An insurer’s duty to pay for settlements

A look at the insurer’s duty to pay for settlements even when it has reserved its rights to deny coverage

2017 September

Liability insurers frequently reserve their rights to deny coverage when their insureds are sued. When they do so, an insured – and often those suing an insured – faces a key question. When must an insurer pay for a settlement of a lawsuit against its insured?

As one court has stated, “[E]ven if the insurer has not denied coverage or refused to defend, the insurer has a duty to accept a reasonable settlement, and the insurer’s refusal to settle may give rise to the insured’s action for reimbursement of the settlement. . . . In such a case, the insured has the burden of showing the settlement was reasonable and if it meets that burden, then again the act of settlement raises two presumptions: that the claim was legitimate and that the amount of the settlement was the amount of the insured’s liability.” (Armstrong World Indus., Inc. v. Aetna Cas. & Sur. Co. (1996) 45 Cal.App.4th 1, 85, 52 Cal.Rptr.2d 690.)

Most insurance policies include a clause commonly referred to as a “cooperation” clause. A typical cooperation clause states, “You and any other involved insured must . . . Cooperate with us in the investigation or settlement of the claim or defense against the ‘suit’ . . . .” Commercial General Liability Coverage Form § IV.2.c(3) (Insurance Services Office, Inc. 2012).

The consent clause

Almost every insurance policy also has a clause typically referred to as a “voluntary payments” or “consent” clause. One common version states, “No insured will, except at that insured’s own cost, voluntarily make a payment, assume any obligation, or incur any expense, other than for first aid, without our consent.” (Supra, § IV.2.d.)

Generally speaking, at least in California (and the law varies from state to state), an insured’s failure to perform a condition will not excuse an insurer’s performance unless the insurer is actually and substantially prejudiced thereby. In fact, this standard has been applied to the cooperation clause. California courts recognize that an insurer “may assert defenses based upon a breach by the insured of a condition of the policy such as a cooperation clause, but the breach cannot be a valid defense unless the insurer was substantially prejudiced thereby.” (Campbell v. Allstate Ins. Co. (1963) 60 Cal.2d 303, 305, 32 Cal.Rptr. 827 [“[P]rejudice is not shown simply by displaying end results; the probability that such results could or would have been avoided absent the claimed default or error must also be explored”]; Clemmer v. Hartford Ins. Co. (1978) 22 Cal.3d 865, 883 n.12, 151 Cal.Rptr. 285 [“[A]n insurer, in order to establish it was prejudiced by the failure of the insured to cooperate in his defense, must establish at the very least that if the cooperation clause had not been breached there was a substantial likelihood the trier of fact would have found in the insured’s favor.”]; Billington v. Interins. Exch. of S. California (1969) 71 Cal.2d 728, 737, 79 Cal.Rptr. 326 [same].)

However, this prejudice test has not been uniformly applied with respect to the consent clause. Indeed, many courts have held that if an insured does not obtain its insurer’s consent before entering into a settlement, or perhaps even before commencing settlement negotiations, then an insurer need not fund a settlement. One court observed:

Ours is the rare case where the insured tenders the defense and negotiates a settlement on its own, leaving the insurer in the dark. . . . [W]e find no case holding that a post tender breach of [a consent] provision is unenforceable. Indeed, language from a leading Supreme Court decision indicates that such a provision is enforceable post tender until the insurer wrongfully denies tender.

(Low v. Golden Eagle Ins. Co. (2003) 110 Cal.App.4th 1532, 1546, 2 Cal.Rptr.3d 761; see also Crowley Mar. Corp. v. Federal Ins. Co. (N.D. Cal. Dec. 1, 2008, No. C 08-00830 SI) 2008 U.S.Dist. LEXIS 97393, at *16 [granting summary judgment for insurer when it was “undisputed that [the insured] began settlement negotiations in December 2006 without [insurer’s] knowledge,” “executed a settlement agreement on March 19, 2007,” and insurer was “first informed of the settlement no earlier than March 22, 2007”], aff’d (9th Cir. 2010) 373 F.App’x 782, 783.

Did the insurer breach the contract?

An insurer’s consent to a settlement is not required if an insurer has denied coverage or otherwise materially breached its duties. As one court explained, “an insurer is not allowed to rely on an insured’s failure to perform a condition of a policy when the insurer has denied coverage because the insurer has, by denying coverage, demonstrated performance of the condition would not have altered its response to the claim.” (Select Ins. Co. v. Superior Court (1990) 226 Cal.App.3d 631, 637, 276 Cal.Rptr. 598; see also Jamestown Builders, Inc. v. Gen. Star Indem. Co. (1999) 77 Cal.App.4th 341, 347-48, 91 Cal.Rptr. 2d 514 [“The no-voluntary-payments provision is superseded by an insurer’s antecedent breach of its coverage obligation.”). Indeed, consent conditions are only enforceable “‘in the absence of . . . insurer breach.’”]; Low v. Golden Eagle Ins. Co. (2003) 110 Cal.App.4th 1532, 1544, 2 Cal.Rptr.3d 761 [quoting Jamestown]; Croskey, Heeseman, Ehrlich, & Klee, California Practice Guide: Insurance Litigation § 7:439.7 (Rutter Group 2017) [consent provisions are enforceable only “in the absence of any breach by the insurer . . . .”].)

Furthermore, in a breach situation, the insured is not obligated to keep the carrier informed of any developments thereafter. (Samson v. Transamerica Ins. Co. (1981) 30 Cal.3d 220, 238, 178 Cal.Rptr. 343 [“[I]f an insurer denies coverage to the insured, the . . . insured is relieved of his obligation to inform the insurance company of the service of summons or the date of trial of the action.”]) As the California Supreme Court explained, the denial of coverage and a defense entitles the insured to make a reasonable, noncollusive settlement without the insurer’s consent and to seek reimbursement for the settlement amount in an action for breach of the covenant of good faith and fair dealing. (Hamilton v. Maryland Cas. Co. (2002) 27 Cal.4th 718, 732, 117 Cal.Rptr.2d 318; see also Pruyn v. Agricultural Ins. Co. (1995) 36 Cal.App.4th 500, 515, 42 Cal.Rptr.2d 295 [“Courts have for some time accepted the principle that an insured who is abandoned by its liability insurer is free to make the best settlement possible with the third party claimant . . . . Provided that such settlement is not unreasonable and is free from fraud or collusion, the insurer will be bound thereby.”])

Even if an insurer refuses to fund more than part of a settlement, “the insured may conclude a favorable settlement by contributing the deficit itself and, assuming the insurer’s breach can be proven, recover the payment in a subsequent action for breach of the covenant of good faith and fair dealing.” (Hamilton, 27 Cal.4th at p. 732.)

[W]hen a primary insurer wrongfully denies coverage, unreasonably delays processing a claim, or refuses to defend an action against the insured as required by the policy, the insured is entitled to make a reasonable settlement of the claim and then sue for reimbursement, even though the policy prohibits settlements without the consent of the insurer. . . . The insurer is deemed to have waived its rights under the ‘no action’ clause by such conduct constituting a breach of its obligations under the policy.

(Diamond Heights Homeowners Ass’n v. Nat’l Am. Ins. Co. (1991) 227 Cal.App.3d 563, 581, 277 Cal.Rptr. 906.)

Indeed, “where – because of the insurer’s reservation of rights based on possible noncoverage under the policy – the interests of the insurer and the insured diverge, the insurer must pay reasonable costs for retaining independent counsel by the insured.” (Hartford Cas. Ins. Co. v. J.R. Mktg., L.L.C. (2015) 61 Cal.4th 988, 997-98, 190 Cal.Rptr.3d 599.) In that circumstance, the insured’s independent counsel “then controls the litigation.” (Assurance Co. of Am. v. Haven (1995) 32 Cal.App.4th 78, 84, 38 Cal.Rptr.2d 25; see Fuller-Austin Insulation Co. v. Highlands Ins. Co. (2006) 135 Cal.App.4th 958, 983, 38 Cal.Rptr.3d 716 [“by reserving their rights instead of acknowledging coverage and assuming the defense of the matter, [the insurers] surrendered their right to rely on any policy provision requiring their consent to a settlement”]; Intergulf Dev. LLC v. Superior Court (2010) 183 Cal.App.4th 16, 20, 107 Cal.Rptr.3d 162 [“Breach of duty to defend also results in the insurer’s forfeiture of the right to control defense of the action or settlement, including the ability to take advantage of the protections and limitations set forth in [Civil Code] section 2860.”])

What triggers the insurer’s duty to pay?

Assuming that an insurer consents, or does not object to a settlement, the question that then must be addressed is: What must an insured show to obtain coverage for a settlement? Can an insured simply point to the allegations against it and the same potential for coverage standard that triggers the insurer’s duty to defend? (See, e.g., Gray v. Zurich Ins. Co. (1966) 65 Cal.2d 263, 276-77, 54 Cal.Rptr. 104 [“An insurer, therefore, bears a duty to defend its insured whenever it ascertains facts which give rise to the potential of liability under the policy”]; Montrose Chem. Corp. v. Superior Court (1993) 6 Cal.4th 287, 300, 24 Cal.Rptr.2d 467 [the duty to defend exists if there is simply “a bare ‘potential’ or ‘possibility’ of coverage.”])

In Axis Surplus Insurance Co. v. Glencoe Insurance Ltd. (2012) 204 Cal.App.4th 1214, 1223, 139 Cal.Rptr.3d 578, the court addressed the settlement of a construction-defect lawsuit when the settling insurer funded and sought contribution from a non-participating insurer. The court rejected the non-participating insurer’s argument that the settling insurer had to prove that the non-participating insurer’s self-insured retention applied only to “covered property damage.” The court explained:

By settling, the parties forgo their right to have liability ‘establish[ed]’ by a trier of fact, and the settlement ‘becomes presumptive evidence of the [insured’s] liability and the amount thereof, which presumption is subject to being overcome by proof. . . . ‘A contrary rule would make the right to settle meaningless . . . .’” (Id., quoting Phoenix Ins. Co. v. United States Fire Ins. Co. (1987) 189 Cal.App.3d 1511, 1526-27, 235 Cal.Rptr. 185.)

The court emphasized that, “By settling, the parties avoid the cost and uncertainty of litigation. The settlement and the amount of the settlement are thus presumptive evidence of the insurer’s liability and the amount of liability. In other words, the settlement is presumed to be made for only damages covered under the applicable policy.” (Id. at p. 1224, citation omitted).

In Advent, Inc. v. National Union Fire Insurance Co. of Pittsburgh, Pa. (2016) 6 Cal.App.5th 443, 211 Cal.Rptr.3d 685, an insurer that had funded a settlement sought contribution from an insurer that refused to participate. The court held that “‘where [the nonparticipating coinsurer’s] duty to defend is undisputed, and where by law the settlements are presumptively reasonable[,] the burden of proof is on [the nonparticipating coinsurer] to establish that there was no coverage (and not on the [settling coinsurer] to prove the opposite.’”). (Id. at p. 455.)

The court further explained: “Certainly, in cases where the plaintiff has not settled the case, liability would often be conclusively established by the ensuing trial’s outcome.” (Id. at p. 457.) However, by settling, parties “forgo their right to have liability ‘established’ by a trier of fact, and the settlement ‘becomes presumptive evidence of the [insured’s] liability and the amount thereof, which presumption is subject to being overcome by proof.’” (Id., citation omitted). The court stated that when an insurer continues to reserve its rights throughout the settlement process, “the settlement cannot be used as presumptive evidence that [the insurer’s] policy applies.” (Id.) In that circumstance, the court is “left” to consider the allegations in the underlying complaint. In that situation, the settling party “can satisfy its initial burden to make a prima facie showing of actual coverage if it can demonstrate a potential for coverage based on known extrinsic facts and the allegations of the complaint . . . [T]his is the same as the potential liability triggering a duty to defend.” (Ibid.) It would then be incumbent upon the insurer to present “affirmative evidence negating an essential element of [the insured’s] claim” or to prove “each element of an affirmative defense . . . , such as the affirmative defense that there was no actual coverage.” (Ibid.)

In other words, the insured must only “show that there is a potential for coverage,” which would then mandate that the insurer prove, at least in opposing a summary judgment motion, that “there is evidence or a reasonable inference showing that there is no actual coverage.” (Id. at p. 458.) By comparison, an insurer could not prevail on a motion for summary judgment disputing coverage for the settlement unless it could “satisfy its initial burden by showing there are undisputed facts supporting each element of its affirmative defense of lack of coverage.” (Id. at p. 458.) In that circumstance, the insured could still defeat the motion if it proved that there was “evidence or a reasonable inference that there is coverage.” (Ibid.)

What if the insurer has denied coverage?

This still leaves another question: What must an insured do if an insurer has denied coverage or failed to respond to notice? In that situation, an insured is not obligated to keep the insurer informed of any developments thereafter. (See Samson v. Transamerica Ins. Co., 30 Cal.3d at p. 238 [noting that an insurer’s denial of coverage relieves the insured of its obligation to keep the insurer informed about developments in the case].) An insurer’s “denial of coverage and a defense entitles the policyholder to make a reasonable, noncollusive settlement without the insurer’s consent and to seek reimbursement for the settlement amount in an action for breach of the covenant of good faith and fair dealing.” (Hamilton v. Maryland Cas. Co. (2002) 27 Cal.4th 718, 728, 117 Cal.Rptr.2d 318.)

In Teleflex Medical Inc. v. National Union Fire Insurance Co. (9th Cir. 2017) 851 F.3d 976, an insurer and insured determined to settle an underlying lawsuit. The excess insurer did not agree, questioning whether the potential liability would reach its layer. It declined to consent to the settlement. The insured proceeded to settle, thereafter suing the excess insurer. The excess insurer argued that it had the “absolute right to veto the settlement under the policy’s ‘no voluntary payments’ and ‘no action’ clauses.” The Ninth Circuit disagreed, stating the fact that “‘no action’ and ‘no voluntary payment’ clauses do not create absolute rights to veto settlements is long established.” It stated: “Many courts have held that ‘when a primary insurer wrongfully denies coverage, unreasonably delays processing a claim, or refuses to defend an action against the insured as required by the policy, the insured is entitled to make a reasonable settlement of the claim in good faith and then sue for reimbursement, even though the policy prohibits settlements without the consent of the insurer.’” (Id. at 984-85. quoting Diamond Heights, 227 Cal.App.3d at p. 581).

In so ruling, the court rejected the excess insurer’s arguments that the insured’s liability was uncertain, that discovery was not complete, that there was no pending trial date, and that the insured had no exposure beyond the total available insurance. The court explained that settlements could be found “to be reasonable assessments of potential liability,” and that “there may be good reasons to settle mid-discovery, such as the risk of disclosing damaging documents, rather than on the eve of trial.” (Id. at p. 986.) The court also pointed out that while the insured might not have had “a strong incentive to achieve the lowest settlement,” and may have had “an incentive to fork over its insurers’ money,” there was substantial evidence before it to support the conclusion that “the settlement was reasonable and not a product of collusion.” (Id. at p. 987.)

Thus, when an insurer has reserved its rights to deny coverage or has denied coverage, the insured may have substantial flexibility in determining whether to settle a suit against it and may, indeed, have the right to do so without its insurer’s consent.

Settlements reached at a mediation

These issues are even more complicated when a settlement is reached at a mediation under California law. This is simply because evidence of communications at a mediation is, by statute, neither discoverable nor admissible in subsequent legal proceedings, absent the written consent of all participants, including the mediator. (Evid. Code, § 1119 [“No evidence of anything said or any admission made for the purpose of, in the course of, or pursuant to, a mediation or a mediation consultation is admissible or subject to discovery, and disclosure of the evidence shall not be compelled . . . .”].)

California courts have enforced these nondisclosure/nondiscovery/nonadmissibility requirements, even in extreme circumstances. For example, in Foxgate Homeowners’ Association v. Bramalea California, Inc. (2001) 26 Cal.4th 1, 108 Cal.Rptr.2d 642], the California Supreme Court considered the appropriateness of a sanctions order based on allegedly dilatory and obstructive conduct during a mediation. The Superior Court had granted sanctions against a party and its counsel, based in part upon a mediator’s declaration that counsel had aborted the mediation session by refusing to participate in good faith. The California Supreme Court ruled that the order for sanctions was inappropriate and constituted an irregularity in the proceedings. (Id. at p. 18.) It ruled that if a sanctions motion were pursued on remand, “‘No evidence of communications made during the mediation may be admitted or considered.’” (Ibid.) It reasoned that because mediation confidentiality is designed to promote “‘a candid and informal exchange regarding events in the past,’” that exchange “‘is achieved only if the participants know that what is said in the mediation will not be used to their detriment through later court proceedings and other adjudicatory processes.’” (Id. at p. 14.) The court expressly stated:

[W]e do not agree . . . that the court may fashion an exception for bad faith in mediation because failure to authorize reporting of such conduct during mediation may lead to ‘an absurd result’ or fail to carry out the legislative policy of encouraging mediation. (Id at p. 17.)

Other decisions following this rule include Eisendrath v. Superior Court (2003) 109 Cal.App.4th 351, 364, 134 Cal.Rptr.2d 716 [“[T]he confidentiality rule in section 1119 sweeps broadly . . . . Thus, the confidentiality rule in section 1119 encompasses communications by participants before the end of mediation that are materially related to the purpose of the mediation, regardless of whether these communications are made in the mediator’s presence”]; Cassel v. Superior Court (2011) 51 Cal.4th 113, 118, 119 Cal.Rptr.3d 437 [“We have repeatedly said that these confidentiality provisions are clear and absolute. Except in rare circumstances, they must be strictly applied and do not permit judicially crafted exceptions or limitations, even where competing public policies may be affected.”])

Furthermore, mediation confidentiality applies not just to communications between adversaries in a mediation and not just to communications in the mediation itself. As the California Supreme Court has explained:

The California Law Revision Commission comment . . . states, in its analysis . . . , that “mediation documents and communications may be admitted or disclosed only upon agreement of all participants, including not only parties but also the mediator and other nonparties attending the mediation (e.g., a disputant not involved in litigation, a spouse, an accountant, an insurance representative, or an employee of a corporate affiliate).

(Cassel, 51 Cal.4th at p.131, final emphasis added.)

Mediation confidentiality provisions have practical implications for parties in several contexts, particularly when an insured is seeking coverage from an insurer. If the insurer is not a participant in the mediation, then under Evidence Code section 1119, it should have no right to compel disclosure or offer into evidence any of the mediation communications. Furthermore, an insurer may not even be able to force the insured to tell the insurer what happened in the mediation, what positions were taken, or the bases for any settlement achieved at the mediation.

Indeed, in at least one lawsuit in California, an insurer’s request for settlement information was rejected by a discovery referee. In American International Specialty Lines Insurance Co. v. Coca-Cola Enterprises Inc. (Cal. San Francisco Super. Ct. Oct. 14, 2003) Case No. 320748, an insurer sought information regarding an underlying settlement that was negotiated at a mediation. The referee rejected the insurer’s request, stating:

The Referee recognizes that access to settlement information is crucial for [the carrier], and one might conclude that it appears inherently unfair that [the insured] can request reimbursement, yet not provide the details by shielding them within the mediation privilege. But the California Supreme Court has made it clear that this is the way the statutory scheme operates, and with very limited statutory exceptions, all communications, negotiations, discussions or findings resulting from a mediation are confidential.

(Slip op. at p. 14.)

Thus, the referee enforced section 1119’s confidentiality requirements, even when the insured had sought coverage for the settlement negotiated at the mediation and even though the insurer’s coverage counsel had attended at least parts of the mediation. Therefore, an insurer may find itself barred from obtaining information it believes necessary to its coverage defenses.

A similar problem is presented for insureds if they would like to use communications or materials from a mediation to establish that there was an opportunity to settle a claim against them, or that an insurer consented to a settlement. In that situation, the materials arguably would not be discoverable or admissible in evidence against an insurer in a subsequent legal proceeding without the consent of all involved in the mediation, or at least without the insurer’s consent if only it and the insured were involved in the communication. Thus, if the underlying plaintiff or the insurer objected to the insured’s disclosure of information, the insured could find itself handicapped in the presentation of evidence against the insurer – for example, to show the course of offers and counter-offers or what was said by the mediator (including any assessments of the insured’s potential liability, a mediator’s recommendation to settle, or a “mediator’s proposal”).

Therefore, the parties to a mediation may wish to address this situation by (1) obtaining, as part of a mediation agreement or a settlement, authorization to use relevant mediation communications in disputes with insurers, (2) ensuring that settlement demands also be made outside the course of the mediation, and/or (3) obtaining an insurer’s agreement to waive the protections afforded by the mediation confidentiality at least as to its consent to settle.

There is one other point that should be kept in mind: Many mediation providers require mediation participants to sign confidentiality agreements. Those agreements typically have provisions barring discovery or use of mediation communications. They, too, may serve to limit the ability to disclose or use mediation communications. Therefore, exceptions should be included in such agreements or via separate writings that ensure the ability to use mediation communications in a coverage dispute.