In an appeal which sheds considerably more heat than light, the defendant-appellant, the Boston Edison Company (Edison), attempts to persuade us that a money judgment which eventuated against it following a nonjury trial in the United States District Court for the District of Massachusetts was improvidently rendered. We are not persuaded.
1. BACKGROUND
Edison is a utility company which, in 1981, needed to accomplish certain block wall modifications at one of its generating facilities, Pilgrim Nuclear (Pilgrim), in Plymouth, Massachusetts. Edison prepared specifications for the work and requested bids on a time-and-materials basis. Several prospective vendors responded. One such bidder was the plaintiff-appellee, RCI Northeast Services Division (RCI). In a transmittal letter, RCI described its quotation as a “cost plus percentage fee” proposal. It submitted a series of enclosures which set forth fee schedules, a rundown of shop facilities, labor costs, and equipment charges. On the page dealing with hourly and overtime rates for the crafts, the following language appeared:
Labor cost rates include all costs, burdens, insurances and taxes applicable, based on current labor rates and are subject to escalation.
Eventually, Edison decided to let the contract to RCI. Though the reasons why the defendant selected plaintiff’s bid are somewhat blurred — the different formats used by those who responded to the solicitation make cross-comparison a chancy exercise— it is clear that the (relatively modest) markup was a factor. When matters came down to the wire, Edison chose to forgo preparation of an integrated contract document. Instead, it issued a series of purchase orders which declared that the work was to be performed “on a cost plus fee basis.” Billings were to be “in accordance with the [RCI] proposal on file with Edison.” That proposal, of course, included the above-quoted escalation clause.
RCI performed the work in 1981-82, and submitted periodic billings — billings which the appellee, at least, viewed as estimates. These were paid. In December 1983, almost eighteen months after the work was completed, the other shoe dropped: RCI billed Edison to the tune of $185,535, claiming that it had underrecovered its applicable workers’ compensation insurance costs. The power company, shocked, refused payment. This suit followed.
II. ISSUE PRESENTED
There is, in the last analysis, but a solitary substantive issue in this case. Nevertheless, in order to achieve a balanced perspective, it is necessary first to examine the derivation of the increased insurance costs.
The workers’ compensation coverage carried by RCI referable to the Pilgrim job was written retrospectively as to rate. Under this sort of arrangement, an estimated premium was deposited when the coverage took effect, and actual premium costs were determined after the fact, based on a formula embodying the employer’s accident experience for the project. The payment was thereafter adjusted to reflect the recal *201 culation. When so determined, the insurer would apply and bill increments to the premiums retrospectively for the years in question (or, if the experience was more favorable than anticipated, refund any overcharges). Typically, the actual “experience” is not ascertainable for months or even years after the work has ended (when all of the job-related injuries can be fully assessed and the extent of the anticipated payments intelligently computed). Often, an appreciable interval elapses before the true insurance cost becomes known.
There was evidence that this type of policy format was not uncommon in the industry. Indeed, RCI had been engaged under similar cost-plus contracts for other utility companies. The plaintiffs practice, it was said, was to make an appropriate adjustment via a credit or request for additional payment, depending on the experience and the premium fluctuation, after receipt of the final premium cost tabulation. In this case, some eighteen months down the road, RCI was confronted with a retrospective premium increase of $185,535 attributable to the Pilgrim project. Based on the clause quoted above, the appellee submitted its adjustment claim to Edison. 1 But, the receptacle was not live.
The issue in this case is a straightforward one. From RCI’s standpoint, the contract documents specifically protected it against mounting insurance costs, thus shifting the burden of the retrospective workers’ compensation premium hike to Edison. The defendant reads the same language quite differently: the base labor rate alone was subject to change, and the associated “burdens,” including compensation insurance, would fluctuate only in direct proportion to, and in the same percentage as, the base labor rate itself. In short, Edison maintains that the contract price was not meant to change in accordance with increases in raw insurance costs.
III. PROCEEDINGS BELOW
When the parties came to loggerheads, RCI (a New York corporation) invoked federal diversity jurisdiction, 28 U.S.C. § 1332(a), and sued Edison in the district court. As originally pleaded and postured, the suit involved several related claims and counterclaims of RCI and Edison, each against the other, but all of them have been dropped or settled and need not be further remarked. What remains is the premium overrun, as to which Edison denied liability. Eventually, the matter was submitted to the district court as a case stated, on a paper record consisting of the pleadings, various affidavits and depositions, and sundry documentary exhibits.
After briefing and oral argument, the district court issued its memorandum of decision.
RCI Northeast Services Division v. Boston Edison Co.,
IV. STANDARD OF REVIEW
In this case, the parties stipulated that submission on the agreed record would empower the district court to decide the controversy as if “a full evidentiary hearing with live testimony ... produced a record essentially identical to the one [proffered].” Edison asserts, nevertheless, that we have here a pure question of law, subject to plenary review, without recourse to Fed.R. Civ.P. 52(a) and the “clearly erroneous” rule contained therein. 2 The appellant ar *202 gues .first, that since the decision below rests on an interpretation of a written document, it is not factbound; second, inasmuch as the dispute turned on a silent record (i.e., no live witnesses testified), the district court’s findings are entitled to no special deference; and third, that the district court operated under a misapprehension as to the law, thereby undermining any findings of fact. Such static need not occupy us for long.
To be sure, it is for the court to determine if a writing is ambiguous.
See Edmonds v. United States,
In this instance, the district court navigated the shoals with commendable circumspection. Although it plainly believed the escalation clause unequivocally to cover RCI’s extra insurance costs,
RCI I,
at 1179, it treated the matter as if the contract were subject to a latent ambiguity on this score. The court proceeded to examine the dealings of the parties, and found that they “never intended that any escalation in plaintiff’s costs would have to be absorbed by plaintiff____”
Id.
This last determination was, on any reasoned view of the fact/law dichotomy, sufficiently factbound to slip comfortably within the integument of Rule 52(a).
See Boston Five Cents Savings Bank v. Secretary of the Dep’t of HUD,
With these principles in mind, Edison’s position becomes completely untenable. To the extent that there is but one “natural” reading of the disputed clause, it favors RCI. The language, on its face, is unelectrifying; it appears, simply put, to make “insurances ... subject to escalation.” Conversely, to the extent — if at all — that there is some latent ambiguity (in the Edmonds phrase, more than one “plausible definition”), the district court’s divination of the parties’ discovered intentions and thus, the clause’s meaning, can be set aside only if shown to be clearly erroneous.
This conclusion is not weakened in any way by the appellant’s remaining pair of arguments. It is by now settled beyond peradventure that findings of fact do not forfeit “clearly erroneous” deference merely because they stem from a paper record.
See Anderson v. City of Bessemer City,
In
Boston Five Cents Savings Bank,
we observed that “an argument between parties about the meaning of a contract is typically an argument about a ‘material fact,’ namely, the factual meaning of the contract.”
V. THE MERITS
We start our discussion of the merits with the disputed sentence itself. We repeat it here for ease in reference:
Labor cost rates include all costs, burdens, insurances and taxes applicable, based on current labor rates and are subject to escalation.
The least forced reading of the language is to the effect that “labor cost rates” include a variety of components — “costs, burdens, insurances and taxes” — and that these components, as well as the labor rates themselves, “are subject to escalation.” Whatever may be said, pro and con, as to whether this is the
only
reasonable construction of the clause, it is surely a plausible interpretation. And Massachusetts law — which governs in this diversity action — favors the construction of terms “according to their ordinary meaning.”
Thomas v. Hartford Accident & Indemnity Co.,
Once the switch of inquiry is thrown, the case is made the clearer. It is hornbook law that,
If the language of the contract is susceptible of more than one interpretation, the court should construe the contract in the light of the situation and relation of the parties at the time it was made, and, if possible, accord it a reasonable and sensible meaning, consonant with its dominant purpose.
Continental Bus System, Inc. v. NLRB,
The intention of the parties is to be gathered, not from [a] single sentence ..., but from the whole instrument read in the light of the circumstances existing at the time of negotiations leading up to its execution.
Miller v. Robertson,
Here, it was Edison that determined to let the contract on the basis of the vendor’s costs, supplemented by a reasonable profit. It was Edison which, in its purchase orders, stated that RCI was to perform the work “on a cost plus fee basis.” The notion that RCI should absorb increases in its insurance costs for the job, rather than pass such increases along to the owner, is at odds with the “dominant *204 purpose” of the cost-plus arrangement. The appellant points out, correctly, that cost-plus contracts can come in varying shapes and sizes, and can be tailored to suit the specific needs of any set of contracting parties. That is certainly true — but it is of marginal relevance on the record before us. Although it would have been possible for the parties to have limited the pass-through of insurance expense in some artificial way, there is nothing in the language of the contract, the relationship of the parties, or the situation as a whole which suggests such was the case. Defining the contours of the escalator in terms of real cost increases rather than in the palpably obscure manner suggested by the appellant fits much more comfortably into the everyday context of the deal.
In the face of this remorseless logic, Edison lights up the sky with a barrage of postulates. The asseveration that the disputed phrase must be construed unfavorably to RCI, as the author, has already blown a fuse, see supra n. 3, and need not be rehashed. Most of the remaining surges are of low voltage and do not require discussion. We will, however, reflect briefly upon four of them.
First, the appellant’s reliance upon its purchasing agent’s professed understanding of RCI’s proposal is unilluminating in view of the factfinder’s refusal to credit that self-serving interpretation. As we have recently noted, “... contracts depend on objective manifestations of consent and not on uncommunicated subjective expectations.”
Bushkin Associates, Inc. v. Raytheon Co.,
Nor does the mere fact that the contract was competitively bid make a difference. That Edison, when comparing competing quotations preliminary to an award of the work, regarded RCI’s labor costs as more or less fixed cannot defeat either the express language of the contract or the legitimate rights of the appellee. This is especially true in a competitive bidding situation such as was undertaken in respect to Pilgrim, a situation wherein Edison — as the owner — had (but chose not to exercise) the power to insist that proposals be made on uniform terms or in a standardized format.
The argument that the risk of fluctuating insurance prices was avoidable— that the workers’ compensation coverage could have been procured for a flat premium — proves perhaps too much. Although the premise is true, the conclusion which Edison would have us draw does not follow. After all, the appellant controlled the solicitation process. If escalation of insurance costs was of concern, it would have been a simple matter to insist that bidders, generally, eschew variable experience-rated policies in favor of guaranteed fixed-dollar premiums. Alternatively, Edison could have conditioned the purchase orders which it issued to RCI in that way. The defendant, of course, elected neither option. It is unbefitting that we accomplish by judicial fiat what Edison neglected to achieve contractually.
Cf New England Structures, Inc. v. Loranger,
Edison also contends that RCI’s billing practices, as illustrated by its worksheets, are consistent with the appellant’s view of the disputed proviso. We readily acknowledge that the conduct of the parties in the course of performing a contract can illumine the meaning of the words employed. See Restatement (Second) of Contracts § 212 comment b (1979). Yet, we have no cause to untangle the filament which Edison’s counsel has so ingeniously *205 spun from the historical record of the progress payments and the underlying worksheets. At best, the inference urged by the appellant is a permissive one which could have been drawn from the interim billings. Be that as it may, the district judge was free to accept, instead, the deposition testimony of RCI’s president, who brushed off these calculations as mere “approximations.” That sort of choice — the choice between conflicting (plausible) inferences — is precisely the sort of judgment call that Rule 52(a) insulates from appellate tampering.
VI. CONCLUSION
There would be scant utility in any further analysis. In the arena of commerce, it avails us little to stand language on its ear in an effort to rescue a firm from a sinkhole of its own design. It is no appropriate part of judicial business to rewrite contracts freely entered into between sophisticated business entities. Rather, the courts must give effect to the language of such agreements and to their discernible meaning. That is exactly what has occurred: the district judge adopted the most natural reading of the disputed sentence, ascertained that the parties intended the clause to operate in precisely that way, and decided the case accordingly. His finding that, under the purchase orders, RCI was entitled to recoup from Edison the augmented costs attributable to the retrospective insurance premium increases enjoys adequate record support.
In sum, we hold that Fed.R.Civ.P. 52(a) enswathes our inquiry on this appeal, and that the decision below was not clearly erroneous. We are left well short of any conviction — or even suspicion — that a mistake has been made. This means, of course, that the lights go out for Edison’s appeal, and that the judgment entered by the district court must be
Affirmed.
Notes
. Edison has expressed considerable skepticism as to whether or not, if the estimates had proven more generous than the actual earned premiums, RCI would have spontaneously issued a credit or volunteered a refund. We shall never know. We do know, however, that the appellant eschews any question concerning the accuracy of the $185,535 figure. It contests only whether or not it is contractually obligated to pay the amount.
. Fed R.Civ.P. 52(a) provides in pertinent part:
In all actions tried upon the facts without a jury ... the court shall find the facts specially *202 and state separately its conclusions of law thereon____ Findings of fact, whether based on oral or documentary evidence, shall not be set aside unless clearly erroneous, and due regard shall be given to the opportunity of the trial court to judge of the credibility of the witnesses.
. Edison argues that the district court should have construed the "doubtful” language of the escalation clause against the appellee, as the draftsperson. Passing the fact that this is a fall-back rule, to be applied "only in the last resort,”
Aldrich
v.
Bay State Construction Co.,
