Defendant-appellant General Accident Insurance Company of America (General Accident) claims that it is entitled to reimbursement from plaintiff-appellee Lexington Insurance Company (Lexington) for a pro rata share of defense costs incurred on behalf of the parties’ mutual insured. The district court disagreed, and General Accident now appeals. We affirm.
The essential facts are not in dispute. General Accident issued a professional responsibility policy with an aggregate limit of liability of $10,000,000 to the law firm of Blank, Rome, Comisky & McCauley (Blank Rome). Desiring extra protection, the insured purchased four separate excess insurance policies. Lexington underwrote the first layer of excess coverage under a policy containing an aggregate Lability limit of $15,000,000. Blank Rome procured second-layer excess policies from Safety National Insurance Company (Safety National), International Surplus Lines Insurance Company (ISLIC), and American Insurance Company (American), respectively. All five policies were in effect for a one-year period beginning April 8, 1984. They collectively afforded Blank Rome $50,000,000 in professional liability coverage.
During the currency of the policies, Blank Rome was rocked by a series of securities fraud suits arising out of the collapse of a failed financial institution. General Accident undertook Blank Rome’s defense. That undertaking lasted until July 28,, 1988, when Blank Rome assumed responsibility for its own defense.1 Gener*45al Accident claims, without contradiction, to have incurred roughly $5,500,000 in legal fees and ancillary expenses while defending Blank Rome.
Unwilling to shoulder the load alone, General Accident sought to compel the four excess carriers to help defray the defense costs it had expended. Using limits of liability as a guide, it calculated each company’s ratable share of the $5,500,000 total. Under this formula, it earmarked $1,650,000 (30% of the total) as Lexington’s responsibility.
Lexington balked at this suggestion. It did, however, enter into an agreement with General Accident to toll the statute of limitations while the insurers “explore[d] the potential for resolution of General Accident’s claims without resort to litigation.” Although General Accident eventually resolved its claims against Safety National, ISLIC, and American, its dispute with Lexington proved intractable.
In an effort to bring matters to a head, Lexington brought a diversity action in the United States District Court for the District of Massachusetts. See 28 U.S.C. § 1332(a). Lexington took the position that its policy unambiguously excluded the payment of defense costs and asked the court to declare that it had no obligation to reimburse General Accident for Blank Rome’s defense.
Lexington relied primarily on language in its policy stating that it would indemnify Blank Rome:
... in accordance with the applicable insuring agreements, terms, conditions and exclusions of the Underlying Policy [i.e., the General Accident policy] ... except as regards the premium, the obligation to investigate and defend and for costs and expenses incident to the same, ... and any other provision therein inconsistent with this policy.
Noting that its policy defined “costs” as “interest on judgments, investigations, adjustments and legal expenses ” (emphasis supplied), Lexington contended that General Accident’s expenditures — which were legal expenses incident to the insured’s defense — were excluded.
General Accident took a different slant. It sought to have Lexington share in the legal expenses both as a matter of contract and under the doctrine of equitable contribution. The linchpin of General Accident’s claim was an apportionment provision contained within the limits of liability section of its policy. That provision, entitled “Apportionment of Claims Expenses,” reads:
In the event payment for damages by the Insured, any other carrier on behalf of the Insured and the Company is in excess of the amount of the limit available under this policy, the Company shall be obligated to pay that proportion of claim expenses as the amount of damages paid by the Company bears to the total amount of damages.
In General Accident’s view, this language imposed an expense-sharing obligation on Lexington both as a matter of contract and as a matter of equity.
In due season, the parties cross-moved for summary judgment. Lexington urged the district court to find that its policy exculpated it from any responsibility to share in the cost of Blank Rome’s defense whereas General Accident exhorted the district court to find Lexington’s policy ambiguous because it neither explicitly disclaimed a duty to reimburse the primary insurer nor specifically negated the applicability of the apportionment provision contained in the underlying policy. In this regard, General Accident posited that Lex*46ington, as the second-in-time insurer, had the obligation to repudiate explicitly any language in the underlying policy that it found to be objectionable and to clarify that it was excepting itself from an otherwise comprehensive and integrated insurance program that followed the form of the underlying policy. As a fallback, General Accident invoked the doctrine of equitable contribution.
The district court resolved this clash in Lexington’s favor. See Lexington Ins. Co. v. Gen. Accid. Ins. Co., Civ. No. 01-11556, slip op. (D.Mass. Dec. 20, 2002) (unpublished). The court concluded that Lexington’s policy was unambiguous; that its language clearly established Lexington’s intent not to participate in the payment of Blank Rome’s defense costs; and that, therefore, Lexington was not obligated to reimburse General Accident for any part of the defense costs that it had incurred. Id. at 8. In addition, the court rejected General Accident’s equitable contribution theory. Id. This timely appeal ensued.
This appeal presents a purely legal question. The facts are undisputed, and the district court had no occasion to engage in differential factfinding. The general rule is that orders granting summary judgment are reviewed de novo, e.g., Houlton Citizens’ Coalition v. Town of Houlton, 175 F.3d 178, 184 (1st Cir.1999), and that standard applies here. In all events, a trial court’s interpretation of an insurance contract is equally subject to de novo review. See Utica Mut. Ins. Co. v. Weathermark Invs., Inc., 292 F.3d 77, 80 (1st Cir.2002); Greenly v. Mariner Mgmt. Group, Inc., 192 F.3d 22, 25-27 (1st Cir. 1999).
Before embarking upon our analysis, we pause to discuss a preliminary matter. It is a black-letter rule that state substantive law supplies the rules of decision for a federal court sitting in diversity jurisdiction. Erie R. Co. v. Tompkins, 304 U.S. 64, 78, 58 S.Ct. 817, 82 L.Ed. 1188 (1938); Crellin Techs., Inc. v. Equipmentlease Corp., 18 F.3d 1, 4 (1st Cir.1994). Thus, the initial question involves which source of law should guide our interpretive efforts.
In determining what state law is relevant, a federal court must apply the choice-of-law framework of the forum state (here, Massachusetts). See Klaxon Co. v. Stentor Elec. Mfg. Co., 313 U.S. 487, 491, 61 S.Ct. 1020, 85 L.Ed. 1477 (1941); Crellin Techs., 18 F.3d at 4. In this instance, the choice is between Pennsylvania (home to General Accident and Blank Rome) and Massachusetts (where Lexington is based). Our research reveals no material differences between the laws of these two jurisdictions with respect to the interpretive task that confronts us. That circumstance renders further inquiry unnecessary.
It is a well-established — and prudential — principle that when the result in a case will not be affected by the choice of law, an inquiring court, in its discretion, may simply bypass the choice. See, e.g., Fashion House, Inc. v. K Mart Corp., 892 F.2d 1076, 1092 (1st Cir.1989). That course is especially attractive where, as here, the parties have taken the position that either state’s law will lead to the same result. Cf. Borden v. Paul Revere Life Ins. Co., 935 F.2d 370, 375 (1st Cir.1991) (explaining that when the parties achieve a satisfactory consensus on choice of law, a federal court sitting in diversity jurisdiction is free, if it so elects, to forgo independent analysis and accept the parties’ agreement). Because the choice-of-law question in this case has been reduced to the point where nothing turns on a more precise refinement, we eschew any further inquiry. In the discussion that follows, we will cite interchangeably to applicable *47Pennsylvania and Massachusetts precedents.
The baseline rule in both Pennsylvania and Massachusetts is that insurance contracts must be interpreted to reflect the intention of the parties as manifested by the policy language. See Travelers Cas. & Sur. Co. v. Castegnaro, 565 Pa. 246, 772 A.2d 456, 459 (2001); Affiliated FM Ins. Co. v. Const. Reins. Co., 416 Mass. 839, 626 N.E.2d 878, 881 (1994). When the policy language is plain and admits of only one reasonable construction, a reviewing court must give effect to that construction. See Riccio v. Am. Repub. Ins. Co., 550 Pa. 254, 705 A.2d 422, 426 (1997); Cody v. Conn. Gen. Life Ins. Co., 387 Mass. 142, 439 N.E.2d 234, 237 (1982); accord GRE Ins. Group v. Metro. Boston Hous. P’ship, Inc., 61 F.3d 79, 81 (1st Cir. 1995) (applying Massachusetts law). If, however, a policy provision is ambiguous, the court may take extrinsic evidence as to intent, and, absent clarification, ordinarily may adopt the interpretation that favors coverage (that is, the reading most beneficial to the insured). See Bateman v. Motorists Mut. Ins. Co., 527 Pa. 241, 590 A.2d 281, 283 (1991); Hazen Paper Co. v. United States Fid. & Guar. Co., 407 Mass. 689, 555 N.E.2d 576, 583 (1990); see also Merchants Ins. Co. v. United States Fid. & Guar. Co., 143 F.3d 5, 8 (1st Cir.1998) (applying Massachusetts law).
We caution that an ambiguity is not created merely because an imaginative reader devises a way to split hairs. Nor is an insurance policy ambiguous simply because its terminology sparks a controversy between two parties, each of whom advocates an interpretation contrary to the other. See Citation Ins. Co. v. Gomez, 426 Mass. 379, 688 N.E.2d 951, 953 (1998); Riccio v. Am. Repub. Ins. Co., 453 Pa.Super. 364, 683 A.2d 1226, 1233 (1996), aff'd, 550 Pa. 254, 705 A.2d 422 (1997). Absent a genuine issue as to meaning, there is no ambiguity. See, e.g., Suffolk Const. Co. v. Banco Scaffolding Co., 47 Mass.App.Ct. 726, 716 N.E.2d 130, 133 (Mass.App.Ct. 1999) (stating that contract language is ambiguous only where “an agreement’s terms are inconsistent on their face or where the phraseology can support reasonable difference of opinion as to the meaning of the words employed and the obligations undertaken”) (quoting Fashion House, 892 F.2d at 1083); Hutchison v. Sunbeam Coal Corp., 513 Pa. 192, 519 A.2d 385, 390 (1986) (similar).
It is against this jurisprudential backdrop that we evaluate the coverage issue in this case. The task is simplified by the clarity of Lexington’s policy, which unambiguously states that Lexington will indemnify Blank Rome in accordance with the underlying (General Accident) policy “except as regards ... the obligation to investigate and defend and for costs and expenses incident to the same.... ” This phraseology admits of only one sensible interpretation: that the Lexington policy follows the form of the underlying policy except as to particular provisions (such as those related to the costs of Blank Rome’s defense). The language hardly could be clearer.
If more were needed' — -and we doubt that it is — nothing in the policy itself obliged Lexington either to assume any responsibility for legal costs or to engage in expense-sharing with Blank Rome’s other insurers. Nor does the fact that General Accident, rather than Blank Rome, was the party seeking reimbursement disguise the reality of events. The costs in question were incurred in the defense of Blank Rome, and, thus, were plainly excluded from the scope of Lexington’s coverage.
*48By like token, the apportionment provision in General Accident’s policy did not transform the legal landscape. That proviso does no more than disclaim any obligation on General Accident’s part to pay more than its proportionate share of Blank Rome’s defense costs; it does not impose on any excess carrier a duty to reimburse General Accident for a pro rata share of those costs.2 Given this circumstance, it was pointless (and, thus, unnecessary) for Lexington expressly to deny a non-existent obligation to indemnify General Accident.
That effectively ends the matter. In construing insurance contracts, courts have no warrant either to convolute the straightforward meaning of policy language or to endow the words with a gloss that is belied by the language itself. See 116 Commonwealth Condo. Trust v. Aetna Cas. & Sur. Co., 433 Mass. 373, 742 N.E.2d 76, 78 (2001); Loomer v. M.R.T. Flying Service, Inc., 384 Pa.Super. 244, 558 A.2d 103, 105 (1989). Like beauty, ambiguity sometimes lies solely in the eye of the beholder. Seaco Ins. Co. v. Davis-Irish, 300 F.3d 84, 87 (1st Cir.2002). So it is here.
In an effort to blunt the force of this reasoning, General Accident musters a plethora of counter-arguments. We have carefully examined this asseverational array. We conclude that none of these counter-arguments carries the day.
We start with General Accident’s contention that its policy and Lexington’s must be read together as a part of a comprehensive, integrated insurance program. This may be true from Blank Rome’s standpoint, but there is nothing in the record that justifies a conclusion that all four excess policies were written to follow the form of General Accident’s underlying policy in all respects. The record merely shows five separate policies, independently issued. Each of the four excess carriers had the opportunity to write its own policy, and the ensuing policies were differently phrased. Each was subject to its own set of exclusions, exceptions, and limitations. It cannot be said that these policies uniformly follow the form of General Accident’s policy. Consequently, General Accident’s attempt to portray them as carbon copies is wholly unconvincing.
Next, General Accident points to the settlement and defense provision within Lexington’s policy. That provision creates an option for participation in the insured’s defense.3 General Accident twists this option like a pretzel and treats it as an obligation. Building on that shaky foundation, General Accident insists that the settlement and defense provision is incompatible with Lexington’s present position. This is palpably incorrect.
The authority upon which General Accident relies does not assist its cause. To support its anfractuous reasoning, General *49Accident cites the decision in a case brought by it against one of Blank Rome’s other excess insurers. See Gen. Accid. Ins. Co. v. Am. Ins. Co., 2000 WL 92097 (E.D.Pa. Jan.27, 2000). Fairly read, this decision strengthens, rather than weakens, Lexington’s hand. Like Lexington’s policy, American’s policy provided specific conditions that had to be met before a duty to defend would arise. Id. at *5. The district court awarded summary judgment to American because these conditions had not been satisfied (and, thus, the contractual language of American’s policy did not give rise to an obligation to contribute to the costs of Blank Rome’s defense). Id. The court also found that principles of equity did not allow it to ignore the plain language of American’s policy in order to impose a reimbursement obligation that American had never committed itself to undertake. Id. This rationale is fully transferable to the case at bar. The result, therefore, should be the same.
General Accident derives an additional argument from the sprawl of previous litigation between it and Blank Rome’s other excess insurers. This time, the emphasis is on the decision in Gen. Accid. Ins. Co. v. Safety Nat’l Cas. Corp., 825 F.Supp. 705 (E.D.Pa.1993). We are satisfied that this emphasis is misplaced.
In Safety National, the district court held that one of Blank Rome’s second-layer excess insurers was obliged to pay General Accident a pro rata share of the defense costs that General Accident had incurred. See id. at 711. The court premised its decision on the fact that Safety National’s policy not only incorporated all the terms of General Accident’s primary policy (including the apportionment of claims provision) but also said nothing about the matter of defense costs. Id. at 708. This is a far cry from the scenario that confronts us, in which (a) Lexington’s policy does not blindly follow the form of the underlying policy, and (b) Lexington took pains to protect itself against any obligation to pay defense costs. Indeed, the Safety National court, in a subsequent decision denying reconsideration, acknowledged that it would have been a different case had Safety National issued a policy along the lines of Lexington’s. See Gen. Accid. Ins. Co. v. Safety Nat’l Cas. Corp., 1993 WL 372279, at *5 n. 3 (E.D.Pa. Aug.30, 1993) (noting that “had Safety National intended a different result, it could have simply issued an endorsement stating that its policy would follow only the terms and conditions of other excess policies, such as Lexington’s”).
The final weapon in General Accident’s armamentarium is an asseveration that it is entitled to reimbursement under the doctrine of equitable contribution. Broadly stated:
Equitable contribution is ... the right to recover ... from a co-obligor who shares such liability with the party seeking contribution. In the insurance context, the right to contribution arises when several insurers are obligated to indemnify or defend the same loss or claim, and one insurer has paid more than its share of the loss or defended the action without any participation by the others.... Equitable contribution permits reimbursement to the insurer that paid on the loss for the excess it paid over its proportionate share of the obligation, on the theory that the debt it paid was equally and concurrently owed by the other insurers.
Truck Ins. Exch. v. Unigard Ins. Co., 79 Cal.App.4th 966, 974, 94 Cal.Rptr.2d 516 (2000); see also Hartford Cas. Ins. Co. v. Argonaut-Midwest Ins. Co., 854 F.2d 279, 283 (7th Cir.1988). The right of equitable contribution does not depend on an express agreement between the parties to *50indemnify each other, but, rather, rests upon equitable principles that imply an obligation to contribute ratably toward the payment of a common obligation. Ohio Cas. Ins. Co. v. State Farm Fire & Cas. Co., 262 Va. 238, 546 S.E.2d 421, 423 (2001). Neither Pennsylvania nor Massachusetts has definitively adopted this doctrine in the insurance context.4
We need not prophesy what treatment the doctrine will receive in these two jurisdictions. For present purposes, it suffices to say that there is no support in the case law of any jurisdiction for the proposition that, in the absence of exceptional circumstances, the doctrine of equitable contribution can override explicit, unambiguous policy language. See Tex. Employers Ins. Ass’n v. Underwriting Members of Lloyds, 836 F.Supp. 398, 408 (S.D.Tex.1983) (refusing to apply equitable contribution doctrine where doing so “would fly in the face of unambiguous policy language”); Truck Ins. Exch., 79 Cal. App.4th at 974, 94 Cal.Rptr.2d 516 (stating that courts ordinarily “should not impose an obligation on an insurer that contravenes a provision in its insurance policy”). That is consistent with the hallowed principle that, absent some amphiboly, a court cannot, in the name of equity, rewrite the language of an insurance contract. See Williams v. Nationwide Mut. Ins. Co., 750 A.2d 881, 885 (Pa.Super.2000); Hakim v. Mass. Insurers’ Insolv. Fund, 424 Mass. 275, 675 N.E.2d 1161, 1164 (1997).
To say more on this point would be supererogatory. There are no exceptional circumstances here and Lexington’s policy is unambiguous in respect to defense costs. Thus, the doctrine of equitable contribution does not come into play.5
We need go no further. For the reasons stated above, we hold that Lexington has no obligation to pay any portion of the defense costs incurred by General Accident in defending Blank Rome. Accordingly, the district court did not err in granting summary judgment in Lexington’s favor.
Affirmed.
. At that time, Blank Rome entered into an agreement with the five insurers. The agreement provided, in substance, for each insurer to contribute its policy limit to Blank Rome in exchange for a release of any and all further liability (including, but not limited to, defense (including, but not limited to, defense costs).
. To be sure, absent the exchange of releases Blank Rome might have been responsible for the balance of the costs — but that determination is well outside the proper purview of this action.
. The setdement and defense provision reads in pertinent part:
Anything in the Underlying Insurance to the contrary notwithstanding, the Company shall not be obligated to assume charge of the settlement or defense of any claim or suit brought or proceeding instituted against the Insured, but the Company, at its option but not being required to, shall have the right and be given the opportunity to associate with the Insured in the defense or control of any claim, suit or proceeding which appears reasonably likely to involve the Company, in which event the Insured and the Company shall cooperate in all things in the defense or control of such claim, suit or proceeding.
. In Massachusetts, however, the Commonwealth’s intermediate appellate court has indicated a willingness to entertain such an action. See Rubenstein v. Royal Ins. Co., 44 Mass.App.Ct. 842, 694 N.E.2d 381, 388 (1998) (suggesting that "there is no bar against an insurer obtaining a share of indemnification or defense costs from other insurers under the doctrine of equitable contribution”).
. We note, as an aside, that "[i]n the context of insurance coverage, proof that the policies insure the same property is not sufficient to establish a common obligation; the policies in question must afford coverage for the same insureds and the same risk.” Ohio Cas., 546 S.E.2d at 423. Using this logic, some courts have held that the doctrine of equitable contribution does not pertain as between a primary insurer and an excess insurer. E.g., Home Indem. Co. v. Gen. Accid. Ins. Co., 213 Ill.App.3d 319, 157 Ill.Dec. 498, 572 N.E.2d 962, 963 (1991) (noting that primary and excess policies perforce cover different risks). We decline to address that question here.