Scott Peckham v. Continental Casualty Insurance Co., Scott Peckham v. Continental Casualty Insurance Co.

*832 SELYA, Circuit Judge. These appeals stem from a tragic highway accident in which plaintiff Scott Peck-ham was rendered a quadriplegic. Andrew Tripp was driving the vehicle in which Peckham was a passenger. The facts of the accident are relatively unimportant. It suffices to say that Tripp’s liability was clear and his coverage, underwritten by defendant Continental Casualty Co. (CNA), was meagre. Peckham and his wife, plaintiff Jo Anne Peckham, thereafter engaged counsel, Brian Corey, who strove mightily to find a silver lining amidst clouds of despair. Counsel’s efforts led, eventually, to a me-gaverdict, an assignment of rights from Tripp, and a suit against CNA alleging inter alia negligence, bad faith, breach of contract, and unfair claim settlement practices. Following a bifurcated trial, all parties appeal. I.BACKGROUND Tripp was insured by CNA under a standard Massachusetts automobile policy, 1983 vintage. The policy provided bodily injury coverage of $10,000 “for injuries to any one person as a result of any one accident” (the “per person” limit), and $20,000 “for injuries to two or more people as a result of any one accident” (the “per accident” limit). Underinsured motorist coverage, subject to the same monetary limits, was afforded. The policy also contained certain ancillary avails to which Peckham was entitled, i.e., $5,000 in medical payments coverage and $2,000 in no-fault personal injury protection benefits. The sequence of events which ensued can be capsulized as follows: 1. In October 1983, Corey spoke to Eileen Kelley, a CNA claim representative. He presented claims for Scott and Jo Anne (who, although not herself involved in the accident, alleged a loss of consortium). CNA promptly offered to settle the claims for the “full policy limits.” 2. A hitch soon developed. Corey believed that JoAnne’s loss of consortium constituted an injury to her person, thus entitling her to $10,000 under the policy’s bodily injury coverage and $10,000 under the underinsured motorist coverage, completely independent of her spouse’s claim. Kelley disagreed, believing that, under the policy, JoAnne’s loss of consortium flowed from the injury to Scott’s person and that, consequently, the per person limits blanketed the plaintiffs’ claims. Translated into dollars, Corey believed CNA’s maximum contractual exposure to be $47,000 (the $10,000 per person liability limit applying separately to Scott’s claim and to JoAnne’s claim, totalling $20,000; the $10,000 under-insured motorist limit applying in the same manner, totalling another $20,000; and $7,000 being payable in ancillary benefits), whereas CNA thought its full exposure on the Peckhams’ claims was capped at $27,-000 (the $10,000 per person liability limit vis-a-vis Scott’s injuries subsuming both his claim and his wife’s derivative claim for loss of consortium; the $10,000 per person underinsured motorist limit operating in the same way; and $7,000 being payable in ancillaries). Because CNA did not consider the consortium claim to be a separate injury under the policy, Kelley’s offer of “full” policy limits translated into a settlement offer of $27,000 for all claims. 3. In February 1984, Corey rejected CNA’s proposal, but offered either to settle Scott’s claim alone for $27,000, or to settle all claims for $47,000. 4. In March, Kelley learned that a case was pending before the Massachusetts Supreme Judicial Court (SJC) which would determine the precise (and unsettled) question of whether a loss of consortium claim was to be treated as an injury to the person of the claimant for coverage purposes. She informed Corey of this development and told him that a judicial decision was expected in the near future. At Kelley’s request, attorney Stephen Paris then drafted a proposed agreement providing that, in exchange for the Peckhams’ general release (discharging Tripp and CNA), the insurer would pay Scott $27,000 and would forthwith pay JoAnne an additional $20,000 if and when the SJC held that a consortium claim was an independent injury for coverage purposes. The proposal (which we *833 shall call a “Bilodeau agreement”) was received by Corey on or about March 23, 1984. 5. Corey never responded directly to this proposal. Indeed, the record does not show that he contemporaneously revealed it to his clients or discussed it with them. Rather, Corey advised the Peckhams that CNA refused to settle Scott’s claim unless JoAnne also signed a release. He neither mentioned the SJC case nor alerted the Peckhams that CNA’s proposition envisioned paying JoAnne an additional $20,000 should the coverage issue be resolved favorably. 6. On April 19, 1984, Corey filed suit against Tripp in state superior court, seeking $8,000,000 in damages for Scott’s injuries and $3,000,000 for JoAnne’s loss of consortium. When Kelley inquired ten days later as to the status of CNA’s settlement offer, Corey did not allude to the suit but advised her that a response would be forthcoming shortly. 7. On May 30, 1984, Christine Cooney, a staff attorney for CNA, learned that Tripp had been sued and was already in default for failure to answer. She called Corey. He said that his offer to settle for $47,000 was no longer open and that he was now seeking payment in excess of the policy limits because of CNA’s “bad faith.” The superior court allowed a motion to remove the default and CNA answered the case on Tripp’s behalf. 8. No CNA representative met with Tripp until Kenneth Roberts interviewed him on June 18, 1984. Prior thereto, the insurer made no effort to apprise Tripp of the status of negotiations or the coverage dispute. The record is also clear that CNA did not, before then, inform Tripp of the pendency of Bilodeau in the SJC or explain to him that he might have the benefit of an additional $20,000 in coverage if the SJC were to hold that loss of consortium was a separate injury for coverage purposes. Roberts did request authority from Tripp to offer the policy limits to Scott Peckham ($27,000), leaving JoAnne’s claim open. Tripp did not authorize Roberts to pursue that course of action. Roberts spoke with Tripp again in late June. He testified that he mentioned “there was a pending case that would deal with the wife’s claim,” and discussed it with Tripp in layman’s terms (although he apparently never spelled out the case’s implications). Roberts again sought Tripp’s acquiescence in a $27,000 offer to Scott Peckham. Tripp, again, was noncommittal. 9. On July 17, 1984, the SJC decided the unresolved coverage question, holding that under a standard automobile policy loss of consortium was an “injury” to the claimant’s person separate and distinct from the spouse’s physical injury. See Bilodeau v. Lumbermens Mut. Cas. Co., 392 Mass. 537, 467 N.E.2d 137 (1984). The record is clear that Bilodeau represented a radical departure from the previously prevailing view within the industry. 10. Within a matter of weeks, Kelley informed Corey that CNA would settle all claims for $47,000 as Corey had earlier demanded. This overture was unceremoniously rejected. Thereafter, Paris and Corey engaged in negotiations which addressed not only the Peckhams’ bodily injury claims but also the potential bad faith claim against CNA. The insurer proposed, for example, that it would pay the full coverage ($47,000); that Tripp would be relieved of all personal liability and would assign any rights he might have against CNA to the Peckhams; and that CNA’s liability for damages in excess of the policy limits, if any, would be determined in subsequent court proceedings. Corey rejected this proposal on January 23, 1985, wanting a jury to reduce the Peckhams’ claims against Tripp to verdict before the Peck-hams went to the mat with CNA. Negotiations stalled. 11. Meanwhile, CNA continued to provide Tripp with a defense as the policy required. By July 1986, Timothy Sterritt had replaced Cooney as Tripp’s counsel of record. Corey told Sterritt that his clients would accept CNA’s policy limits and release Tripp on condition that Tripp assign his rights against CNA to them. Sterritt informed Tripp of this proposition and recommended that he engage private counsel. *834 Tripp took the advice and retained attorney Steven Chananie. 12. In late 1986, Chananie sent a letter to Corey confirming Tripp’s acceptance of an offer to hold Tripp harmless for any excess judgment in exchange for an assignment of rights against CNA. Chananie’s acceptance of a pending offer seems to have clinched a deal. Corey claims, however, that he held matters in abeyance at that point (although agreeing to keep the offer open until after a verdict had been returned). The trial then took place in March 1987. The jury awarded $3,000,000 to Scott Peckham and $75,000 to JoAnne Peckham. CNA paid its full policy limits, together with interest and costs as required. 1 13. In May 1987, Tripp and the Peck-hams formalized their pretrial agreement. Tripp executed and delivered an assignment; the Peckhams executed and delivered a release. The Peckhams, in their capacity as Tripp’s assignees, then sued CNA in state court. CNA removed the case to federal district court. II. PROCEEDINGS BELOW Plaintiffs filed a five count complaint. We are concerned with only three statements of claim: count II (which alleged that CNA acted in bad faith in failing to settle the Peckhams’ claims against Tripp); count IV (which alleged breach of contract); and count V (which alleged that CNA had engaged in unfair and deceptive acts in violation of Mass.Gen.L. ch. 93A and ch. 176D). The district court bifurcated the trial. In the first phase, a jury was empaneled to consider counts II and IV. After denying several motions, including plaintiffs’ motion for a partial directed verdict on causation, the district court submitted the case through special questions. In response, the jury found that CNA, prior to the rendition of Bilodeau, acted in bad faith by failing to settle Scott Peckham’s claim within the policy limits. The jury also found, however, that such bad faith did not cause the excess judgment against Tripp. Moreover, the talesmen absolved CNA of any bad faith after July 17, 1984. Consequently, the jury neither recommended the imposition of damages nor fixed an amount. The court discharged the panel and entered judgment for defendant on counts II and IV. Because the special questions and the answers thereto assume critical importance on appeal, we attach the three page verdict form as an appendix. Count V was thereafter tried to the bench. The district court determined that CNA acted in bad faith — thereby violating chapter 93A — in one limited respect. The court went on to find that CNA’s blunder caused Tripp no harm and awarded Tripp’s assignees (the Peckhams) nominal damages on their statutory claim, together with attorneys’ fees. III. SOME GENERAL OBSERVATIONS To put these appeals into proper perspective, we write briefly about the relationship of insurer to insured. Under Massachusetts law, an insurance carrier has a duty to make settlement decisions in good faith. Murach v. Massachusetts Bonding & Ins. Co., 339 Mass. 184, 158 N.E.2d 338, 341 (1959). Where, as here, the policy limits are much lower than the insured’s potential exposure, the insurer cannot put its own interests first, but must negotiate as it would if its liability limits were unbounded. See id. As we have phrased it, “the duty to negotiate in good faith would require the carrier to give ‘the interest of the insured’ consideration ‘equal to that consideration given its own interest,’ or ‘to treat the claim as if it were alone liable for the entire amount’.” Voccio v. Reliance Ins. Cos., 703 F.2d 1, 2 (1st Cir.1983) (citations omitted). Moreover, in such straitened circumstances, the insurer must inform the insured of its conflicting interests, advise him of his rights, and keep him abreast of settlement offers and meaningful developments. See Murach, 158 N.E.2d at 342. *835 The thorny problem faced by an insurer in an excess-limits case is complicated logarithmically when multiple claims exist, each likely to outstrip the coverage. Apprising the insured becomes even more important in that circumstance, for payment to one claimant, exhausting or unreasonably depleting the available fund, may leave the insured unprotected — or nearly so — in respect to other claimants. The insurer has both the right and the duty to exercise its professional judgment in settling, or refusing to settle, such claims — but it must do so mindful of the insured’s best interests and in good faith. The insurer’s goal should be to try to effect settlement of all or some of the multiple claims so as to relieve its insured of so much of his potential liability as is reasonably possible, considering the paucity of the policy limits. 2 See Liberty Mut. Ins. Co. v. Davis, 412 F.2d 475, 481 (5th Cir.1969). So long as it acts in good faith, the insurer is not held to standards of omniscience or perfection; it has leeway to use, and should consistently employ, its honest business judgment. See Steele v. Hartford Fire Ins. Co., 788 F.2d 441, 447 (7th Cir.1986) (mistake in judgment alone does not constitute bad faith); see also Gulezian v. Lincoln Ins. Co., 399 Mass. 606, 506 N.E.2d 123, 127 (1987). The carrier, in fine, “will not be held to prophesy.” Murach, 158 N.E.2d at 341. The insurer’s duty, of course, is to its insured. It owes no correlative contractual duty to third-party claimants. Correspondingly, such claimants — and their counsel— are not obliged to ease the insurer’s bind. A claimants’ attorney owes his allegiance to his clients and must strive, within ethical and legal constraints, to maximize their recovery. Nevertheless, we must not lose sight of the fact that the doctrinal impetus for insurance bad faith claims derives from the idea that the insured must be treated fairly and his legitimate interests protected. Claimants have no right to insist upon any punctilio in the insurer’s observance, or not, of its obligations toward its insured. In other words, the justification for bad faith jurisprudence is as a shield for insureds — not as a sword for claimants. Courts should not permit bad faith in the insurance milieu to become a game of cat- and-mouse between claimants and insurer, letting claimants induce damages that they then seek to recover, whilst relegating the insured to the sidelines as if only a mildly curious spectator. IV. THE JURY TRIAL Plaintiffs challenge the conduct and outcome of the jury phase of the bifurcated trial, complaining that the district court erred in submitting the issue of causation for jury resolution. The plaintiffs contend that, as a matter of law, defendant’s bad faith must be deemed the cause of the excess judgment against Tripp and that the jury should not have been afforded the prerogative to decide otherwise. A. Certain Preliminary Matters. Before entering these embattled precincts, we point out that plaintiffs are effectively foreclosed from attacking (1) the special questions, (2) the need for a showing of proximate cause in a bad faith case, or (3) the admission of expert testimony anent the issues. We explain briefly. Plaintiffs cannot be heard to argue either ambiguity in the framing of the spe *836 cial questions or possible inconsistency in the answers thereto. Fed.R.Civ.P. 49 governs special verdicts and interrogatories. The parties agree that in this instance Rule 49(a) controls. 3 The rule ensures that, if submitted questions omit material issues of fact and no timely objection is lodged, the district court may itself make the findings which are necessary to cure the omission. See Anderson v. Cryovac, Inc., 862 F.2d 910, 915-16 (1st Cir.1988). Curative findings are implied even when not expressly made. Id. Any claim of inconsistency in the answers is also outlawed at this stage. When the jury returned with replies to the special questions, plaintiffs’ counsel did not seasonably claim that the answers were contradictory and request that they be reconciled. Rather, he permitted the verdict to be returned and the panel dismissed before voicing any concerns. The law is perfectly clear that plaintiffs thereby waived any claim of internal inconsistency “by failing to object after the verdict was read and before the jury was discharged.” See McIsaac v. Didriksen Fishing Corp., 809 F.2d 129, 134 (1st Cir.1987). Thus, the special verdict must “be upheld if there is a view of the case which makes the jury’s answers consistent.” Id. at 133. We have similar problems with plaintiffs’ assertion that an insurer’s bad faith refusal to settle, once made manifest, is conclusively presumed to be the cause of an ensuing excess judgment. For one thing, this contention directly contradicts the position which they took at trial 4 — and we are not prone to let litigants so easily play the chameleon. For another thing, we think it is settled beyond serious question that, in order to warrant recovery for an insurer’s bad faith anent settlement negotiations, it must be shown that the insurer’s conduct caused the ensuing excess judgment. See Steele, 788 F.2d at 449; Voccio, 703 F.2d at 2. The question seems to be a “simple question of causation, like any other causation question, for example, in any garden variety tort case,” McNally v. Nationwide Ins. Co., 815 F.2d 254, 263 (3d Cir.1987), and thus, ordinarily, for the jury rather than the court. Causation is binary, comprising causation in fact and proximate (or “legal”) causation. A defendant’s breach of a legal duty is a cause in fact of the plaintiffs’ harm if that harm would not have occurred “but for” the breach. Restatement (Second) of Torts § 432(1) (1965). Nevertheless, but-for causality is only half the battle. The second half is whether the breach proximately caused the harm, that is, whether the defendant should bear legal responsibility for the injury. What is more, the inquiry into proximate cause is independent of that into actual cause. Swift v. United States, 866 F.2d 507, 509 (1st Cir.1989) (applying Massachusetts law). “The touchstone is foreseeability: [conduct results in] liability if, and to the extent that, a foreseeable risk of harm materializes.” Id. at 510. *837 Causation questions of this sort are normally grist for the jury’s mill. As we have written: “Application of the legal cause standard to the circumstances of a particular case is a function ordinarily performed by, and peculiarly within the competence of, the factfinder.” Id.; see also Marshall v. Perez Arzuaga, 828 F.2d 845, 850-51 & n. 8 (1st Cir.1987) (similar), cert. denied, 484 U.S. 1065, 108 S.Ct. 1027, 98 L.Ed.2d 991 (1988); Springer v. Seamen, 821 F.2d 871, 876 (1st Cir.1987) (“[n]ot only ordinary fact questions, but also ‘evaluative applications of legal standards ... to the facts’ are properly [for the factfinder]”) (citations and footnote omitted). The Massachusetts cases are to like effect. See, e.g., Irwin v. Town of Ware, 392 Mass. 745, 467 N.E.2d 1292, 1305 (1984); Mullins v. Pine Manor College, 389 Mass. 47, 449 N.E.2d 331, 338 (1983); Zezuski v. Jenny Mfg. Co., 363 Mass. 324, 293 N.E.2d 875, 878-79 (1973); see also DiMarzo v. American Mut. Ins. Co., 389 Mass. 85, 449 N.E.2d 1189, 1199 (1983) (in insurance bad faith setting, “the question whether the requisite causal connection has been proven is one of fact”). That causation is ordinarily a fact question does not make it invariably a fact question. But, the standard of appellate review is strict. To warrant withdrawal of the issue from the jury, the evidence and the inferences reasonably extractable therefrom, seen in the light most hospitable to the verdict-winner, must be such as to permit thoughtful factfinders to reach but one reasoned conclusion. See Wagenmann v. Adams, 829 F.2d 196, 200 (1st Cir.1987). In making this inquiry, an appellate court “may not consider the credibility of witnesses, resolve conflicts in testimony, or evaluate the weight of the evidence.” Id. Placing these tenets together with the Mclsaac rule, see supra at p. 836, we perceive our present task as limited to reviewing the record and the jury’s findings to determine whether there is any rational construction of the evidence which suggests that, although CNA may have improvidently refused to settle Scott Peck-ham’s claim early on, such bad faith was not necessarily the proximate cause of the excess judgment against Tripp. Phrased differently, is there a view of the evidence which can harmonize the jury’s answers to questions No. 1 and No. 3? Before reaching the heartland, we must decide what facts were properly to be considered. On this score, plaintiffs urge us to disregard entirely the expert testimony which CNA presented. We do not believe that course is open to us. We have said so often that it scarcely bears repeating that: “A trial judge has wide discretion anent the admission and exclusion of expert testimony, and his decisions are to be sustained unless his discretion has been abused.” Allied Int’l, Inc. v. International Longshoremen’s Ass’n, 814 F.2d 32, 40 (1st Cir.), cert. denied, 484 U.S. 820, 108 S.Ct. 79, 98 L.Ed.2d 41 (1987); see also United States v. Ladd, 885 F.2d 954, 959 (1st Cir.1989); Freeman v. Package Machinery Co., 865 F.2d 1331, 1338 (1st Cir.1988); United States v. Hoffman, 832 F.2d 1299, 1310 (1st Cir.1987); Marshall, 828 F.2d at 851. In this case, defendant presented two well-credentialed experts— attorneys versed in the nuances of insurance law — who offered opinion evidence as to proximate cause. Insurance is a complicated subject and the industry, over time, has developed a patina of custom and usage. Arcana abound. Defendant’s proffered experts could reasonably be expected to shed some light in a shadowy domain. Indeed, plaintiffs started this particular ball rolling by presenting expert testimony in their case in chief, and one of their experts, Ashley, gave an opinion on causation. Below, plaintiffs did not challenge either the generic propriety of expert testimony or the qualification of CNA’s witnesses. Their sole objection was on the ground that the matter of causation was “the ultimate issue” for jury resolution and thus unfit for expert testimony. That objection directly contradicted Fed.R.Evid. 704(a), which stipulates that, with exceptions not pertinent here, “testimony in the form of an opinion or inference otherwise admissible is not objectionable because it embraces an ultimate issue to be decided by the trier *838 of fact.” We are satisfied that the district court acted within its proper province in admitting the challenged testimony over plaintiffs’ voiced objection. B. The Central Question. With this preface, we need not linger long in examining whether the evidence on causation was so overwhelming as to eliminate any need for a jury to pass upon the question. In the first place, two experts (out of three who opined on the issue) said categorically that proximate cause was lacking. This evidence, backed by concomitant explanations, in and of itself seems sufficient to preserve the issue for the venire. Despite plaintiffs’ sarcastic assault upon the experts’ opinions, it was for the jury to determine what weight and value, if any, should be assigned to them. In the second place, the evidence, viewed as needs must in the light most congenial to CNA, admits of several possible reasons why the excess judgment against Tripp might not be causally related to bad faith on the insurer’s part. We cite some (but not necessarily all) of the possibilities: 1. The jury could reasonably have believed that CNA breached its duty to Tripp by failing to offer a Bilodeau agreement earlier than March 1984, 5 but that this failure had no causal effect because, given Corey’s intransigence, he would have spurned such a proposal (as indeed happened when the carrier tendered a Bilo-deau agreement in March 1984). 2. The jury could have believed that CNA’s bad faith refusal to settle Scott Peckham’s claim occurred in the winter of 1983-84 and stemmed from its failure to acknowledge the possibility of JoAnne Peckham’s entitlement to an independent application of the “per person” limits. The causal chain could have snapped, however, because CNA made a curative offer in March 1984, when it proposed entering into a Bilodeau agreement — and it was never proved that Corey told the Peckhams about that proposal. The jury could, therefore, have believed that the Peckhams, if apprised, would have accepted the agreement, thus ending the matter. On this theory, and on certain of the scenarios which follow, it must be remembered that where an independent, intervening event contributes in some unexpected and significant manner to the harm which eventuates, a defendant’s conduct cannot be the legal cause of the injury. Restatement (Second) of Torts § 442 (1965). The proper inquiry is whether the defendant is to be held liable for an injury to which the defendant has in fact made a substantial contribution when a later cause of independent origin, for which the defendant is not responsible, actually brings about the harm. W. Keeton, Prosser & Keeton on Torts, § 44, at 301 (5th ed. 1984). In this connection, an intervening cause sufficient to insulate the original actor from liability is one which comes into active operation in producing the result after the defendant’s actionable conduct has occurred. In other words, the first actor is to be held liable if, but only if, the intervening cause is foreseeable. Id. at 302. In this instance, for example, the jury could have visualized Corey’s failure to inform his clients as an unforeseeable intervening act which became the efficient cause of the ensuing harm and insulated CNA from liability for its earlier laggardness. 3.The jury could have concluded that CNA’s bad faith consisted of its failure to settle Scott Peckham’s claim alone for $27,-000 in February 1984. But, a prudent insurer would not have consummated such a settlement without the insured’s assent (for under the prevailing policy interpretation, using the whole $27,000 to assuage Scott would have left the insured completely unprotected as to JoAnne Peckham’s claim). Even if CNA, in good faith, should have pressed to settle at that time and for that amount, the jury could well have believed, based on Tripp’s dealings with Roberts later that year, that Tripp’s assent would not have been forthcoming. If that were true, *839 the carrier’s bad faith refusal might be stripped of the requisite causal effect. 4. An insurer’s obligation to exercise good faith toward its insured does not require it to roll over and play dead vis-a-vis the claimants; nor does the looming of an excess judgment debar an insurer from employing conventional negotiating stratagems in good faith (although the carrier uses such tactics at greater risk where its limits are inadequate). Negotiation is an art, not a science. The jury could have believed that CNA fully intended to settle Scott Peckham’s claim for the full policy limit; that it exhibited bad faith in failing to do so at the very beginning; but it was not reasonably foreseeable that, in response to a fairly prompt offer of a Bilo-deau agreement — and one was made here — the plaintiffs would refuse to settle. 5. In much the same vein, the jury could plausibly have found that it was not reasonably foreseeable that, in response to an immediate post-Bilodeau offer of the full policy limits — as was made here — the plaintiffs would refuse to settle. 6. There was evidence that, before the case against Tripp went to trial, Corey proposed an agreement whereby the Peck-hams would not look to Tripp to satisfy any judgment rendered. There was also evidence that Tripp accepted the offer prior to trial. We think that the jury could reasonably have inferred that a bargain was struck and that the ensuing trial, leading to the excess judgment, was not really adversarial, that is, Tripp was unconcerned with the trial’s outcome since he would not bear any financial onus. Fair play is a two-way street. Even if an insurer breaches its obligation of good faith, it is not then at the mercy of an insured’s infidelity. On these peculiar facts, the jury was free to believe that Corey’s backroom deal with Tripp, not CNA’s bad faith, was the competent producing cause of the excess judgment. We need go no further. On appeal, plaintiffs have challenged only the legal effect of the evidence on causation, not its weight. 6 But as we have shown, the matter was not cut and dried. We recognize that some of the described possibilities are relatively remote — but they remain possibilities. We cannot reject possibilities rooted in the record merely because, if sitting as factfinders, we would likely have drawn a different set of conclusions. In sum, there was no warrant to withhold the issue of proximate cause from the jury — particularly since the plaintiffs had the burden of proving it. See Swift, 866 F.2d at 509; Michnik-Zilberman v. Gordon’s Liquor, Inc., 390 Mass. 6, 453 N.E.2d 430, 433-34 (1983); Irwin, 467 N.E.2d at 1305. V. THE BENCH TRIAL In the second phase of the trial, the statutory claims were tried to, and decided by, the court. All parties express dissatisfaction with the outcome. A. The Violation. The applicable Massachusetts consumer protection statute prohibits “unfair or deceptive acts or practices in the conduct of any trade or commerce....” Mass. Gen.L. ch. 93A, § 2(a) (1985). Commonwealth law contains a similar ban against such conduct in the insurance industry. See Mass.Gen.L. ch. 176D, § 2 (1987). That prohibition extends to “unfair claim settlement practices,” which the statute defines as including “[f]ail[ure] to effectuate prompt, fair and equitable settlements of claims in which liability has become reasonably clear.” Id. at § 3(9)(f). Those injured by insurance practices proscribed under chapter 176D may sue under chapter 93A. *840 Equitable Life Assurance Society of the United States v. Porter-Englehart, 867 F.2d 79, 88 (1st Cir.1989); Whyte v. Connecticut Mut. Life Ins. Co., 818 F.2d 1005, 1010-11 (1st Cir.1987); Van Dyke v. St. Paul Fire & Marine Ins. Co., 388 Mass. 671, 448 N.E.2d 357, 360 (1983). The district court's reasoning, as expressed in a lengthy, unpublished rescript replete with findings of fact, is the starting point for our analysis. The court expressly found that “CNA’s interpretation of coverage for loss of consortium prior to Bilo-deau was a plausible interpretation of the policy made in good faith.” The court also found that “CNA did not act in bad faith in managing [the policy] proceeds in the face of two liability claims, each of which was likely to exceed the insured’s coverage.” In the court’s view, CNA was guilty of bad faith in only a single respect: not “keeping its insured Tripp fully informed of settlement offers and ... failing to advise him of the pendency and implications of Bilodeau, between the filing of the Peckhams’ suit against Tripp on April 19, 1984, and the SJC’s decision in Bilodeau on July 17, 1984_” The court went on to find, however, that CNA’s bad faith during this interval “was not the cause of Tripp’s exposure to the excess judgment.” Absent a mistake of law — and we see none here — we review these findings of fact only for clear error under Fed.R.Civ.P. 52(a). This means, of course, that: If the district court’s account of the evidence is plausible in light of the record viewed in its entirety, the court of appeals may not reverse it even though convinced that had it been sitting as the trier of fact, it would have weighed the evidence differently. Where there are two permissible views of the evidence, the factfinder’s choice between them cannot be clearly erroneous. Anderson v. City of Bessemer City, 470 U.S. 564, 573-74, 105 S.Ct. 1504, 1511, 84 L.Ed.2d 518 (1985); see also Anderson v. Cryovac, 862 F.2d at 916. We have read the entire record with care and see no material error. In part, defendant’s appeal challenges the finding that a chapter 93A violation occurred. For reasons which we believe are self-evident, we deem the challenge meritless and reject it out of hand. The carrier — whatever else it did right — ignored Tripp (although not necessarily his interests) during the embryonic stages of negotiation. The steps which it took to enlighten and inform Tripp were well short of a textbook model. That conduct, at the trier’s discretion, was fairly characterizable as constituting an unfair claim settlement practice. After all, chapter 176D decrees that an insurer cannot “[m]isrepresent[ ] pertinent facts or insurance policy provisions relating to coverages at issue.” Mass.Gen.L. ch. 176D, § 3(9)(a). In such a context, silence in the face of a duty to speak can be as harmful to an insured’s interests as outright prevarication. Accordingly, good faith requires that an insurer keep its insured informed of facts material to his exposure, including information relating to coverage and settlement negotiations. See, e.g., Kivi v. Nationwide Mut. Ins. Co., 695 F.2d 1285, 1288 (11th Cir.1983) (good faith obligates insurer to tell insured of settlement opportunities); Young v. American Casualty Co., 416 F.2d 906, 910 (2d Cir.1969) (under New York law, finding of insurer’s bad faith may be based on failure to inform insured of offers in compromise); Murach, 158 N.E.2d at 342 (nondisclosure “may be some evidence of the insurer’s lack of good faith”); see also Ward v. State Farm Mut. Auto Ins. Co., 539 F.2d 1044, 1049 (5th Cir.1976). In this situation, we cannot say that the trial court’s view of the evidence was clearly erroneous. From this point forward, the battle shifts to another front. Plaintiffs protest the district court’s conclusion that, while an unfair practice was proven, they were entitled only to nominal damages. We think the court could permissibly have concluded, as it did, that CNA’s failure to keep Tripp properly and seasonably informed was not the cause of the excess verdict. There is simply no cogent proof that, had Tripp known of Bilodeau a few months earlier or had he been kept better advised of the negotiations, matters would have turned out differently. There is nothing which *841 convincingly suggests, for instance, that if Tripp knew the full ramifications of the situation, he would have insisted that $27,-000 be offered outright to Scott Peckham, leaving JoAnne’s claim open; or that he would have committed additional funds to the settlement pot; or that Corey, at that point in time, was other than intransigent. And in the absence of cognizable harm, nominal damages were appropriate. See Carey v. Piphus, 435 U.S. 247, 266-67, 98 S.Ct. 1042, 1053-54, 55 L.Ed.2d 252 (1978); Maldonado Santiago v. Velazquez Garcia, 821 F.2d 822, 829 (1st Cir.1987); Aubin v. Fudala, 782 F.2d 280, 286 (1st Cir.1983). We need not repastinate ground already thoroughly ploughed by the district judge. We conclude unhesitatingly that the court below did not err in finding a statutory violation or in assessing only nominal damages in respect thereto. B. Attorneys’ Fees. We turn next to defendant’s contention that the district court abused its discretion in awarding plaintiffs $56,000 in attorneys’ fees in respect to Count V. 7 We start with fundamentals. The applicable statute provides that, if chapter 93A is transgressed, “the petitioner shall, in addition to other relief provided for by this [statute] and irrespective of the amount in controversy, be awarded reasonable attorneys’ fees and costs incurred in said action.” Mass.Gen.L. ch. 93A, § 11. Hence, the absence of a substantial recovery on a chapter 93A claim does not preclude awarding reasonable attorneys’ fees under the statute. See Shapiro v. Public Service Mut. Ins. Co., 19 Mass.App.Ct. 648, 477 N.E.2d 146, 153 (1985). All that need be shown to support a fee award is that the “defendant committed an unfair or deceptive act, [violative of the statute,] irrespective of whether that act caused a loss of money or property.” Id. Thus, plaintiffs, as Tripp’s assignees, were entitled to reasonable counsel fees. From that point on, the waters grow much murkier. In a diversity case, state rather than federal law controls the question of attorneys’ fees. Northern Heel Corp. v. Compo Indus., Inc., 851 F.2d 456, 475 (1st Cir.1988); Shelak v. White Motor Co., 636 F.2d 1069, 1072 (5th Cir. 1981). In such a situation, “state law also controls the method of calculating the size of the award.” Blanchette v. Cataldo, 734 F.2d 869, 878 (1st Cir.1984). While Massachusetts prescribes no pat formula for computation of fee-shifting awards, we deem it elementary that the amount due should represent “fair and reasonable compensation for the services rendered.” First National Bank v. Brink, 372 Mass. 257, 361 N.E.2d 406, 410 (1977). The amount of time expended is a relevant consideration, as are “the complexity of the litigation, the amount of work reflected by the record, ... and the quality of the services rendered.” Northern Heel, 851 F.2d at 476 (applying Massachusetts law); see also Linthicum v. Archambault, 379 Mass. 381, 398 N.E.2d 482, 488 (1979); McInerney v. Massasoit Greyhound Ass’n, 359 Mass. 339, 269 N.E.2d 211, 219 (1971). A fee-setting court may also consider “the amount and importance of the matter involved, ... the value of the property affected by the controversy, and the results secured.” Id. (quoting Cummings v. National Shawmut Bank, 284 Mass. 563, 188 N.E. 489, 492 (1934)). In a chapter 93A case, it would follow that a prevailing party would only be entitled to recover for additional fees caused by the breach of statutory duty. See Equitable Life, 867 F.2d at 91 (where chapter 93A violation proven, fees recoverable only for “any meaningful amount of legal work ... independently required” by the dereliction); Whyte, 818 F.2d at 1011 n. 20 (similar). The caselaw suggests that, if a plaintiff prevails on some, but not all, of his claims, and there is no interconnection, the fee award might be limited to the time spent regarding the proven chapter 93A violation. See, e.g., Hanner v. Classic *842 Auto Body, Inc., 10 Mass.App.Ct. 121, 406 N.E.2d 686, 688 (1980); cf. Wagenmann, 829 F.2d at 225 (under federal law, if unsuccessful claims are “sufficiently interconnected” with claims on which plaintiff prevails, fees may be awarded for all work reasonably undertaken). In the last analysis, the fee-shifting anodyne focuses on “what [counsel’s] services were objectively worth.” Heller v. Silverbranch Constr. Corp., 376 Mass. 621, 382 N.E.2d 1065, 1071 (1978). In the case at bar, the district court made a gross award of $56,000, without setting forth specific findings of fact or explicating its reasoning to any extent. We can only guess how the court may have reached the figure — and we are uneasy at the prospect. Appellate review of fee awards ordinarily requires that concrete findings be made and that the court below supply a clear explanation of the reasons undergirding a particular fee award. See Jacobs v. Mancuso, 825 F.2d 559, 560 (1st Cir.1987); see generally Kelley v. Everglades Drainage Dist., 319 U.S. 415, 422, 63 S.Ct. 1141, 1145, 87 L.Ed. 1485 (1943) (“there must be findings ... sufficient to indicate the factual basis for the ultimate conclusion”); In re Rare Coin Galleries of America, Inc., 862 F.2d 896, 900 (1st Cir.1988) (same). Except where the basis for a fee calculation is apparent on the record’s face — a circumstance not present here — the trier must at least offer, from the bench or in a brief memorandum, some indication of his thought processes and how he structured the award. While fee awards in chapter 93A cases are “largely discretionary,” Linthicum, 398 N.E.2d at 488, the trial court’s discretion is not unbridled. Cf, e.g., Independent Oil and Chemical Workers of Quincy, Inc. v. Procter & Gamble Mfg. Co., 864 F.2d 927, 929 (1st Cir.1988) (elucidating test for abuse of discretion). Under the circumstances, we have little choice but to vacate the fee award and remand for further findings. See, e.g., Pearson v. Fair, 808 F.2d 163, 165-66 (1st Cir.1986) (per curiam); Myers v. Gulf Oil Corp., 731 F.2d 281, 284 (5th Cir.1984); SquirtCo v. Seven-Up Co., 628 F.2d 1086, 1092 (8th Cir.1980); Complaint of Ithaca Corp., 582 F.2d 3, 4 (5th Cir.1978) (per curiam); Nickerson v. Travelers Ins. Co., 437 F.2d 113, 114 (5th Cir.1971) (per curiam); cf. Segal v. Gilbert Color Systems, Inc., 746 F.2d 78, 87 (1st Cir.1984) (remanding for reeomputation of fee award which, although reasonable on its face, was calculated in accordance with erroneous standard). We do not in any way suggest that busy trial courts must write polished opinions exhaustively explicating counsel fee awards. Cf. United States v. Metropolitan Dist. Comm’n., 847 F.2d 12, 16 (1st Cir.1988) (district courts need not “drown in a rising tide of fee-generated minutiae”). Nor do we suggest that the fees awarded in this case were reasonable or unreasonable, too large, too small, or just right. Rather, we confess that, given the many variables which enter into the fee-setting calculus, we are unable meaningfully to review the district court’s determination on the record before us. VI. CONCLUSION To recapitulate, we find the jury’s responses to the special questions to be reconcilable with each other and supportable on the basis of the evidence presented at trial. In particular — albeit reserving comment on the weight of the evidence, see supra note 6 — we find that the question of proximate cause was appropriately submitted for jury consideration. As to the second phase of the bifurcated trial, we see no clear error in the district court’s findings or conclusions as to liability or damages. The court’s determination that chapter 93A was infracted appears sustainable, as does its determination that no actual harm resulted therefrom and that only nominal damages should accrue. Lastly, although the record justifies the lower court’s decision to grant counsel fees on plaintiffs’ statutory claim, the award itself is so bereft of findings that it cannot stand. The case must, therefore, be returned to the district court to reconsider the fee award, refigure it if necessary, and elaborate upon the basis for it. *843 Accordingly, we affirm the judgment below in all respects except as to the quantum of attorneys’ fees; we vacate that award and remand for further proceedings consonant with this opinion. All parties shall bear their own costs on appeal. So ordered. APPENDIX [[Image here]] *844 [[Image here]] *845 [[Image here]] Footnotes [1] Because of sums paid by another alleged tort-feasor (the cafe where Tripp and Scott Peckham had been imbibing shortly before the accident), JoAnne's claim was fully satisfied and Tripp was exposed to no excess liability on her account. [2] The district court charged the jury in this regard: If the insured's coverage is needlessly exhausted on one claim when they might cancel out other claims as well, the insured suffers from the company's readiness to settle; thus, the insured's coverage limits should not be exhausted without an attempt to settle as many claims as possible. If the policy limits are so slight compared to the total of the injured person's claims as to rule out any possibility of using the policy limits to settle all of the injured person’s claims, the insurer's insistence upon settling all the claims might not benefit the insured. The insured might be better protected if the leverage of his insurance money is applied to at least some of the claims so as to reduce his ultimate judgment debt. Bearing in mind the existence of two claims and limited coverage and the insured’s exposure to heavy damages, the jury must determine whether the insurer acted in good faith in managing the insurance proceeds in a manner reasonably calculated to protect the insured by minimizing his total liability. [3] The rule stipulates: The court may require a jury to return only a special verdict in the form of a special written finding upon each issue of fact. In that event the court may submit to the jury written questions susceptible of categorical or other brief answer or may submit written forms of the several special findings which might properly be made under the pleadings and evidence; or it may use such other method of submitting the issues and requiring the written findings thereon as it deems most appropriate. The court shall give to the jury such explanation and instruction concerning the matter thus submitted as may be necessary to enable the jury to make its findings upon each issue. If in so doing the court omits any issue of fact raised by the pleadings or by the evidence, each party waives the right to a trial by jury of the issue so omitted unless before the jury retires the party demands its submission to the jury. As to an issue omitted without such demand the court may make a finding; or, if it fails to do so, it shall be deemed to have made a finding in accord with the judgment on the special verdict. Fed.R.Civ.P. 49(a). [4] For example, one of plaintiffs' experts, Ashley, was asked for an opinion on causation. Moreover, in objecting to opinion testimony from one of CNA’s experts, Meehan, plaintiffs’ counsel told the court; “I object to the opinion on proximate causation, Your Honor. Again, / believe that’s an issue that the jury should determine, and not Mr. Meehan.” (Emphasis supplied). [5] Effectuating a Bilodeau agreement would have entailed, among other things, paying Scott Peck-ham the full $27,000 which he originally demanded and securing a general release from him. [6] Plaintiffs made a motion for new trial following the special verdict. Such a motion, unlike a motion for a directed verdict or judgment n.o.v., allows the verdict-loser to test the weight of the evidence. See Wagenmann, 829 F.2d at 200-01; Hubbard v. Faros Fisheries, Inc., 626 F.2d 196, 200 (1st Cir.1980). The district court denied the new trial motion. In designating issues on appeal, plaintiffs eschewed any articulated challenge to this ruling. They have, therefore, waived appellate review of the weight of the evidence. Notwithstanding, plaintiffs preserved their objection to the lower court’s denial of their motion for partial directed verdict and judgment n.o.v. (which addressed the issue of causation); the sufficiency of the evidence, as opposed to the weight, is thus properly before us. [7] Although the fees awarded were less than 60% of what plaintiffs requested, plaintiffs have not asked us to review the amount for inadequacy.