Trist v. FIRST FEDERAL S. & L. ASS'N OF CHESTER

*580 OPINION JOSEPH S. LORD, III, Chief Judge. This is a suit under the Truth in Lending Act [TILA] and the Sherman Act against certain savings and loan associations in the Greater Philadelphia area that made home mortgage loans to plaintiffs. The antitrust count, being maintained as a class action against twenty banks, 1 alleges that defendants violated Section 1 of the Sherman Act, 15 U.S.C. § 1, by agreeing between 1968 and 1975 to collect an “interest charge” at settlement on a loan. The TILA count, certified as a class action as to three defendants and maintained as individual actions against three others 2 , alleges that the “interest charge” collected at settlement was mislabeled in violation of the disclosure requirements of 15 U.S.C. § 1601 et seq. and Regulation Z promulgated thereunder, 12 CFR 226.1 et seq. Defendants have moved for summary judgment on both counts pursuant to F.R.Civ.P. 56. We have jurisdiction under 28 U.S.C. § 1331. I. THE PRACTICE: Home mortgage loan funds usually are distributed to the borrower at closing or settlement and applied by him to pay a portion of the purchase price. Commonly, the title company conducting the settlement collects certain charges from the mortgagor and remits them to the lender. Among the charges collected, from plaintiffs, for the twenty defendants in this case was a sum equal to interest on the loan from the date of settlement to the first monthly payment. Defendants’ Brief In Support of Their Motion for Summary Judgment as to Counts I and II of the Third Amended Complaint at 2. For example 3 , assume a loan of $30,000 at 8% annual interest, with a January 1 settlement date, monthly installment payments to begin on February 1. At the closing, a charge of $200 ($30,000 times 8% divided by 12 months), labeled “interest to first payment” 4 , would be collected for defendant lender. Then, on February 1, the plaintiff-borrower would begin paying monthly installments, the number and amount of which defendants determined by consulting mathematical tables prepared by the Financial Publishing Company (FPC) of Boston, Massachusetts. Reply Brief of Defendants in Support of Their Motion for Summary Judgment as to Counts I and II of the Third Amended Complaint at 7. To illustrate, if the loan were for twenty years, payments due monthly, then the mortgagor would make 240 installments of $250.93. *581 Financial: Compound Interest and Annuity Tables, 5th Edition # 376 Financial Publishing Company (Sixth Printing 1975). This collection of both “interest to first payment” and a monthly installment determined by FPC tables — referred to as “the practice” — is the focal point of plaintiffs’ challenge. II. THE ANTITRUST COUNT: Plaintiffs seek to spin a web of facts supporting an inference that defendants conspired to engage in the practice of using FPC tables in conjunction with an “interest to date of first payment” charge. There are four major strands to this web: (1) the practice is idiosyncratic; (2) it is improbable that all twenty defendants independently adopted this idiosyncratic practice; (3) defendants, through common membership in two trade associations, had ample opportunity to conspire and demonstrated an inclination to act as a group; (4) defendants had an economic motive for agreeing. We will consider each strand to see if plaintiffs’ allegations raise a “genuine issue of material fact” precluding summary judgment. Our inquiry must be animated not only by the general requirement that all evidence and inferences are to be viewed in a light most favorable to the party opposing the motion, Bishop v. Wood, 426 U.S. 341, 347 n.11, 96 S.Ct. 2074, 48 L.Ed.2d 684 (1975), but also by several rules peculiar to Sherman Act conspiracy suits. Summary judgment is to be used sparingly in complex antitrust eases involving questions of motive and intent. Poller v. Columbia Broadcasting System, 368 U.S. 464, 82 S.Ct. 486, 7 L.Ed.2d 458 (1962). However, where plaintiffs’ opposition to a motion for summary judgment fails to counter defendants’ proof that the facts are not susceptible to plaintiffs’ interpretation, summary judgment is appropriate. First National Bank v. Cities Service Co., 391 U.S. 253, 88 S.Ct. 1575, 20 L.Ed.2d 569 (1968); Cf. Tripoli Co. v. Wella Corp., 425 F.2d 932 (3d Cir.) cert. denied, 400 U.S. 831, 91 S.Ct. 62, 27 L.Ed.2d 62 (1970) (facts stipulated; only legal question presented). Each fact in an alleged conspiracy must be viewed in the totality of circumstances, not in isolation. Continental Ore Co. v. Union Carbide, 370 U.S. 690, 82 S.Ct. 1404, 8 L.Ed.2d 777 (1962). And formal agreement need not be proven; circumstantial evidence supporting a reasonable inference of conspiracy is sufficient. American Tobacco Co. v. United States, 328 U.S. 781, 809-10, 66 S.Ct. 1125, 90 L.Ed. 1575 (1946). It is with these legal yardsticks that we must measure plaintiffs’ case. A. The practice is idiosyncratic. There are three standard methods of collecting interest: interest in arrears; interest in advance at the same effective rate; and interest in advance at a higher effective rate. Assuming a $3,000 loan at 10% annual interest, settled on January 1, 1978 with $1,000 annual principal repayments, the following charts numerically illustrate each of the standard techniques: 5 I. Interest In Arrears. Total Payment Interest Principal 1/1/78 [0] 1/1/79 [1000] 1/1/80 [1000] 1/1/81 [1000] *582 II. Interest In Advance. Same Effective Rate as In Arrears. Total Payment Interest Principal 1/1/78 [250] 250 6 [0] 1/1/79 [1000] 1/1/80 [1000] 1/1/81 [1000] III. Interest In Advance. Effective Rate Increased. Total Payment Interest Principal 1/1/78 [300] _0_ 1/1/79 [1000] 1/1/80 [1000] 1/1/81 [1000] Fundamental to plaintiffs’ case is their allegation that on these facts the practice of collecting an “interest charge” at settlement is an aberrant accounting procedure that irrationally merges two of these standard collection techniques and thereby charges the first interest payment twice. The two accounting techniques allegedly confused by defendants’ practice are collection of interest in advance at a higher effective rate (III) and collection of interest in arrears (I). The deviant marriage of these collection methods is worked, they say, by conjoining in a single technique the FPC tables, which plaintiffs contend are prepared on an in-arrears basis, and the “interest” charge at settlement, which defendants aver is based on an in-advance theory. The issue of this purportedly illicit union is alleged to be the following illegitimate technique: IV. The Practice According to Plaintiffs. Total Payment Interest Principal 1/1/78 [300] _0_ 1/1/79 [1000] 1/1/80 [1000] 1/1/81 [1000] Plaintiffs offer the affidavits of Robert C. Collett, vice president of Financial Publishing Company, and John S. deCani, a Wharton economic statistics professor, as *583 proof that the FPC tables, because prepared on an in-arrears basis, include as part of the first installment the interest on the entire loan principal for one month immediately prior to the date of the first payment. Plaintiffs’ affiants conclude that a charge, imposed at closing by a lender using FPC tables, computed as though it were interest in advance from settlement to the first monthly installment is an improper duplication of the interest charge and therefore mathematically irrational. Affidavit of Robert C. Collett, February 22, 1978; Affidavit of John S. deCani, November 20,1974. It is in these affidavits that plaintiffs ground their allegation that the practice is idiosyncratic — a contention that is in turn the foundation of plaintiffs’ inference of conspiracy. Defendants insist that their practice of charging at settlement an amount equal to interest on the loan from closing to the first monthly installment was simply collection of interest in advance at a higher effective rate — Chart III — and that plaintiffs’ position is “absurd”. Whether Chart III or Chart IV illustrates the controversial collection method is thus squarely disputed. We conclude that this dispute raises a genuine issue of material fact preclusive of summary judgment. That the dispute is a factual one is obvious: it is for the jury to decide whether the practice was standard or aberrant. That the disputed fact is material is also clear: as discussed in B and C below, plaintiffs anchor their inference of conspiracy in the idiosyncratic nature of the common practice and we cannot at this stage dismiss that inference as insupportable. Finally, the issue of material fact is genuine. Defendants do not explicitly challenge the Collett and deCani statements that the FPC tables, admittedly used by all twenty lenders, are prepared on an in-arrears (Chart I) basis, although they do categorically deny any knowledge of the theoretical foundation of the FPC calculation, Reply Brief of Defendants at 7 — a denial plaintiffs reject as implausible given the sophisticated nature of defendants’ lending business. Defendants do, however, emphatically challenge plaintiffs’ contention that the practice was not interest in advance, albeit at a higher effective rate. Again and again, in their briefs and at oral argument, defendants maintain that they were using a standard accounting procedure legal under all relevant statutes and regulations. The collection of interest in advance, they concede through the May 15, 1978 affidavit of Professor Eli Schwartz, can increase the effective yield of a loan (compare Chart II with Chart III), but the technique, defendants contend, is nonetheless a conventional one. Cited as proof that the practice is not irregular is the admitted fact that many defendants granted an interest refund to a borrower who paid off his debt in mid-month. Defendants’ argument is that this refunding of part of the monthly interest charge is theoretically consistent only with collection of interest in advance, which defendants say they were doing, and is conceptually antithetical to collection of interest in arrears, the method of which plaintiffs accuse them. Thus, defendants urge that the only reasonable inference to be drawn from the fact of refunds is that interest was being collected in advance, at the beginning of each installment. We are not persuaded that the existence of a refund policy rebuts plaintiffs’ position as a matter of law. First, not all defendants granted a refund to a borrower who finished repayment in mid-month. According to defendants’ own presentation, of the twenty banks involved here, five institutions did not grant a refund at all, four did so only upon demand by the mortgagor, nine granted an automatic refund and the policies of two defendants are unknown. Defendants’ Brief — Schedule C. These mixed statistics are inconclusive as to the existence of a consistent refund policy, let alone as to the urged inference from the alleged policy that defendants’ practice was interest in advance. Further, even if a consistent refund pattern were established, its rebuttal significance would be unimpressive. Only if the *584 final installment paid by a borrower were all principal and no interest would the grant of a refund to the mortgagor who made that final payment before it was due be theoretically consistent with collection of interest in advance alone. The refund in that case would be probative of in-advance collection because the lender would not offer the grant unless he had already collected interest on the outstanding principal for the entire installment period that the borrower shortened by early repayment. The shortening of the final period during which any principal was unpaid would entitle the mortgagor to a refund of that portion of the already-collected interest that covered use of the principal for the time between the early repayment and the scheduled due date. If interest had not been collected in advance, the repayment in mid-term would not only not result in a refund but would require a charge of interest for the time between the previous interest collection and the date of repayment — again assuming that the repayment sum was all principal. However, if the final installment consists of both principal and interest, then the grant of a refund to a borrower who repays early is consistent with collection of interest in arrears. The interest component of the early repayment would be the lender’s charge for use of the outstanding principal for the final installment period. If that period were abbreviated by early repayment, then the mortgagor theoretically would be eligible for a refund of that part of the interest that represented the time by which he had shortened the life of his debt. Thus, whether a refund policy is probative of collection of interest in advance or in arrears depends on whether the early repayment triggering that grant consists of principal only or of principal and interest. In this case, the final installment payment, because computed on an in-arrears basis by the Financial Publishing Company, included both an interest and principal component. Given the admitted use by defendants of the FPC calculations, the policy of granting a refund to a borrower who made his final installment payment in mid-term is neither theoretically inconsistent with collection of interest in arrears nor probative of collection of interest in advance. 7 Faced as we are here with defendants’ motion for summary judgment, we must ask if there are two alternative reasonable answers to the question whether combined use of an in-arrears FPC table with an interest in advance charge at settlement creates an idiosyncratic double collection. Because this factual issue is an obscure one not yet fully illuminated by the affidavits and briefs, we cannot say as a matter of law that either the idiosyncratic or the interest in advance interpretation of the practice is unreasonable. Accordingly, the material fact question is genuine and must go to the jury. Our decision that plaintiffs’ contentions raise a genuine issue of material fact *585 compels denial of defendants’ motion for summary judgment. Were we presented with a routine case our inquiry perhaps would be concluded. However, because this is an antitrust conspiracy action, we think it important to discuss the totality of plaintiffs’ allegations. Only by viewing the idiosyncracy strand in the context of plaintiffs’ full web of averments can the materiality of that issue to the inference of unlawful agreement be fairly appreciated. Also, plaintiffs’ three remaining fibers may enhance the persuasiveness of the idiosyncracy argument and, as stated in our preface to this part, plaintiffs are entitled to the possibilities of interplay between the various elements of their case. Continental Ore Co. v. Union Carbide, 370 U.S. at 699, 82 S.Ct. 1404. We therefore reject defendants’ divide-and-conquer approach to plaintiffs’ contentions, Reply Brief of Defendants at 1-2, and will consider plaintiffs’ allegations in their entirety. B. Independent Adoption by Twenty Defendants of Idiosyncratic Practice is Improbable. Plaintiffs urge that the independent adoption by twenty separate savings and loan associations of a deviant accounting practice is too improbable to be believed. Agreement, they say, is the only plausible inference to be drawn from common use of an aberrant technique. This argument is of course dependent on the primary contention that the practice is in fact idiosyncratic. Plaintiffs, however, allege at least two other facts that buttress the urged inference of conspiracy. First, the practice is not logically suggested by the language of the mortgage documents, they say, and it therefore is suspect that twenty-two separate lenders endorsed the uncommon technique. Second, plaintiffs contend that only part of one discrete segment — savings and loan associations — of the tri-corner residential lending market in the Greater Philadelphia area engaged in the practice. Neither commercial banks nor savings societies used the practice, another fact suggesting agreement, according to plaintiffs. Defendants counter the allegation of improbability with insistent assertions that the practice was a “carry over” from predecessor banks. Defendants’ Brief at 7. They offer not only the statements of the officers of eighteen of twenty defendants that the origins of the practice are unknown because venerable, but also the affidavits of Edmund Jones, president of defendant First Federal Savings & Loan of Chester, and of David Shapiro, chairman of the board of defendant Independence Federal Savings & Loan Association, that the practice was being used as early as 1934, when First Federal of Chester and Independence were incorporated as federal associations. Affidavit of Edmund Jones, May 10, 1978; affidavit of David Shapiro, May 18, 1978. In further rebuttal of plaintiffs’ contention of improbability defendants categorically deny that they knew of one another’s practice of charging interest at settlement, Defendants’ Brief — Schedule A, and point to the non-parallel collection and refund procedures of several alleged co-conspirators, Defendants’ Brief at 9 — 11; Reply Brief of Defendants at 9, as evidence that the lenders acted independently. Also vigorously denied is knowledge of the in-arrears basis of the FPC tables. Finally, defendants dispute at length plaintiffs’ claim that the practice is not logically suggested by the loan documents, Defendants’ Brief at 28-31, and argue that plaintiffs’ proof that neither of the other sectors of the Philadelphia home loan market used the practice is inconclusive. We think that the issue of improbability, like that of idiosyncracy in which it is rooted, presents a jury question. The carryover rebuttal urged by defendants is genuinely challenged by plaintiffs’ affiant, Clelland L. Mitchell, a banking lawyer with four decades experience with Philadelphia savings and loan associations, although never with any of the defendants. See also Plaintiffs’ Brief at 44 — 45. Plaintiffs’ allegation that only lenders in the savings and loan segment of the Philadelphia home *586 mortgage market used the practice is likewise genuinely disputed. Although plaintiffs have affidavits from nine major lenders in the commercial and savings sectors of the loan market that they never engaged in the controversial practice, Exhibits to Plaintiffs’ Brief at 4A-AH, we cannot conclude as a matter of law that this fact — material because probative of improbability, which in turn is probative of conspiracy — is not subject to a reasonable difference of opinion. Resolution of these fact questions is for the jury. Also for the jury is the assignment of importance to those of the above allegations that are found to be proven. This responsibility is one that we must be especially cautious not to usurp here. A jury may find that the practice was idiosyncratic but that defendants were entirely unaware of its aberrant nature and that the common adoption of the method was therefore neither improbable nor suggestive of agreement. Or a jury could determine that the practice was idiosyncratic and known to be, but was not conspiratorily adopted by these defendants because carried over from earlier institutions. These questions of knowledge and intent, critical to a conspiracy case, are precisely the sort of issues that the factfinder is charged with resolving. Poller v. CBS, 368 U.S. 464, 82 S.Ct. 486, 7 L.Ed.2d 458 (1962). The possibilities sketched here, and the more elaborate ones discussed immediately below in C, preclude summary disposition of plaintiffs’ claim. C. Defendants, through common memberships in two trade associations, had ample opportunity to conspire and demonstrated an inclination to act as a group. The Insured Savings Association of the Delaware Valley (ISA) is a trade association formed by the 1966 merger of a trade organization that had represented the many federally-chartered institutions since the late 1930’s and a trade group that long had served the state-chartered institutions in the Greater Philadelphia area. Throughout the period at issue in this case, the ISA held regular meetings of the Board of Directors and of the membership. All twenty defendants belonged to the Insured Savings Association of the Delaware Valley during the eight years in which the conspiracy is alleged to have occurred. 8 Affidavit of John Stewart, May 11, 1978. In 1967 a new trade association, the Society of Savings and Loan Mortgage Officers (the Society), was formed. The Society membership consisted of savings and loan officers with responsibility for mortgage loans. Monthly meetings of the Board and of the members were held. Further, the Society circulated a monthly mortgage survey to its members inquiring about various lending policies. Deposition of Angelo Amoroso, April 27, 1977 at 48-50, 57, 63. Seventeen of the twenty antitrust defendants belonged to the Society from 1969 through 1975. 9 Defendants’ Brief at 16. Plaintiffs contend that “through these two trade associations alone, defendants had regular and frequent opportunities to exchange information, make agreements, give signals, etc. without having to make any extraordinary effort to communicate and without leaving a paper trail.” Plaintiffs’ Brief at 21. Citing extracts from the minutes of the ISA (and its predecessor associations) and of the Society, plaintiffs further argue that there is evidence of discussion by defendants of “every conceivable aspect of their business” and “numerous examples of invitations for concerted action.” Plaintiffs’ Brief at 25. This evidence of opportunity and discussion, plaintiffs urge, presents a factual base from which a jury can infer an inclination on the part of defendants to act as a group. This inference of a predisposition to agree is presumably offered as quasi-habit evidence *587 of defendants’ alleged conspiracy to use the practice. Defendants’ response to this argument is several-fold. First, the categorical denial by defendants that there was any discussion among ranking officers of the practice or that there was any knowledge on the part of defendants of one another’s interest policies is repeated. Second, defendants point out that the ISA and Society minutes in which plaintiffs divine such revealing discussions contain not the vaguest mention of the practice challenged here. Finally, the urged inference that the trade associations were crucibles of conspiracy is contrary, defendants say, to the undisputed facts that not all members of the ISA and the Society engaged in the practice and not all of those who did use the practice were members of the ISA and the Society. 10 The departure point for our evaluation of the “opportunity and inclination to agree” fiber of plaintiffs’ web is the obvious rule that mere membership in trade associations is not an antitrust violation. United Shoppers Exclusive v. Broadway-Hale Stores, Inc., 1966 CCH Trade Cases, ¶ 71,727 (N.D.Cal.1965). However, membership in a trade association coupled with regular exchange of business information, while again not violative of the antitrust laws, is admissible evidence because circumstantially probative, to whatever small extent, of an alleged agreement in restraint of trade. Standard Oil Co. of California v. Moore, 251 F.2d 188, 209 (9th Cir. 1958); C-O-Two Fire Equipment Co. v. United States, 197 F.2d 489, 493 (9th Cir.), cert. denied, 344 U.S. 892, 73 S.Ct. 211, 97 L.Ed. 690 (1952). Where, as here, the exchange of information was on matters other than the subject of the alleged conspiracy, such evidence would not, without more, support an inference of agreement. Harlem River Consumers Co-Op. v. Associated Grocers, 408 F.Supp. 1251, 1278, 1280 (S.D.N.Y.1976); United Shoppers Exclusive v. Broadway-Hale Stores, Inc., 1966 CCH Trade Cases, ¶ 71,727 at 82,270. Plaintiffs’ proof of membership and exchange is nonetheless some evidence of agreement because of its “tendency to picture a business setting and relationship which invites inferences favorable to [plaintiffs’] position.” Standard Oil Co. of California v. Moore, 251 F.2d at 209. 11 It is precisely this contextual use that plaintiffs seek to make of their evidence of membership and discussion. They argue that their picture of a “business setting and relationship” is one from which a jury reasonably could infer an agreement to engage in the practice and thus, plaintiffs say, summary judgment is unavailable. Defendants respond that such an inference, drawn from the facts of common membership and exchange of information about matters other than the practice, is illegitimate as a matter of law and therefore not preclusive of summary disposition. Although we agree with defendants that the “opportunity and inclination” evidence without more would be insufficient to support an inference of conspiracy, 12 we again decline their constant invitation to unravel plaintiffs’ factual mesh and consider each strand in isolation. *588 This “divide and conquer” approach is contrary to the Supreme Court’s decision that “[pjlaintiffs should be given the full benefit of their proof without tightly compartmentalizing the various factual components and wiping the slate clean after scrutiny of each.” Continental Ore Co. v. Union Carbide, 370 U.S. 690, 699, 82 S.Ct. 1404, 1410, 8 L.Ed.2d 777 (1962). We therefore must view the “opportunity and inclination” evidence in conjunction with the proof of idiosyncracy and improbability discussed above, and vice versa. On the one hand, plaintiffs’ proof of an idiosyncratic, improbable practice used by twenty different institutions, if believed, could illuminate the evidence of an opportunity and inclination to agree so vividly as to make an inference of conspiracy, insupportable on the business setting facts alone, not only reasonable but irresistible. On the other hand, plaintiffs’ proof of trade association activity might play on the jury’s appreciation of the evidence of idiosyncracy or improbability in such a way that an argument unconvincing in isolation becomes in context more probable than not. Given these possibilities of interplay, summary judgment is unavailable. The way in which evidence that is impotent when isolated can ripen into a genuine issue of material fact when planted in context is well illustrated by allegations concerning a mortgage questionnaire circulated by the Society of Savings and Loan Mortgage Officers. For at least four years 13 the Society conducted a monthly survey of its members. The questionnaire requested information about “interest rates, discounts, service fees, appraisal fees and credit reports.” 14 The survey was conducted anonymously and only twenty-five percent of the Society membership responded. Plaintiffs offered two discrete facts concerning the survey from which they say a jury could reasonably infer that the questionnaire was used to further the conspiracy to use the practice. First, plaintiffs point to the fact that despite service on all defendants and on the Society of requests to produce copies of the survey, not a single questionnaire has been produced. 15 Second, *589 plaintiffs extract from the minutes of the May 25, 1971 board meeting of the ISA an exchange in which a question asked by Mr. Amoroso, president of the Society, whether the survey should be discontinued in view of general advice of counsel that the questionnaire was of suspicious propriety under the antitrust laws was answered with a unanimous resolution that the survey be continued absent a more detailed objection. 16 The conspiratorial implications of these facts, plaintiffs argue, are for the jury to determine. Plaintiffs’ Brief at 25-26. Defendants again respond with several points. The non-production is explained, they say, by the 4V2 years that elapsed between the last survey and the request for copies. The surveys were unreliable even when current, because anonymous, and were useless 4V2 years later. Thus none was retained. The unanimous vote to continue the questionnaire despite vague antitrust concerns is no proof whatsoever of a conspiracy, defendants contend, because of the twenty alleged conspirators only three were represented at the meeting and only one participated in the vote. Reply Brief of Defendants at 17; Exhibit 11- — Defendants’ Brief. 17 Further, defendants uniformly deny that the practice was ever mentioned in the survey. In sum, they say, it would be illegitimate for a jury to conclude that an anonymous, poorly-received survey circulated alike among banks that followed the practice and those that did not was part of a conspiracy. Were we confronted in a vacuum with plaintiffs’ interpretation of the facts about the survey we would find their case untenable. Although the ISA Board vote undercuts a bit defendants’ argument that the survey was useless and therefore neither relied on nor saved, this vote alone hardly supports an inference of agreement. Similarly, the reasons for the non-production of the questionnaire may raise a genuine issue of fact but viewed in isolation the fact question is not material because proof of spoliation, although sufficient to support an adverse inference, II Wigmore on Evidence § 291, would not alone support a verdict of conspiracy. Thus, in isolation, the non-production of the survey does not save plaintiffs from defendants’ motion. Here again, however, we must evaluate the survey issue with plaintiffs’ idiosyncracy and improbability allegations in mind.’ If either or both of those contentions is accepted by the jury, the facts about the survey could assume a new significance and could become integral to a supportable inference of agreement to use the practice. Alternatively, the questionnaire issue, although harmless to defendants when isolated, could color the idiosyncracy or improbability question just enough to yield a conclusion on one or both of those issues favorable to plaintiffs. Or perhaps even the combination of the three elements discussed so far — idiosyncracy, improbability and trade association activity — would not persuade the jury, but that conglomerate of facts, taken in conjunction with the motive allegations treated below, may move the *590 factfinders to conclude that a conspiracy more likely than not existed. Plaintiffs must be given the chance to realize these psychological possibilities and others too numerous and subtle for us to list or even perceive. Conspiracy, by its very nature, is almost never susceptible to direct proof. It is most often a matter of inference, apprehended and proven circumstantially. American Tobacco Co. v. United States, 328 U.S. at 809-10, 66 S.Ct. 1125; Milgram v. Loew’s, Inc., 192 F.2d 579, 583 (3d Cir. 1951). We must not permit an inference of agreement to be conjured up from thin air. 18 But neither may we insist that the relevant circumstances compel an inference of conspiracy with the directness and ineluctability of a logician’s proof. Plaintiffs’ web of allegations need not be impregnable to withstand a motion for summary judgment; it need only be elastic enough to support the interpretation that plaintiffs urge. Although brittle, plaintiffs’ fibers are at least this strong. Despite our skepticism, we would be usurping the function of the jury were we to conclude that conspiracy is an impossible inference on these facts. Walsh v. Butcher & Sherrerd, 452 F.Supp. 80, 82-83 (E.D.Pa.1978). Thus, we reaffirm our conclusion that summary judgment is unwarranted. 19 D. Motive. We will briefly consider plaintiffs’ final area of evidence — motive—to complete the *591 context in which the allegations discussed above must be viewed. Although the exact figures are not ascertainable until the issue in II A above is resolved, plaintiffs and defendants agree that by using the practice a lender was able to obtain a greater yield on a loan than he would have received by collecting interest in arrears. 20 This increase, both parties agree, is the motive behind an individual lender adopting the practice and defendants say this simple explanation rebuts the allegation of a group decision. Plaintiffs, however, argue that there were at least three economic reasons for defendants to engage in the practice as a group. 21 First, plaintiffs claim, by using the practice as a group defendants reduced the likelihood that a customer could avoid paying the increased charge. Second, by engaging in the practice as a group, defendants could represent the “interest from settlement” charge as standard and thereby reduce the likelihood that customers would seek the loan elsewhere. And third, plaintiffs contend that defendants, by acting in concert with one another, greatly reduced the risk of adverse public exposure because any accusations by an irate borrower of unfair dealing would be diluted by the fact that so many lenders used the practice. Plaintiffs rely almost exclusively on the testimony of Barry Schechter, an economist, who concluded that participation in a price fixing scheme by a significant, if not dominant, number of lenders furthers the conspiracy by reducing consumer alternatives and decreasing the likelihood of detection. Affidavit of Barry Schechter, June 8, 1978. Plaintiffs offer only two other pieces of evidence: the statements of two borrowers from defendants who discovered that the practice was “not proper.” Deposition of Charles Bender at 24; affidavit of Sheldon F. Neubauer, July 13, 1978. This three-pronged allegation of economic motive is, we think, an especially vulnerable *592 part of plaintiffs’ case. As defendants so forcefully point out, there is not a scintilla of empirical evidence that the use of the practice affected in any way whatsoever a lender’s position in the home mortgage market. The experience of Independence Federal, which continued the practice after eighteen other defendants quit and lost no customers, is an example of the inconsequential impact of the controversial method. Further, plaintiffs have admitted that there are over 100 savings and loan associations in the geographic area in which the conspiracy allegedly occurred. There are also numerous commercial banks, savings fund societies, savings banks and mortgage brokers. In such an unconcentrated market, a conspiracy of twenty lenders in one sector would be of small value to the participants. In the face of such facts, the abstract propositions of the Schechter affidavit are unpersuasive. However, the Schechter testimony must of course be weighed in the context of plaintiffs’ total case. With this conclusion we are brought full circle. Plaintiffs’ evidence of motive, however paltry, must be allowed to exercise its influence on the proof of idiosyncracy, improbability and trade association activity. It is for the jury, not us, to evaluate this entirety. 22 Consequently, defendants’ motion for summary judgment as to the antitrust count will be denied. III. TILA: We preface our consideration of plaintiffs’ TILA count with the observation of Judge Newman that “[l]ike most issues that arise in this highly technical field, the issue here approaches analysis of heavenly pinhead terpsichore.” Grey v. European Health Spas, Inc., 428 F.Supp. 841, 846 (D.Conn.1977). While portions of the TILA labyrinth do seem to us only slightly less inscrutable than angelic choreography, the worldly import of the Truth in Lending Act is considerably more immediate than that of the seraphic dancers. The purpose of TILA is to prevent the uninformed use of credit. 15 U.S.C. § 1601; 12 C.F.R. 226.1. “Let the Seller disclose” is the Act’s basic philosophy, Mourning v. Family Publications Service, Inc., 411 U.S. 356, 93 S.Ct. 1652, 36 L.Ed.2d 318 (1973), and the detailed disclosure requirements are to be liberally construed and strictly enforced. Eby v. Reb Realty Inc., 495 F.2d 646, 650 (9th Cir. 1974). Easy comparison of the terms and cost of credit is the intended result of the TILA scheme and generous civil liability is imposed on *593 lenders who frustrate the congressional design. 15 U.S.C. § 1640. Section 1631(a) of TILA states the general requirement that all disclosures be “clear and conspicuous.” Part 226.6(a) of Title XII of the Code of Federal Regulations restates this provision as a requirement that disclosures shall be made “clearly . in the terminology prescribed in applicable sections.” There is some suggestion in TILA literature that the requirement of “clear” disclosure is to be read in strict conjunction with the “conspicuous” requirement and thus mandates simply that any statement be easily legible, see R. Clontz, Truth in Lending Manual 2-77 (4th ed. 1976), but the courts have interpreted § 1631(a) to impose a substantive requirement that the meaning of a disclosure be clear to the average borrower. Turoff v. May Co., 531 F.2d 1357, 1361 (6th Cir. 1976); Doggett v. Ritter Finance Co. of Louisa, 384 F.Supp. 150, 155-57 (W.D.Va.1974); Peritz v. Liberty Loan Corp., 4 CCH Consumer Credit Guide (N.D.Ill. March 5,1973); 9th Liberty Loan Corp. v. Hardy, 53 Ill. App.3d 601, 11 Ill.Dec. 363, 366, 368 N.E.2d 971, 974 (1977). Although plaintiffs are frustratingly curt about their TILA argument, we consider this imperative of clarity to be the standard at issue here. 23 As discussed in I above, defendants collected at settlement a charge equal in amount to the interest on the loan from the date of closing to the date of the first monthly installment. Defendants, on their TILA disclosure statements, labeled this charge “interest to date of first payment.” 24 The heart of plaintiffs’ claim is that this charge was not interest because, as above outlined, the first monthly payment included interest from date of settlement. The controversial charge, plaintiffs contend, was an extra fee. Thus, the charge was mislabeled. This mislabeling, plaintiffs say, violated the clear disclosure requirements of TILA. Defendants respond to this allegation with a lengthy justification of the legality under TILA of the collection of interest in advance. Because this discussion is almost entirely addressed to issues not raised by plaintiffs’ third amended complaint, we will not detail it. 25 Again, here, as with the antitrust count, the main line of defendants’ TILA defense is that the practice was the standard accounting method of interest in advance. Thus, defendants say, labeling the prepaid finance charge “interest” was perfectly clear and not in violation of any disclosure requirements. The question whether labeling the disputed charge “interest” violates the clarity re *594 quirement of § 1631(a) involves two separate issues. The first is whether the charge was or was not interest. If the charge was interest then it was not mislabeled and the disclosure was unobjectionable. Only if the charge was not interest is the second question reached, namely, whether labeling as “interest” a settlement fee that is actually a charge in addition to the interest due is impermissible because unclear. Thus, for the TILA count to be ripe for summary judgment, we must be able to answer, as a matter of law, either the first or second question in defendants’ favor. The first question, whether the settlement charge was interest, cannot yet be answered. This question itself requires two separate determinations: (a) a definition of interest; and (b) a determination whether the charge collected here is within that definition. We confess some uncertainty whether these are issues of law or fact. Although used in several TILA provisions, e. g., 15 U.S.C. § 1605(a)(1); 12 C.F.R. 226.4(a)(1), the term “interest” is nowhere defined in the statute or regulations, nor is it defined in any of the loan documents used in this case. Thus, for our purposes, we are confronted with an undefined term. 26 To the extent that the disclosure statement in which the “interest” label appears is considered to be part of a contractual transaction, the meaning intended by the parties should control and both the definition and the determination whether this charge is within that definition are questions of fact. Cf. Maes v. Motivation for Tomorrow, Inc., 356 F.Supp. 47 (N.D. Cal.1973) (contractual intent is fact question in TILA case). However, the disclosure statement is not actually part of the contract between plaintiffs and defendants; only the bond agreement is. Rather, the statement is an ancillary document included because required by federal statute. To the extent that the disclosure statement is thus considered to be part of the TILA scheme, the definition of any term used therein arguably is an issue of law for the court to decide, with the determination whether the charge here is interest also being a matter for the bench. Because the controversial term is one supplied by the parties, albeit on a required form, rather than one literally prescribed by the statute or Regulation Z, 12 C.F.R. 226.1 et seq., we on balance conclude that the issue is one for the jury to decide. This conclusion, of course, compels us to deny defendants’ motion for summary judgment. We need not, however, rely solely on resolution of the law/fact dilemma to deny defendants’ motion because summary judgment here is unavailable on an alternative ground as well. Prerequisite to a determination, either by the court or the jury, whether the settlement fee was mislabeled “interest” is an answer to the question discussed in II A above — whether the practice of collecting that charge was or was not “interest in advance at a greater effective rate.” As we stated there, that question is a “genuine issue of material fact” unresolved by the briefs and exhibits and unanswered at the hearing. Therefore, because we cannot determine whether the practice was interest in advance, we cannot decide whether the charge was interest. The case must proceed to trial of that issue, unless we are able to conclude, as a matter of law, that even if the charge was not interest the clear disclosure requirement of TILA was not violated. This we are unable to do. Without resolution of the underlying issue whether the practice was interest in advance we cannot begin to measure the disclosure statement *595 against the clarity requirement. 27 Given the divergent conclusions that courts have reached on disclosure questions, compare Lewis v. Walker-Thomas Furniture Co., Inc., 416 F.Supp. 514, 517 (D.D.C.1976) (consistent and uniform terminology prerequisite to informed use of credit) with Matter of Black, 411 F.Supp. 749 (S.D.Ohio 1975) (TILA satisfied where cost of credit is clear regardless of precision of terminology), we will not enter the thicket until the underlying facts are settled. 28 Accordingly, we deny defendants’ motion for summary judgment as to the TILA count. 29 Footnotes [1] Plaintiff class includes "all borrowers from defendants who ‘obtained mortgage loans on or after January 1, 1968, which were secured by conventional mortgages on one-family homes where [the alleged illegal] charge was made.’ ” Chevalier v. Baird Savings Association, 72 F.R.D. 140, 144 (E.D.Pa.1976). The twenty defendants are: Abraham Lincoln Federal Savings & Loan Association; Benjamin Franklin Federal Savings & Loan Association; Colonial Federal Savings & Loan Association; Commonwealth Federal Savings & Loan Association; Community Federal Savings & Loan Association of Philadelphia; East Girard Savings Association; Elmwood Federal Savings & Loan Association; First Federal Savings & Loan Association of Philadelphia; First Federal Savings & Loan Association of Chester; Founders Federal Savings & Loan Association; Garfield Federal Savings & Loan Association; Independence Federal Savings & Loan Association; Inter-county Savings Association; Key Savings & Loan Association; Liberty Federal Savings & Loan Association; Metropolitan Savings & Loan Associations; Mid-City Federal Savings & Loan Association; Penn Federal Savings & Loan Association; Polonia Federal Savings & Loan Association; and Second Federal Savings & Loan Association. [2] The TILA class is a subset of the antitrust class. It encompasses ail borrowers who settled home mortgages with Metropolitan Federal Savings & Loan or Liberty Federal Savings & Loan after August 12, 1971 and all borrowers who settled a home mortgage with First Federal of Chester Savings & Loan after October 12, 1971. Individual actions are being maintained against Abraham Lincoln, Community and Polonia Federal Savings & Loan Associations. Chevalier v. Baird Savings Association, 72 F.R.D. at 151-158. [3] We use the relatively simple figures of this example because the actual numbers involved in plaintiffs’ loans require calculations too detailed to be the clearest illustration of the controversial practice. But see note 5 infra. [4] The charge was variously labeled “interest to first payment”, “interest to date of first installment”, “interest to February 1”, etc. [5] We abandon here our earlier example of a $30,000 loan at 8% interest for 20 years. Although, as stated above in n.3, the $30,000 example is easier to comprehend than many of the actual figures involved in plaintiffs’ mortgages, even the $30,000 example involves unnecessarily complex calculations when used to illustrate the various methods of collection of interest. We accordingly will opt for a more elementary example, offered by plaintiffs and not repudiated by defendants, in this section. Plaintiffs’ Brief in Opposition to Defendants’ Motion for Summary Judgment as to Count I and Count II at 14. [6] These examples are from plaintiffs’ brief. We do not pretend to understand how $250 can be interest at 10% on a $3,000 loan, but since this method was not used, it is irrelevant and not worth explaining. [7] At the November 22 hearing, defendants stated a final argument, grounded in the refund grants, against plaintiffs’ contention that the practice is idiosyncratic. Defendants asserted that the total amount collected under the controversial practice is the same as that collected under the interest in advance at a greater effective rate method. Defendants’ counsel, referring in terms to the sample charts discussed above, stated that a lender using the practice would collect the $600 interest total of Chart III, not the $900 total of Chart IV as argued by plaintiffs, “the reason, of course, [being] the refund in the end.” Therefore, defendants urge, the practice must be interest in advance. This position is unconvincing for two reasons. First, as we observed above, not all defendants followed a refund policy and counsel’s statement, even if accurate, is thus not universally applicable. Second, and more fundamental, defendants have made an invalid comparison. The amounts compared by defendants’ counsel are the total interest collected using the practice with a refund and that collected using interest in advance without a refund. The relevant comparison of course, is the practice without a refund and interest in advance without a refund. According to the logic of defendants’ comparison, the interest collected using the practice without a refund must be more than that collected by interest in advance without a refund because the amount collected using the practice with a refund is the same as that yielded by interest in advance without a refund. This comparison argument is thus at best unhelpful to defendants and perhaps detrimental. [8] Polonia Federal did not become a member of the ISA until 1969, one year after the illegalities alleged in this suit began. The other nineteen defendants had been members since at least 1966. Affidavit of John Stewart, May 11, 1978. [9] Elmwood Federal, Key Savings and Polonia Federal did not belong to the Society. See note 22, infra for implications of this. [10] As we observed in note 9 supra, defendants Elmwood Federal, Key Savings and Polonia Federal did not belong to the Society. Polonia Federal was charging “interest from settlement” before joining the ISA in 1969. Defendants’ Brief at 20. Independence Federal was also. Reply Brief of Defendants at 11. Home Unity, Baird, Provident and Trevose Federal all belong to the ISA, Affidavit of John Stewart, May 11, 1978, but none used the practice. Defendants’ Brief at 20. [11] Defendants seem to contend that the Harlem River and United Shoppers cases cited above state the conclusion that membership in trade associations, even coupled with exchange of information, is not evidence of an agreement. Defendants’ Brief at 45^48. If this is indeed their position, they have confused the admissibility of evidence with its sufficiency to support a verdict. Proof of membership and exchange is admissible, because probative, but rarely sufficient, because too circumstantial. [12] The “business setting and relationship” evidence, standing alone, cannot support an inference of agreement because there is no suggestion whatsoever in any of the trade association documents of any discussion of the practice. Further, defendants have categorically denied any mention of the practice at either ISA or Society meetings. Even if plaintiffs could discredit these denials, they have no affirmative proof of discussion of the practice. [13] Plaintiffs and defendants disagree whether the survey ran for four or five years. Compare Plaintiffs’ Brief at 25 with Reply Brief of Defendants at 16. Because the longevity of the survey is not material we need not evaluate this issue to see if the dispute is genuine. [14] Plaintiffs and defendants both cite the April 27, 1977 deposition of Angelo Amoroso as the source of their statements about the survey. We also will rely on that testimony. [15] Defendants are accused of another non-production from which plaintiffs argue an inference of cover-up can be drawn. Leigh Bauer, a partner in the law firm representing plaintiffs, testified at his deposition that Walter Bannister, an officer of Olney Federal Savings & Loan Association (not a defendant) told him in a 1972 conversation at the home of Bauer’s parents in Avalon, New Jersey, that “a group of savings and loan associations” in the industry circulated a memorandum concluding that “the practice was proper and should be continued.” Bauer further testified that in 1974, when he visited Bannister’s office to obtain a promised copy of the memorandum, Bannister could find “neither the memorandum nor a relevant letter from Bill Bender, counsel for the Federal Home Loan Bank of Pittsburgh.” Bannister’s interest in finding the memo had “cooled off very substantially,” Bauer said. Deposition of Leigh W. Bauer, at 35, 36, 54-56; Exhibit 13 to Plaintiffs’ Brief. Plaintiffs argue that failure of any defendant to produce this memo is evidence of a cover-up. To this defendants reply simply that the alleged Bannister memo never existed. They point to the deposition of Bannister, at which he testified that he had never told anyone that there was in existence a memorandum concerning interest in advance, Deposition of Walter Bannister at 83, and to the categorical denials of all defendants examined that they had ever heard of or received any such memo. Defendants’ Brief — Schedule B. Without proof of existence of the memorandum, defendants say, there is no foundation on which to base an inference of cover-up. Because we conclude that plaintiffs have no admissible evidence of the Bannister memo, we agree with defendants that the non-production issue is a spurious one. First, we note in passing the dubious propriety of a lawyer testifying for plaintiffs that his firm is representing. We need not however pursue that question because the Bauer testimony is in any event based on inadmissible hearsay. That the statement attributed to Bannister is hearsay under the Federal Rules of Evidence is clear. It is squarely within the § 801c definition and is not saved by the exclusionary provisions of § 801d(2) because Bannister is not a *589 party-opponent (§ 801d2A), Olney Federal not having been named a defendant. Even if Olney Federal is alleged to have been a co-conspirator, the supposed statement to Bauer was not made “during the course and in furtherance of the conspiracy” (§ 801d2E) but rather was made outside the scope of the alleged agreement. Nor is the Bannister statement on which the Bauer testimony is based within any of the § 803 or § 804 hearsay exceptions. We thus conclude that the Bauer testimony would be inadmissible at trial to prove the existence of a memorandum. Hearsay not admissible at trial cannot be considered on a summary judgment motion, Daily Press, Inc. v. UPI, 412 F.2d 126 (6th Cir. 1969), and we therefore disregard plaintiffs’ non-production allegation here. The preceding discussion is also dispositive of plaintiffs’ reliance on the statement of Wilbert Myers, Plaintiffs’ Brief at 32; Exhibit 11, which has the additional infirmity of questionable relevance. Reply Brief of Defendants at 14. [16] Text of Minutes of the Society of Savings and Loan Mortgage Officers at 25-26 — Exhibit 7 — Plaintiffs’ Brief. [17] We are unable to determine from the minutes of the ISA board meeting whether only one representative of defendants participated in the vote and do not mean by repeating the statement here to endorse it as accurate. [18] That we are unwilling to allow plaintiffs to concoct a jury issue from implausible inferences is demonstrated by our refusal to consider in this summary judgment analysis several allegations that are too weak to carry any weight. For example, plaintiffs contend that a mortgage course conducted for two years by Angelo Amoroso, an officer of defendant Colonial Federal Savings & Loan, in which the practice was taught as the “local” procedure, is evidence of conspiracy. Plaintiffs’ Brief at 37. This suggestion is frivolous. Amoroso’s predecessor taught the “in-arrears” method for ten years, Defendants’ Brief at 22, and the text used by Amoroso included that method. No policy makers attended the course; only ministerial personnel were enrolled. And there is no evidence of any attendance by any of defendants. Reply Brief of Defendants at 20. The Amoroso course raises neither a genuine nor a material issue. Similarly lame is plaintiffs’ allegation that the fact that 18 of 20 defendants ceased the practice between April 1974 and April 1975, all offering the 1974 passage of a Pennsylvania statute regulating maximum interest rates as the reason, is probative of agreement. The 1974 statute, 41 Pa. C.S.A. § 101 et seq., referred to as Act VI, prohibited collection by a residential mortgage lender of a loan yield in excess of the maximum lawful rate of interest. Nearly all defendants interpreted Act VI as banning the use of interest in advance where interest was collected at the maximum lawful rate because, as discussed above in II A, collecting interest in advance increases the effective yield. Defendants’ Brief at 10. This explanation is the only plausible interpretation of 18 of 20 defendants ceasing the practice within a year of Act VI. Finally, the inferences of conspiracy that plaintiffs seek to draw from non-production of the minutes of a meeting at which First Federal of Chester stopped the practice and from the pre-1969 minutes of the Society, Plaintiffs’ Brief at 39-40, are untenable. [19] Modern Home Institute, Inc. v. Hartford Accident & Indemnity Co., 513 F.2d 102 (2d Cir. 1975), much relied on by defendants, is not antagonistic to this conclusion. That case involved the allegedly conspiratorial refusal of defendant insurance companies to purchase plaintiff’s list of holders of expired policies. Plaintiff sought to raise an inference of agreement with evidence that the list was useful and that defendants rejected its proposal after being initially enthusiastic about it. The court, presented with convincing proof of several business reasons for refusing to deal with plaintiff, held plaintiff’s inference unreasonable as a matter of law and granted defendants’ motion for summary judgment. Plaintiffs’ evidence in this suit, while far from overwhelming, is critically stronger than that in Modern Home Institute in at least two respects: (a) plaintiffs ground their inference of conspiracy in the detailed pattern of facts discussed above, a pattern involving more than parallel conduct alone; and (b) although defendants contend that the profitable nature of the practice proves that independent adoption by 20 banks is probable, this defense may be undercut by plaintiffs’ idiosyncracy and improbability arguments whereas the several business explanations for similarity of conduct in Modern Home Institute were wholly unrebutted. 513 F.2d at 108-110. Our denial of summary judgment here is therefore in harmony with the Second Circuit’s treatment of Modern Home Institute. [20] Plaintiffs say that the extra amount is equal to interest on the full loan principal for one month. Defendants say that the amount is that increment in effective yield gained by using the interest in advance at an increased rate method; for example, the increase from 7% nominal to 7.065 percent effective described in the Schwartz affidavit. Reply Brief of Defendants at 8. This disagreement is simply the fact question of II A above recast. For purposes of discussion, the parties have agreed to a figure of $150 extra per mortgage loan but this is only a compromise number. [21] Defendants contend that plaintiffs’ allegations of motive are incomplete without a companion averment that the challenged action was against defendants’ economic interest. This requirement is attributed to the opinion in Venzie Corp. v. U. S. Mineral Products Co., Inc., 521 F.2d 1309 (3d Cir. 1975), a refusal to deal case where the court affirmed a judgment n. o. v. for defendants with the conclusion that “[Plaintiffs] evidence does not, however, include two elements generally considered critical in establishing conspiracy from evidence of parallel business behavior: (1) a showing of acts by defendants in contradiction of their own economic interests . and (2) satisfactory demonstration of a motivation to enter an agreement . . . ” 521 F.2d at 1314. Defendants say that without this proof, which is impossible here, plaintiffs’ case is “meaningless.” Defendants’ Brief at 43. We do not agree with defendants’ invocation of Venzie on these facts. The requirement of proof of action in contradiction to self-interest is logical in a case, like Venzie, of parallel refusals to deal because such evidence makes the refusals suspect and raises a question of agreement. Evidence of conduct against interest could also be expected of plaintiffs alleging a conspiracy to drive prices down — to force a competitor out of business, for example — because, again, proof of such unprofitable behavior undermines a defense of independent albeit parallel decisions. However, in a case of an alleged agreement to increase prices, the traditional antitrust category in which this suit falls, the Venzie requirement is illogical. A conspiracy to raise prices is of course not contrary to the economic interest of its participants and the showing defendants say is required by Venzie could never be satisfied. If Venzie proof is “critical” in a price-increase case, then such conspiracies are immune from antitrust liability. This is absurd on its face. We thus conclude, as plaintiffs urge, that the Venzie reasoning is inapplicable here and no evidence of conduct contrary to economic interest is required (as defendants implicitly acknowledge by letting plaintiffs’ arguments against the Venzie point go unrebutted in their Reply Brief). [22] As stated above in note 9, Elmwood, Key and Polonia have never been members of the Society and never received the mortgage questionnaire. These three defendants contend that for this reason summary judgment must be granted as to them even if denied as to the other seventeen. Because membership in the Society is just one of a complex of circumstances on which the allegation of conspiracy is based, we think that plaintiffs should be allowed to take their case against Elmwood, Key and Polonia to the jury. We will bear in mind, though, that our charge to the jury may have to include separate instructions as to these three defendants. Defendant Polonia also points to the fact that it was using the practice before joining the ISA in 1969, see note 10 supra, and argues that because plaintiffs allege that the ISA was a vehicle for the conspiracy the case against Polonia is deficient. Similarly, defendant East Girard contends that because it did not adopt the practice until 1969, after all the others, plaintiffs cannot connect them with the alleged conspiracy. Reply Brief of Defendants at 24— 25. Given the scattered set of facts and inferences on which plaintiffs rely, we must reject the separate arguments for summary judgment, noting again that we may have to include these differences in our statement to the jury, if the case gets that far. It is of course possible that issues of fact that appear to be genuine and material at this stage of the case will, after all of the evidence is in, admit of only one reasonable interpretation and thus not merit jury consideration. Especially is this true in a suit like this one where certain probative facts have yet to be alleged, let alone decided — for example, what percentage of ISA and Society members used the practice. We note this possibility merely to underline that our decision here is that the case cannot be terminated without trial, not that plaintiffs’ allegations must go to the jury regardless of what develops as the proof unfolds. [23] Also relevant here is 12 C.F.R. 226.6(c): “Section 226.6 — General disclosure requirements (c) Additional Information. At the creditor’s . . . option, additional information or explanations may be supplied with any disclosure required by this Part, but none shall be stated, utilized, or placed so as to mislead or confuse the customer . . .” (Emphasis supplied) A creditor is not required to itemize elements of the finance charge for a loan, secured by a first lien on real estate, made in connection with the purchase of a dwelling. 12 C.F.R. 226.8(d)(3). The controversial item labeled “interest to date of first payment” was thus “additional information” and subject to the 226.6(c) standards. However, because we consider the clarity imperative to be primary and because the clarity and misleading requirements share a common analysis, we will not discuss 226.6(c) in terms. [24] Often, the actual date of first payment was stated. Thus, for example, the charge may have been labeled “interest to March 1, 1974” etc. [25] Defendants extensively demonstrate two points: that the annual percentage rate (APR) was accurately stated on the disclosure forms and that interest in advance is a permissible collection method. Although plaintiffs’ initial complaint did maintain that the APR had been understated, this claim is no longer pressed. Plaintiffs have never challenged the legality of collection of interest in advance; rather, they have simply maintained that defendants did not use that method. Thus, the bulk of defendants’ TILA response is irrelevant to plaintiffs’ averments. [26] The dictionary definition of “interest” is not only of no legal significance but is also exquisitely ambiguous here: “the price paid for borrowing money generally expressed as a percentage of the amount borrowed paid in one year.” Webster’s Third New International Dictionary of the English Language Unabridged (1963). [27] The recent case of Dzadovsky v. Lyons Ford Sales, Inc., 452 F.Supp. 606 (W.D.Pa.1978) is not contrary. Plaintiffs 'there claimed, inter alia, that a $15 fee labeled “prepaid finance charge” on the disclosure statement was actually a “service charge” and was thus misleadingly labeled. Chief Judge Weber concluded that “[t]he phrase ‘prepaid finance charge’ and ‘service charge’ are not mutually exclusive and both can correctly be used to describe the same payment without causing the whole document to fail under 12 C.F.R. § 226.6(a) which requires disclosure to be ‘clear and conspicuous’ ” 452 F.Supp. at 609. Further, plaintiff complained that the “interest expense” was mislabeled as “regular charge.” This claim was dismissed as “meritless,” because “even the unsophisticated consumers would interpret ‘regular charges’ as the interest expense.” Id. These conclusions superficially suggest that even if the controversial settlement charge here was not technically interest the imprecision in defendants’ disclosure does not violate the TILA clarity requirement. However, the Dzadovsky holdings are inapposite in this case for at least two reasons: (a) the Dzadovsky court knew the exact nature of the fees involved there and thus could determine how accurate defendants’ labels were; this we cannot do; and (b) plaintiffs in Dzadovsky alleged no deception or injury but merely a technical inaccuracy in the labels, whereas plaintiffs here aver that they were affirmatively misled and suffered pecuniary harm. Third Amended Complaint, 1| 21. Thus, we need not challenge Dzadovsky to decide that we cannot conclude as a matter of law that the mislabeling here, if any, was not violative of TILA. [28] We observe further that there is some small doubt whether the clarity issue is one of law or fact. Although the weight of authority, correctly we think, calls for the court to decide issues of clear disclosure, see, e. g., Turoff v. May Co., 531 F.2d 1357 (6th Cir. 1976); Doggett v. Ritter Finance Co. of Louisa, 384 F.Supp. 150 (W.D.Va.1974); Andrucci v. Gimbel Brothers, 365 F.Supp. 1240 (W.D.Pa.1973), the cases are not unanimous. See Peritz v. Liberty Loan Co., 4 CCH Consumer Credit Guide, (N.D.Ill. March 5, 1973). On the law/fact line in TILA cases generally, see Eby v. Reb Realty Inc., 495 F.2d 646, 648 (9th Cir. 1974). [29] Finally, we note that the TILA statute includes a good faith defense provision, 15 U.S.C. § 1640(f). Because defendants have not raised this issue we express no opinion on it.