SELYA, Circuit Judge.
Invoking diversity jurisdiction, 28 U.S.C. § 1332, appellants Maury A. Ryan, James H. Ryan and Stanley Klimek, trading as the Ryan, Klimek, Ryan Partnership (collectively, “Ryan”), brought suit in the United States District Court for the District of Rhode Island against appellee Royal Insurance Company of America (“Royal”). Appellants alleged
inter alia
that Royal had failed to defend and indemnify according to the tenor of a series of insurance policies issued by it. The district court granted summary judgment in Royal’s favor on all counts of the amended complaint.
Ryan, Klimek, Ryan Partnership v. Royal Ins. Co.,
728 F.Supp. 862 (D.R.I.1990). We affirm.
I. BACKGROUND
Ryan owned real property in Henrietta, N.Y. Until 1987, the site was leased to Stuart-Oliver-Holtz, Inc. (“SOH”), a corporation owned by the individual appellants. SOH conducted a plating and painting business there. Its operations contaminated the site with trichlorethylene (“TCE”) and other hazardous chemicals.
On December 20, 1974, a fire severely damaged the premises. Appellant had $553,000 in insurance coverage available at the time. Arm’s-length negotiations between them and Royal, meticulously detailed by the district court,
id.
at 864-65, resulted in a settlement of all insurance claims for $474,929. Though the list of lost materials included two drums of TCE, neither side raised the issue of potential groundwater contamination and no claim for pollution cleanup costs was advanced. Following consummation of the settlement, the structural damage was repaired and SOH resumed full operations, apparently discontinuing its TCE usage at that juncture.
In 1986, SOH sought the protection of Chapter 11 of the Bankruptcy Code. Ryan, glimpsing the handwriting on the wall, began attempting to sell the property. Concerned about toxic waste contamination, Ryan hired Lozier Architects/Engineers (“Lozier") to conduct an environmental study. Lozier’s site assessment report (“SAR”) confirmed appellants’ fears of pollution, indicating the presence of TCE in the groundwater.
Ryan reported Lozier’s findings to both the New York Department of Environmental Conservation (“NYDEC”) and the federal Environmental Protection Agency (“EPA”), sending each agency a copy of the SAR. Ryan’s counsel discussed the situation with NYDEC personnel over the telephone and in person during the spring of 1987. On April 1, NYDEC sent Ryan a letter confirming the substance of these discussions. Pointing out that federal law required correction of hazardous waste contamination, NYDEC wrote that the “EPA retains primary responsibility for the implementation authority of the corrective action provision,” and elaborated the kinds of corrective activities usually required in EPA consent orders. As to state action, NYDEC indicated only that (1) it would place the Henrietta site on the New York State Registry, an informational listing of all sites known or suspected to contain hazardous wastes; and (2) the state super-fund program would address the site in the event that the EPA did not do so. The letter explained what the state’s superfund program customarily entailed, remarked deficiencies in Lozier’s SAR, and requested Ryan to submit plans for any proposed remedial work.
In two subsequent letters, dated April 27 and June 2 respectively, NYDEC first advised SOH of the need to submit a complete closure plan and thereafter told SOH that it (NYDEC) would likely not pursue certain treatment, storage, and disposal (“TSD”) violations at the site if SOH accomplished closure in an approved manner.
In none of these letters did NYDEC demand that SOH or Ryan undertake to clean up the site. Throughout, the agency’s mien was conciliatory rather than belligerent. It called for voluntary cooperation “in addressing the site contamination,” stating, in its final communique, that it was seeking to explore ways of meeting this goal that would not “cause undue hardship to [SOH].” To this date, neither NYDEC nor EPA has ever insisted upon decontamination of the site or demanded reimbursement of cleanup costs.
And as the district court observed, “none of the parties involved have expended any sums for the cleanup of the ... property.”
Ryan, Klimek, Ryan Partnership,
728 F.Supp. at 865.
Before and during the events in question, Royal insured Ryan under a series of comprehensive general liability (“CGL”) policies. Soon after initiating the dialogue with NYDEC, appellants requested that Royal “defend” them against the state agency’s “claim,” and “indemnify” them for the “damages” sustained by reason of the site contamination.
Royal not only refused but cancelled the policy, effective July 13, 1987. To put appellants’ requests and appellee’s demurrer into proper perspective, we quote the pertinent policy language:
The Company will pay on behalf of the insured all sums which the insured shall become legally obligated to pay as damages because of bodily injury or property damage to which this insurance applies, caused by an occurrence, and the Company shall have the right and duty to defend any suit against the insured seeking damages on account of such bodily injury or property damage, even if any of the allegations of the suit are groundless, false or fraudulent, and may make such investigation and settlement of any claim or suit as it deems expedient, but the Company shall not be obligated to pay any claim or judgment or to defend any suit after the applicable limit of the Company’s liability [under the CGL policy] has been exhausted by payment of judgments or settlements.
In November 1987, Ryan entered into an agreement to sell the Henrietta property for $987,900. The sale was closed in due course. Taking the position that, absent toxic waste contamination, the value of the property was $2,100,000, appellants sued Royal, charging breach of contract and arguing that their “loss” from the sale, namely, the difference in property value with and without pollution, was the appropriate measure of the liability against which Royal insured.
II. PROCEEDINGS BELOW
Appellants’ amended complaint contained five counts. The gravamen of counts I and II was that defendant failed to indemnify' or defend the insureds with respect to what plaintiffs characterized as NYDEC’s claim. The district court, describing these counts as charging “that Royal willfully, maliciously, and in bad faith refused to defend the NYDEC’s cleanup order and refused to reimburse [the insureds] for cleanup costs,”
Ryan, Klimek, Ryan Partnership,
728 F.Supp. at 865, granted summary judgment for the defendant. It ruled that Royal’s duty to defend was not triggered because NYDEC’s participation never “rose to the level of [making] coercive, adversarial demands,” pointed out that the insureds had never “expended] any money towards a cleanup,” and reasoned that the duty to
indemnify could not accrue “until actual costs or damages arise.”
Id.
at 868.
In counts III and IV, appellants sought to set aside the fire loss settlement,
see supra
p. 732, on equitable grounds, e.g., fraud or mutual mistake, presumably believing that, if the 1974 fire loss were reopened, appellants could claim some of the cleanup expenses under the debris removal coverage. The district court also granted summary judgment on these two statements of claim.
See id.
at 869-71. Ryan has not pursued this aspect of the lower court’s order and we do not remark the topic further.
In the amended complaint’s last count, appellants prayed unsuccessfully for an award of consequential damages referable to Royal’s cancellation of the CGL policy in an “untimely” fashion and without "proper explanation.” The district court, on what it termed “the ‘cancellation’ issue,” granted summary judgment.
See id.
at 871-72.
On appeal, Ryan has briefed and argued the viability of counts I and II while ignoring counts III and IV. The status of count V is muddled. In briefing the argument that Royal was liable for consequential damages, appellants did allude to the “wrongful” termination of the .CGL policy. Yet, they never explain how or why the cancellation was at odds with the applicable policy provisions or governing law. It is settled in this circuit that issues adverted to on appeal in a perfunctory manner, unaccompanied by some developed argumentation, are deemed to have been abandoned.
See United States v. Zannino,
895 F.2d 1, 17 (1st Cir.),
cert. denied,
— U.S. -, 110 S.Ct. 1814, 108 L.Ed.2d 944 (1990);
Collins v. Marina-Martinez,
894 F.2d 474, 481 n. 9 (1st Cir.1990);
Brown v. Trustees of Boston Univ.,
891 F.2d 337, 352 (1st Cir.1989),
cert. denied,
— U.S. -, 110 S.Ct. 3217, 110 L.Ed.2d 664 (1990). Appellants’ passing reference did not succeed in preserving the question of wrongful cancellation for review. Appellants’ reply brief is a bit more focused, but it does not supply the missing argumentation. At any rate, it constitutes too little and comes too late.
See Sandstrom v. Chemlawn Corp.,
904 F.2d 83, 86-87 (1st Cir.1990) (appellants cannot instate omitted ground of appeal merely by raising it in a reply brief);
Pignons S
.A.
de Mecanique v. Polaroid Corp.,
701 F.2d 1, 3 (1st Cir.1983) (same).
We need not paint the lily. Whether or not the cancellation issue was properly raised appears, as the court below perspicaciously observed, to be an academic matter. Because no “contingencies insured against ever occurred during what would have been the remaining term of the policy,”
see Ryan, Klimek, Ryan Partnership,
728 F.Supp. at 872, the district court did not err in granting
brevis
disposition anent count V.
In the following pages, therefore, we concentrate almost exclusively on the lower court’s rulings with regard to counts I and II, commenting briefly, however, on the (arguably preserved) bad-faith claim contained in count V.
III. LEGAL PRINCIPLES AFFECTING REVIEW
Inasmuch as the issue of whether Royal was in breach depends upon the scope of the parties’ agreement, the task which confronts us is, prima facie, one of contractual interpretation. The parties acknowledge that New York law controls in this diversity case, so we must attempt to construe the CGL policy as would New York’s highest tribunal.
See Erie R. Co. v. Tompkins,
304 U.S. 64, 78, 58 S.Ct. 817, 822, 82 L.Ed. 1188 (1938);
Moores v. Greenberg,
834 F.2d 1105, 1107 (1st Cir.1987). When, as here, the question presented has not been determined by the state’s court of last resort, we look to “such sources as analogous state court de
cisions, adjudications in cases elsewhere, and public policy imperatives.”
Kathios v. General Motors Corp.,
862 F.2d 944, 949 (1st Cir.1988);
see also Moores,
834 F.2d at 1107 & n. 3.
In approaching the appellate task, we are also cognizant that the district court resolved this case on summary judgment. It was thus incumbent upon the judge to consider the record in the light most amiable to the plaintiffs, indulging all reasonable inferences in their favor.
See Griggs-Ryan v. Smith,
904 F.2d 112, 115 (1st Cir.1990);
Garside v. Osco Drug, Inc.,
895 F.2d 46, 48 (1st Cir.1990);
see generally
Fed.R.Civ.P. 56(c). The court of appeals must apply the same standard, vacating a grant of summary judgment if it determines that genuine issues of material fact, adequately raised and suitably documented below, need to be resolved at a trial before the dispositive legal issues can be adjudicated.
See Garside,
895 F.2d at 48;
Mack v. Great Atlantic & Pacific Tea Co.,
871 F.2d 179, 181 (1st Cir.1989).
IV. WHAT THE POLICY MEANS
The obligations to defend and indemnify under a CGL policy are correlative, not coterminous. At the outset, we discuss the insuring agreement as a whole, without differentiating between the duties. We then direct the lens of our inquiry primarily to the duty to defend — a duty which, in general, is broader and possesses a more elongated reach. Finally, we apply our rationale to the duty of indemnification.
A
By its express terms, the policy requires the insurer to defend the insured in respect to “any suit ... seeking damages” and to indemnify the insured for amounts the insured is “legally obligated to pay.” Royal argues for a bright-line rule, linking the obligation to defend to some pending judicial (or at least judicial-type) proceedings, and the obligation to indemnify to the existence of an actual judgment. Since Ryan’s losses stemmed from a private sale of property rather than from a suit and an imposed judgment, the insurer contends there was nothing to be defended against and no basis for any indemnification.
We believe that so literal an approach understates the sweep of the insuring agreement. While there is respectable authority conducing to a fairly rigorous benchmark under New York law,
see, e.g., State Farm Mut. Auto Ins. Co. v. Westlake,
35 N.Y.2d 587, 591, 324 N.E.2d 137, 138, 364 N.Y.S.2d 482, 485 (1974) ("[ljiability of the insurer attaches when there is a final judgment against the insured as a result of an obligation imposed by law”);
Podolsky v. Devinney,
281 F.Supp. 488, 494 (S.D.N.Y.1968) (obligation to indemnify cannot “arise until a judgment has been entered against the insured”);
cf. Medical Malpractice Ins. Ass’n. v. Medical Liab. Mut. Ins. Co.,
86 A.D.2d 476, 479-82, 450 N.Y.S.2d 191, 194-95 (1982) (hospital, vicariously liable, not entitled to subrogate against individual doctors because no judgment was ever entered against them), there is some play in the joints.
See generally
8 J. Appleman,
Insurance Law and Practice
§ 4851 (1981) (citing cases both ways).
To argue that the word “suit” is to be accorded talismanic significance brings to the language of the policy a precision that the drafter omitted and that the parties were not bound to anticipate.
See Avondale Indus. Inc. v. Travelers Indem. Co.,
887 F.2d 1200, 1206 (2d Cir.1989) (espousing “broad construction of the word ‘suit’ ”) (construing New York law),
cert. denied,
— U.S. -, 110 S.Ct. 2588, 110 L.Ed.2d 269 (1990);
Aire Frio, S.A. v. United States Fid. & Guar. Co.,
309 F.Supp. 1388 (D.Canal Zone 1970) (proceeding before industry board of contract appeal held to be a “suit” within the policy language);
Madawick Contracting Co. v. Travelers Ins. Co.,
307 N.Y. 111, 118, 120 N.E.2d 520, 523 (1954) (defining “suit” to include arbitration proceedings);
Community Unit School Dist. No. 5 v. Country Mut. Ins. Co.,
95 Ill.App.3d 272, 278-79, 50 Ill.Dec. 808, 812, 419 N.E.2d 1257, 1261 (1981) (holding unfair employment practice complaint filed with state administrative agency to trigger duty to defend and stating that there is nothing in the standard
policy language which limits suits “to legal actions in the common law courts”); 7C J. Appleman,
supra,
§ 4682 at 25 (1979) (suggesting that a variety of matters, such as “proceedings before an administrative body,” may trigger the obligation to defend);
cf. Simon v. Maryland Cas. Co.,
353 F.2d 608, 612 (5th Cir.1965) (formalities are less important than whether the situation satisfies the “sense, purpose and true meaning of a ‘suit against the insured’ ”);
but see, e.g., Aetna Cas. & Sur. Co. v. Gulf Resources,
709 F.Supp. 958, 960 (D.Idaho 1989) (“suit” means court litigation, not administrative claim or proceeding). Logic dictates that if a proceeding is the functional equivalent of a traditional suit, then coverage may inhere.
Cf. Clarke v. Fidelity and Cas. Co.,
55 Misc.2d 327, 336, 285 N.Y.S.2d 503, 511 (1967) (focusing upon substance rather than form in determining whether insured, who had been impleaded rather than sued directly, was being “sued” within meaning of policy). By the same token, the policy does not wed the phrase “legally obligated to pay” to judgments imposed by common law courts; so long as the obligation is sufficiently certain and precise, other constructions may at times be permissible.
See, e.g., Chemical Applications Co. v. Home Indem. Co.,
425 F.Supp. 777, 779 (D.Mass.1977) (allowing insured to recover, under liability policy, cleanup expenditures, notwithstanding absence of judgment) (applying Massachusetts law).
In New York, as elsewhere, the objectively reasonable expectation of the typical policyholder, here the ordinary businessman, must be considered when construing an insurance contract.
See, e.g., Ace Wire & Cable Co. v. Aetna Cas. & Sur. Co.,
60 N.Y.2d 390, 398, 457 N.E.2d 761, 764, 469 N.Y.S.2d 655, 658 (1983);
Clarke,
285 N.Y.S.2d at 511;
see also Titan Holdings Syndicate, Inc. v. Keene,
898 F.2d 265, 270 (1st Cir.1990). We are unprepared to say that the cryptic phrases “any suit” and “legally obligated to pay,” unqualified by clear and unmistakable language in the policy’s text,
cf., e.g., Seaboard Surety Co. v. Gillette Co.,
64 N.Y.2d 304, 311, 476 N.E.2d 272, 275, 486 N.Y.S.2d 873, 876 (1984) (exceptions from policy coverage “must be specific and clear in order to be enforced”);
Clarke,
285 N.Y.S.2d at 511 (“in case of doubt policies of insurance are construed against the insurer”), necessarily require (1) that a civil case be commenced in a court of law before the duty to defend arises and (2) that a money judgment be entered against the insured before the insurer’s payment obligation vests.
B
Recognizing, as we do, that the challenged phrases may, in certain limited circumstances, cross frontiers beyond those staked out by the formal institution of judicial proceedings and the reduction of awards to judgment, we are still left with the more specific question of whether the insuring agreement in this case can plausibly be construed to encompass the situation confronting, and the losses alleged by, the insureds. This inquiry produces a negative answer.
We commence our odyssey by focusing first on Royal’s duty to defend. While conceding the absence of any suit or formal administrative proceeding, appellants argue that the duty to defend arises whenever a state agency takes the position that an environmental condition requires remediation and notifies the policyholder to that effect. This hypothesis, we think, bends the policy language well past the breaking point.
Under New York law, liability insurance is characterized as "litigation insurance.”
Servidone Constr. Corp. v. Security Ins. Co.,
64 N.Y.2d 419, 423, 477 N.E.2d 441, 444, 488 N.Y.S.2d 139, 142 (1985);
Seaboard Surety,
476 N.E.2d at 275, 486 N.Y.S.2d at 876. Whereas the duty to defend has been broadly construed to arise “whenever the allegations in a complaint against the insured fall within the scope of the risks undertaken by the insurer, regardless of how false or groundless these allegations might be,”
id.
at 275, 486 N.Y.S.2d at 876;
see also A. Meyers & Sons Corp. v. Zurich Am. Ins. Group,
74 N.Y.2d 298,
302, 545 N.E.2d 1206, 1208, 546 N.Y.S.2d 818, 820 (1989) (liability insurer’s duty “to defend an action brought against an insured is determined by the allegations in the complaint”), existence of the duty presupposes that a complaint has been filed and a suit commenced — or that the functional equivalent of a complaint-cum-suit has emerged. The ramifications for the instant case are clear: the issue before us boils down to whether the New York Court of Appeals would likely treat the somewhat desultory correspondence Ryan received from NYDEC as constituting the functional equivalent of a “suit” sufficient to trigger the duty to defend under the terms of the policy. As the district court accurately observed,
Ryan, Klimek, Ryan Partnership,
728 F.Supp. at 868, the state’s highest tribunal has never decided this specific issue and, to the extent that other courts interpreting New York law have addressed the question, they have reached opposite, though not contradictory, conclusions. We review the three most significant decisions.
In
Technicon Electronics Corp. v. American Home Assur. Co.,
141 A.D.2d 124, 533 N.Y.S.2d 91 (1988), the Appellate Division of the New York Supreme Court held that a letter from the EPA designating the insured as a “potentially responsible party” (PRP) was not a “suit” sufficient to trigger the duty to defend. Inasmuch as the EPA letter “merely informed” the company of its “potential liability” under the Comprehensive Environmental Response, Compensation and Liability Act of 1980 (CERCLA), 42 U.S.C. § 9601
et seq.,
and alerted it that the EPA was “interested in discussing ... voluntary participation in remedial measures,” the notice “was an invitation to voluntary action on Techni-con’s part and ... not the equivalent of the commencement of a formal proceeding within the meaning of [a CGL policy].”
Technicon,
533 N.Y.S.2d at 105. The New York Court of Appeals subsequently affirmed on the ground that the policy’s pollution exclusion precluded coverage, declining to answer the question of whether the notice letter was a suit. 74 N.Y.2d 66, 542 N.E.2d 1048, 544 N.Y.S.2d 531 (1989).
In
Avondale,
887 F.2d 1200, the Second Circuit, interpreting New York contract law, concluded that a letter from the Louisiana Department of Environmental Quality (“LDEQ”) constituted a suit for purposes of crossing the coverage threshold. There, LDEQ made a formal “demand” for “immediate action” to clean up a hazardous waste site and ordered the insured to attend a meeting or face the institution of suit.
Id.
at 1202. Contrasting the adver-sariness of LDEQ’s demand letter to the meekness of EPA’s “invitation to voluntary action” in
Technicon,
the Second Circuit found the cases easily distinguishable: “A request to participate voluntarily in remedial measures is not the same as the adversarial posture assumed in the coercive demand letter that Avondale received in the instant case.”
Id.
at 1206.
The scope of the duty to defend under New York law was addressed most recently in an opinion authored by a single justice of the state supreme court in
County of Niagara v. Fireman’s Fund Ins. Co.,
4 Mealey’s Litigation Reports-Insurance No. 10, p. C — 1 (N.Y.Sup.Ct., Niagara Cty., Mar. 6, 1990). In
Niagara,
the insured received a PRP letter from the EPA. The letter “encourag[ed]” the insured to undertake “voluntary clean-up activities” and gave it 14 days to reply.
Id.
at C-4. Absent a positive response, the EPA said that it would (1) assume the company had declined to take ameliorative steps and (2) “act accordingly.”
Id.
Drawing upon the
Avon-dale
analysis, but stopping well shy of uncritical acceptance of
Avondale’s,
reasoning,
the court found the PRP letter “sufficiently ‘adversarial’ to constitute a suit.”
Id.
at C-5.
Despite the different conclusions they reach, the three cases have an important
common denominator: all of them stand for the proposition that, to find agency conduct to have created a suit counterpart sufficient to trigger the duty to defend under New York law, there must be some cognizable degree of coerciveness or adver-sariness in the administrative body’s actions.
C
We can, of course, look to authorities elsewhere for guidance in our attempt accurately to forecast New York law.
See Moores,
834 F.2d at 1107;
Murphy v. Erwin-Wasey, Inc., 460
F.2d
661, 663 (1st
Cir.1972). The results of such a search are, in one sense, inconclusive. Courts stand divided on the issue of whether a PRP letter, for example, is a fair congener to a suit.
Compare, e.g., Ray Indus., Inc. v. Liberty Mut. Ins. Co.,
728 F.Supp. 1310, 1314 (E.D.Mich.1989) (duty to defend triggered by PRP letter),
U.S. Fid. & Guar. Co. v. Specialty Coatings Co.,
180 Ill.App.3d 378, 388-89, 129 Ill.Dec. 306, 314, 535 N.E.2d 1071, 1079 (1989) (similar), and
Hazen Paper Co. v. U.S. Fid. & Guar. Co.,
407 Mass. 689, 696, 555 N.E.2d 576, 581 (1990) (consequences of receipt of EPA letter “so substantially equivalent” to commencement of suit that duty to defend arises “immediately”)
with Maryland Cas. Co. v. Armco, Inc.,
822 F.2d 1348, 1354 (4th Cir.1987) (mere possibility of liability under CERCLA does not trigger duty to defend),
cert. denied,
484 U.S. 1008, 108 S.Ct. 703, 98 L.Ed.2d 654
(1988), Arco Indus. Corp. v. Travelers Ins. Co.,
730 F.Supp. 59, 68 (W.D.Mich.1989) (PRP letter is not the functional equivalent of suit), and
Detrex Chem. Indus., Inc. v. Employers Ins. of Wausau,
681 F.Supp. 438, 446 (N.D.Ohio 1987) (EPA letter informing company that it might be liable for cleanup costs, penalties and punitive damages under CERCLA does not trigger duty to defend).
On the issue of whether a letter from a state environmental agency is a suit, the authorities are similarly divided.
Compare, e.g., Higgins Indus., Inc. v. Fireman’s Fund Ins. Co.,
730 F.Supp. 774, 776 (E.D.Mich.1989) (insurance company must defend all governmental claims and demands in environmental context),
Polkow v. Citizens Ins. Co.,
180 Mich.App. 651, 657, 447 N.W.2d 853, 856 (1989) (state environmental agency letter requiring insured to investigate and remedy contaminated site triggered duty to defend),
appeal granted,
435 Mich. 862 (1990), and
C.D. Spangler Constr. Co. v. Industrial Crankshank & Eng’g Co.,
326 N.C. 133, 153-54, 388 S.E.2d 557, 570 (1990) (issuance of state compliance order directing remedial action constituted “suit” within meaning of policies)
with Hazen Paper,
555 N.E.2d at 580 (letter from state environmental agency describing extent of release of hazardous materials from overpacked drums and ordering removal not equivalent to suit) and
Patrons Oxford Mut. Ins. Co. v. Marois,
573 A.2d 16, 20 (Me.1990) (state environmental agency directive aimed at compelling cleanup not a “suit seeking damages”).
In a more salient respect, however, the authorities are virtually unanimous and provide helpful guidance: our survey reveals that, notwithstanding idiosyncratic differences in nomenclature, case after case insists that, shy of an actual suit, a substantial entry-level burden must be carried before the duty to defend under a CGL policy arises. Most of the cases describe the entry-level burden in terms of coerciveness or adversariness.
See, e.g., Avondale,
887 F.2d at 1206. Almost all of them require, at a bare minimum, some showing of
probable
and
imminent
governmental action as a condition precedent to coverage.
In a variation on this theme, some recent decisions seem to suggest that the duty to defend arises whenever there has been a serious governmental effort to force the insured to take action (or, given a slightly different spin, whenever it becomes clear that there will be serious consequences if the insured fails to act).
See, e.g., Fireman’s Fund Ins. Co. v. Ex-Cell-O Corp.,
662 F.Supp. 71, 75 (E.D.Mich.1987) (“ ‘suit’ includes any effort to impose on the policyholders a liability ultimately enforceable by a court”);
Hazen Paper,
555 N.E.2d at 581 (consequences of receipt of EPA letters, which are “dangerous for the alleged polluter to ignore because they often result in
dispositive, extrajudicial solutions,” make them substantially the equivalent of lawsuits). These eases are not qualitatively different from the
Technicon/Avondale
line: they, too, focus on probabilities, not possibilities; on imminent consequences, not pessimistic speculation about a highly uncertain chain of future events. As a limiting principle, we find the test of serious efforts/serious consequences essentially synonymous with the test of adversariness and we believe that, in application, the formulations are effectively interchangeable. Cf
., e.g.,
W. Shakespeare,
Romeo and Juliet,
Act II, sc. ii (1595) (“that which we call a rose by any other name would smell as sweet”).
There is, of course, one particularly telling point. Precedent always maintains the presumption, either explicitly or implicitly, that “defense” implies a serious “offense.”
See, e.g., Avondale,
887 F.2d at 1206;
Continental Cas. Co. v. Cole,
809 F.2d 891, 898-99 (D.C.Cir.1987) (duty to defend arises only when insured is “legally threatened” because of covered activities). Without discernible exception, all the reported cases, whether focusing on adversariness, coer-civeness, seriousness of efforts, or seriousness of results, require more to implicate a CGL policy than an insured’s communication to a government agency of information suggesting that site decontamination activities may be exigible.
D
Where unsettled questions are involved, we can assume that New York’s highest court would adopt the view which, consistent with its precedent, seems best supported by the force of logic and the better-reasoned authorities.
See Moores,
834 F.2d at 1107 n. 3;
Stool v. J.C. Penney Co.,
404 F.2d 562, 563 (5th Cir.1968). Thus, we believe it is wise to look at the origins and purpose of the duty to defend in attempting to determine the gloss which the New York Court of Appeals would most likely place on the policy language. In this instance, such an exercise convinces us that the standard of adversariness is altogether consonant with the conceptual underpinnings of the duty to defend.
The root purpose of insurance is indemnity.
See
3 G. Couch,
Cyclopedia of Insurance Law,
§ 24:12 (2d ed. 1982). It follows, then, that insurance contracts ought to be interpreted and enforced consistent with the aim of conferring a benefit on the insured equal to — but certainly no greater than — the actual loss sustained.
See
R. Keeton & A. Widess,
Insurance Law: A Guide to Fundamental Principles, Legal Doctrines, and Commercial Practices,
§ 3.1(a) (1988); W. Vance,
Handbook on the Law of Insurance,
§ 14 (3d ed. 1951); 4 J. Appleman,
supra,
§ 2125, at 47. Put another way, the legitimate object of insurance is to provide reimbursement for loss— and nothing more. To the extent that an insured may obtain insurance proceeds greater than the actual loss sustained, the insurance contract runs afoul of public policy.
See
R. Keeton & A. Widess,
supra,
§ 3.1(c); W. Vance,
supra,
§ 14, at 102; 4 J. Appleman,
supra,
§ 2121, at 27-28.
A second public policy imperative concerns the need to guard against intentional destruction of property — -a form of fraud. To the extent that reimbursement might exceed the actual value of the loss, policyholders will have a powerful incentive to rid themselves of the insured property.
Cf., e.g.,
Statute of George II, 19 Geo. 2, c. 37 (1746) (preamble) (noting that lucrative insurance contracts had proven “productive of many pernicious practices, whereby great numbers of ships, with their cargoes, have ... been fraudulently lost and destroyed”). When insurance proceeds offer no net gain, any such incentive disappears.
In the realm of property insurance, the concept of insurable interest mirrors these concerns. An insurable interest encompasses a broadly conceived factual expectancy.
See generally
R. Keeton & A. Widess,
supra,
§ 3.2 (reviewing history of doctrine of insurable interest); 3 G. Couch,
supra,
§ 24:3; 4 J. Appleman,
supra,
§ 2125. Just as the principle of indemnity guided the evolution of the doctrine of insurable interest in property law, we believe it can serve as a beacon in plotting the
outermost limits of the duties imposed by, and the liability risks subsumed under, conventional CGL policies.
The risk against which liability insurance protects, ultimately, is the loss of assets through liability, say, through toxic waste liability. Whether the loss eventuates from a lawsuit or takes the shape of a judgment is of scant consequence to the insured. Other methods of proceeding, such as by administrative channels or imposing liens, can be just as devastating as civil litigation. Other forms of loss, such as incurment of cleanup costs or reimbursement to the government for such costs, will have the same economic impact.
Nevertheless, this does not mean that loss, in and of itself, can be the sole determinant. For two reasons, the principle of indemnity in the liability insurance context demands that the relevant focus be on avoidance of liability rather than conservation of assets. First, the extent of the insurer’s exposure is a function of the liability claim against the insured, not of any decrease in the value of the insured’s property. Whereas in property insurance models the underlying asset limits and defines the insurer’s obligation and serves to prevent the insured from claiming more than was lost, the underlying asset affords no such similar check in the liability insurance context. Second, in contrast to property insurance — where the existence of a claim against an insurer depends on actual destruction of, or harm to, the insured’s assets — liability insurance permits claims to be mounted against an insurer without such antecedent loss or harm. The third party’s claim against the insured, and thus, the insurer’s duty to the insured, bears no necessary quantitative relation to the insured’s personal assets. Because the opportunities for wagering and fraud are, therefore, correspondingly great, public policy suggests that liability insurance contracts be interpreted in such a way as to neutralize, or at least minimize, these concerns.
Admittedly, requiring a suit and an ensuing judgment as conditions precedent to insurance reimbursement will serve to implement the principle of indemnity and limit discernible opportunities for wagering and fraud. It can plausibly be argued, however, that so rigid a rule jettisons the baby with the bath water. Insurance is meant, by and large, to afford coverage, not to exclude it — and a suit-cum-judgment formulation by no means represents the broadest scope of coverage a liability insurer may provide without endangering the principle of indemnity. Even short of judicial proceedings, let alone entry of judgment, the fundamental purposes underlying the principle of indemnity will not be placed in mortal jeopardy so long as the factual expectancy of ultimate liability is sufficiently high and its quantification sufficiently precise.
The junction where environmental law meets insurance law provides especially fertile ground for these equivalencies. Where pollution coverage is concerned, there is precious little to commend an inflexible suit-cum-judgment rule as opposed to a standard anchored in the probability that a potential liability will actually materialize in the immediate future. For one thing, liability under many hazardous waste laws is strict.
See, e.g., Dedham Water Co. v. Cumberland Farms Dairy,
889 F.2d 1146, 1150 (1st Cir.1989); T. Gordon & R. Westendorf,
Liability Coverage
for Toxic Tort, Hazardous Waste Disposal and Other Pollution Exposures,
25 Idaho L.Rev. 567, 568-72 (1988-89). The probability of liability upon the commencement of enforcement proceedings is therefore higher than in many other situations. For another thing, government is a major force behind the cleanup of hazardous waste sites. Given the regulatory scheme, a considerable part of the government’s energies vis-a-vis PRPs may be spent prior to the initiation of suit, for example, in negotiations or administrative hearings.
See, e.g., United States v. Cannons Eng’g Corp.,
899 F.2d 79 (1st Cir.1990) (tracing extensive litany of pre-suit dealings). Consequently, the likelihood that PRPs will endure hazardous waste liability does not always — or even often — hinge upon the filing of a formal complaint in court.
To sum up, the origins and purpose of the duty to defend seem best accommodated neither by a restrictive suit-cum-judgment rule nor by an expansive “any contact with a government agency is enough” rule, but by focusing instead on the data most relevant to the probability of actual toxic waste liability: coereiveness, adversariness, the seriousness of the effort with which the government hounds an insured, and the gravity of imminent consequences. Since the law holds PRPs to so strict a liability standard, the degree of compulsion the government wields in pursuing an insured seems an apt proxy for measuring factual expectancy according to the actual probability and immediacy of toxic waste liability.
Cf., e.g., Liberty Mut. Ins. Co. v. Continental Cas. Co.,
771 F.2d 579, 586 (1st Cir.1985) (pre-suit services may be considered part of defense against liability, for insurance purposes, when suit is certain and need for services virtually inevitable). Equally as important, a requirement that courts look to adversariness to establish a sufficient factual expectancy of imminent liability works to avoid the twin evils that imperil the principle of indemnity. If an insured may claim reimbursement from its insurer only when the government actually becomes the insured’s foe and undertakes a serious effort to enforce liability, the opportunities for profiteering and fraud will be substantially diminished because, in the last analysis, the insured can plausibly claim only that which the government plausibly seeks. As a theoretical matter, then, allowing certain narrowly defined legal relationships to serve as proxies for suits and judgments seems to have considerable merit.
E
In this case, all roads lead to Rome. The policy language, the decisions applying New York law, the case law elsewhere, and the theoretical underpinnings of the duty to defend converge to instruct us on the one hand that a rigid suit-cum-judgment interpretation of language such as is contained in Royal’s policy is unwarranted; and on the second hand that something more than an invitation voluntarily to initiate cleanup activities is required to animate the insurer’s duty. The “something more,” we suggest, must relate to the seriousness of purpose which characterizes the government’s role.
If government assumes an adversarial posture, making sufficiently clear that the force of the State will be brought promptly to bear in a way that threatens the insured with probable and imminent financial consequences, then the functional equivalent of a suit may be in progress and the insured might reasonably expect the insurer to defend.
Here, however, there is no “something more.” Evidence of coerciveness or a serious state enforcement effort is completely
wanting. The sense and flavor of the insureds’ dealings with NYDEC cannot realistically be termed adversarial. We are confident that, no matter where the New York Court of Appeals might ultimately choose to draw the precise line, Royal would be home free on the facts
sub judi-ce.
After all, “the power to bind an insurer must be found in the written contract of insurance.”
Clarke,
285 N.Y.S.2d at 511. Whatever ambiguity may lurk in the policy language, it is simply too much of a stretch to bind Royal by the CGL policy’s duty-to-defend language to the acceptance of NY-DEC’s implied invitation to voluntary action.
To be sure, appellants do not go gently into this dark night. They argue strenuously that, because environmental liability is strict, a property owner whose land is polluted becomes,
ipso facto,
“legally obligated to pay,” so that
any
governmental involvement is tantamount to a suit. We do not agree. Even though environmental liability may be strict, it is only when the government actually purposes to enforce the law against a property owner that the latter will bear the consequences of strict liability. If the government decides for any reason (e.g., shortage of funds) not to pursue public rights, the property owner will avoid liability, no matter how dim his prospects on the law and the facts. Thus, absent serious pursuit of the public interest by the agency charged — what the district court, and other authorities, term “adver-sariness” — the factual expectancy of liability is too low to satisfy either the principle of indemnity or any plausible construction of the policy language. Put another way, the “reasonable expectation and purpose of the ordinary businessman,”
Ace Wire,
457 N.E.2d at 764, 469 N.Y.S.2d at 658, would not envision coverage in the absence of adversariness.
Appellants also argue that, even on a standard of adversariness, the road confronting us leads not to Rome, but perhaps to El Dorado. They say that NYDEC’s overtures, like the LDEQ letter in
Avon-dale,
constituted a coercive demand for action sufficient to trigger the duty to defend. We have read NYDEC’s letters in the light most hospitable to the appellants and indulged all reasonable inferences in their favor. The communications patently lack any significant indicia of adversariness. In none of the letters does NYDEC use hortatory terminology. No mention is made of a “demand” or an “order.” The sternest note struck in the agency’s initial letter was a request that appellants “should submit ... for approval a work-plan for any further work anticipated at the site.” The agency’s second letter likewise placed the insureds under no compulsion to begin a cleanup of the site. It cited possible violations of environmental regulations and warned of potential penalties only insofar as appellants might fail to supply required documentation concerning the closure plan. NYDEC’s final letter was cut from much the same uncontroversial cloth, making clear that NYDEC had no intention of pursuing certain TSD violations or causing appellants any “undue hardship” as long as closure was achieved in an approved manner.
These mild statements are in marked contrast to the adversarial posture adopted by the state agency in
Avondale,
887 F.2d at 1202, where LDEQ said it planned to take “immediate action” holding the company responsible. Furthermore, the laws of New York, unlike those of Louisiana, require both notice and an opportunity to be heard before the state agency gains authority to implement a cleanup order.
Compare
N.Y. Environmental Conservation Law § 27-1313 (McKinney 1984)
with
La.Rev.Stat.Ann. §§ 30:2025(C), (G) (West 1989). Insofar as the record reflects, in this instance no such proceedings were ever convened, or scheduled, or even mentioned. Thus, NYDEC’s hope for voluntary compliance could not possibly be construed as carrying an implied threat that, if
appellants did not willingly capitulate, they would soon be confronted with a mandatory order.
Appellants, of course, had cause for concern — but that cause was, at the time material hereto, more theoretical than real. Once they put NYDEC on notice of underground contamination at the site, they may have been wise to assume that, if they did not comply to the extent necessary to gain closure approval, there was the possibility that NYDEC might in the future resort to available statutory procedures. They might also have assumed that, eventually, EPA or NYDEC — if either thought the site contamination to be significant — might initiate some corrective action. But the mere possibility of future litigation, indefinite and unfocused, cannot trigger the duty to defend under a CGL policy. Reading the contract so liberally would effectively rewrite it, producing a very different type of agreement than the parties bargained for or reasonably could have expected.
In the absence of any genuine issue of material fact, the district court appropriately entered summary judgment in Royal’s favor on count I of the complaint. Fed.R. Civ.P. 56(c).
F
Consideration of count II (charging breach of the duty to indemnify) need not detain us long.
Under New York law, it is a well-established principle that the duty to defend is broader than the duty to indemnify.
See Seaboard Surety,
476 N.E.2d at 275, 486 N.Y.S.2d at 876;
International Paper Co. v. Continental Cas. Co.,
35 N.Y.2d 322, 326, 320 N.E.2d 619, 621, 361 N.Y.S.2d 873, 876 (1974). That is because “[t]he duty to defend is measured against the allegations in the pleadings but the duty to pay is determined by the actual basis for the insured’s liability to a third person.”
Servidone, 477
N.E.2d at 444, 488 N.Y.S.2d at 142.
We have found on the facts of this case that Royal had no duty to defend under the terms of the policy and New York insurance law. It follows
a fortiori
that Royal had no obligation to indemnify. After all, there was no “suit” against the insureds. They never became “legally obligated to pay as damages” any “sums” on account of NYDEC’s “claim.” The “damages” for which indemnity is sought did not result from NYDEC’s demands or its assumption of an adversarial stance; rather the damages arose in consequence of a private market transaction, albeit one influenced indirectly, perhaps, by potential public liability. Such losses, whether or not economically real, cannot form the basis for an indemnity claim against an insurer under the provisions of a CGL policy.
For these reasons, the district court’s grant of summary judgment on count II was unimpugnable.
V. CONCLUSION
Other issues raised by appellants in their somewhat discursive briefing do not require comment. To the extent that any such issues are material, we reject them out of hand. We mention only appellants’ repeated exhortation that Royal may somehow be liable for consequential damages because it acted in bad faith.
We need not decide, however, whether Ryan, in opposing summary judgment, adduced
enough proof of Royal’s alleged bad faith to raise a genuine question of material fact. Whatever theoretical liability an insurer may have for actions undertaken in bad faith, if Royal had no obligation under the policy to defend against NYDEC’s “suit” or to indemnify Ryan with respect thereto, then its corporate behavior in turning a deaf ear to Ryan’s plaints, “no matter how rude or unseemly,” could not serve as a predicate for a bad-faith claim.
See, e.g., Gleason v. Merchants Mut. Ins. Co.,
589 F.Supp. 1474, 1477 (D.R.I.1984);
see also Peckham v. Continental Cas. Ins. Co.,
895 F.2d 830, 836 (1st Cir.1990) (emphasizing need to show that the insurer’s bad-faith conduct caused the claimed harm) (citing representative cases from several jurisdictions);
Harris v. Standard Accident and Ins. Co.,
297 F.2d 627, 633 (2d Cir.1961) (under New York law, insured must prove harm attributable to insurer’s bad-faith conduct before recovery can be had),
cert. denied,
369 U.S. 843, 82 S.Ct. 875, 7 L.Ed.2d 847 (1962). Put another way, unless appellants can demonstrate that Royal breached a duty owed to them under the CGL policy, their claim for consequential damages will not lie. As we have already determined that no breach of duty occurred in this case,
see supra
Part IV, the bad-faith claim is meritless.
We need go no further. Appellants elected to sue in the district court rather than in the New York state courts. We have warned, time and again, that litigants who reject a state forum in order to bring suit in federal court under diversity jurisdiction cannot expect that new trails will be blazed.
See, e.g., Porter v. Nutter,
913 F.2d 37, 41 (1st Cir.1990);
Croteau v. Olin Corp.,
884 F.2d 45, 46 (1st Cir.1989);
Kassel v. Gannett Co.,
875 F.2d 935, 950 (1st Cir.1989);
Cantwell v. University of Massachusetts,
551 F.2d 879, 880 (1st Cir.1977). In effect, appellants have asked us, as they asked the district court, to stretch New York law to reach an unknown and unexplored frontier. They have, however, given us no well-plotted roadmap showing an avenue of relief that the state’s highest court would likely follow. In such circumstances, appellants’ supplication must be rejected. Here, again, “[w]e may, perhaps, be unadventurous in our interpretation of [state] law, but a plaintiff who seeks out a federal venue in a diversity action should anticipate no more.”
Porter v. Nutter,
913 F.2d at 41 (footnote omitted).
Affirmed.