on disputed facts, but
    not when the dispute is confined to the rules governing liability. [Citations.]” (Olson v.
    Cory (1983) 
    35 Cal.3d 390
    , 402.)
    In Continental’s view, this test sets up a dichotomy between liability and damages
    — i.e., a dispute over damages precludes certainty, but a dispute over liability does not.
    37
    And there is language in the case law that supports this view. For example, the Supreme
    Court has stated: “‘Damages are deemed certain or capable of being made certain within
    the provisions of subdivision (a) of section 3287 where there is essentially no dispute
    between the parties concerning the basis of computation of damages if any are
    recoverable but where their dispute centers on the issue of liability giving rise to damage.
    [Citation.]’ [Citations.]” (Leff v. Gunter (1983) 
    33 Cal.3d 508
    , 519-520, italics added.)
    In the State’s view, however, it sets up a dichotomy between legal issues and
    factual issues — i.e., a dispute over a factual issue precludes certainty, but a dispute over
    a legal issue does not. Once again, language can be found in the case law to support this
    view. For example, in Collins v. City of Los Angeles (2012) 
    205 Cal.App.4th 140
    , the
    court stated: “[A] legal uncertainty concerning the measure of damages rather than a
    factual uncertainty . . . does not prevent damages from being ascertainable.” (Id. at
    p. 152; see also Benson v. City of Los Angeles (1963) 
    60 Cal.2d 355
    , 366.)
    The State has the better side of the argument. At least in insurance cases, what has
    been treated as controlling is whether the uncertainty is legal or factual. What portion of
    a loss will be allocated to a given insurer has been treated as a matter of the extent of
    liability, not a matter of damages — at least when it does not turn on disputed factual
    issues.
    For example, in Hartford Accident & Indemnity Co. v. Sequoia Ins. Co. (1989)
    
    211 Cal.App.3d 1285
    , Hartford had issued both a primary policy and an umbrella policy
    to the owners of a car that was involved in an accident (id. at pp. 1290-1292); Sequoia
    38
    had issued a primary policy and Transamerica had issued an umbrella policy to the driver
    of the car. (Id. at pp. 1290-1291, 1293-1296.) After Hartford settled the claims of the
    injured passengers for an amount in excess of $1.8 million, it sued Sequoia and
    Transamerica for contribution. (Id. at p. 1291.) The trial court granted summary
    judgment for Hartford; it awarded Hartford the full amount of the Sequoia policy and a
    pro rata amount of the Transamerica policy. (Id. at pp. 1291, 1307; see also 
    id. at p. 1293
    .)
    The appellate court held that Hartford was entitled to prejudgment interest on the
    award against Transamerica. (Hartford Accident & Indemnity Co. v. Sequoia Ins. Co.,
    supra, 211 Cal.App.3d at pp. 1305-1307.) Transamerica argued that prejudgment interest
    was not available because “the amount of liability, if any, remain[ed] in dispute until
    judgment.” (Id. at p. 1307.) The appellate court responded that although “Hartford’s
    right to recover damages from Transamerica . . . was in issue, the amount of damages
    recoverable was ‘certain, or capable of being made certain by calculation’ and was
    ‘vested’ in Hartford on . . . the day Hartford exhausted its primary policy limit and first
    paid out money under its umbrella policy. Assuming Hartford was entitled to recover
    damages, the only question remaining was how the trial court would prioritize the
    policies. In this respect, the trial court had only two options: (1) to hold, as it did, that
    the Sequoia policy was second in order and that the Hartford and Transamerica policies
    share the excess liability on a prorata basis, or (2) that the Hartford [u]mbrella policy and
    the combined limits of the Sequoia and Transamerica policies be prorated. This was
    39
    purely a question of law since the amount of damages under either formula was readily
    ascertainable by mathematical calculation. Thus, the amount of damages were never
    ‘unliquidated’ or ‘contingent’ but rather, only the legally proper order of priority of the
    respective policies was uncertain. Under these circumstances, Hartford is entitled to
    prejudgment interest.” (Id. at p. 1307.)
    Continental argues that Hartford is not on point because there, “there was no
    dispute as to the methodology to be applied to the allocation . . . .”   In light of the
    court’s own discussion of the two alternative methodologies, this is clearly inaccurate.
    Rather, what was crucial in Hartford was that the choice of methodology was a legal
    issue, not a factual issue.
    Similarly, in Fireman’s Fund Ins. Co. v. Allstate Ins. Co. (1991) 
    234 Cal.App.3d 1154
    , Fireman’s, Allstate, and Northbrook each admittedly insured at least one of the
    parties involved in a truck-car collision. (Id. at pp. 1158-1159.) After settling with the
    injured passengers (id. at p. 1158), all three insurers filed cross-actions for declaratory
    relief. (Id. at pp. 1159-1160.) The trial court ruled that Fireman’s was required to
    contribute its full policy limits. (Id. at pp. 1160-1161.)
    The appellate court upheld this ruling. (Fireman’s Fund Ins. Co. v. Allstate Ins.
    Co., supra, 234 Cal.App.3d at pp. 1161-1172.) It further held that Allstate and
    Northbrook were entitled to prejudgment interest against Fireman’s. (Id. at pp. 1172-
    1174.) It explained: “Here, the principal dispute was one of liability . . . . Once liability
    was established, the extent of that liability was no longer a question. . . . [¶] Whatever
    40
    uncertainty about the extent of Fireman’s liability may have been fostered by the
    alternative theories Fireman’s proposed, we do not view that uncertainty as an
    impediment to the award of prejudgment interest. While Fireman’s proposed a general
    formula based on four inapt theories of lesser liability, it also suggested a specific amount
    due Allstate and Northbrook under each theory. Through it all, the extent of Fireman’s
    exposure remained purely a question of law. . . . By pursuit of its various suggested
    alternatives, Fireman’s could neither confuse the magnitude of its obligation nor alter
    applicable law.” (Fireman’s Fund Ins. Co. v. Allstate Ins. Co., supra, 234 Cal.App.3d at
    pp. 1173-1174.)
    Continental argues that Fireman’s is not controlling because there, according to
    Continental, the only disputed issue was whether Fireman’s had validly canceled its
    policy. Once again, the court’s own discussion of Fireman’s four alternative theories
    demonstrates that this is incorrect.
    Finally, in Shell Oil Co. v. National Union Fire Ins. Co. (1996) 
    44 Cal.App.4th 1633
    , an injured worker sued Shell as well as S.I.P., an engineering firm that was doing
    work at a Shell refinery. (Id. at pp. 1637-1638.) National had issued a policy to S.I.P.
    (Id. at p. 1638.) National paid its policy limits to settle the case against S.I.P., but it
    denied that its policy afforded any coverage to Shell. Shell then sued National. The trial
    court ruled that Shell was covered as an additional insured. It awarded Shell half the
    policy limits as damages; it also awarded prejudgment interest. (Id. at p. 1639.)
    41
    On appeal, National argued that Shell was not entitled to prejudgment interest
    “because [Shell had] proposed three different measures of damages to the [trial] court, all
    of which were in dispute.” (Shell Oil Co. v. National Union Fire Ins. Co., supra, 44
    Cal.App.4th at p. 1651, fn. omitted.) The appellate court disagreed, stating: “Shell’s
    alternative theories required only the court’s legal determination of which was
    appropriate; the amount of damages would thereby be fixed. The present case thus
    resembles Hartford . . . .” (Ibid.)
    Continental asserts that in Shell, “the issue with respect to prejudgment interest did
    not concern damages. Instead, it turned on whether the insurer was liable to cover Shell.
    The distinction was not between legal issues and factual issues, but between coverage
    liability and damages.” Admittedly, there was a dispute over coverage. However, that
    was not why National claimed that damages were uncertain. Rather, it claimed that
    damages were uncertain because Shell had proposed three alternative measures of
    damages. Moreover, the appellate court rejected this argument expressly because the
    correct measure of damages presented only a legal issue.
    It has been said that “[a] legal dispute concerning the defendant’s liability or the
    proper measure of damages . . . does not render damages unascertainable. [Citations.]”
    (Collins v. City of Los Angeles, supra, 205 Cal.App.4th at p. 151, italics added; accord,
    Uzyel v. Kadisha (2010) 
    188 Cal.App.4th 866
    , 919.) This formulation is consistent with
    Hartford, Fireman’s, and Shell, and we agree with it.
    42
    Continental nevertheless argues that St. Paul Mercury Ins. Co. v. Mountain West
    Farm Bureau Mut. Ins. Co. (2012) 
    210 Cal.App.4th 645
     supports the proposition that
    whenever “the amount of damages . . . depends upon a resolution in court, prejudgment
    interest is not available . . . .” We disagree.
    St. Paul was the liability insurer for a general contractor. (St. Paul Mercury Ins.
    Co. v. Mountain West Farm Bureau Mut. Ins. Co., supra, 210 Cal.App.4th at p. 650.)
    After settling a construction defect action against its insured (id. at pp. 651-652), St. Paul
    sued Mountain West — the insurer of a subcontractor — for contribution. (Id. at pp. 650,
    652.)
    The appellate court upheld the judgment against Mountain West. (St. Paul
    Mercury Ins. Co. v. Mountain West Farm Bureau Mut. Ins. Co., supra, 210 Cal.App.4th
    at pp. 653-664.) Among other things, it held that the trial court did not abuse its
    discretion by allocating 43 percent of the defense costs to Mountain West, based, in part,
    on its factual findings about the parties’ relative responsibility for the construction
    defects. (Id. at pp. 662-664.) It further held that the trial court did not abuse its
    discretion by allocating settlement costs using the time-on-the-risk theory rather than the
    premiums-paid theory or any of the other available theories. (Id. at p. 664.)
    It then reversed the award of prejudgment interest. (St. Paul Mercury Ins. Co. v.
    Mountain West Farm Bureau Mut. Ins. Co., supra, 210 Cal.App.4th at pp. 665-666.) It
    acknowledged that “[g]enerally, nonparticipating coinsurers can be ‘liable for interest on
    their prorated share of the loss, computed from the date of settlement’ because that is the
    43
    date that the loss is certain or capable of being made certain by calculation. [Citation.]”
    (Id. at p. 665.) However, it continued: “Here, no dispute existed about the amounts paid
    in settlement; the parties stipulated to the sum. Indeed, . . . ‘[t]he settlement and the
    amount of the settlement are . . . presumptive evidence of the insurer’s liability and the
    amount of liability. [Citation.]’ [Citation.] However, the issues at trial included not only
    legal questions about whether Mountain West was responsible for contribution, but also
    the allocation of responsibility and hence the amount of that contribution. The trial court
    was asked to choose the method of allocation, i.e., the basis for computation, and to
    calculate the exact portion of the stipulated defense and settlement costs Mountain West
    would have to contribute. Thus, Mountain West was not able to compute the damages
    [citation] and its share of the loss was not ‘“certain, or capable of being made certain by
    calculation”‘ [citations] until the trial court determined what method of allocation was
    most equitable. Consequently, the trial court erred in awarding prejudgment interest.”
    (Id. at pp. 665-666.)
    We perceive no conflict between St. Paul and the cases we have cited (Hartford,
    Fireman’s, and Shell). Collectively, they stand for one simple test: When the allocation
    of indemnity among insurers turns on factual issues, damages are uncertain and pretrial
    interest is unavailable; when it turns exclusively on legal issues, damages are certain and
    pretrial interest is available.8
    8       In addition to St. Paul, Continental also cites Wisper Corp. v. California
    Commerce Bank (1996) 
    49 Cal.App.4th 948
    . Wisper is not an insurance case; however, it
    held that prejudgment interest was not available because comparative fault as between the
    [footnote continued on next page]
    44
    Continental complains that it could not know how much, if anything, it should pay
    the State until the trial court made (1) the no-annualization ruling, (2) the all-sums ruling,
    (3) the no-stacking ruling, (4) the one-occurrence ruling, and (5) the vertical exhaustion
    ruling.9 This argument, based on fairness, has a certain superficial appeal. However, it
    cannot be squared with the well-established principle that damages can be deemed certain
    even if the defendant wholly denies liability for them. (Leff v. Gunter, supra, 33 Cal.3d
    at pp. 519-520; accord, Howard v. American Nat. Fire Ins. Co., supra, 187 Cal.App.4th
    at p. 535; Boehm & Associates v. Workers' Comp. Appeals Bd. (1999) 
    76 Cal.App.4th 513
    , 517.) This is the rule, despite the fact that the defendant can hardly be expected to
    pony up until its liability has been finally adjudicated.
    What is critical is not whether the defendant actually knows how much it should
    pay; rather, it is whether the defendant could have calculated how much it should pay, if
    it had known how a court would ultimately rule on the legal issues. (See, e.g., Olson v.
    [footnote continued from previous page]
    plaintiffs and the defendant had been “hotly disputed” and thus “[t]he amount of damage
    could not be determined until after trial.” (Id. at p. 962.) From the court’s references to
    the “factual environment” of the comparative negligence claim and to the jury as the trier
    of fact” (ibid.), it seems clear that the court viewed comparative negligence as an issue of
    fact, not an issue of law, under the circumstances before it. Thus, it is consistent with our
    reading of the insurance cases.
    9      Indeed, Continental asserts that some of these issues remained open, and
    thus precluded certainty, until this court’s opinion in 2009 or the Supreme Court’s
    opinion in 2012. Presumably, in Continental’s view, even though we are affirming the
    trial court’s vertical exhaustion ruling (see part III, ante), damages are still uncertain and
    will remain uncertain until the Supreme Court affirms, reverses, or denies review.
    45
    Cory, supra, 35 Cal.3d at p. 402 [parties’ dispute over legal issue of whether city charter
    amendment was unconstitutional did not prevent damages from being certain].)
    We now apply these principles to Continental’s various claims of uncertainty.
    B.     Uncertainty Regarding Vertical Exhaustion.
    First, Continental contends that the damages were uncertain as long as there was a
    dispute as to whether vertical or horizontal exhaustion applied.
    This dispute, however, presented a purely legal question, turning on the
    interpretation of the relevant policy language. The trial court made its ruling that vertical
    exhaustion applied as a matter of law, in the context of cross-motions for summary
    judgment.    In part III, ante, we reviewed that ruling as a matter of law, and we affirmed
    it.
    Continental nevertheless argues that this issue goes to damages rather than
    liability, and hence damages were uncertain until it was finally resolved. As already
    discussed, however, the liability-damages dichotomy is not controlling; rather, what is
    controlling is the legal-factual dichotomy.
    If only out of an excess of caution, however, we note that the exhaustion issue
    actually does go to liability. It must be remembered that the liability at issue is not the
    State’s liability for remediation costs, which continues to mount; rather, it is
    Continental’s contractual liability to the State. Moreover, for purposes of prejudgment
    interest, the key question was whether, at the moment the Rule 54(b) judgment was
    entered, all applicable retentions and lower-level policies were exhausted. If vertical
    46
    exhaustion applied, then the answer was yes; Continental’s policies not only attached, but
    were fully exhausted, so that Continental was liable for its policy limits. On the other
    hand, if horizontal exhaustion applied, then, as Continental concedes, “$80 million would
    not have begun to implicate [Continental’s] policies, much less exhaust them.” I.e., at
    that moment, Continental was not liable at all.
    C.     Uncertainty Regarding the Number of Occurrences.
    Next, Continental contends that the damages were uncertain as long as there was a
    dispute as to whether there was one covered occurrence or five covered occurrences.
    Each insurer’s policy limits applied on a “per occurrence” basis. (See State of
    California v. Continental Ins. Co., supra, 88 Cal.Rptr.3d at pp. 295-296.) Continental
    argues that, if there were five occurrences, “that would have resulted in five times the
    underlying limits that would need to be exhausted before [Continental’s] excess policies
    were triggered.”
    Again, this dispute presented a purely legal question that turned on the
    interpretation of the relevant policy language. The parties presented this issue in the trial
    court in the form of a motion for summary judgment and, in the alternative, a motion in
    limine. (State of California v. Continental Ins. Co., supra, 88 Cal.Rptr.3d at p. 314.) The
    trial court resolved it by applying the law to the undisputed facts.
    In the previous appeal, we affirmed the trial court’s one-occurrence ruling. We
    held that the five occurrences listed by the State (reduced to four by the time of the
    47
    appeal) actually constituted only one occurrence as a matter of law. (State of California
    v. Continental Ins. Co., supra, 88 Cal.Rptr.3d at pp. 313-317.)
    Continental nevertheless argues that this issue goes to damages rather than
    liability. Once again, however, it actually does go to liability. We may assume, without
    deciding, that Continental’s analysis of the impact of the one-occurrence ruling is correct.
    Even if so, however, that ruling determines whether, upon entry of the Rule 54(b)
    judgment, Continental was or was not liable. Continental has not shown that there was
    any scenario in which it was liable for some amount between $0 and $12 million, but the
    exact amount turned on the one-occurrence ruling.
    D.     Uncertainty Regarding Annualization.
    Continental also contends that the damages were uncertain as long as there was a
    dispute as to whether policy limits applied per year or per multi-year policy period. It
    argues that, if annualization applied, there would be a larger pool of money that would
    have to be exhausted before Continental’s policies would be triggered.
    Once again, this dispute presented a purely legal question. In its no-annualization
    ruling, the trial court ruled that policy limits applied just once per policy period.
    Although it held a bench trial and took evidence, it indicated it reached this particular
    ruling as a “conclusion[] of law,” based on “the plain meaning rule” as applied to the
    relevant policy provisions. We affirmed this ruling, also as a matter of law. (State of
    California v. Continental Ins. Co., supra, 88 Cal.Rptr.3d at pp. 317-319.)
    48
    Also, like the one-occurrence ruling, the no-annualization ruling, if relevant at all,
    goes to whether Continental’s policies attached, which is a question of liability.
    E.     Uncertainty Regarding “All-Sums” and Stacking.
    Finally, Continental contends that the damages were uncertain as long as there
    were disputes over the all-sums approach and stacking.
    Yet again, these disputes presented purely legal questions. The trial court
    rendered its all-sums ruling on motion, as a question of law, based on case law that it
    deemed to be controlling. The Supreme Court affirmed this ruling; it held as a matter of
    law that the all-sums approach follows from standard policy language. (State of
    California v. Continental Ins. Co., supra, 55 Cal.4th at pp. 196-200.) For similar reasons,
    it further held that stacking applied. (Id. at pp. 200-202.)
    And again, these rulings go to whether Continental’s policies attached at all; this is
    a question of liability, not damages. As Continental itself states, “under a ‘pro rata’
    approach, even $300 million in damages would not have impacted [Continental’s]
    policies because the damages would have been spread over the period of continuing
    property damage — here, a period of almost 50 years. . . . [T]hat would be about $6
    million each year, which would not implicate [Continental’s] policies even on a vertical
    basis.” Indeed, because all of Continental’s policies were excess to retentions of at least
    $16 million, it would have taken $800 million in damages over 50 years to trigger them
    on a pro rata basis.
    49
    VII
    FAILURE TO APPLY OFFSETS
    Continental contends that the trial court erred by failing to treat approximately
    $160 million that the State had received from other sources as offsets. It asserts that:
    “The $160 million in offsets the State has received . . . must be applied to the State’s
    damages[] before any determination can be made as to [Continental’s] share of those
    damages.”10
    By the time of trial, the State had received approximately $150 million in
    settlements with other insurers and approximately $10 million in settlements with other
    potentially responsible parties. Continental argued that the State’s damages had to be
    offset by this $160 million before prejudgment interest could be calculated. The trial
    court rejected this argument because the State had not yet been fully compensated:
    “While settlements with other insurers over the years have compensated the State for a
    large portion of its damages, a substantial portion has remained unpaid.”
    As discussed in part V, ante, the Rule 54(b) judgment established that, as of 1998,
    the State was liable for at least $80 million. However, this was a floor, not a ceiling. It
    represented only the State’s liability as of February 29, 1992. By September 1998, the
    10      Continental is not arguing that the question of whether to treat the $160
    million as offsets made the State’s damages uncertain. (Cf. part VI, ante.) Presumably
    this is because it is hornbook law that “only the claimant’s damages themselves must be
    certain. Damages are not made uncertain by the existence of unliquidated counterclaims
    or offsets interposed by defendant. [Citation.]” (Howard v. American Nat. Fire Ins. Co.,
    supra, 187 Cal.App.4th at p. 536.)
    50
    United States’s remediation costs had increased to $116 million; meanwhile, the State
    was incurring remediation costs of its own, which by 2015 totaled over $200 million.11
    At trial, the parties stipulated that “the State is going to incur more money as to the
    Stringfellow site.”
    By December 1998, when the Rule 54(b) judgment was entered, the State had
    recovered only a relatively trivial amount ($3,050,000) in settlements. Thus, as already
    discussed (see part V, ante), the $80 million liability established by the Rule 54(b)
    judgment not only caused Continental’s policies to attach but also fully exhausted them,
    and thus made Continental liable to the State for its $12 million policy limits. Thereafter,
    the State began to collect more and larger settlements; but at the same time, its liability
    continued to grow. Continental has not shown, and on this record it cannot show, that the
    settlements exceeded the State’s liability.
    Continental is not entitled to any offset unless it can prove that an offset is
    necessary to prevent a double recovery. “As the party seeking the offset, [Continental]
    11    The trial court ruled that Continental was liable for the State’s own
    remediation costs. In a footnote, Continental contends that this was error.
    It has forfeited this contention by failing to raise it under a separate heading or
    subheading and by failing to support it with reasoned argument and citation of authority,
    as required. (Cal. Rules of Court, rule 8.204(a)(1)(B); Parisi v. Mazzaferro (2016) 
    5 Cal.App.5th 1219
    , 1226, fn. 10.) We need not consider arguments “asserted
    perfunctorily on appeal in a footnote as an afterthought [citation].” (Bollay v. California
    Office of Administrative Law (2011) 
    193 Cal.App.4th 103
    , 111.)
    51
    had the burden of proving the facts essential to it.” (Conrad v. Ball Corp. (1994) 
    24 Cal.App.4th 439
    , 444.)
    “The fact that several insurance policies may cover the same risk does not . . . give
    the insured the right to recover more than once. Rather, the insured’s right of recovery is
    restricted to the actual amount of the loss. Hence, where there are several policies of
    insurance on the same risk and the insured has recovered the full amount of its loss from
    one or more, but not all, of the insurance carriers, the insured has no further rights against
    the insurers who have not contributed to its recovery. Similarly, the liability of the
    remaining insurers to the insured ceases, even if they have done nothing to indemnify . . .
    the insured.” (Fireman’s Fund Ins. Co. v. Maryland Cas. Co. (1998) 
    65 Cal.App.4th 1279
    , 1295, first italics added, second italics in original.)
    “[I]n some circumstances, an insurer may recover funds paid to the insured by a
    legally responsible third party, even though the insurer did not participate in the insured’s
    legal action against the third party. [Citations.] [¶] A determining factor on this matter
    is whether the insured has been ‘made whole.’ [Citation.]” (Plut v. Fireman’s Fund Ins.
    Co. (2000) 
    85 Cal.App.4th 98
    , 104.)
    Continental failed to show that the State had been made whole. Thus, the State is
    entitled to every dollar from any settlement or similar recovery and to $12 million from
    Continental, unless and until Continental can show that the State’s uncompensated loss
    has been reduced below $12 million.
    52
    Continental relies heavily on the trial court’s 2006 offset ruling. At that time,
    based on the no-stacking ruling (which we later reversed), the most the State could
    recover was $48 million; the trial court ruled that the State’s settlements with other
    insurers had to be offset against this, leaving the State with zero damages. Continental
    argues: “[T]hat decision was not reversed on appeal. [Citation.] As a result, the trial
    court was required to apply offsets.”
    “If an order is appealable, however, and no timely appeal is taken therefrom, the
    issues determined by the order are res judicata. [Citation.]” (In re Matthew C. (1993) 
    6 Cal.4th 386
    , 393.) Generally, this principle limits a litigant’s ability to challenge a prior
    unappealed order: “An appeal from the most recent order . . . may not challenge earlier
    orders for which the time for filing an appeal has passed. [Citation.]” (Sara M. v.
    Superior Court (2005) 
    36 Cal.4th 998
    , 1018 [dependency].) We are not aware of any
    case holding that this principle limits a trial court’s ability to reconsider a prior
    unappealed order that has not become part of a final judgment; certainly Continental cites
    none. To the contrary, as a general rule, a trial court has the inherent authority “on its
    own motion, to reconsider its prior interim orders so it may correct its own errors.” (Le
    Francois v. Goel (2005) 
    35 Cal.4th 1094
    , 1107.)
    In any event, the offset ruling was appealed. The State expressly challenged it in
    the prior appeal. (State of California v. Continental Ins. Co., supra, 88 Cal.Rptr.3d at
    p. 319.) We declined to reach this issue, because we determined that it was moot. (Id. at
    p. 320.) We relied on the State’s representation that, provided we reversed the no-
    53
    stacking ruling, there would be no remaining issue as to offset. (Ibid.) Even assuming
    that was a mistake, the fact remains that we did not affirm or reverse the offset ruling.
    That left its correctness an open question.
    wFinally — and again, separately and alternatively — the offset ruling was made
    under significantly different factual circumstances. At that time, because it appeared that
    the most the State could recover was $48 million, the State’s settlements with other
    insurers were more than sufficient to make it whole. Thus, Continental would be entitled
    to an offset.12
    Once we reversed the no-stacking ruling, however, it became apparent that the
    State could recover more than $48 million — it could recover up the full extent of its
    liability (as long as that was within the applicable policy limits). Accordingly, for the
    reasons just discussed, Continental is not entitled to an offset. The offset ruling simply
    does not apply.
    12      Continental asserts that, in making the offset ruling, the trial court “rejected
    the State’s arguments that offsets cannot be applied until the State was ‘made whole.’” In
    support of that statement, Continental cites our opinion in the previous appeal; however,
    the pages cited do not actually support it.
    54
    VIII
    DISPOSITION
    The judgment is affirmed. The State is awarded costs on appeal against
    Continental.
    CERTIFIED FOR PARTIAL PUBLICATION
    RAMIREZ
    P. J.
    We concur:
    McKINSTER
    J.
    SLOUGH
    J.
    55
    Attachment A
    56
    Attachment B
    57
    Attachment C
    58