Justice PARRISH, opinion of the Court:

INTRODUCTION

¶ 1 The Ohio Casualty Insurance Company (Ohio Casualty) insured Cloud Nine, LLC1 under a commercial general liability (CGL) policy from June 10, 2001 to June 10, 2002. Unigard Insurance Company (Unigard) insured Cloud Nine under a CGL policy from December 12, 2002 through December 12, 2005. Edizone, LC sued Cloud Nine in federal district court, alleging injuries that began during the last three months of Ohio Casualty's policy period and continued throughout Unigard's policy period. The federal district court ruled that the insurers must equally share the total defense costs they incurred in defending Cloud Nine against the Edizone suit. Ohio Casualty appealed that ruling to the Tenth Circuit Court of Appeals, which certified to us the following question regarding the apportionment of defense costs:

¶ 2 We conclude that the "other insurance" clauses do not apply to successive insurers. Accordingly, defense costs should be apportioned using a modified version of the Sharon Steel method that divides responsibility for defense costs between the two insurers in proportion to their time on the risk.

BACKGROUND

¶ 3 We state the facts as described by the Tenth Circuit in the Order of Certification. Ohio Casualty insured Cloud Nine from June 10, 2001, to June 10, 2002. From June 10, 2002, through December 12, 2002, Cloud Nine was uninsured. Unigard then insured Cloud Nine from December 12, 2002, through December 12, 2005.

¶ 4 In their respective policies, Ohio Casualty and Unigard both agree to "pay those sums that the insured becomes legally obligated to pay as damages because of `personal and advertising injury' . . . caused by an offense arising out of [the insured's] business but only if the offense was committed in the `coverage territory' during the policy period."

¶ 5 Both policies also contain "other insurance" clauses that provide as follows:

¶ 6 Edizone is a product and technology developer that licensed patents and other intellectual property to Cloud Nine for the manufacture and sale of an elastometer gel technology and a product known as "Gelastic" and "GellyComb." In its federal case against Cloud Nine, Edizone alleged that Cloud Nine continued to manufacture, use, and sell its products after Edizone terminated Cloud Nine's license agreement on March 11, 2002.

¶ 7 Cloud Nine requested that both Ohio Casualty and Unigard provide a defense to Edizone's federal suit. Unigard agreed to defend, but Ohio Casualty refused. Ohio Casualty then filed a declaratory judgment action in federal district court alleging that it had neither a duty to defend nor indemnify Cloud Nine. Unigard intervened as a plaintiff and moved for partial summary judgment, arguing that Ohio Casualty had a duty to defend the Edizone suit, and that Ohio Casualty was obligated to share defense costs equally with Unigard.

¶ 8 The federal district court ruled in favor of Unigard on both issues. It held that Ohio Casualty did have a duty to defend Cloud Nine and that the two insurance companies should share equally in paying defense costs. In ruling on defense costs, the district court relied on the "other insurance" provision of Ohio Casualty's policy and the broad scope of an insurer's duty to defend under Utah law.

¶ 9 Ohio Casualty appealed the portion of the ruling regarding the allocation of defense costs to the U.S. Court of Appeals for the Tenth Circuit. The Court of Appeals determined that the disposition of Ohio Casualty's appeal turned on important and unsettled principles of Utah law. Accordingly, it certified the following question to this court:

We have jurisdiction under Utah Code section 78A-3-102(1) to answer a question of law certified by a federal court.

STANDARD OF REVIEW

¶ 10 The Tenth Circuit has asked us to rule on a certified question of Utah law. "Accordingly, we are not presented with a decision to affirm or reverse, and traditional standards of review do not apply." Robert J. DeBry & Assocs., P.C. v. Qwest Dex, Inc., 2006 UT 41, ¶ 11, 144 P.3d 1079.

ANALYSIS

¶ 11 Under the framework we announced in Sharon Steel Corp. v. Aetna Casualty & Surety Co., when determining how to apportion defense costs among insurers, we "apply equitable principles . . . unless express policy language decrees the method of apportionment." 931 P.2d 127, 140 (Utah 1997) (alteration in original) (internal quotation marks omitted).

¶ 12 Ohio Casualty argues that the "other insurance" clause found in both its own policy and Unigard's policy does not constitute "express policy language that decrees the method of apportionment." It accordingly urges us to follow the time-on-the-risk method that we determined was the most equitable means of apportionment in Sharon Steel.

¶ 13 In contrast, Unigard argues that the "other insurance" clauses expressly decree the method of apportionment and require that it and Ohio Casualty should equally share the cost of defending Qoud Nine. In the alternative, it argues that a provider's time on the risk is not the most equitable method of apportioning defense costs and further that the method undermines insurers' broad duty to defend under Utah law.

¶ 14 We conclude that the "other insurance" clauses do not apply to successive insurers and therefore do not control the apportionment of costs in this case. In accordance with Sharon Steel, we hold that costs should be apportioned using the time-on-the-risk method. But on the facts of this case, that method must be modified so that the portion of defense costs attributable to Cloud Nine for the time it was uninsured is divided proportionally between the two insurers.

I. THE "OTHER INSURANCE" CLAUSES IN THE TWO INSURANCE POLICIES APPLY ONLY TO CONCURRENT INSURERS AND THUS DO NOT CONTROL THE APPORTIONMENT OF DEFENSE COSTS IN THIS CASE

¶ 15 We first consider whether the identical "other insurance" clauses in the Ohio Casualty and the Unigard policies expressly control the apportionment of defense costs between the two insurers.

¶ 16 Insurance policies are contracts between the insurer and the insured and must be analyzed according to principles of contract interpretation under Utah law. "[I]f the language within the four corners of the contract is unambiguous, the parties' intentions are determined from the plain meaning of the contractual language." Benjamin v. Amica Mut. Ins. Co., 2006 UT 37, ¶ 14, 140 P.3d 1210 (internal quotation marks omitted). We "afford[ ] the policy terms their usually accepted meanings and giv[e] effect to and harmoniz[e] to the extent possible all policy provisions." S.W. Energy Corp. v. Cont'l Ins. Co., 1999 UT 23, ¶ 12, 974 P.2d 1239.

¶ 17 The policies' "other insurance" clauses state that when both insurance policies in question are primary and both "permit[ ] contribution by equal shares," then "if other valid and collectible insurance is available to the insured for a loss [the insurer] cover[s]. . . [the insurer] will share with all that other insurance by . . . equal shares."

¶ 18 The parties agree that both policies are primary and that both permit contribution by equal shares. Thus, under the policies' language, the "other insurance" clause can apply only if there was "other valid and collectible insurance" available to Cloud Nine "for a loss [Ohio Casualty] cover[ed]." Unigard correctly notes that both insurers covered the same type of loss. But it does not follow, as Unigard contends, that "other valid and collectible insurance" was available to Cloud Nine for the loss covered by Ohio Casualty. To the contrary, Ohio Casualty's coverage of Cloud Nine was expressly limited to losses that arose out of offenses "committed. . . during the policy period," which terminated on June 10, 2002. Similarly, Unigard's policy did not cover losses that occurred before its effective date, December 12, 2002. As Ohio Casualty's coverage and Unigard's coverage did not overlap, Unigard did not provide valid and collectible insurance for a loss that Ohio Casualty covered or vice-versa. As a result, the "other insurance" clause with its "equal shares" provision is inapplicable.

¶ 19 Though we find the policy language to be unambiguous and therefore controlling, we note that our conclusion is also consistent with the purpose and function of "other insurance" clauses. Courts have recognized that "other insurance" clauses "serve to prevent multiple recoveries" when "two or more policies provide coverage during the same period." Consol. Edison Co. of N.Y. v. Allstate Ins. Co., 98 N.Y.2d 208, 746 N.Y.S.2d 622, 774 N.E.2d 687, 694 (2002).2 But such "other insurance" provisions do not apply to successive insurers. For example, the Supreme Judicial Court of Massachusetts held that "other insurance" clauses "simply reflect a recognition of the many situations in which concurrent, not successive, coverage would exist for the same loss," for instance "where one insurer issued an umbrella liability policy to the lessor of a vehicle involved in a motor vehicle accident and another insurer issued a liability policy to the lessee." Bos. Gas Co. v. Century Indem. Co., 454 Mass. 337, 910 N.E.2d 290, 308-09 (2009). And the U.S. Court of Appeals for the Seventh Circuit has expressed skepticism that an "other insurance" provision could apply in a case where "two policies, each with an `other insurance' clause, insure merely the same kind of risk, but not the same risk because the policies are successive." Taco Bell Corp. v. Cont'l Cas. Co., 388 F.3d 1069, 1079 (7th Cir.2004) (Posner, J.).3

¶ 20 Having concluded that the policies in question do not address the allocation of defense costs in cases such as this, we turn to Unigard's argument that we should follow Federal Insurance Co. v. Cablevision Systems Development Co. and apply the equal shares apportionment method because the "other insurance" clauses "demonstrate[ ] an intent to apportion indemnity loss equally." 836 F.2d 54, 57 (2d Cir.1987). We fail to see any general intent to apportion loss equally under the facts of this case, especially in light of the express limitation in both policies limiting covered losses to those resulting from offenses committed during the policy period. Furthermore, the Second Circuit's reasoning is incompatible with our decision in Sharon Steel, which requires an equitable distribution of defense costs absent "express policy language [that] decrees the method of apportionment," rather than general manifestations of intent. 931 P.2d at 140 (internal quotation marks omitted).

¶ 21 In summary, "other insurance" was not available to Cloud Nine for the loss Ohio Casualty covered. Consequently, the "other insurance" clause in Ohio Casualty's policy does not constitute "express policy language [that] decrees the method of apportionment" and therefore does not govern the apportionment of defense costs in this case.

II. DEFENSE COSTS SHOULD BE APPORTIONED PURSUANT TO A MODIFIED VERSION OF THE SHARON STEEL FORMULA

¶ 22 Having determined that the policy provisions do not control the apportionment of defense costs, we turn to equitable principles to determine how to apportion defense costs between Ohio Casualty and Unigard. Sharon Steel Corp. v. Aetna Cas. & Sur. Co., 931 P.2d 127, 140 (Utah 1997). In Sharon Steel, the underlying injury involved the release of toxic material into the environment that, like the injury here, spanned many years and triggered coverage under multiple consecutive insurance policies provided by several insurance companies. Id. at 130. We held that an insurer can compel contribution for defense costs from a coinsurer that is equally obligated to defend. Id. at 139. Because we found that one insurer was "entitled to be reimbursed for those defense expenses it paid in excess of its fair share, we deem[ed] it prudent to offer guidance to the trial court in apportioning those defense costs." Id. at 140.

¶ 23 In the continuous injury, successive insurer context presented in Sharon Steel, we considered and expressly rejected apportionment based on the equal shares method that Unigard asks us to employ in this case. Id. at 140 n. 19. We rejected the equal shares method because we were "unpersuaded that [it] . . . reflect[ed] the most equitable method of allocating the defense costs." Id. We instead applied the "time on the risk" apportionment method that considers the time each insurer spent "on the risk" and each insurer's policy limits. Id. at 140-41. We concluded that this method was the most equitable because it fairly related both to the time each insurer spent on the risk and the degree of risk each insurer contracted to assume. Id.4 We also found that "the property owners must be prepared to pay their fair share of defense costs for those years that they were without insurance coverage." Id. at 141 (internal quotation marks omitted).

¶ 24 Unigard asks us to disregard this holding. Specifically, Unigard argues that our imposition of time-on-the-risk apportionment was non-binding "guidance" to the trial court. Unigard also asserts that applying time-on-the-risk apportionment is inconsistent with its and Ohio Casualty's contractual duty to defend and with our precedent regarding the duty to defend under Utah law. It argues that applying time-on-the-risk apportionment is tantamount to holding that an insurer has the duty to defend only the claims covered by its own policy, rather than all claims "until it can limit the suit to those claims outside of the policy coverage." Benjamin v. Amica Mut. Ins. Co., 2006 UT 37, ¶ 25, 140 P.3d 1210 (internal quotation marks omitted).

¶ 25 We first reject the notion that the discussion of defense cost apportionment in Sharon Steel constituted mere "guidance" without precedential value. In Sharon Steel, we "deem[ed] it prudent to offer guidance to the trial court," and remanded the case to the trial court with instructions to fashion the precise allocation formula because it, in its capacity as fact finder, was best suited procedurally to allocate costs to the parties based upon our time-on-the-risk standard. 931 P.2d 127 at 140-42. But our conclusion that costs must be apportioned using our time-on-the-risk formula was the holding of the case and was not a holding the district court was free to ignore.

¶ 26 Moreover, we see no reason to drastically deviate from our holding in Sharon Steel. Unigard correctly notes that both insurers had a duty to defend based on both the policy language and our precedent. See Benjamin, 2006 UT 37, ¶ 25, 140 P.3d 1210. But it does not follow, as Unigard contends, that defense costs must be apportioned equally. The duty to defend and the apportionment of defense costs between two insurers that have an equal duty to defend are distinct issues. See Sharon, 931 P.2d at 137-42 (evaluating apportionment methods and applying a time-on-the-risk formula despite finding that each insurer had an equal duty to defend). The U.S. Court of Appeals for the Fifth Circuit has acknowledged this distinction, "approv[ing of] the concept of apportioning the cost of the insured's defense among those liable," but not "limit[ing] the duty of defending the insured." Gulf Chem. & Metallurgical Corp. v. Associated Metals & Minerals Corp., 1 F.3d 365, 372 (5th Cir. 1993).

¶ 27 Given this distinction, there is no logical conflict between our duty to defend precedent and the time-on-the-risk formula we adopted in Sharon Steel. The time-on-the-risk method fairly allocates costs between insurers based on the amount of risk each contracted to undertake and the premiums each received without compromising the rights of the insured. It also comports with our policy of encouraging prompt and effective defense of the insured by the insurer. See Benjamin, 2006 UT 37, ¶¶ 22-25, 140 P.3d 1210 (construing an insurance policy liberally to promote the purposes of insurance and requiring an insurer to defend until uncertainties can be resolved against coverage). Under the time-on-the-risk method, the insurer facing larger indemnity costs has a greater stake in controlling choice of counsel and settlement negotiations. This insurer can more practically and efficiently take the lead in defending the suit without interference from the insurer with less indemnity cost at stake while still receiving contribution from that insurer for a benefit conferred. Alternatively, if the insurer with more indemnity cost at stake fails to defend, the insurer with less time on the risk can defend vigorously knowing that it can recoup a proportionate share of the costs from the insurer with more time and resources on the risk.

¶ 28 We therefore follow Sharon Steel to apportion defense costs in this case. But we decline to follow that portion of Sharon Steel that apportioned defense costs to the insured for those periods of time when the insured was without coverage. In this case, there is language in both polices that expressly gives each insurer control over its defense of the insured. Ohio Casualty reserved the right to, "at [its] discretion, investigate any offense and settle any claim or `suit' that may result." And Unigard reserved the right to "conduct and control the defense of the indemnitee." Given the insurers' legal and contractual duties to defend, they often, as Ohio Casualty and Unigard did here, reserve the exclusive right to control any litigation and make important decisions regarding the course of the litigation, including the hiring and firing of counsel and whether or not to settle. In light of this practice, it would be inequitable to apportion any defense costs to an insured who has no power to select counsel or negotiate rates and no voice in deciding whether to settle the suit. Accordingly, we conclude that it would be inequitable to hold the insured responsible for the share of defense costs attributable to the time period during which it was uninsured.

¶ 29 This conclusion is consistent with insurers' duty to defend under Utah law. Where an insured holds coverage from a single insurer for part of a period of continuous injury and is then without coverage for the remainder of the injury period, the insurer may not recover defense costs from the insured for the period of noncoverage because the insurer must "provide a defense to the entire suit, at least until it can limit the suit to those claims outside of the policy coverage." Id. ¶ 25 (internal quotation marks omitted); see also Sharon Steel, 931 P.2d at 133 (acknowledging that "an insurer's duty to defend is broader than its duty to indemnify"). Where, as in this case, there are multiple insurers, the broad duty to defend also prevents the insurers from recovering defense costs from the insured for any periods of non-coverage.

¶ 30 In accordance with Sharon Steel and consistent with the policy language specific to this case that provides the insurance companies with complete control over the litigation, we conclude that defense costs in this case should be apportioned by a modified version of the Sharon Steel formula. This formula begins by apportioning the defense costs between successive insurers according to their time on the risk and the amount of their policy limits. It then divides the portion of defense costs attributable to any periods during which the insured lacked coverage in the same proportions.

CONCLUSION

¶ 31 In response to the question certified by the Tenth Circuit, we hold that the "other insurance" provisions in the policies in question do not control the apportionment of defense costs. Instead, defense costs should be allocated using the Sharon Steel "time on the risk" formula modified to proportionally apportion to the insurers any defense costs attributable to periods of noncoverage. This is the most equitable method of apportionment because under it each insurer's allocation of defense costs is a function of the amount of time each insurer spent "on-the-risk" and each insurer's policy limits. And this method ensures that the insured is not responsible for defense costs related to litigation over which its insurer has full control and an absolute duty to defend.

Justice PARRISH authored the opinion of the Court, in which Chief Justice DURHAM and Justice LEE joined.

Associate Chief Justice DURRANT filed a dissenting opinion, in which Justice NEHRING joined.

Associate Chief Justice DURRANT dissenting:

¶ 32 I concur in the majority's conclusion that the "other insurance" policy provisions do not apply to successive insurers and, therefore, do not control the apportionment of defense costs in this case. In addition, I agree that, absent controlling contractual language, we apply equitable principles to apportion defense costs. But I disagree that the "most equitable method of apportionment" allocates defense costs according to "the amount of time each insurer spent `on-the-risk' and each insurer's policy limits."1

¶ 33 In my view, defense costs should be allocated in equal shares between each insurer who has a duty to defend. Indeed, the apportionment of equal shares is consistent with the broad scope of an insurance provider's duty to defend under Utah law. Because the duty to defend obligates each insurer whose policy is triggered to provide the insured with a full defense, I would allocate the costs associated with that duty equally. And because the duty to defend is not tied to the insurance provider's time on the risk, or to its policy limits, I believe it is inappropriate to apportion defense costs according to these factors.

¶ 34 Under Utah law, an insurance provider owes its insureds two independent duties: (1) a duty to indemnify and (2) a duty to defend.2 It is axiomatic, however, that the duty to defend is broader than the duty to indemnify.3 For example, the duty to indemnify is limited to damage caused by acts within the defined period of insurance coverage, and is further limited by the insurance provider's policy limits.4 But there is no such limitation for the insurer's duty to defend.5 Unlike the duty to indemnify, the duty to defend is broad in three respects: (1) the duty to defend is triggered whenever a complaint "alleges a risk within the coverage of the policy";6 (2) the duty to defend one claim creates "a duty to defend all of the claims brought" against the insured, even claims outside the period of insurance coverage;7 and (3) the duty to defend exists regardless of the merits of the underlying claims.8

¶ 35 Consistent with the broad scope of an insurer's duty to defend, each insurance provider whose duty is triggered owes the insured an independent obligation to defend the entire suit.9 Thus, where there is only one insurance provider, that insurer bears the full obligation to pay for the defense costs. Where there are multiple insurance providers whose duties are triggered, they each have an independent duty to defend the entire lawsuit. Because each insurer has an independent duty to defend, and because that duty is not tied to the insurer's time on the risk, I see no reason to apportion defense costs based on the majority's formula. Instead, because each insurance provider bears the obligation to defend the entire suit, I would apportion the costs associated with that duty equally.10

¶ 36 In this case, both Ohio Casualty and Unigard had a duty to defend Cloud Nine. Because of the broad scope of this duty, if either Ohio Casualty or Unigard had refused to defend, the other still would have been obligated to provide a full defense. Given that they had an equal duty to defend the entire suit, I would apportion the defense costs associated with that duty equally because I believe that creates the most logical and equitable result.

¶ 37 I recognize that equal apportionment of defense costs is at odds with our holding in Sharon Steel Corp. v. Aetna Casualty & Surety Co.11 In that case, we limited an insurer's duty to defend by apportioning defense costs to insurance providers and insureds based on their time on the risk.12 But I believe that the test for overturning that precedent has been satisfied. It is appropriate to overturn precedent "if we are clearly convinced that the rule was originally erroneous or is no longer sound because of changing conditions and that more good than harm will come by departing from precedent."13 Regarding the rule announced in Sharon Steel, I believe that each of these requirements is satisfied.

¶ 38 The rule announced in Sharon Steel was erroneous in two respects. First, our limitation of an insurer's duty to defend in that case conflicted with our prior statements about the breadth of that duty.14 In Deseret Federal Savings & Loan Ass'n v. U.S. Fidelity & Guaranty Co., a case decided before Sharon Steel, we stated that the duty to defend was broad and "arises whenever the insurer ascertains facts which give rise to the potential of liability under the policy."15 Thus, we had recognized that insurers had to provide their insureds with a full defense when there was even the potential for liability.16 Without reference to this prior statement, in Sharon Steel we limited an insurer's duty to defend to only those occurrences that took place within the policy period.17

¶ 39 Second, in Sharon Steel we set forth a rule without receiving argument on the merits of the various methods of allocation or argument on how each method would impact an insurer's duty to defend. Specifically, we adopted the time on the risk allocation even though "[n]either party ha[d] provided a thorough briefing on th[e] issue."18 In fact, we recognized that although one party had "summarily set[ ] forth the different allocation methods employed by various courts," no party had "address[ed] the merits of each method, . . . [or] explain[ed] how these methods might apply to the instant case."19 Because we departed from precedent without satisfying the test for doing so, and selected a method of apportionment without receiving argument from the parties, I believe that the rule was erroneous.

¶ 40 In addition to being erroneous, I believe that our holding in Sharon Steel is no longer sound because, in a subsequent case, we retreated from any limitation on an insurance provider's duty to defend. In Benjamin v. Amica Mutual Insurance Co., a case decided after Sharon Steel, we returned to our position that the duty to defend is broad and, accordingly, requires an insurer to provide a defense to the entire suit, even to claims that fall outside the period of insurance coverage.20 By recognizing the broad scope of the duty to defend and the corresponding broad obligations associated with that duty, we have already moved away from the limitation announced in Sharon Steel. Thus, we have already indicated that the rule adopted in Sharon Steel is no longer sound.

¶ 41 Similarly, and most importantly, I believe that the majority's rationale for overturning a portion of the holding in Sharon Steel also supports the conclusion that the entire rule should be overturned. In overturning the portion of our holding that allocates defense costs to the insured, the majority recognizes that "[w]here an insured holds coverage from a single insurer . . . and is then without coverage . . ., the insurer may not recover defense costs from the insured for the period of non-coverage because the insurer must provide a defense to the entire suit."21 Thus, the majority reasons that allocation to the insured for periods of noncoverage would be inequitable because the insurance provider has a broad duty to defend that requires it to provide a defense for the entire suit, even for periods of noninsurance.22

¶ 42 While I agree with the majority on this point, I believe that this same logic applies to overturning the entire rule put forth in Sharon Steel. Just as the breadth of the duty to defend makes it inappropriate to apportion defense costs to the insured for periods of non-coverage, the breadth of the duty also makes it inappropriate to apportion defense costs to insurance providers based on their periods of coverage. Because an insurer has a duty to defend the entire suit regardless of whether there are other insurance providers with a similar duty, it would be inequitable to apportion defense costs based on a pro rata formula. Thus, for the same reason that the majority overturns a portion of the holding in Sharon Steel, I would overturn the entire rule.

¶ 43 Finally, I believe that more good than harm will come from overturning our holding in Sharon Steel. Specifically, overturning the rule would reduce disputes between insurance providers about the proper apportionment of defense costs and would induce insurers to more promptly defend their insureds. If insurers know that defense costs will be allocated equally among those whose duties are triggered, each insurer will have the same incentive to provide a prompt and efficient defense. Because each insurer would share the defense costs equally from the moment the lawsuit is filed, they would have shared incentives to choose counsel, initiate a defense, and engage in settlement negotiations. They would not need to wait for any judicial determination of their respective responsibilities.

¶ 44 In contrast, adherence to the pro rata formula adopted in Sharon Steel, and supported by the majority in this case, actually encourages insurance providers to dispute defense costs or take a "wait and see" approach.23 In fact, the time-on-the-risk formula encourages delay from the moment the lawsuit begins. Although an insured's defense costs accrue from the moment any suit is filed, at that moment there likely will not have been a judicial determination for how multiple insurance providers are to allocate the defense costs. And until the court has determined how each insurer's time on the risk and policy limits impact the apportionment of defense costs, insurers will be motivated to delay in providing a defense to their insureds. Indeed, if an insurer anticipates that it will be responsible for only a small proportion of the defense costs, it will have no incentive to quickly select counsel or initiate settlement negotiations. And while the insurance providers delay, the insured will be left without a defense.

¶ 45 For the foregoing reasons, I believe that our holding in Sharon Steel should be overturned not just in part, as the majority concludes, but in its entirety. In overturning the rule in Sharon Steel, we are free to determine the most equitable method of apportionment of defense costs.

¶ 46 In my view, the most equitable and logical way to allocate defense costs is in equal shares between each insurer who has a duty to defend. Equal share apportionment is consistent with the broad scope of an insurer's duty to defend under Utah law and comports with our policy of encouraging a prompt and effective defense by the insurer. Because Ohio Casualty and Unigard have a coextensive duty to defend Cloud Nine for the entire suit, I would allocate the costs associated with their duty on an equal basis. Accordingly, I would affirm the district court's holding.