649 F. 2d 175 - Sharp v. Coopers & Lybrand

This appeal from verdicts in favor of the three named plaintiffs in a securities fraud class action raises several important questions under the antifraud provisions of the Securities Exchange Act of 1934. Specifically, we are asked to decide whether the district judge properly instructed the jury on imputed or secondary liability, on § 20(a) of the Act, on reliance, and on the measure of damages. For the reasons below, we affirm on all issues except the measure of damages, which we reverse and remand for a new trial.

The factual setting of this case1 combines Caribbean intrigue, creative accounting, and high finance against the backdrop of investors attempting to limit their tax obligations. Westland Minerals Corporation was the promoter of a venture, the "Ohio Program," in which multiple limited partnerships were formed for the purpose of drilling for oil and gas. WMC sold interests in each of these partnerships from July 22, 1971, through early July, 1972. Each limited partnership was to spend $140,000 on drilling and developing a well. Of this amount, $65,000 was to be raised from contributions by investors in the partnership, and the balance from "suitable banks or other lending institutions."

As part of its sales presentation, WMC used a tax opinion letter issued by the accounting firm of Coopers & Lybrand, the appellant. WMC requested the letter in April, 1971, in behalf of one investor, Muhammed Ali, and on July 22, 1971, an opinion letter signed by a Coopers & Lybrand partner in the firm name was sent to the president of WMC. The letter stated that "based solely on the facts contained (in the WMC Limited Partnership Agreement) and without verification by us," a limited partner who contributed $65,000 in cash could deduct approximately $128,000 on his 1971 tax return. Herman Higgins, a tax supervisor employed by Coopers & Lybrand who worked directly under the supervision of four partners, drafted the letter. In October, 1971, Higgins told David Wright, a partner at Coopers & Lybrand, that WMC was showing copies of the letter to investors as part of its sales program, and Wright decided that a more complete letter should be drafted for those purposes. On October 11, 1971, Wright sent to WMC a revised letter, which he signed in the name of Coopers & Lybrand.

Under the investment plan, investors would purchase partnership shares in WMC's Ohio Program, with the money raised to be used in oil drilling. The investors would be limited partners in the sense that their liability would be limited, but their participation in the profits would be quite extensive once the oil started to flow. The plan's main attraction was the investors' ability to deduct twice the amount invested from their federal income tax returns in the first year of investment. The reason for this double deduction was that WMC would borrow on a nonrecourse basis, secured only by the oil wells, an amount of money approximately equal to that invested. Because the outside investors received their interests as limited partners, they did not subject themselves to liability on the loans merely by being partners; because the loans were nonrecourse, they were also not subject to them by explicit agreement.

The tax benefit arose because oil drilling requires large initial noncapital expenditures before any recovery can occur. These expenditures would be incurred in 1971, the year of investment. The partnership agreement allowed the profits and losses to be passed through the partnership to the partners. Not only would they be able to claim a total loss for the money actually invested, but they would also be able to claim a loss for the money borrowed and spent on the oil wells. The result was that each investor was able to deduct twice what he actually invested as ordinary (noncapital) losses, and could recover his investment in the first year by tax savings if he was in a fifty percent or higher tax bracket.

The practical problem with this is that banks are seldom if ever willing to lend money for oil drilling when their only security is the oil well. The reason is obvious: the well may not produce, and the bank will then have no security. Obtaining large amounts of loans was crucial to the scheme. Higgins, the Coopers & Lybrand associate, proposed a solution. He suggested that WMC borrow the money from a bank, and then use the money borrowed to purchase savings certificates from the bank, which the bank would hold as collateral. The net result of this arrangement was a paper transaction, by which WMC obtained loans on paper without actually receiving the money.2 WMC decided to transact these loans through a bank in the Bahamas that it acquired in September, 1971, before Coopers & Lybrand issued the second opinion letter. The scheme began to unravel when WMC was indicted for securities violations. The IRS began denying the deductions taken by the investors, the wells were frequently dry, and WMC collapsed.

Stanley Sharp filed this suit on behalf of himself and 210 other WMC investors similarly situated on May 8, 1975, alleging violations of § 10(b) of the Securities Exchange Act of 1934, 15 U.S.C. § 78j(b), and Securities Exchange Commission rule 10b-5, 17 C.F.R. § 240.10b-5.3 The trial judge granted class certification, and a jury trial of the class issues was held from February 27 through March 10, 1978. By answers to special interrogatories, the jury found that the October 11, 1971, tax opinion letter contained material misrepresentations and omissions, and that Herman Higgins had acted recklessly and intentionally. Although the jury found that no partner of Coopers & Lybrand had acted with scienter in causing the omissions and misrepresentations, the district court concluded as a matter of law that the firm was liable under the common law doctrine of respondeat superior for the actions of its employee, Higgins. Sharp v. Coopers & Lybrand, 457 F.Supp. 879, 891 (E.D.Pa.1978). The court also concluded as a matter of law that Coopers & Lybrand was liable for Higgins' conduct by virtue of § 20(a) of the 1934 Act, 15 U.S.C. § 78t(a). 457 F.Supp. at 893-94.

A second jury trial on the issue of damages sustained by the three named plaintiffs was held from May 12 through May 19, 1980. This jury found, also by special interrogatories, that the three class representatives had exercised due diligence, had filed their claims within one year of the date on which the fraud reasonably could have been discovered, and would not have invested in the partnerships absent the opinion letter. The jury also found that the "actual value" of a one-eighth limited partnership share in 1971 was $1,240.00. Based on these findings, the trial judge entered judgment in favor of Stanley L. Sharp in the amount of $6,885, Sam Geftic in the amount of $3,442.50, and H. James Conaway, Jr., in the amount of $6,193. Each representative also received prejudgment interest at the rate of six percent from December 31, 1971.

In this appeal, Coopers & Lybrand advances four principal arguments.4 First, it argues that the trial judge erred in concluding after the first trial that Coopers & Lybrand was secondarily liable for the acts of its employee, Herman Higgins, under the common law doctrine of respondeat superior. Second, it argues that the trial judge erred by failing to give proper jury instructions on § 20(a) of the 1934 Act. Third, it argues that the trial court erred during the second trial in creating a presumption that the class representative relied on the misrepresentations and omissions of the second opinion letter. Finally, it attacks the measure of damages used by the district court in the second trial.

Appellant's primary argument is that it may not be held liable for the actions of Higgins in light of the jury's finding, in answer to a special interrogatory, that no Coopers & Lybrand partner possessed the necessary scienter to violate rule 10b-5. See Ernst & Ernst v. Hochfelder, 425 U.S. 185, 201-11, 96 S.Ct. 1375, 1384-1389, 47 L.Ed.2d 668 (1976). The trial court rejected this argument, holding as a matter of law that Coopers & Lybrand was liable for the securities law violations of Herman Higgins committed in the scope of his employment. The district court noted that this court has not recognized respondeat superior as a basis for establishing secondary liability under rule 10b-5, see Rochez Brothers, Inc. v. Rhoades, 527 F.2d 880, 884 (3d Cir. 1975) (Rochez II ), cert. denied, 425 U.S. 993, 96 S.Ct. 2205, 48 L.Ed.2d 817 (1976), but reasoned that this case presents a sufficiently egregious situation to distinguish it from the prevailing case law. It relied on a "broker-dealer" exception alluded to in Rochez II, id. at 886, reasoning that an accounting firm issuing a tax opinion letter to be used in selling securities is laden with the same public trust as a broker-dealer. 457 F.Supp. at 891.

As an alternative basis for imposing liability on Coopers & Lybrand, the district court relied on § 20(a) of the 1934 Act. The district court declined appellant's invitation to instruct the jury to determine whether Coopers & Lybrand had "culpably participated" in the securities law violations. See Rochez II, 527 F.2d at 884-85. The court reasoned that an adequate instruction on culpable participation would have been difficult to construct, that the facts regarding the issue were not in dispute, and that the proper manner of resolving the issue was by the court as a matter of law. 457 F.Supp. at 893.

The starting point in this court for analysis of secondary liability under rule 10b-5 is Rochez II. In that case we held that material omissions in violation of rule 10b-5 by a purchaser serving as Chairman of the Board of Directors, Chief Executive Officer, and President of a corporation, to the seller of a substantial block of stock in the corporation, could not be attributed to the corporation involved under principles of respondeat superior. Rochez II, 527 F.2d at 885-86; see also Rochez Brothers, Inc. v. Rhoades, 491 F.2d 402, 405-06 (3d Cir. 1974) (Rochez I ). The court examined § 20(a), which imposes liability on certain controlling persons, and concluded that imposition of respondeat superior liability would circumvent the good faith defense provided in that section. 527 F.2d at 885. The court noted:

If we were to apply respondeat superior as appellant wishes, we would in essence impose a duty on a corporation to supervise and oversee the activities of its directors and employees when they are dealing with their own corporate stock as individuals, and not for the corporation or for the benefit of the corporation. To impose such a duty would make the corporation primarily liable for any security law violation by any officer or employee of the corporation. We believe that Congress did not intend to expand liability to this degree when it passed the Securities Exchange Act.

Id. Later, in Gould v. American-Hawaiian Steamship Co., 535 F.2d 761, 779 (3d Cir. 1976), the court declined to impose secondary liability on two corporations, the joint director of which had failed to disclose a conflict of interest in a proxy statement issued prior to a merger. In doing so, the court stated that "(t)he liability of Litton and Monroe for the acts of Casey cannot be upheld on the agency theory utilized by the district court." Id. at 779. This statement operates as an explicit reaffirmation of Rochez II.

Relying on Rochez II and Gould, Coopers & Lybrand argues that the district court erred by imposing secondary liability on it under a respondeat superior theory. The Securities and Exchange Commission, as amicus curiae, argues that Rochez II should be narrowly construed to allow the use of respondeat superior in private damage actions under rule 10b-5, or alternatively should be overruled.5 Appellees, although suggesting that Rochez II is no longer viable, propose a less drastic course of retaining the general prohibition against the use of respondeat superior in rule 10b-5 actions while adopting an exception for accounting firms that render investment-oriented opinion letters. We conclude that the rule announced in Rochez II is and should be viable, that Rochez II envisioned exceptions to that rule, and that this record supports the district court's determination that Coopers & Lybrand is liable for the wrongful acts of Higgins under the precise exceptions set forth therein.

Our examination of the case law regarding employer liability under rule 10b-5 discloses that other courts have emphasized the special duties that certain employers assume under the federal securities laws when their conduct is likely to exert strong influence on important investment decisions. For example, the second circuit has recently relied on respondeat superior to impose liability on a brokerage house for the fraudulent conduct of a trainee in its employ. Marbury Management, Inc. v. Kohn, 629 F.2d 705 (2d Cir.), cert. denied, -- U.S. --, 101 S.Ct. 566, 66 L.Ed.2d 469 (1980). In holding that the plaintiff could not recover from the employer, the district court found that the employer had exercised due care in its supervision of the employee. The second circuit reversed the judgment in favor of the employer and remanded for a new trial, adopting principles of respondeat superior in this limited context. Id. at 716. Prior second circuit decisions had expressly declined to apply respondeat superior in other contexts,6 however, and the only distinction between those decisions and Marbury Management seems to be the court's conclusion that a brokerage firm owes a higher duty to its customers than do other employers.7 The implicit reasoning in Marbury Management that brokers have a higher public duty under the securities laws than do other persons leads to imposition of a duty to exercise a high standard of supervision. This duty is enforceable through imposition of secondary liability based on respondeat superior.

Other courts of appeals that have employed respondeat superior have done so in cases in which the employee was a high level officer or director of the employer,8 the employer was a brokerage firm,9 or both.10 In many of these decisions, the courts have emphasized the public trust of the firms involved, and the duty to supervise arising therefrom.11 These decisions are consistent with our decision here. We draw support from them for our conclusion but emphasize strongly that the doctrine of respondeat superior, though applicable in this case, has not been and should not be widely expanded in the area of federal securities regulation. See Rochez II, 527 F.2d at 885.

Aaron v. SEC, 446 U.S. 680, 100 S.Ct. 1945, 64 L.Ed.2d 611 (1980), decided soon after Marbury Management, does not compel a different result. In Aaron, the Court reversed the second circuit's decision holding that the SEC need prove only negligence to obtain an injunction for violations of rule 10b-5. Aaron, a management employee at a brokerage firm, had been informed that two employees under his supervision were making gross misrepresentations in their sales of a particular security. Aaron did nothing about the complaint other than inform one of the employees about it and direct him to contact the complainant. The fraudulent conduct continued. The district court held that Aaron had "intentionally failed to discharge his supervisory responsibility," and issued an injunction. The second circuit affirmed, holding that Aaron had acted negligently in supervising the two employees, and declining to reach the question whether the record supported a finding of scienter. The Supreme Court reversed, holding that the SEC, like private litigants, must prove scienter as an element when it sues for an injunction under § 10(b) and rule 10b-5. In Aaron, the Court dealt with a supervisory employee rather than an employer, and this distinction is important for purposes of applying respondeat superior. See SEC v. Coffey, 493 F.2d 1304, 1315 (6th Cir. 1974), cert. denied, 420 U.S. 908, 95 S.Ct. 826, 42 L.Ed.2d 837 (1975). We conclude that Aaron does not foreclose imposition of secondary liability in the situation before us. We now address the adjudicative facts upon which the decision in this case turns.

Here, two opinion letters were issued by Coopers & Lybrand. The first went out on July 21, 1971, to WMC and was signed in the firm name by a partner. The letter was in response to a request by WMC in behalf of one of its investors. The second letter went out in October, 1971, after a partner, Wright, had learned that WMC was showing copies of the letter to investors generally as part of WMC's sales program. With full knowledge of the letter's intended use a tool to be used by a securities seller as part of a sales program the partnership, through a partner, made the calculated decision to send out a more complete letter. Moreover, it was also decided that the letter be signed, not in the name of a partner, but in the partnership name. These facts are central to the important inquiry, whether this activity propelled Coopers & Lybrand into a position in which the investing public would place their trust and confidence in it. We determine that it did ascend to that position, and the ultimate issue turns on this determination.

In Chiarella v. United States, 445 U.S. 222, 100 S.Ct. 1108, 63 L.Ed.2d 348 (1980), the Supreme Court emphasized that the petitioner, a printer who had abused inside information obtained from financial statements delivered to him for printing, "had no prior dealings with (the investors). He was not their agent, he was not a fiduciary, he was not a person in whom the (investors) had placed their trust and confidence." Id. at 232, 100 S.Ct. at 1116. By contrast here, a realization, even an expectation, by Coopers & Lybrand that buyers of partnership interests would "place( ) their trust and confidence" in the information provided them required appellant to exercise a high degree of care in preparation of the letter, and this care included close supervision of its employee, Higgins. We recognized in Rochez II that situations can arise in which corporations owe a duty of careful supervision to the public:

We recognize that corporations do have certain duties imposed on them for protection of public interest. To exact this duty on a mere showing of a principal-agent relationship would violate the legislative purpose and effectively nullify the "controlling person" provision of the Act. See also Lanza v. Drexel & Co., 479 F.2d (1277, 1299 (2d Cir. 1973) (in banc)).

We are not faced with the type of relationship that prevails in the broker-dealer cases where a stringent duty to supervise employees does exist. This duty is imposed to protect the investing public and make brokers aware of the special responsibility they owe to their customers. We can find no reason to impose this same duty in a situation like the one presently before us where the parties were dealing for themselves and for their own accounts.

527 F.2d at 885-86 (footnotes omitted). In this situation, the absence of actual knowledge of or reckless disregard for material omissions or misrepresentations should not insulate Coopers & Lybrand from liability because the expectation that investment decisions would be made on the basis of the opinion letter required the firm to exercise a "stringent duty to supervise" its employees in drafting and issuing the letter. Under the circumstances of this case viz, the existence of actual knowledge that the firm's letter would influence the investing public, we cannot distinguish the responsibility of the accounting firm here from the broker-dealer discussed in Rochez II.

A simple garden variety master-servant relationship is not what activates the doctrine of respondeat superior in this case. Rather, the presence of what we described in Rochez II as a "stringent" or high duty to supervise employees triggers the doctrine. At common law, the misrepresentations of an employee, even those made fraudulently without the employer's knowledge, will render any employer liable for damages to persons experiencing harm as a result. Restatement (Second) of Agency § 249; id., comment c; § 257; § 261; § 264. Our decision does not go so far.

When the firm's public representations are designed to influence the investing public, the firm should not be shielded from compensating persons who suffered from reckless or knowing acts by its employees. Otherwise, it could immunize itself from liability by constructing a "Chinese wall" between its employees and partners, allowing only the former to draft opinion letters. Partners, with their greater experience and knowledge, would have a strong incentive to avoid using their expertise to benefit the investors to whom opinion letters are directed. See Johns Hopkins University v. Hutton, 297 F.Supp. 1165, 1213 (D.Md.1968), aff'd in part, rev'd in part, 422 F.2d 1124 (4th Cir. 1970). This incentive can be reversed only by recognizing an absolute duty on the part of the firm, which acts through its partners, to supervise employees closely whenever its representations are designed to influence the investing public. Protection of investors is, after all, the primary purpose of the securities laws. Ernst & Ernst, 425 U.S. at 195, 96 S.Ct. at 1382.12

We recognize that accounting firms perform a valuable service in evaluating, synthesizing, and explicating complex financial data. In recognition of these important services, we emphasize the limited scope of our holding. See Ernst & Ernst, 425 U.S. at 216 n.33, 96 S.Ct. at 1392 n.33 (expressing reluctance to "extend to new frontiers the 'hazards' of rendering expert advice" under the 1934 Act). Coopers & Lybrand contracted to draft and issue an opinion letter with notice that it would be used to influence the decisions of investors. In this limited situation, the accounting firm owes a responsibility to the investing public to exercise stringent or high supervision of its employees, and failure to perform this duty will expose it to liability for their violations of rule 10b-5 under the doctrine of respondeat superior.

The trial court also held Coopers & Lybrand liable for the acts of Higgins under § 20(a) of the 1934 Act:

Every person who, directly or indirectly, controls any person liable under any provision of this chapter or of any rule or regulation thereunder shall also be liable jointly and severally with and to the same extent as such controlled person to any person to whom such controlled person is liable, unless the controlling person acted in good faith and did not directly or indirectly induce the act or acts constituting the violation or cause of action.

15 U.S.C. § 78t(a). We have previously interpreted this section as imposing "secondary liability on one who controls a violator of the securities laws, and who fails to show he acted in 'good faith.' " Rochez II, 527 F.2d at 889. One element of any case imposing liability under § 20(a) is "culpable participation" in the securities violation. Id. at 885.13 To impose secondary liability on a controlling person for his inaction, the plaintiff must prove that the inaction "was deliberate and done intentionally to further the fraud." Id. at 890.

In this case, the trial judge declined to give a jury instruction on "culpable participation," instead resolving the issue himself as a matter of law. 457 F.Supp. at 893. The court relied on our decision in Straub v. Vaisman & Co., 540 F.2d 591 (3d Cir. 1976), in which we affirmed a trial judge's findings of fact and conclusions of law holding a securities brokerage firm liable for the fraudulent acts of an employee. We there stated that "(t) he company's active participation in the scheme, its receipt of the benefits and its status as a broker-dealer fully justify imposition of liability. It is clear that VaisCo's role was not merely that of a facade for fraud, but rather one of a culpable confederate." Id. at 596 (citing Rochez II ).

The procedural context of Straub, as compared to the present appeal, is significant. Straub was an appeal from a bench trial in which the judge had found sufficient facts to meet the test of "culpable participation." In contrast, this case arises from a jury trial, and even though the trial judge found culpable participation, he invaded the province of the jury by doing so. The proper arbiter of this factual dispute was the jury. We hold that the trial court erred in failing to instruct the jury in accordance with the law of this court.

Although we find error, it does not require retrial of the liability issue. The preceding analysis imposing a duty of careful supervision on Coopers & Lybrand, enforceable through respondeat superior, adequately justifies affirming the district court's determination that Coopers & Lybrand was liable. The error occasioned by improper application of § 20(a) is therefore harmless.

Appellant next argues that the trial judge erred in his instruction to the jury on reliance and in formulating the special interrogatory to the jury on that issue. The trial judge instructed the jury that the appellees were entitled to a rebuttable presumption of reliance.14 In the special interrogatories to the jury regarding reliance, the trial court asked with respect to each named plaintiff: "Would (the plaintiff) have purchased his investment in the 1971 WMC Ohio Program even if the truth had been told in defendant's opinion letter?" App. at 2814a. The jury answered in the negative with regard to each class representative. Appellant argues first that the court erred in creating a presumption of reliance favoring the appellees, and second that the effect of the interrogatory was to create an irrebuttable presumption of reliance. We conclude that in the circumstances presented by this case, which involves both misrepresentations and omissions, the district court correctly allowed a presumption of reliance. In addition, we find no support for appellant's assertion that the trial court employed an irrebuttable presumption. We therefore reject both arguments.

Reliance is an element of a plaintiff's action for damages under rule 10b-5. See Straub v. Vaisman & Co., 540 F.2d at 596; Landy v. FDIC, 486 F.2d 139, 170 (3d Cir. 1973), cert. denied, 416 U.S. 960, 94 S.Ct. 1979, 40 L.Ed.2d 312 (1974); see also Thomas v. Duralite Co., 524 F.2d 577, 583 (3d Cir. 1975).15 The obvious reason for this requirement is that a plaintiff in a rule 10b-5 action should not be allowed to recover damages when the defendant's wrongful action had no relationship to the plaintiff's loss. See 3 A. Bromberg, Securities Law § 8.7, at 213 (1979). Reliance is therefore one aspect of the ubiquitous requirement that losses be causally related to the defendant's wrongful acts. Rochez I, 491 F.2d at 410; see Note, The Reliance Requirement in Private Actions Under SEC Rule 10b-5, 88 Harv.L.Rev. 584, 587 (1975); see also Prosser, The Law of Torts § 41, at 236 (4th ed. 1971).16 This precept is manifest in the Securities Exchange Act of 1934 in § 28(a), 15 U.S.C. § 78bb(a), which states in part that "no person permitted to maintain a suit for damages under the provisions of this chapter shall recover, through satisfaction of judgment in one or more actions, a total amount in excess of his actual damages on account of the act complained of." Normally, a plaintiff suing under rule 10b-5 bears the burden of proving all the elements of his case. See McLean v. Alexander, 599 F.2d 1190, 1196-97 (3d Cir. 1979). Nevertheless, the necessity of an element to a valid claim does not determine the allocation of the burdens of going forward and persuasion with respect to that element. See McCormick's Handbook of the Law of Evidence § 337, at 785-86 (Cleary ed. 1972).

The Supreme Court authoritatively addressed the requirement of proving reliance in rule 10b-5 actions in Affiliated Ute Citizens of Utah v. United States, 406 U.S. 128, 92 S.Ct. 1456, 31 L.Ed.2d 741 (1972), stating:

Under the circumstances of this case, involving primarily a failure to disclose, positive proof of reliance is not a prerequisite to recovery. All that is necessary is that the facts withheld be material in the sense that a reasonable investor might have considered them important in the making of this decision. This obligation to disclose and this withholding of a material fact establish the requisite element of causation in fact.

Id. at 153-54, 92 S.Ct. at 1472 (citations omitted). The Court has subsequently defined a material omission in the context of proxy statements under rule 14a-9 as "a substantial likelihood that a reasonable (investor) would consider it important (T)here must be a substantial likelihood that the disclosure of the omitted fact would have been viewed by the reasonable investor as having significantly altered the 'total mix' of information made available." TSC Industries, Inc. v. Northway, Inc., 426 U.S. 438, 449, 96 S.Ct. 2126, 2132, 48 L.Ed.2d 757 (1976). We have held this standard of materiality applicable to rule 10b-5 actions as well. See Healey v. Catalyst Recovery of Pennsylvania, Inc., 616 F.2d 641, 647 (3d Cir. 1980); id. at 653 (Aldisert, J., dissenting).17 Affiliated Ute makes clear that in at least some situations a presumption of reliance in favor of the rule 10b-5 plaintiff is proper. See Rochez I, 491 F.2d at 410. Our present task is to determine whether that presumption was properly applied in this case.

Both parties in this case cite decisions indicating that the presumption of reliance is proper in cases of alleged omissions, whereas no presumption arises in cases of alleged misrepresentations. See, e. g., Vervaecke v. Chiles, Heider & Co., 578 F.2d 713, 717-18 (8th Cir. 1978); Rifkin v. Crow, 574 F.2d 256, 262-63 (5th Cir. 1978); Schlick v. Penn-Dixie Cement Corp., 507 F.2d 374, 380-81 (2d Cir. 1974), cert. denied, 421 U.S. 976, 95 S.Ct. 1976, 44 L.Ed.2d 467 (1975). This distinction has led Coopers & Lybrand to argue that its wrongful conduct in this case arose from misrepresentations in the opinion letter, whereas the class representatives argue that the violation resulted from the appellant's failure to disclose certain material facts.

We have concluded that both misrepresentations and omissions are present in this case. The jury heard evidence that Coopers & Lybrand had misrepresented certain crucial facts in the letter, such as its disclaimer of verification of the facts on which the opinion letter was based and its assertion that cash supplemental to the limited partner's contributions would be borrowed "from suitable banks or other lending agencies " App. at 2892a, 2894a. The jury also heard evidence that Coopers & Lybrand had failed to disclose certain material facts, such as the affiliation between the putative lender, the Bahamian bank, and WMC. App. at 1202a, 938-39a. A strict application of the omissions-misrepresentations dichotomy would require the trial judge to instruct the jury to presume reliance with regard to the omitted facts, and not to presume reliance with regard to the misrepresented facts. Although this resolution would have great appeal to graduate logicians in a classroom, we are not persuaded to adopt it for use in a courtroom.

We begin by embracing the obvious proposition recently stated by the second circuit: "We therefore presume reliance only 'where it is logical' to do so." Lewis v. McGraw, 619 F.2d 192, 195 (2d Cir. 1980) (action under § 14(e) of the Williams Act, 15 U.S.C. § 78n(e)), cert. denied, -- U.S. --, 101 S.Ct. 354, 66 L.Ed.2d 214 (1980). A steadfast rule requiring the defendant to refute a presumption of reliance would be neither equitable nor logical. The plaintiff traditionally assumes the burden of demonstrating causation. See Eriksson v. Galvin, 484 F.Supp. 1108, 1126 (S.D.N.Y.1980); see also 3 A. Bromberg, Securities Law § 8.7, at 213 (1979); Prosser, Handbook of the Law of Torts § 41, at 241 (4th ed. 1971); but see Collins v. Signetics Corp., 605 F.2d 110, 114 (3d Cir. 1979) (§ 11(e) of the 1933 Act, 15 U.S.C. § 77k(e), shifts burden on causation to defendant). Only in unusual circumstances is this burden shifted from the plaintiff to the defendant. The reason for shifting the burden on the reliance issue has been an assumption that the plaintiff is generally incapable of proving that he relied on a material omission. Lewis v. McGraw, 619 F.2d at 195; Thomas v. Duralite Co., 524 F.2d at 585.18 This incapacity arises from the difficulty of proving a speculative state of facts: Had the facts not been omitted, would plaintiff have acted on the information made available and thereby averted his loss? But this observation does not justify a clear distinction between the treatment of misrepresentations and omissions. First, the defendant confronts the same problem of speculation in trying to refute the presumption of reliance because he possesses no more information on the plaintiff's hypothetical behavior than does the plaintiff. Second, the problem of speculation is not unique to situations in which omissions have occurred. In misrepresentation actions as well, proof of reliance requires a degree of speculation on the action that the plaintiff would have taken had no misrepresentation occurred. Therefore, we are unpersuaded that the existence of misrepresentations and omissions, without more, necessitates any particular treatment of the reliance issue.

A further consideration guides our decision. If we were to follow blindly the omission-misrepresentation distinction in this case, we would be compelled to require a dual jury instruction. The jury would be instructed to search for proof of reliance by the plaintiff with regard to the misrepresentations, and to search for proof of nonreliance by the defendant with regard to the omissions. The problems with such a complicated approach before a lay jury are legion. We conclude that the proper approach to the problem of reliance is to analyze the plaintiff's allegations, in light of the likely proof at trial, and determine the most reasonable placement of the burden of proof of reliance. See Affiliated Ute, 406 U.S. at 153-54, 92 S.Ct. at 1472 ("positive proof of reliance" not required "under the circumstances of this case") (emphasis added). Such a flexible approach avoids the potential problems of a broad judicial pronouncement of a precept governing reliance.19

We agree with the district court that the burden in this case should fall on Coopers & Lybrand. The opinion letter issued by Coopers & Lybrand was intended to influence the investment decisions of persons interested in WMC partnerships. The appellant undoubtedly foresaw that it would have that effect.20 As in Affiliated Ute, Coopers & Lybrand by its action facilitated the transactions at issue but failed to disclose certain facts. Its misrepresentation of other facts should not alleviate its burden of proving nonreliance. Considering the likelihood that investors would rely on the opinion letter, we conclude that the trial judge properly placed the burden of refuting a presumption of reliance on the appellant.

Although we agree with appellant that the presumption of reliance, when appropriate, is rebuttable, see Thomas v. Duralite Co., 524 F.2d at 585, we reject its argument that the interrogatories made the presumption irrebuttable. The interrogatories on reliance did not, to be sure, emphasize this point, but if the trial judge had intended to make the presumption irrebuttable, no interrogatories on reliance would have been submitted. In addition, the jury instructions made crystal clear that the presumption was rebuttable. See n. 14, supra. We therefore find no error in the trial court's treatment of reliance.21

The fourth issue is the proper measure of damages. The trial judge adopted an "out-of-pocket" theory of damages, which he defined in his opinion as "the difference between the price (plaintiffs) paid for their investment and what it was actually worth at the time of purchase." Sharp v. Coopers & Lybrand, 83 F.R.D. 343, 347 (E.D.Pa.1979). He explained this formulation in the jury instruction by stating:

We are asking you to find what was the actual value of this investment in 1971. You heard Mr. Wilson testify that he would have projected in 1971 400 million cubic feet of gas and oil, and that that would give a return of somewhere around $7900 over a nine-year period.

However, when he adopted the same method of Mr. Templeton, who took the average production over five years, he came out with 250 million cubic feet of gas and oil.

I am now instructing you to disregard Mr. Wilson's testimony as to the 400 million cubic feet because the test is: What was the actual value in 1971, not its theoretical value in 1971.

So, you will disregard that, and you will make your choice on the actual value of these investments between the testimony of Mr. Whitman who in effect said it had no value because it was not saleable and that he would not advise anybody to touch it, the testimony of Mr. Templeton who gave you a figure of $1240, and the testimony of Mr. Wilson who gave you a figure of $4,080 for a 1/8 interest but who also said that if he knew of fraud, if he knew that the General Partner was dishonest, that he wouldn't touch it with a 10-foot pole.

App. at 2691a-92a (emphasis added). The interrogatory submitted to the jury on the issue of damages asked "(w)hat was the actual value, if any, of a 1/8 limited partnership interest in a 1971 WMC Ohio Producers Program as of late December, 1971?" App. at 2815a. The jury adopted Mr. Templeton's figure of $1,240. Id. We conclude that this formulation of damages was incorrect, and we therefore reverse and remand for a new trial on the damages issue.

The Supreme Court addressed the measure of damages in rule 10b-5 actions in Affiliated Ute, concluding that "the correct measure of damages under § 28 of the Act, 15 U.S.C. § 78bb(a), is the difference between the fair value of all that the seller received and the fair value of what he would have received had there been no fraudulent conduct " 406 U.S. at 155, 92 S.Ct. at 1473 (citations omitted). To paraphrase our previous interpretation of this passage, "the measure of damages is the difference between what the (buyer paid) for his stock and what he would have (paid) had there been no fraudulent conduct." Thomas v. Duralite Co., 524 F.2d at 586. See also Huddleston v. Herman & McLean, 640 F.2d at 554-555 (1981); Glick v. Campagna, 613 F.2d 31, 36 (3d Cir. 1979). Appellant argues that the trial court's formulation places it in the position of insuring the appellees' investment. It argues that the proper measure of damages is the additional tax that the appellees had to pay when the IRS denied the deductions that the opinion letter forecast.

We agree with the appellant's argument in part. By requiring the jury to find the "actual value" of the investments in 1971, and by allowing them to consider subsequent production data unknown to anyone at the time the investment was made, the court neglected the speculative nature of the investment at the time the appellees bought it. Appellees found the investment attractive for two reasons. First, it offered a potential for future appreciation and income if the oil wells were successful. The nature of the petroleum exploration business necessarily renders this potential uncertain. Second, it offered an opportunity for rapid deduction of twice as much money from the investor's tax return as he actually invested. This aspect of the investment was less speculative, based as it was on the advice rendered in the October, 1971, opinion letter.

Our goal in formulating a damage instruction must be to compensate appellees precisely for the damage directly resulting from the appellant's wrongful acts. Investors viewing the oil drilling opportunities offered by WMC in 1971 might have considered the probability of a substantial monetary return to have been great, even if they had access to all information available at that time. But by allowing the jury to consider subsequent production figures from the wells, the risk inherent in the investments when made in 1971 was removed. If the opinion letter contained omissions and misrepresentations impacting only on the validity of the tax opinion, the proper measure of damages is the amount of money invested minus the value of the speculative investment at the time of purchase, derived from all information available to the investors at the time, without considering the predicted tax benefits.22 The record before us does not demonstrate conclusively that this formula would encompass all components of allowable recovery. Accordingly, we remand for a new trial on damages.

We believe that on remand the parties should be more discriminating in advancing or defending various damage theories. Moreover, the parties should request the court to submit special interrogatories that break down the damage award more specifically than the ones used in the previous trial. Thus, if the appellees can establish at the damages retrial that Coopers & Lybrand knew, by knowledge imputed from Higgins, that the money mentioned in the opinion letter would not actually be borrowed by WMC or would not actually be invested, that is a fact that would affect the speculative value of the investment. They would be able to demonstrate that without the expenditures for drilling, the risk of no long term return was greatly increased. The proper measure of damages in this situation would be more than the tax loss; it would be the amount of money paid by each investor minus the value the investment would have had if Coopers & Lybrand had disclosed all it knew through Higgins. The subsequent returns on the investment, of course, would have to be deducted from the difference obtained. On the other hand, the investors may have been provided with the most accurate information on the oil production potential of their partnership shares. If so, subsequent events rendering that information invalid should not allow the investors to deem a party who had no connection with the investment information the effective insurer for risks that the investors knowingly took.

Thus, in our view, it is central to a proper determination of damages that the evidence control the specific interrogatories to be submitted to the jury. In any event, it is important to separate concepts of possible investment loss from the possibility that the only loss sustained related to tax advantages.

Appellant also raises a number of issues that require only abbreviated consideration.

The first of these issues is that the action is barred by the statute of limitations. The three named class representatives reside in different states. Sharp resides in Pennsylvania, Geftic in New Jersey, and Conaway in Delaware. At trial, the parties and the trial judge apparently assumed that the one year limitation in § 504(a) the Pennsylvania Securities Act, Pa.Stat.Ann. tit. 70, § 1-504(a) (Purdon Supp.1980), applied to this action. The appellees' theory on statute of limitations was "reasonable discovery," and they date their discovery as sometime after May 7, 1974. Suit was filed on May 4, 1975. The jury answered the interrogatories relevant to the three individual plaintiffs favorably to the plaintiffs, and Coopers & Lybrand moved for judgment n. o. v. We do not reach appellant's argument that the jury's answer to the interrogatory on the reasonable discovery issue was against the manifest weight of the evidence. Instead, we conclude that a six year statute of limitations is applicable.

This court has recently examined the applicable statute of limitations for rule 10b-5 cases in two decisions. The first, Roberts v. Magnetic Metals Co., 611 F.2d 450 (3d Cir. 1979), held that a rule 10b-5 action arising in New Jersey would be governed by the six year statute of limitations for common law fraud rather than the statute applicable to actions under the New Jersey Uniform Securities Act. In Biggans v. Bache Halsey Stuart Shields, Inc., 638 F.2d 605 (3d Cir. 1980), we held that the six year statute of limitations for common law fraud,23 rather than the one year limitation in § 504 of the Pennsylvania Securities Act, applied to a rule 10b-5 action for "churning" that arose in Pennsylvania. The court reasoned that the Pennsylvania securities statute grants a private remedy to a buyer only against his seller. It concluded that an action against a broker would not lie under the statute, and that the most analogous state action was an action for common law fraud. 638 F.2d at 610.

Biggans is not distinguishable from this case. As in Biggans, the defendant here is not the seller, and a state securities action will not lie. The only remedy under state law would appear to be common law fraud, and the six year statute governs. Thus, Coopers & Lybrand's argument that the plaintiffs were on notice by October, 1973, is irrelevant, for suit was filed well within the statutory period.24

Coopers & Lybrand next argues that the trial judge abused his discretion in certifying a class action. It argues that hundreds of individual actions will be necessary even though the trial judge granted certification. This argument misses the point of class actions. If Coopers & Lybrand had been found not liable in the class action, all 240 potential suits would have been resolved in its favor. Since it lost that suit, the only remaining issue to be resolved on an individual basis is damages. The time savings are undeniable, and we see no abuse of discretion. See Wetzel v. Liberty Mutual Insurance Co., 508 F.2d 239, 245 (3d Cir.), cert. denied, 421 U.S. 1011, 95 S.Ct. 2415, 44 L.Ed.2d 679 (1975).

Appellant's attack on the district court's decision to bifurcate the trial is fully contingent on the class certification argument. If class certification were proper, subsequent trials would be required to determine the individual issues. In addition, the standard of review on this issue is abuse of discretion, see Idzojtic v. Pennsylvania R.R. Co., 456 F.2d 1228, 1230 (3d Cir. 1972), and we find none.

Appellant's next argument is that the trial court erred in awarding prejudgment interest at the rate of six percent to the individual plaintiffs. It contends that the interest award reflected a double recovery because in establishing damages the plaintiff's expert discounted the losses to reflect a 1971 value. Beginning with a calculation of $10,932, Elmer Templeton applied a discount factor of eight percent "(s)o that $10,932 which the investor in a whole well would receive but in drips and drabs over a period of five years was worth at the beginning of 1972 or the middle of 1972 $9,926." App. at 2129a. Templeton then divided $9,926 by eight to obtain $1,240 as the 1971 value of a one-eighth partnership share. Application of the discount factor was necessary to standardize the value of dollars actually invested in 1971 and the value of dollars that the investors could reasonably expect to receive in subsequent years; it was merely mathematical. In contrast, the prejudgment interest award was compensatory; it gave the plaintiffs some compensation for losing the use of the money prior to judgment. Subsequent application of a prejudgment interest rate, therefore, did not give plaintiffs a windfall; it merely allowed them some return on the money tied up in the partnerships.

With respect to the general standards for award of prejudgment interest, this court noted in Gould v. American-Hawaiian Steamship Co., 535 F.2d at 784-85, that the determination is to be based on fairness and is within the discretion of the district court. See Thomas v. Duralite Co., 524 F.2d at 589. Although Coopers & Lybrand is correct in arguing that it did not have the use of plaintiffs' money during the relevant time period, it is incorrect in arguing that this fact is determinative. To the extent Coopers & Lybrand was responsible for plaintiffs' losses, the trial judge did not abuse his discretion in awarding interest against it.

Coopers & Lybrand next argues that appellees failed to establish the requisite scienter, required in rule 10b-5 actions by Ernst & Ernst. This issue differs from the secondary liability issue in that Coopers & Lybrand is here arguing that the trial court applied an improper standard of scienter to the actions of Herman Higgins. In this circuit, the standard for scienter includes recklessness. See Coleco Industries, Inc. v. Berman, 567 F.2d 569, 574 (3d Cir. 1977) (per curiam), cert. denied, 439 U.S. 830, 99 S.Ct. 106, 58 L.Ed.2d 124 (1978). In McLean v. Alexander, we adopted the definition of recklessness applied to rule 10b-5 actions by the seventh circuit in Sundstrand Corp. v. Sun Chemical Corp., 553 F.2d 1033, 1045 (7th Cir.), cert. denied, 434 U.S. 875, 98 S.Ct. 224, 54 L.Ed.2d 155 (1977). We stated:

In Sundstrand Corp. v. Sun Chemical Corp., the Seventh Circuit stated the minimum threshold for liability under § 10(b) as follows:

"(R)eckless conduct may be defined as highly unreasonable (conduct), involving not merely simple, or even inexcusable negligence, but an extreme departure from the standards of ordinary care, and which presents a danger of misleading buyers or sellers that is either known to the defendant or is so obvious that the actor must have been aware of it."

In context, the Sundstrand formula was applied only to omissions, but the standard of liability proposed there is, we think, equally applicable to misstatements, and we approve it in both contexts.

McLean v. Alexander, 599 F.2d at 1197 (citation and footnote omitted). The trial judge tracked this language in his instruction to the first jury, which tried the class issues:

Reckless conduct may be defined as a highly unreasonable statement or omission involving not merely negligence or inexcusable negligence, but an extreme departure from standards of ordinary care, presenting a danger of misleading buyers that is either known to the defendant or is so obvious that the defendant must have known of it.

App. at 1614a. In the interrogatories submitted to the first jury, he asked: "(D)o you find that Herman Higgins, acting in the scope and course of his duties as an employee of defendant, caused with scienter, that is, intent to defraud or recklessness as I have defined it to you: (a) the material misrepresentations; (b) the material omissions." App. at 1648a (emphasis added). The jury answered both questions "yes."

We interpret appellant's other argument on scienter to be a convoluted adjunct to its secondary liability argument. Appellant predicates the scienter argument on the jury's finding in the first trial that "it (was not) foreseeable to the defendant that purchasers of Westland Securities would be likely to suffer injury or damages as a result of misrepresentations " Referring to the definition of recklessness in McLean, Coopers & Lybrand argues that recklessness was not shown. It is absolutely correct, but only if appellees had to show that appellant acted recklessly in causing the rule 10b-5 violations. As we have concluded previously, although Coopers & Lybrand did not participate in the violations, it owed a duty of careful supervision to the investors, enforceable by the common law doctrine of respondeat superior, and therefore it may be liable for the reckless actions of its employee.

The argument that Higgins did not possess the requisite scienter is frivolous. Coopers & Lybrand relies on a deposition taken from Higgins after he had testified before the grand jury. It concedes that many of the statements made in the deposition are inconsistent with statements made before the grand jury. See Appellant's Reply Brief at 6. In the deposition, Higgins attested to his own good faith and pure motives, explaining away the inconsistencies by stating that he did not have his diary present when he testified before the grand jury. A purer question of credibility has rarely been formulated, and the jury simply found Higgins' grand jury testimony more credible.

Appellant's due diligence argument essentially states that the plaintiffs acted unreasonably in failing to discover the misrepresentations and omissions and that this failure should absolve it of liability. The trial judge propounded interrogatories to the jury on this issue with regard to each class representative, and the jury resolved them against appellant. App. at 2814-15. The interrogatories belie the further assertion that the trial judge improperly shifted the burden of proof on the due diligence issue to appellant.

For the foregoing reasons, we will vacate the award of damages and remand to the district court for a new trial on that issue. The judgment of the district court in all other respects will be affirmed.

Each side to pay its own costs.