While commonly conflated, the doctrines are different, and there’s no good reason for lawyers for either insurers or policyholders to wrap their arguments in the wrong garb. As the Hawai'i Supreme Court recently reiterated:

In the context of insurance law, . . . the terms “waiver” and “estoppel” have often been used without careful distinction, and thereby abused and confused. . . . The fact that these doctrines are closely akin and often may coexist does not mean they are identical in connotation.

Enoka v. AIG Hawai'i Ins. Co., Inc. (Haw. Feb. 23, 2006), slip op. at 32 (citations omitted). In the Enoka case, the Hawai‘i Supreme Court was forced to “attempt to extricate” the two arguments, which were used “interchangeably throughout [the] opening brief, ” and in doing so the Hawai‘i Supreme Court helpfully collected cases nationwide on both waiver and estoppel in the insurance context. In addition to waiver and estoppel, insurance cases also may involve the less commonly invoked “mend the hold” doctrine, which like the others is a principle whose effect is to preclude an insurer from interposing an otherwise valid ground for avoidance.

By way of background, it is important to recognize that insurers have duties to investigate claims submitted by insureds, may only reasonably assert the grounds they interpose for refusing to perform, and must assist their insureds in perfecting claims against them. Let’s assume that an insurer has done all this and thereafter asserts two valid grounds for refusing to perform, each of which is sufficient to bar coverage. In such circumstances, the insurer quite obviously has no obligation to pay and has breached no duty to the insured.

Let’s say that instead, the insurer asserts two grounds for refusing to pay, but only one is valid. It is still possible (at least in some jurisdictions, including Hawai'i) for the insurer to be liable for first-party bad faith – or flipping the matter around, a valid ground for denial of coverage is not a license to commit bad faith. E.g., Enoka. But in those circumstances, the insurer will not be liable for breach of contract.

But where the situation is reversed, that is, where the invalid ground is asserted, and the valid ground is not, questions of waiver, estoppel, and mend-the-hold arise. Really, the question is whether we should allow the insurer belatedly to assert the valid ground for denial. (The use of the word “belatedly” in the previous sentence is not a question-beg but rather is a matter of definition: these issues do not arise if the assertion of the valid coverage defense is timely.)

There is nothing wrong with an insurer’s electing not to press one defense or another. Moreover, one does not wish to encourage insurers to change tack after they learn that a claim may be expensive (and covered) and seek to refuse to perform on a ground they initially thought to be a trifle. Precluding insurers from resuscitating defenses thus forecloses opportunistic behavior.

Similarly, the law does not wish to induce in insureds a false sense of security, that is, even if the insurer asserts an invalid ground, one can suppose the insured would have understood the ground not to be valid, and therefore have assumed there will be coverage (once the misunderstanding is cleared up and since the insured does not know the insurer will (later) rely on the unasserted, valid ground). Allowing insurers then to raise seriatim defense after defense undermines the objective of security that the insurance is meant in part to provide.

Further, where an insurer interposes only the invalid ground, we do not want to create a situation where an insured acts in a manner different from how it would have acted had the insurer articulated the valid ground for denying coverage; in other words, if an insured changes its position in a fashion based on the nonassertion of the coverage defense, the insurer should not later be permitted to revive the coverage defense that it could have asserted earlier.

A countervailing consideration is that absent unusual circumstances a policyholder that incurred a loss never comprehended within the insurance contract should not be able to manufacture coverage merely from an insurer’s misstep.

These are all principles that animate how insurance law deals with the situation of the untimely assertion of a valid coverage defense by insurers, many of which are explored by the Hawai'i Supreme Court in Enoka.

The Enoka court first addressed whether the insurer in that case “waived” its coverage defense. Sometimes the case law in this regard is confused in drawing a distinction between express and implied waivers: it is not the “waiver” that is express or implied; rather, it is the insurer’s conduct that may be said to waive a defense expressly or by implication. Either way there is a waiver, and the only question is how that waiver is proved at trial. (The idea is similar to the “difference” between an express and implied contract – there is no difference, the only distinction is one of proof.) So it is with “waiver,” and the question is whether the insurer mouthed or wrote the words that it was waiving a particular defense or whether though silent the carrier’s conduct indicates its waiver.

There is no doubt that insurers are free to waive their valid coverage defenses or to make a payment to an insured for a loss that was never covered by the insuring agreement to begin with. (Insurers may give such succor deliberately to foster positive business relationships with the particular insured or may make payment for other reasons, such as obtaining the favor of politicians or insurance regulators.)

Waiver can be evidenced by conduct inconsistent with the assertion of the coverage defense, and courts are more willing to find such waivers when the defense concerns “technical” or “forfeiture” grounds, such as timely notice or proof of loss; courts are less likely to find a waiver if that which is being claimed to have been waived goes to the non-existence of coverage in the first place (i.e., that a loss was not covered by the insuring agreement, without regard to exclusions, etc.).

Accordingly, a demand by an insurer for more information after a deadline for submitting claims whose expiration is readily apparent will be deemed to be a waiver of that deadline. Enoka, slip op. at 37. (One can also conceive of this as an estoppel claim based on the insured’s going to the effort to respond to this request, but its aptness is questionable given that the expiration of the deadline should be as apparent to the policyholder as it is to the insurer which leads to questions of the reasonableness of the insured’s reliance on the insurer’s request for more information, etc. etc.)

But where the carrier’s defense relates to a “coverage issue”, waiver will be less likely to be found from conduct alone. The notion largely goes to the burden of proof on each party’s prima facie case: the insurer bears the burden of showing exclusions and limitations on coverage and as with any affirmative defense it can be waived (in this context classed as a “technical” defense); this is different, however, from an insurer’s “waiving” its objection to the policyholder’s failure to sustain its prima facie case for coverage (classed as a “coverage” issue).

The other common fount for argument that an insurer cannot revive a defense to performance is estoppel, which prevents an insurer from switching positions in its dealings with the particular insured. (One should also be aware of the notion of “judicial estoppel,” which prevents litigants from taking inconsistent positions from case to case where the party had prevailed on a prior ground, Iowa v. Duncan (Iowa Feb. 17, 2006); Whiteacre Partnership v. Biosignia Inc. (N.C. Feb. 6, 2004) (containing a lengthy discussion of estoppel and judicial estoppel).)

Many if not most courts say that estoppel may not be used to “broaden the coverage” granted to begin with, Enoka slip op. at 39. Nevertheless, estoppel can apply to broaden coverage in three circumstances according to the Enoka court:

1. Where the insurer or its agent made a misrepresentation at policy inception that, if corrected at that time, would have enabled the insured to protect itself by buying back an exclusion or purchasing a different policy form;
2. Where the insurer undertook and controlled the insured’s defense in a liability matter without mentioning the coverage issue (and thus prevented the insured from protecting its interest in the conduct of the defense or in settling the matter, a topic on which I have previously commented .
3. Where the insurer has acted in bad faith or relatedly where allowing the change of position would be manifestly unjust.

“Estoppel” requires detrimental, reasonable reliance by the “innocent” party and injury that would result from allowing the change in position. In the first exception, the principle applied is meant to prevent “sharp practices” and is more a pure invocation of equity. (There is also the related idea of estoppel in pais, which applies more broadly in that the party claiming the benefit of the estoppel need not be the same as the person to whom the representation was made (i.e., there is no “mutuality” limitation), as in Free v. Sluss, 87 Cal.App.2d Supp. 933 (1948) and Morton Int'l Inc. v. General Accident Ins., 134 N.J. 1, 629 A.2d 831 (1993) (unnecessarily called “regulatory estoppel” when estoppel in pais is sufficient).) It is also related to a rule of contract construction whereby an promisor will be held to a meaning of a contract in the sense it knew the promisse had understood it at the time of contract formation.

The third situation again is more a broad principle of equity such that where the conduct involved is sufficiently unreasonable, equity will prevent a party from changing its position in order to prevent manifest injustice – i.e., detrimental, reasonable reliance is not needed. (Injury is implicit, otherwise there would be no fight over the issue: de minimis non curat lex.) This type of circumstance may fall also under “quasi estoppel” principles but mutuality is required for quasi-estoppel to apply. See Whiteacre.

Waiver and estoppel are not confined to contracts or insurance contracts, but there is one additional doctrine that is derived from contract law whose operation is similar, which is the “mend the hold” principle. The most important recent articulation of this rule is the opinion for the Seventh Circuit authored by Judge Posner in Harbor Ins. Co. v. Continental Bank Corp., 922 F.2d 357, 362 (7th Cir. 1990). And while notions of estoppel and waiver are sprinkled into analyzes of the rule, the core notion, as stated by the US Supreme Court well more than a century ago, is that:

Where a party gives a reason for his conduct and decision touching any thing involved in a controversy, he cannot, after litigation has begun, change his ground, and put his conduct upon another and a different consideration. He is not permitted thus to mend his hold.

Railway Co. v. McCarthy, 96 US 258, 267-68 (1877). The phrasing “mend the hold” hearkens to wrestling.

The mend-the-hold idea is grounded in contract law (substantive law) rather than being a mere implementation of a procedural rule requiring the articulation of claims or defenses (or elections of remedies). It effectuates ex ante decisionmaking and conduct (which undergirds all contract law), whether that prospective private ordering takes place at the time of contract formation or at the time of contract performance (or non-performance). As the Kansas Supreme Court held, “[s]ince the defendant here, before litigation was commenced, gave only as its reason for nonperformance a ground which was inadequate, it could not, after suit was filed, ‘mend its hold’ and rely upon other and different defenses. It was limited in the trial to the single defense it asserted at the time of breach.” Heidner v. Hewitt Chevrolet Co., 199 P.2d 481, 484 (Kan. 1948); see also Coporacion de Mercadeo Agricola v. Mellon Bank Int’l, 608 F.2d 43, 348 (2d Cir. 1979); Western Grocer Co. v. New York Oversea Co., 28 F.2d 518, 520-21 (N.D. Cal. 1928).

Mend-the-hold should have particular force in insurance cases, as the Seventh Circuit found in Continental Bank, but courts sometime dilute its effect by importing (in error) some of the limits to estoppel more generally, Design Data Corp. v. Maryland Cas. Co., 503 N.W.2d 552, 560 (Neb. 1993). But it is in the insurance context and real-estate-brokerage contexts that the doctrine has most typically been applied. See Sitkoff, “Mend the Hold” and Erie, 65 U. Chi. L. Rev. 1059 (1998). It also prevents insurers from insulating against the application of waiver or estoppel by putting in general language in disclaimer/denial letters or reservation-of-rights letters whereby they purport to reserve generally other grounds that have not been articulated, a practice that if validated by the courts undermines the whole purpose of requiring insurers to investigate claims and state their coverage positions.

Insurers are supposed to be the masters of the contracts they sell, and all three of these doctrines create incentives for insurers to do their job right the first time. (They also give insurers incentives to over-articulate defenses to coverage, an incentive that (one hopes) may yield self-correction in the marketplace due to consumer revolt from being greeted with interminable denial-of-coverage letters and regulatory displeasure from such market conduct.) On balance, forcing the prompt articulation of coverage defenses, a corollary of the rule that insurers have the duty to investigate claims, is a good thing, and courts should not hesitate in the individual case to apply waiver, estoppel, or mend the hold, even though an “undeserving” insured obtains coverage from the carrier’s mistake. In the long run everyone – insurers too – benefits from a system that requires insurers, which are responsible for the terms of coverage, sell peace of mind and promptness of performance in the time of need, and have trained forces of claims handlers (paid for by policyholders through the collection of premiums), to both put up and shut up.