Full opinion text
MEMORANDUM OPINION AND ORDER TEITELBAUM, District Judge. Plaintiff, Aluminum Company of America (ALCOA), brought the instant action against defendant, Essex Group, Inc. (Essex), in three counts. The first count requests the Court to reform or equitably adjust an agreement entitled the Molten Metal Agreement entered into between ALCOA and Essex. The second count alleges that the Molten Metal Agreement was modified by oral amendment and that Essex has breached the amended agreement. The second count seeks a declaratory judgment that the alleged breach by Essex excuses ALCOA’s further performance and seeks as well an award of damages caused by the alleged breach of Essex. The third count asks for a declaratory judgment that ALCOA’s prior notice of termination of the Molten Metal Agreement was proper or, in the alternative, that ALCOA may terminate the Molten Metal Agreement if it be determined by this Court to be a contract for the sale of goods. Essex denies all of ALCOA’s material allegations. Essex further counterclaims that ALCOA is liable to it for damages based on ALCOA’s failure to deliver to Essex the amounts of molten metal ALCOA is contractually obligated to deliver under the Molten Metal Agreement and seeks entry of an order specifically enforcing its right to receive molten aluminum from ALCOA in the amounts requested. Jurisdiction is based upon diversity of citizenship and amount in controversy and is one of the few issues in the case sub judice not in dispute. In 1966 Essex made a policy decision to expand its participation in the manufacture of aluminum wire products. Thus, beginning in the spring of 1967, ALCOA and Essex negotiated with each other for the purpose of reaching an agreement whereby ALCOA would supply Essex with its long-term needs for aluminum that Essex could use in its manufacturing operations. By December 26, 1967 the parties had entered into what they designated as a toll conversion service contract known as the Molten Metal Agreement under which Essex would supply ALCOA with alumina which ALCOA would convert by a smelting process into molten aluminum. Under the terms of the Molten Metal Agreement, Essex delivers alumina to ALCOA which ALCOA smelts (or toll converts) into molten aluminum at its Warrick, Indiana, smelting facility. Essex then picks up the molten aluminum for further processing. The price provisions of the contract contained an escalation formula which indicates that $.03 per pound of the original price escalates in accordance with changes in the Wholesale Price Index-Industrial Commodities (WPI) and $.03 per pound escalates in accordance with an index based on the average hourly labor rates paid to ALCOA employees at the Warrick plant. The portion of the pricing formula which is in issue in this case under counts one and two is the production charge which is escalated by the WPI. ALCOA contends that this charge was intended by the parties to reflect actual changes in the cost of the non-labor items utilized by ALCOA in the production of aluminum from alumina at its Warrick, Indiana smelting plant. In count one of this suit ALCOA asserts that the WPI used in the Molten Metal Agreement was in fact incapable of reasonably reflecting changes in the non-labor costs at ALCOA’s Warrick, Indiana smelting plant and has in fact failed to so reflect such changes. It is ALCOA’s contention in count one of its complaint that the shared objectives of the parties with respect to the use of the WPI have been completely and totally frustrated, that both ALCOA and Essex made a mutual mistake of fact in agreeing to use the WPI to escalate non-labor costs at Warrick. ALCOA is seeking reformation or equitable adjustment of the Molten Metal Agreement so that pursuant to count one of its complaint, the pricing formula with respect to the non-labor portion of the production charge will be changed to eliminate the WPI and substitute the actual costs incurred by ALCOA for the non-labor items used at its Warrick, Indiana smelting plant. Essex opposes relief under count one contending that: 1) ALCOA cannot obtain reformation of the Molten Metal Agreement on the grounds of mutual mistake since ALCOA has failed to establish any antecedent agreement on pricing not expressed in the Molten Metal Agreement; 2) ALCOA assumed the risk that its prediction as to future costs would be incorrect; 3) ALCOA has failed to prove that enforcement of the Molten Metal Agreement would be unconscionable. ALCOA alleges in the second count of its complaint that when it became evident that the WPI was not accomplishing the objectives of the parties under the Molten Metal Agreement, discussions were begun between ALCOA and Essex which culminated in a meeting held between Mr. Krome George, Chief Executive Officer of ALCOA, and others of ALCOA and Mr. Paul O’Malley, President of Essex, on July 21, 1975. At that time, Mr. George and Mr. O’Malley allegedly orally agreed to reform the Molten Metal Agreement to reflect the original objectives of the parties. The oral agreement allegedly replaced the WPI with the actual costs incurred by ALCOA. Essex denies entering into the alleged oral agreement and has accordingly refused to perform consistent with its terms. In its second count, ALCOA requests that declaratory judgment be entered whereby Essex’s breach of its alleged oral agreement to reform the Molten Metal Agreement be declared sufficient grounds to excuse any further performance by ALCOA of that Agreement. In addition, ALCOA seeks that it be awarded damages in excess of $11,900,000 accruing to the date of judgment resulting from Essex’s breach, plus interests and costs. ALCOA asks in its third count that it be excused from further performance of the Molten Metal Agreement. ALCOA alleges that its performance is excused by a clause which is contained in a document referred to as the December 27, 1967 Letter Agreement (the Side Letter Agreement). That clause provides that ALCOA and Essex, acting in good faith, entered into the Molten Metal Agreement with the understanding that it was a contract for the furnishing of services by ALCOA to Essex. The clause further provides that in the event a final decision of a court construed the Molten Metal Agreement as a contract for the sale of goods it could be terminated by either party. The Side Letter Agreement was a product of concern that an admittedly preferential price to Essex would threaten a violation of the RobinsonPatman Act if the various transactions could be lumped together and considered to be in substance the sale of aluminum rather than what appears as a matter of form, the sale of services. ALCOA argues it should be permitted to terminate the Molten Metal Agreement under the terms of the Side Letter Agreement. ALCOA urges that this Court should determine whether the Molten Metal Agreement is a contract for the sale of goods. As previously indicated, Essex has filed a counterclaim to the ALCOA complaint. The original counterclaim of Essex contends that under the terms of the Molten Metal Agreement as implemented during the years 1977, 1978 and the first six months of 1979, ALCOA has, on numerous occasions, breached the Molten Metal Agreement by improperly failing to deliver the amounts of molten aluminum required by the contract. This first counterclaim asks that Essex be awarded damages in an amount as to fully compensate it for the failure of ALCOA to deliver molten aluminum in accordance with the terms of the Molten Metal Agreement. The amended counterclaim of Essex arises as a result of a letter dated June 4, 1979, in which ALCOA informed Essex that it was reducing by 15% the amount of its deliveries of molten aluminum requested by Essex. ALCOA claims to have this authority under the terms of the Molten Metal Agreement. Essex contends that ALCOA does not have any such authority and its amended counterclaim additionally asks for an order enforcing the Molten Metal Agreement and awarding damages accordingly. Simply put, if possible, ALCOA seeks relief from this Court in a three count complaint, while Essex opposes ALCOA’s requests and itself seeks relief via a counterclaim. The Court finds, based upon consideration of all the evidence, that ALCOA is entitled to reformation of the Molten Metal Agreement. At the same time, ALCOA’s requests for relief in counts two and three are denied as is the request for relief by Essex in its counterclaim. COUNT ONE ALCOA’s first count seeks an equitable modification of the contract price for its services. The pleadings, arguments and briefs frame the issue in several forms. ALCOA seeks reformation or modification of the price on the basis of mutual mistake of fact, unilateral mistake of fact, unconscionability, frustration of purpose, and commercial impracticability. A. The facts pertinent to count one are few and simple. In 1967 ALCOA and Essex entered into a written contract in which ALCOA promised to convert specified amounts of alumina supplied by Essex into aluminum for Essex. The service is to be performed at the ALCOA works at War-rick, Indiana. The contract is to run until the end of 1983. Essex has the option to extend it until the end of 1988. The price for each pound of aluminum converted is calculated by a complex formula which includes three variable components based on specific indices. The initial contract price was set at fifteen cents per pound, computed as follows: A. Demand Charge $0.05/lb. B. Production Charge (i) Fixed component ' ,04/lb. (ii) Non-labor production cost component .03/lb. (iii) Labor production cost component .03/lb. Total initial charge $0.15/lb. The demand charge is to vary from its initial base in direct proportion to periodic changes in the Engineering News Record Construction Cost-20 Cities Average Index published in the Engineering News Record. The Non-labor Production Cost Component is to vary from its initial base in direct proportion to periodic changes in the Wholesale Price Index-Industrial Commodities (WPI-IC) published by the Bureau of Labor Statistics of the United States Department of Labor. The Labor Production Cost Component is to vary from its initial base in direct proportion to periodic changes in ALCOA’s average hourly labor cost at the Warrick, Indiana works. The adjusted price is subject to an over-all “cap” price of 65% of the price of a.specified type of aluminum sold on specified terms, as published in a trade journal, American Metal Market. The indexing system was evolved by ALCOA with the aid of the eminent economist Alan Greenspan. ALCOA examined the non-labor production cost component to assure that the WPI-IC had not tended to deviate markedly from their non-labor cost experience in the years before the contract was executed. Essex agreed to the contract including the index provisions after an examination of the past record of the indices revealed an acceptable pattern of stability- ALCOA sought, by the indexed price agreement, to achieve a stable net income of about 4$ per pound of aluminum converted. This net income represented ALCOA’s return (i) on its substantial capital investment devoted to the performance of the contracted services, (ii) on its management, and (iii) on the risks of short-falls or losses it undertook over an extended period. The fact that the non-labor production cost component of ALCOA’s costs was priced according to a surrogate, objective index opened the door to a foreseeable fluctuation of ALCOA’s return due to deviations between ALCOA’s costs and the performance of the WPI-IC. The range of foreseeable deviation was roughly three cents per pound. That is to say that in some years ALCOA’s return might foreseeably (and did, in fact) rise to seven cents per pound, while in other years it might foreseeably (and did, in fact) fall to about one cent per pound. See Table I. Essex sought to assure itself of a long term supply of aluminum at a favorable price. Essex intended to and did manufacture a new line of aluminum wire products. The long term supply of aluminum was important to assure Essex of the steady use of its expensive machinery. A steady production stream was vital to preserve the market position it sought to establish. The favorable price was important to allow Essex to compete with firms like ALCOA which produced the aluminum and manufactured aluminum wire products in an efficient, integrated operation. In the early years of the contract, the price formula yielded prices related, within the foreseeable range of deviation, to ALCOA’s cost figures. Beginning in 1973, OPEC actions to increase oil prices and unanticipated pollution control costs greatly increased ALCOA’s electricity costs. Electric power is the principal non-labor cost factor in aluminum conversion, and the electric power rates rose much more rapidly than did the WPI-IC. As a result, ALCOA’s production costs rose greatly and unforeseeably beyond the indexed increase in the contract price. Table I illustrates the relation between the WPI-IC and ALCOA’s costs over the years of the contract, and the resulting consequences for ALCOA: During the most recent years, the market price of aluminum has increased even faster than the production costs. At the trial ALCOA introduced the deposition of Mr. Wilfred Jones, an Essex employee whose duties included the sale of surplus metal. Mr. Jones stated that Essex had resold some millions of pounds of aluminum which ALCOA had refined. The cost of the aluminum to Essex (including the purchase price of the alumina and its transportation) was 36.35 cents per pound around June of 1979. Mr. Jones further stated that the resale price in June 1979 at one cent per pound under the market, was 73.313 cents per pound, yielding Essex a gross profit of 37.043 cents per pound. This margin of profit shows the tremendous advantage Essex enjoys under the contract as it is written and as both parties have performed it. A significant fraction of Essex’s advantage is directly attributable to the corresponding out of pocket losses ALCOA suffers. ALCOA has sufficiently shown that without judicial relief or economic changes which are not presently foreseeable, it stands to lose in excess of $75,000,000 out of pocket, during the remaining term of the contract. B. ALCOA’s Warrick Works, located in Indiana, are the designated source of supply. The Essex plant, where the bulk of the aluminum is used, is also located in Indiana. Essex takes delivery at the Warrick Works. The contract declares “This Agreement shall be governed and interpreted in accordance with the laws of the State of Indiana.” The parties surely have sufficient contacts with the State of Indiana that Pennsylvania courts would enforce their agreement respecting the application of Indiana law. Restatement 2d Conflict of Laws § 187; cf. 13 Pa.C.S.A. § 1105(1) (U.C.C.). This Court must enforce it as well. Klaxon Co. v. Stentor Electric Mfg. Co., 313 U.S. 487, 61 S.Ct. 1020, 85 L.Ed. 1477 (1941). This case presents many issues which are governed by common law principies. Most fall within the interstices of the reported decisions of Indiana courts. Some touch principles announced in hoary Indiana decisions. Where the Indiana law remains undeclared, or where the declaration is far from current, the obligation of this Court is to discern the most probable state of current Indiana law for “the outcome of the litigation in the federal court should be substantially the same, so far as legal rules determine the outcome of a litigation, as it would be if tried in a State court.” Guaranty Trust Co. v. York, 326 U.S. 99, 65 S.Ct. 1464, 89 L.Ed. 2079 (1945); see Commissioner v. Estate of Bosch, 387 U.S. 456, 465, 87 S.Ct. 1776, 1782, 18 L.Ed.2d 886, 893-94 (1967); Bernhardt v. Polygraphic Co., 350 U.S. 198, 76 S.Ct. 273, 100 L.Ed. 199 (1956); McKenna v. Ortho Pharmaceutical Corp., 622 F.2d 657 (3rd Cir. 1980). In connection with these observations, the Court notes that the appellate courts of Indiana appear to join in the habits of thought and in the assessments of policy which have lately prevailed in most of the fine courts in this nation. The Indiana courts have joined the throng of state courts in (i) declaring that residential landlords are bound by an implied warranty of habitability. Old Town Development v. Langford, 349 N.E.2d 744 (Ind.App.1976); (ii) adopting the rule of strict products liability from the Restatement 2d of Torts § 402A, Perfection Paint & Color Co. v. Konduris, 258 N.E.2d 681 (Ind.App.1970); (iii) adopting the increasingly prevalent view that harsh or unconscionable provisions in contracts of adhesion may be refused enforcement, Weaver v. American Oil Co., 276 N.E.2d 144 (Ind.1971). C. ALCOA initially argues that it is entitled to relief on the theory of mutual mistake. ALCOA contends that both parties were mistaken in their estimate of the suitability of the WPI-IC as an objective index of ALCOA’s non-labor production costs, and that their mistake is legally sufficient to warrant modification or avoidance of ALCOA’s promise. Essex appropriately raised several defenses to these claims. Essex first argues that the asserted mistake is legally insufficient because it is essentially a mistake as to future economic events rather than a mistake of fact. Essex next argues that ALCOA assumed or bore the risk of the mistake. Essex finally argues that the requested remedy of reformation is not available under Indiana law. The late Professor Corbin wrote the best modern analysis of the doctrine of mutual mistake. Corbin took pains to show the great number and variety of factors which must be considered in resolving claims for relief founded on the doctrine of mistake, and to show the inappropriateness of any single verbal rule to govern the decision of mistake cases. Corbin on Contracts § 597 at 582 -83 (1960). The present case involves a claimed mistake in the price indexing formula. This is clearly a mistake concerning a factor affecting the value of the agreed exchange. Of such mistakes Corbin concluded that the law must consider the character of the risks assumed by the parties. Id. at § 605. He further concluded: In these cases, the decision involves a judgment as to the materiality of the alleged factor, and as to whether the parties made a definite assumption that it existed and made their agreement in the belief that there was no risk with respect to it. Opinions are almost sure to differ on both of these matters, so that decisions must be, or appear to be, conflicting. The court’s judgment on each of them is a judgment on a matter of fact, not a judgment as to law. No rule of thumb should be constructed for cases of this kind. 3 Corbin on Contracts § 605 at p. 643 (1960). The new Restatement 2d of Contracts follows a similar approach. After defining “mistake” as “a belief not in accord with the facts,” § 293, the Restatement declares: § 294. WHEN MISTAKE OF BOTH PARTIES MAKES A CONTRACT VOIDABLE. (1) Where a mistake of both parties at the time a contract was made as to a basic assumption on which the contract was made has a material effect on the agreed exchange of performances the contract is voidable by the adversely affected party unless he bears the risk of the mistake under the rule stated in § 296. (2) In determining whether this mistake has a material affect on the agreed exchange of performances, account is taken of any relief by way of reformation, restitution, or otherwise. Both Professor Corbin and the Restatement emphasize the limited place of the doctrine of mistake in the law of contracts. They, along with most modern commentators, emphasize the importance of contracts as devices to allocate the risks of life’s uncertainties, particularly economic uncertainties. Where parties to a contract deliberately and expressly undertake to allocate the risk of loss attendant on those uncertainties between themselves or where they enter a contract of a customary kind which by common understanding, sense, and legal doctrine has the affect of allocating such risks, the commentators and the opinions are agreed that there is little room for judicial relief from resulting losses. Corbin on Contracts § 598 and authorities there cited. The new Restatement agrees, § 296. This is, in part, the function of the doctrine of assumption of the risk as a limitation of the doctrine of mistake. Whether ALCOA assumed the risk it seeks relief from is at issue in this case. The doctrine of assumption of the risk is therefore considered below. The important point to note here is that the doctrine of assumption of the risk is not the only risk allocating limitation on the doctrine of mistake. Other important risk allocating limitations are inherent in the doctrine of mistake itself. They find expression in the cases and treatises in declarations that there has been no mistake, or no legally cognizable mistake, or a mistake of the wrong sort. ALCOA claims that there was a mutual mistake about the suitability of the WPI— IC as an index to accomplish the purposes of the parties. Essex replies that the mistake, if any, was not a mistake of fact, but it was rather a mistake in predicting future economic conditions. Essex asserts that such a mistake does not justify legal relief for ALCOA. The conflicting claims require the Court to resolve three questions: (1) Was the mistake one of “fact” as the cases and commentators use that word? (2) If so, was it of the sort of fact for which relief could be granted? (3) If the mistake was not one of “fact”, is relief necessarily foreclosed? The initial question requires the characterization, as a matter of fact rather than of law, of the claimed mistake. The cases and commentaries contain useful thoughts and analagous problems which aid in this characterization. But the characterization is itself a question of fact. That it may have ultimate legal significance, and that it requires the exercise of judgment does not distinguish this determination from other determinations of fact. The distinction between questions of law and questions of fact is old. Its resolution is often doubtful. No simple and mechanical verbal formula can capture the distinction and resolve the hard cases. Factors affecting the characterization of a question include its suitability for jury or other fact finder determination and its analytical separability from the final determination of legal consequences. The separation of fact from things which are not fact-opinion, prediction, desire, and the like-is principally a question of common sense or epistemology rather than of law, even when the separation must be done by courts. So it is here. ALCOA calls the mistaken assumption that the index was suitable a factual assumption. Essex calls it a prediction. This is a dispute of facts, not law. Its resolution will affect the decision of this case as factual determinations usually do. The law must be applied to it to yield a result. Neither is this question beyond the usual function and capacity of a jury or other fact finder. The first restatement of Contracts notes, and the published Tentative Draft No. 10 of the Restatement Second stresses, the distinction between “existing fact” and prediction. See Restatement of Contracts § 502, comment a; Restatement 2d of Contracts § 293. The approved final form of § 293 modifies the emphasis by deleting the word “existing”. The Reporter, Professor Farnsworth of the Columbia University School of Law, related the circumstances of that change to the Court when he appeared on behalf of ALCOA. Q Professor Farnsworth, is there any expression or trend expressed in the new sections on mistake that you have been describing that provide any help in trying to determine the age-old problem of what is an existing fact? A There has been a change, I don’t know how much help it offers. Let me simply tell you the history of it. In 1975 in May at the annual meeting, the material on mistake was presented, and it led off with a Section 293 which defined the mistake as a belief that is not in accordance with existing facts. That section was I think not the subject of any discussion, at least not the subject of any discussion involving the word ‘existing’ in that year. The following year there was presented a chapter that is not in the documents that are here in court, but a chapter dealing with, among other things, misrepresentation, and in defining a misrepresentation, there was also a comparable phrase that what you had to misrepresent was an existing fact. At that meeting, there was one speaker who thought that the word ‘existing’ was not desirable and should be deleted. I confess I have reread the speaker’s statement or the transcript of it, and I can provide it if you want it, but I don’t find it clear enough that it was helpful to me. I did not think that was a desirable change. I think partly because the preceding year when we had considered ‘mistake,’ the Institute had approved ‘existing facts,’ and I felt rather bound by that earlier decision. At this point I lose any record in the transcript. My clear recollection is that following the discussion of misrepresentation, a number of people came up to me and later saw me in the hallway and said that they agreed with the speaker that ‘existing’ should be dropped. It would be fair to say that there were probably as many reasons for dropping it given to me as there were people who had advanced the opinion. I would suppose at the end at least a dozen people had said they didn’t like ‘existing,’ and nobody had defended it. The reporter has the authority to change even the black letter when it is a matter of style, and since I did not bring it back to the Institute for a vote as a matter of substance, I think one would have to say that any change made was considered by the reporter to be a matter of style. In any event, in response to the small but unanimous body of opinion that didn’t like ‘existing,’ it was deleted in the draft that I finally sent off to the editor and it now reads ‘Belief that is not in accord with the facts.’ I think that there is in the comment still a statement with respect to ‘existing,’ but the deletion from the black let ter is at least a change that perhaps permits more flexibility with respect to the line between what is an existing fact or what is a fact and what is a pure presumption which is an extremely difficult line to draw in both cases. Testimony of E. Allan Farnsworth 20-22 (Emphasis added). The Comment Professor Farnsworth mentioned declares: [T]he erroneous belief must relate to the facts as they exist at the time of the making of the contract. A party’s prediction or judgment as to events to occur in the future, even if erroneous, is not a ‘mistake’ as that word is defined here. Corbin and Williston agree in passing, though their analysis emphasizes the various subjects of mistake and the circumstances which influence risk allocation. Cf. Corbin on Contracts § 598; Williston on Contracts § 1541 at 67 (3rd ed. [Jaeger] 1970). Both cite cases declaring erroneous prediction to be beyond the reach of the doctrine of mutual mistake. See Corbin on Contracts § 598; Williston on Contracts § 1543A. The Court finds the parties’ mistake in this case to be one of fact rather than one of simple prediction of future events. Plainly the mistake is not wholly isolated from predictions of the future or from the searching illuminations of painful hindsight. But this is not the legal test. At the time the contract was made both parties were aware that the future was unknown, and their agreed contract was intended to bind them for many years to come. Both knew that Essex sought an objective pricing formula and that ALCOA sought a formula which would cover its out of pocket costs over the years and which would yield it a return of around four cents a pound. Both parties to the contract carefully examined the past performance of the WPI-IC before agreeing to its use. The testimony was clear that each assumed the Index was adequate to fulfill its purpose. This mistaken assumption was essentially a present actuarial error. The parties took pains to avoid the full risk of future economic changes when they embarked on a twenty-one year contract involving services worth hundreds of millions of dollars. To this end they employed a customary business risk limiting device-price indexing-with more than customary sophistication and care. They chose not a single index formula but a complex one with three separate indices. Their care to limit the risk of the future distinguishes this case from cases like Leasco Corporation v. Taussig, 473 F.2d 777 (2nd Cir. 1972). In Leasco the plaintiff corporation contracted to sell to Taussig a subsidiary which engaged in civil engineering and consulting. Taussig was, at the time, the vice-president of the subsidiary. The parties fixed the sales price by capitalizing the anticipated $200,000 earnings of the subsidiary in the year of the sale. Both parties knew that the subsidiary’s earnings were volatile. “The civil engineering business is personalized, highly technical, and extremely risky.” Id. at 781. But the parties made no provision to limit their risk. In fact the projected $200,000 earnings turned into a $12,000 loss due to a design error in a construction project. The court held that this loss did not make the contract voidable: “[W]e hold that there was no mutual mistake. Both Taussig and Leasco may have hoped, but surely could not have been certain, that [the subsidiary] would earn $200,000 in fiscal 1971.. . . Neither party could safely assume the projected earnings would be realized.” Id. The “fact” which led to the dispute was a prediction. The court characterized the situation as one where the parties realized there was doubt about an important fact and assumed, or more accurately placed on the purchaser, the risk of its existence. See Restatement of Contracts § 502, comment f. The Taussig decision rested on two legs: the absence of a mistake of “fact” and the assumption of the risk of future uncertainties. The decision was plainly correct. The parties bottomed their agreement on a naked prediction without the protection of conditions or limitations. The only protection for the parties lay in practical matters. Both were well familiar with the business. And the contract was a short term one for the outright sale of the business. In the short term the parties might think they could sensibly risk uncertainties without specific contractual limits. Having made no attempt to limit future uncertainties, the disappointed purchaser could point to no fact which existed when the contract was made which would justify an award of judicial relief. The contrast between Taussig and the present case is striking. Here the practical necessities of the very long term service contract demanded an agreed risk limiting device. Both parties understood this and adopted one. The capacity of their selected device to achieve the known purposes of the parties was not simply a matter of acknowledged uncertainty like the Taussig prediction. It was more in the nature of an actuarial prediction of the outside limits of variation in the relation between two variable figures-the WPI-IC and the non-labor production costs of ALCOA. Its capacity to work as the parties expected it to work was a matter of fact, existing at the time they made the contract. This crucial fact was not known, and was scarcely knowable when the contract was made. But this does not alter its status as an existing fact. The law of mistake has not distinguished between facts which are unknown but presently knowable, e. g. Raffles v. Wichelhaus, 2 H. & C. 906 (1864), and facts which presently exist but are unknowable, e. g. Sherwood v. Walker, 66 Mich. 568, 33 N.W. 919 (1887). Relief has been granted for mistakes of both kinds. To conclude that the parties contracted upon a mistake of fact does not, by itself, justify an award of judicial relief to ALCOA. Relief can only follow if the mistake was mutual, if it related to a basic assumption underlying the contract, and if it caused a severe imbalance in the agreed exchange. The doctrine of mistake has long distinguished claims of mutual mistake from claims of unilateral mistake. Corbin on Contracts § 608. The standards for judicial relief are higher where the proven mistake is unilateral than where it is mutual. Compare, e. g. Restatement 2d of Contracts § 294 with § 295. Essex asserts that ALCOA’s mistake was unilateral. Mr. O’Malley, Chairman of the Board of Essex Corporation, testified at trial that he had no particular concern for ALCOA’s well-being and that in the negotiations of the contract he sought only Essex’s best interests. Essex claims this testimony tends to rebut any possible mutual mistake of fact between the parties. The Court disagrees. The cases clearly establish that mutual mistake lies in error concerning mutually understood material facts. Leasco Corp. v. Taussig, supra; Baumann v. Florance, 267 App.Div. 113, 114, 44 N.Y.S.2d 706 (3d Dept. 1943). The law of mutual mistake is not addressed primarily to motivation or to desire to have a good bargain, such as that credibly testified to by Mr. O’Malley. As Mr. O’Malley struck the bargain for Essex, he understood the function of the Wholesale Price Index, as part of the pricing formula, to be the protection of ALCOA from foreseeable economic fluctuations. He further had every reason to believe that the formula was selected on the factual prediction that it would, within tolerable limits, serve its purpose. While he did not share the motive to protect ALCOA, he understood the functional purposes of the agreement. He therefore shared this mistake of fact. And his mistake was Essex’s. The Court recognizes that Mr. O’Malley and Essex would cheerfully live with the benefit of their mistake, but the law provides otherwise. As a matter of law Mr. O’Malley’s testimony of Essex’s indifference concerning ALCOA’s motivation for the use of the Wholesale Price Index as a gauge for tracking non-labor costs is immaterial. Is it enough that one party is indifferent to avoid a mutual mistake? The Court thinks not. This situation resembles that in Sherwood v. Walker, supra, the celebrated case of Rose of Aberlone. There the owner of a prize breeding cow sold her for slaughter at the going rate for good slaughter cattle. The owner had unsuccessfully tried to breed her and had erroneously concluded she was sterile. In fact she was pregnant at the time of the sale and she was much more valuable for breeding than for slaughter. There as here, the buyer was indifferent to the unknown fact; he would have been pleased to keep the unexpected profit. But he understood the bargain rested on a presumed state of facts. The court let the seller avoid the contract because of mutual mistake of fact. In Sherwood, the buyer didn’t know the highly pedigreed Rose was with calf. He probably could not have discovered it at the time of the sale with due diligence. Here the parties could not possibly have known of the sudden inability of the Wholesale Price Index to reflect ALCOA’s non-labor costs. If, over the previous twenty years, the Wholesale Price Index had tracked, within a 5% variation, pertinent costs to ALCOA, a 500% variation of costs to Index must be deemed to be unforeseeable, within any meaningful sense of the word. Essex has not seriously argued that the mistake does not relate to an assumption which is basic to the contract. The relation is clear. The assumed capacity of the price formula in a long term service contract to protect against vast windfall profits to one party and vast windfall losses to the other is so clearly basic to the agreement as to repel dispute. While the cases often assert that a mistake as to price or as to future market conditions will not justify relief, this is not because price assumptions are not basic to the contracts. Instead, relief is denied because the parties allocated the risk of present price uncertainties or of uncertain future market values by their contract. Where a “price mistake” derives from a mistake about the nature or quantity of an object sold, the courts have allowed a remedy; they have found no contractual allocation of that sort of risk of price error. Indiana cases hold that where land is sold as a tract for a set price, and it later appears that there was a material error in the parties’ estimate of the quantity of land conveyed, the court will correct the error by adjusting the price, McMahan v. Terkhorn, 67 Ind.App. 501, 116 N.E. 327 (1917), or by allowing rescission, Earl v. VaNatta, 29 Ind. App. 532, 64 N.E. 901 (1902). See Corbin on Contracts §§ 604-05. Similarly many cases allow relief from unilateral price errors by construction contractors. An Indiana decision reached this result. Board of School Comm’rs v. Bender, 36 Ind.App. 164, 72 N.E. 154 (1904). See Corbin on Contracts § 609. Restatement 2d of Contracts § 295, comment a. These cases demonstrate that price assumptions may be basic to the contract. Essex concedes that the result of the mistake has a material effect on the contract and that it has produced a severe imbalance in the bargain. See Restatement 2d of Contracts § 294, comment c. The most that Essex argues is this: ALCOA has not proved that enforcement of the contract would be unconscionable. Essex correctly points out that at the time of the trial ALCOA had shown a net profit of $9 million on the contract. Essex further argues that ALCOA has failed to prove that it ever will lose money on the contract, and that such proof would require expert testimony concerning future economic values and costs. These arguments are insufficient. The evidence shows that during the last three years ALCOA has suffered increasingly large out of pocket losses. If the contract were to expire today that net profit of $9 million would raise doubts concerning the materiality of the parties’ mistake. But even on that supposition, the court would find the mistake to be material because it would leave ALCOA dramatically short of the minimum return of one cent per pound which the parties had contemplated. But the contract will not expire today. Essex has the power to keep it in force until 1988. The Court rejects Essex’s objection to the absence of expert testimony concerning future costs and prices. The objection is essentially based on the traditional refusal of courts to award speculative damages. But Essex presses the argument too far. The law often requires courts to make awards to redress anticipated losses. The reports are filled with tort and contract cases where such awards are made without the benefit of expert testimony concerning future economic trends. Awards are commonly denied because they are too speculative where there is a claim for lost future profits and there is insufficient evidence of present profits to form a basis for protecting future profits. Similarly the courts often decline to speculate concerning future economic trends in calculating awards for lost future earnings. Many states refuse to consider any possibility of future inflation in calculating such awards despite the presence of expert testimony and the teachings of common experience. Compare Havens v. Tonner, 243 Pa. Super. 371, 365 A.2d 1271 (1976) and Vizzini v. Ford Motor Co., 569 F.2d 754, 768 (3rd Cir. 1977), with Feldman v. Allegheny Airlines, 382 F.Supp. 1271 (D.Conn.1974), aff’d in part, rev’d in part, 524 F.2d 384 (2nd Cir. 1975). This demonstrates the law’s healthy skepticism concerning the reliability of expert predictions of economic trends. Where future predictions are necessary, the law commonly accepts and applies a prediction that the future economy will be much like the present (except that inflation will cease). Since some prediction of the future is inescapable in this case, that commonly accepted one will necessarily apply. On that prediction, ALCOA has proved that over the entire life of the contract it will lose, out of pocket, in excess of $60 million, and the whole of this loss will be matched by an equal windfall profit to Essex. This proof clearly establishes that the mistake had the required material effect on the agreed exchange. Indeed, if this case required a determination of the conscionability of enforcing this contract in the current circumstances, the Court would not hesitate to hold it unconscionable. Essex next argues that ALCOA may not be relieved of the consequences of the mistake because it assumed or bore the risk of the market. Essex relies on both Restatements and on a variety of cases fairly typified by Leasco Corp. v. Taussig, supra; McNamara Const. of Manitoba Ltd. v. United States, 509 F.2d 1166 (Ct.Cl.1963), and Flippin Materials Co. v. United States, 312 F.2d 408 (Ct.Cl.1963). McNamara Construction involved a fixed price construction contract which was upset by extensive labor strife. The Court of Claims denied the contractor relief from its extra costs, holding, inter alia, that the contractor had been aware of the risk of labor trouble when it entered the contract; that the contract expressly provided for delay caused by strikes; and that in the absence of some further express provision the contractor bore the risk of the cost of strikes and labor cost increases. The court declared: “What we have in this case is a risk which is known to both parties and results from human inability to predict the future. The authorities are unanimous in distinguishing such risks from bona fide mutual mistakes of fact.” Id. at 1168. In Flippin Materials Co. v. United States, supra, the plaintiff sought relief from unexpected costs it suffered in a contract to quarry limestone from a mountain and to process it into aggregate. The costs derived from the presence of unexpected amounts of waste material at the quarry site. The Court of Claims held that the plaintiff could not claim relief on a mutual mistake theory because the contract clearly put the risk of unknown subsurface conditions on the plaintiff. The contract offer made the government’s geological findings available to the bidders and it further provided: GC-2 SITE INVESTIGATION AND REPRESENTATIONS: The contractor acknowledges that he has satisfied himself as to the nature and location of the work, the general and local conditions, particularly those bearing upon ... conformation and condition of the ground, the character, quality and quantity of surface and subsurface materials to be encountered . . . and all other matters which can in any way affect the work or the cost thereof under this contract. Any failure by the contractor to acquaint himself with all the available information concerning these conditions will not relieve him from responsibility for estimating properly the difficulty or cost of successfully performing the work. Id. at 415. The court declared that relief for mutual mistake is not available “if the contract puts the risk of such a mistake on the party asking reformation ... or normally if the other party, though made aware of the correct facts, would not have agreed at the outset to the change.. .. ” Id. The Restatements and these cases reveal four facets of risk assumption and risk allocation under the law of mistake. First, a party to a contract may expressly assume a risk. If a contractor agrees to purchase and to remove 114,000 cubic yards of fill from a designated tract for the landowner at a set price “regardless of subsurface soil and water conditions” the contractor assumes the risk that subsurface water may make the removal unexpectedly expensive. Customary dealing in a trade or common understanding may lead a court to impose a risk on a party where the contract is silent. Often the result corresponds to the expectation of both parties, but this will not always be true. See Berman, Excuse for Nonperformance in the Light of Contract Practices in International Trade, 63 Colum.L.Rev. 1413, 1420 24 (1963). At times legal rules may form the basis for the inferred common understanding. Equity traditionally put the risk of casualty losses on the purchaser of land while the purchase contract remained executory. This allocation was derived from the doctrine of equitable conversion. “Equity regards as done that which ought to be done.” The rule could always be modified by express agreement. It survives today where it does survive largely by reason of its acceptance as part of the common expectations of real estate traders and their advisors. Third, where neither express words nor some particular common understanding or trade usage dictate a result, the court must allocate the risk in some reasoned way. Two examples from the Restatement 2d of Contracts § 296 illustrate the principle. A farmer who contracts to sell land may not escape the obligation if minerals are discovered which make the land more valuable. And in the case of the sale of fill stated above, if there is no express assumption of the risk of adverse conditions by the contractor, he may still bear the risk of losing his expected profits and suffering some out of pocket losses if some of the fill lies beneath the water table. These cases rest on policies of high generality. Contracts are-generally-to be enforced. Land sales are-generally-to be treated as final. Fourth, where parties enter a contract in a state of conscious ignorance of the facts, they are deemed to risk the burden of having the facts turn out to be adverse, within very broad limits. Each party takes a calculated gamble in such a contract. Because information is often troublesome or costly to obtain, the law does not seek to discourage such contracts. Thus if parties agree to sell and purchase a stone which both know may be glass or diamond at a price which in some way reflects their uncertainty, the contract is enforceable whether the stone is in fact glass or diamond. If, by contrast, the parties both mistakenly believe it to be glass, the case is said not to be one of conscious ignorance but one of mutual mistake. Consequently the vendor may void the contract. In this case Essex raises two arguments. First, it asserts that ALCOA expressly or by fair implication assumed the risk that the WPI-IC would not keep up with ALCOA’s non-labor production costs. Second, it asserts that the parties made a calculated gamble with full awareness that the future was uncertain, so the contract should be enforced despite the mutual mistake. Both arguments are correct within limits, and within those limits they affect the relief ALCOA may receive. Both arguments fail as complete defenses to ALCOA’s claim. Essex first asserts that ALCOA expressly or implicitly assumed the risk that the WPI-IC would not track ALCOA’s non-labor production costs. Essex asserts that ALCOA drafted the index provision; that it did so on the basis of its superior knowledge of its cost experience at the Warrick Works; and that ALCOA’s officials knew of the inherent risk that the index would not reflect cost changes. Essex emphasizes that, during the negotiation of the contract, it insisted on the inclusion of a protective “ceiling” on the indexed price of ALCOA’s services at 65% of a specified published market price. Essex implies that ALCOA could have sought a corresponding “floor” provision to limit its risks. Essex’ arguments rely on two ancient and powerful principles of interpretation. The first is reflected in the maxim “expressio unius est exclusio alterius.” The second is the principle that a contract is to be construed against its drafter. To agree to an indexed price term subject to a ceiling but without a floor is to make a deliberate choice, Essex argues. It is to choose one principle and to reject another. The argument is plausible but not sufficient. The maxim rules no farther than its reason, and its reason is simply this: often an expression of a rule couched in one form reflects with high probability the rejection of a contradictory rule. Less often it reflects a probable rejection of a supplementary rule. To know if this is true of a particular case requires a scrupulous examination of the thing expressed, the thing not expressed, and the context of the expression. The question here is precisely this: By omitting a floor provision did ALCOA accept the risk of any and every deviation of the selected index from its costs, no matter how great or how highly improbable? The course of dealing between the parties repels the idea. Essex and ALCOA are huge industrial enterprises. The management of each is highly trained and highly responsible. The corporate officers have access to and use professional personnel including lawyers, accountants, economists and engineers. The contract was drafted by sophisticated, responsible businessmen who were intensely conscious of the risks inherent in long term contracts and who plainly sought to limit the risks of their undertaking. The parties’ laudable attention to risk limitation appears in many ways: in the complex price formula, in the 65% ceiling, in the “most favored cüstomer” clause which Essex wrote into the contract, and in the elaborate “force majeur” clause favoring ALCOA. It appears as well in the care and in the expense of the negotiations and drafting process. Essex negotiated with several aluminum producers, seeking a long term assured supply, before agreeing to the ALCOA contract. Its search for an assured long term supply for its aluminum product plants itself bespeaks a motive of limiting risks. Essex settled on ALCOA’s offer rather than a proffered joint venture on the basis of many considerations including the required capital, engineering and management demands of the joint venture, the cost, and the comparative risks and burdens of the two arrangements. When ALCOA proposed the price formula which appears in the contract, Essex’s management examined the past behavior of the indices for stability to assure they would not cause their final aluminum cost to deviate unacceptably from the going market rate. ALCOA’s management was equally attentive to risk limitation. They went so far as to retain the noted economist Dr. Alan Greenspan as a consultant to advise them on the drafting of an objective pricing formula. They selected the WPI-IC as a pricing element for this long term contract only after they assured themselves that it had closely tracked ALCOA’s non-labor production costs for many years in the past and was highly likely to continue to do so in the future. In the context of the formation of the contract, it is untenable to argue that ALCOA implicitly or expressly assumed a limitless, if highly improbable, risk. On this record, the absence of an express floor limitation can only be understood to imply that the parties deemed the risk too remote and their meaning too clear to trifle with additional negotiation and drafting. The principle that a writing is to be construed against its maker will not aid Essex here. That principle once sounded as a clarion call to retrograde courts to pervert agreements if they could. Today it is happily domesticated as a rule with diverse uses. In cases involving issues of conscience or of strong policy, such as forfeiture cases, the principle complements the familiar doctrine of strict construction to favor lenient results. In other cases it serves as an aid in resolving otherwise intractable ambiguities. This case presents neither of these problems. The question of defining the risks ALCOA assumed is one of interpretation. It implicates no strong public policy. Neither does it present an intractable ambiguity. Neither is this a case of “conscious ignorance” as Essex argues. Essex cites many cases which establish the general rule that mistaken assumptions about the future are not the sort of mistaken assumptions which lead to relief from contractual duties. Leasco Corp. v. Taussig, supra, is typical of these cases. The general rule is in fact as Essex states it. But that rule has limited application. The new Restatement notes that the rule does not apply where both parties are unconscious of their ignorance-that is, where both mistakenly believe they know the vital facts. See § 296 Comment C. Compare White v. Stelloh, 74 Wis. 435, 43 N.W. 99 (1889) and Backus v. MacLaury, 278 App.Div. 504, 106 N.Y.S.2d 401 (1951) with Sherwood v. Walker, supra. See Corbin on Contracts §§ 598, 605. This distinction is sufficient to settle many cases but it is framed too crudely for sensible application to cases like the present one. The distinction posits two polar positions: certain belief that a vital fact is true and certain recognition that a vital fact is unknown. Such certainties are seldom encountered in human affairs. They are particularly rare in the understanding of sophisticated businessmen. In Taussig the parties anticipated the subsidiary would earn $200,000 in the year of the sale. Had anyone asked them whether they were certain of that prediction, they would surely have answered that such predictions are always made with the recognition of a range of uncertainty. The prediction indicates that the parties believe there is a good chance that the earnings will lie between $175,000 and $225,000 and a very high probability that they will lie between $100,000 and $300,000. If pressed, the parties might agree that there could be a new loss for the year. But they would regard a loss as very highly unlikely. Of course predictions of future earnings must be viewed skeptically because the people who make them are often vitally interested in their contents and in their uses. The same notion of a range of uncertainty is not unknown to Indiana law. In McMahan v. Terkhorn, supra, the parties contracted to purchase and to sell a tract of land which they thought to contain 133 acres for $15,000. Before the date for performance the purchaser had the land surveyed. The surveyor reported the tract contained 104.52 acres. The parties then adjusted the purchase price to $12,000 and completed the conveyance on that basis. Later the vendor learned the survey was wrong; the tract contained 129 acres. He then sued for and won the value of the “excess” land conveyed. The court distinguished between the normal range of survey error which parties are deemed to expect and to risk, and gross errors for which a remedy is available. Accord, Harrison v. Talbot, 32 Ky. 258, 2 Dana 258 (Ky.1834); Nelson v. Matthews, 12 Va. 164 (1808); Quesnel v. Woodlief, 10 Va. (5 Call.) 218 (1796). Once courts recognize that supposed specific values lie, and are commonly understood to lie, within a penumbra of uncertainty, and that the range of probability is subject to estimation, the principle of conscious uncertainty requires reformulation. The proper question is not simply whether the parties to a contract were conscious of uncertainty with respect to a vital fact, but whether they believed that uncertainty was effectively limited within a designated range so that they would deem outcomes beyond that range to be highly unlikely. In this case the answer is clear. Both parties knew that the use of an objective price index injected a limited range of uncertainty into their projected return on the contract. Both had every reason to predict that the likely range of variation would not exceed three cents per pound. That is to say both would have deemed deviations yielding ALCOA less of a return on its investment, work and risk of less than one cent a pound or of more than seven cents a pound to be highly unlikely. Both consciously undertook a closely calculated risk rather than a limitless one. Their mistake concerning its calculation is thus fundamentally unlike the limitless conscious undertaking of an unknown risk which Essex now posits. What has been said to this point suffices to establish that ALCOA is entitled to some form of relief due to mutual mistake of fact. But the stakes in this case are large, and the chances of review by higher courts are high. Therefore the Court thinks it appropriate to rule on two other theories which ALCOA presented in support of its first count. D. ALCOA argues that it is entitled to relief on the grounds of impracticability and frustration of purpose. The Court agrees. In broad outline the doctrines of impracticability and of frustration of purpose resemble the doctrine of mistake. All three doctrines discharge an obligor from his duty to perform a contract where a failure of a basic assumption of the parties produces a grave failure of the equivalence of value of the exchange to the parties. And all three are qualified by the same notions of risk assumption and allocation. The doctrine of mistake of fact requires that the mistake relate to a basic assumption on which the contract was made. The doctrine of impracticability requires that the non-occurrence of the “event”, Restatement 2d of Contracts § 281, or the non-existence of the “fact”, Id. § 286, causing the impracticability be a basic assumption on which the contract is made. The doctrine of frustration of purpose similarly rests on the same “non-occurrence” or “non-existence”, “basic assumption” equation. Id. §§ 285, 286. The three doctrines further overlap in time. There may be some residual notion that the doctrine of frustration and impracticability relate to occurrences after the execution of the contract while the doctrine of mistake relates to facts as they stand at the time of execution. But that view has never won general acceptance. The first Restatement does not specifically limit the mistakes of fact for which relief may be granted to facts existing at the time of the contract. §§ 500, 502. Corbin and Williston do not suggest such a limitation. And the new Restatement equivocates on the point. Section 293 defines “mistake” as “a belief that is not in accord with the facts.” The word “existing” modified the word “facts” in Tentative Draft Number 10 but was deleted by the Reporter. Comment a to the section does declare: [T]he erroneous belief must relate to the facts as they exist at the time of the making of the contract. A party’s prediction or judgment as to events to occur in the future, even if erroneous, is not a ‘mistake’ as that word is defined here. This declaration is anomalous and unexplained. The Court believes the definition rather than the comment expresses the better rule. The denial of relief for mistakes of future facts is better understood to rest on policies of risk allocation discussed above than to rest on the definition of “mistake.” National Presto Industries, Inc. v. United States, 338 F.2d 99 (Ct.Cl.1964), is a prime example of the application of the doctrine of mistake to developments after the execution of the contract. There the corporation contracted to produce artillery shells for the Army at a fixed price, using a new and only partially proven production method. The method was contrived to reduce wasted steel by eliminating the need for shaving excess metal from the shells. After the contract was signed, the corporation spent large sums of money in unsuccessful attempts to make the method work. Eventually it became clear that some shaving would be required. The corporation purchased the necessary equipment and paid for the materials and labor. Then it sought and obtained relief in the Court of Claims for its added expense. The court held that there had been an actionable mutual mistake of fact. The assumed capacity of the new method to produce shells without a shaving step proved to be mistaken. The court based its decision solely on mistake of fact. It does not appear that frustration or impracticability were considered. Conversely the notion that the doctrines of frustration and impracticability apply only to events occurring after the execution of a contract appear to be drawn more from common experience with their application than from any inherent limitation of t