This action, brought by employees who had participated in individual account pension plans maintained by their employer Unisys Corporation ("Unisys"), charged essentially that Unisys and the individual defendants1 had breached their fiduciary duties prescribed by ERISA2 by investing in Executive Life Insurance Guaranteed Investment Contracts ("GICs"). After a bench trial, the District Court granted judgment for all defendants, ruling in their favor on all issues related to the breach of fiduciary duty claims, thus denying all relief and damages to the plaintiff class of Unisys employees.3

We have reviewed the trial record independently and conclude that in all material respects, the District Court's findings of fact were not clearly erroneous, the District Court did not abuse its discretion in excluding Dr. Gottheimer's testimony, and the District Court did not err in making its conclusions of law. We also conclude that while we do not agree with some of the holdings of the District Court, those disagreements have no effect on the District Court's overall holding of prudence, and that therefore, the orders entered by the District Court on November 24, 1997 and January 9, 1998 will be affirmed.

This is the second appeal from the District Court's rulings in this case. Initially, summary judgment was entered in favor of Unisys, but on review, we remanded for trial. In re Unisys Savings Plan Litig., 74 F.3d 420 (3d Cir.), cert. denied, 519 U.S. 810, 117 S. Ct. 56, 136 L. Ed. 2d 19 (1996) ( "Unisys I "). Almost all of the background facts and details are found in our earlier opinion and we refer here only to those matters particularly relevant to the arguments made by the parties on appeal.

Now that the District Court has rendered its findings of fact and conclusions of law in favor of the Unisys,4 Meinhardt complains that the standard used by the District Court was improper, that Meinhardt's expert's testimony was improperly excluded, and that Meinhardt suffered damages that were not recognized by the District Court. As indicated above, we hold that the District Court's essential factual findings were not clearly erroneous and that measured by the appropriate prudence standard of ERISA, the District Court properly concluded that Unisys did not breach any fiduciary duty. This threshold holding makes it unnecessary for us to discuss in detail the subsidiary issues raised on appeal by Meinhardt.

At the outset we call attention to the fact that the contours of the factual record have changed significantly since we last addressed the GIC issue presented in this case. At the earlier summary judgment stage, the prior panel of this court was obliged to assume that various facts presented by Meinhardt were true and that all inferences had to be drawn in Meinhardt's favor. Moreover, we assumed the report of Dr. Gottheimer, Meinhardt's proposed expert witness, would be admitted into evidence. As we discuss infra, the District Court properly excluded Dr. Gottheimer from testifying under Fed.R.Evid. 702.

As we have mentioned, at the conclusion of a ten-day bench trial, the District Court entered judgment for Unisys and issued extensive findings of fact and conclusions of law. We review the District Court's findings of facts for clear error. Application of legal precepts to historical facts receives plenary review. Feder v. Evans-Feder, 63 F.3d 217, 222 n. 9 (3d Cir.1995). The District Court's decision to qualify an expert is reviewed for abuse of discretion. General Elec. Co. v. Joiner, 522 U.S. 136, 118 S. Ct. 512, 139 L. Ed. 2d 508 (1997); In re Paoli R.R. Yard Litig., 35 F.3d 717, 741-46 (3d Cir.1994) ("Paoli II ").

Meinhardt participated in a 401(k) savings plan ("Plan")5 administered by Unisys. Two companies, Sperry and Burroughs, had merged in 1986 to create Unisys, and thus their retirement plans also merged. See Unisys I, 74 F.3d at 425-26. Meinhardt and the class he represents elected to invest their money in two of the funds offered by the Plan, the Fixed Income Fund ("FIF ") and the Insurance Contract Fund ("ICF "). Investments in the FIF/ICF were exclusively restricted to Guaranteed Investment Contracts ("GICs").6 For ease of reference throughout this Opinion we will refer to the FIF/ICF as the "Fund."

Unisys described the Fund as one of the more conservative funds that was intended to return principal with interest.7 Plan participants could invest as much or as little of their money as they desired. In order to prevent interest rate arbitrage by the participants and in order to receive a higher yield from the GICs, Fund assets could not be transferred to the money market fund directly; Fund assets had to be held for at least a year in one of the four equity stock funds. In 1990, Unisys shortened this "equity wash" to a period of six months after receiving approval from Executive Life.App. 1702.

In 1987 and 1988, Unisys purchased three Executive Life GICs as investments for the Fund. The Executive Life GICs themselves were comprised of assets that potentially had high rates of return, some of which were high yield bonds, or "junk bonds." The junk bonds did not fare well in the late 1980's markets. In 1991, California regulators seized Executive Life and temporarily froze all payments from the Executive Life GICs. By 1996, however, the Fund reimbursed the principal to Meinhardt and had paid some interest, but at a lower rate than had been guaranteed. FF p 80.

Meinhardt accuses Unisys of breaching its fiduciary duties of prudence, diversification, and disclosure with respect to the Executive Life contracts that had been purchased for the Fund. ERISA prescribes that fiduciaries must adhere to a standard of prudence. ERISA requires that a fiduciary's duty shall be discharged:

with the care, skill, prudence, and diligence under the circumstances then prevailing that a prudent man acting in a like capacity and familiar with such matters would use in the conduct of an enterprise with like character and with like aims;....

29 U.S.C. § 1104(a)(1)(B).

To determine whether Unisys had satisfied the ERISA standard of prudence, the district court found the following facts from the evidence.

Unisys delegated the responsibility of investing for the Fund to David White and Leon Level. White and Level have educational and practical backgrounds in financial matters. FF pp 17-18; App. 643-45, 1136-37. At the direction of White and Level, Unisys purchased three Executive Life GICs as investments for the Fund through bidding processes in June 1987, December 1987 (both for the FIF), and January 1988 (for the ICF). Of the many GICs purchased for the Fund, only three were Executive Life GICs, constituting between 15 and 20 percent of the Fund's assets. FF pp 14-15; App. 4450. Unisys, through White and Level, considered many firms through a process of competitive bidding, and Unisys heard in-person presentations from the insurance companies. FF p 20; App. 523, 1154.

Unisys presented evidence that for the first bid (June 1987), White and Level hired an experienced investment consultant, Murray Becker, who evaluated many different insurance firms.8 FF p 20. In evaluating potential insurance companies from which to purchase GICs, Becker obtained information and ratings from Standard and Poor's and A.M. Best, ratings services that evaluate the stability and potential profitability of various types of companies including insurance companies. Both ratings services gave Executive Life their highest rating.9 Becker and others testified that the ratings services were quite thorough because they analyzed raw data and interviewed investment managers. App. 535. Becker testified that he and other professionals had a high confidence in the thoroughness of the ratings services, which led to their respective ratings. FF p 21; App. 658-62, 702-03, 708-12, 1160-61. As detailed in the District Court's findings of fact, Becker was very familiar with how the ratings services evaluated the GICs. Id.

Prior to the first bid, Unisys reviewed Becker's bid specifications. FF p 20 n. 2; App. 1267-68, 3659. White testified that he carefully considered the use of high yield bonds and knew the risks involved. The prevailing view in the investment world at that time was that high yield guaranteed insurance contracts were good risks. FF p 27; App. 1038-39.

In addition to understanding the risks associated with Executive Life GICs, White chose Executive Life GICs for their other unique features. He selected the three Executive Life GICs to balance other Fund investments--the Executive Life GICs had longer maturity dates, App. 656-57, 692, and their portfolios lacked real estate mortgages or derivatives and had a low proportion of commercial real estate investments. App. 1010, 1156-58. Furthermore, Executive Life used the "barbell" approach in making investments for the GICs; that is, the higher risk associated with the high yield bonds was balanced with low risk, lower yield government bills. FF p 25; App. 1156-58.

For the second two bids and upon completion of the merger between Burroughs and Sperry into Unisys, Unisys decided not to utilize Becker's services. White, Level, and three other employees under their direction had sufficient professional experience to select GIC issuers. Moreover, Becker had charged the Plan $25,000 per bid. FF p 31; App. 1162-65. White and Level expanded the list of potential bidders for the second and third bids and maximized the market information available to them. FF p 32; App. 1054-55.

In the months between the bids, White and Level "engaged in an ongoing process of reviewing and updating the information on the potential bidders." FF p 33; App. 615-22. They testified that they also kept abreast of developments in the GIC industry by reading trade publications and journals. FF p 33 n. 6; App. 617, 632. They analyzed the portfolio and risk of the insurance carriers, using the Standard and Poor's and A.M. Best ratings as a source of information about both the insurance companies' asset composition and their creditworthiness. FF p 35; App. 1056. They testified that they would not have been able to replicate the analyses done by the ratings services inhouse. App. 711. They had available to them SEC forms 10K and 10Q to review prior to making their selection. FF p 35; App. 1065-66, 631. They had also consulted with a firm that Unisys had engaged to advise it on its defined benefit pension plan. That firm advised Unisys on its selection of issuers for the GIC funds. App. 608, 626-27.

In Unisys I, 74 F.3d at 436, we directed the District Court to determine whether the ratings could be used reliably by Unisys. As mentioned earlier, Level and White both testified that the ratings services were respected for their competence and thoroughness throughout the investment and finance community. App. 1160-61, 658-623, 702, 708-11. The District Court found that the ratings services provided reliable information. FF p 21. We are satisfied that this finding is not clearly erroneous. See also Bussian v. RJR Nabisco, 21 F. Supp. 2d 680, 686-87 (S.D.Tx.1998) ("Nabisco's use of the consultant and rating agencies [in selecting an Executive Life annuity] does not demonstrate imprudence.").

The District Court also resolved at trial four other issues of fact that we identified in Unisys I. First, evidence in the summary judgment record may have indicated that Becker had recommended the purchase of a three-year GIC, rather than a five-year GIC. Unisys I, 74 F.3d at 427. At trial, however, Becker testified that he had not recommended the shorter maturity (three-year) Executive Life GIC. Rather, his discussion with Unisys officers about maturity dates was very general, leading to the conclusion that the investment in five-year GICs was appropriate. FF p 30; App. 555-558.

Second, at the summary judgment stage, the evidence suggested that Becker told White that Executive Life would lose its AAA rating from Standard & Poor if its investments of junk bonds exceeded 35%, and White believed that 40-50% of Executive Life's investments were junk bonds. Unisys I, 74 F.3d at 427. At trial, Becker stated that it was not his view that Executive Life would lose their Standard and Poor's AAA rating if high yield bonds comprised more than 35% of Executive Life's portfolio. Rather, he testified that Executive Life thought its AAA rating might be in jeopardy if their high yield bond holdings exceeded 35%. However, Standard and Poor's had access to the actual percentage of high yield bonds in Executive Life's portfolio, and it still issued a AAA rating to Executive Life. Thus, White's reliance on the AAA rating was neither unreasonable nor imprudent. As discussed infra, even if White had an erroneous view of the percentage of high yield bonds in Executive Life's portfolio, a hypothetical prudent fiduciary would have known that Executive Life's high yield bond percentage was within an acceptable range in Executive Life's and Standard and Poor's analyses. FF p 28; App 501.10

[A] reasonable factfinder could infer from this evidence that Unisys failed to analyze the bases underlying [Becker's] opinion of Executive Life's financial condition and to determine for itself whether credible data supported [Becker's] recommendation that Unisys consider investing plan assets with the insurer. A reasonable factfinder could also conclude that Unisys passively accepted its consultant's positive appraisal of Executive Life without conducting the independent investigation that ERISA requires.

Unisys I, 74 F.3d at 435-36 (emphasis added). However, after taking evidence at trial, the District Court did not find that Unisys had failed to make its own evaluation of Executive Life's financial condition, nor did it find that Unisys "passively" had accepted Becker's appraisal. On the contrary, the District Court made numerous findings that White and Level were aware of the composition of assets of Executive Life GICs and that they understood the risks associated with each of those assets. FF PP 25-27.

Fourth, in Unisys I, we advised the District Court that it should determine the significance, if any, of Executive Life's substantially higher interest rates. We suggested that dramatically higher interest rates might have prompted Unisys to conduct extra investigation into Executive Life's creditworthiness. As stated earlier, at trial, White testified that he carefully had considered the use of high yield bonds and knew the risks involved. Unisys recognized that higher rates could, over time, yield dramatic differences in income, but that the trustees were "constrained by their standards of risk tolerance." FF p 25; App. 1250. The District Court credited the testimony of the Unisys fiduciaries and was satisfied and concluded that Unisys had made a reasonable and thorough investigation of Executive Life GICs. CL p 3. We will not disturb that holding here.11

The District Court's factual findings support its legal conclusions that Unisys was prudent under the standard articulated in ERISA and that a hypothetical prudent investor would have purchased each of the three GICs. As we stated in Unisys I, ERISA requires that a fiduciary shall discharge his duties

with the care, skill, prudence, and diligence under the circumstances then prevailing that a prudent man acting in a like capacity and familiar with such matters would use in the conduct of an enterprise with like character and with like aims;....

Unisys I, 74 F.3d at 434 (quoting 29 U.S.C. § 1104(a)(1)(B)). We also stated that the prudence requirement focuses on "a fiduciary's conduct in arriving at an investment decision, not on its results, and asking whether a fiduciary employed the appropriate methods to investigate and determine the merits of a particular investment." Id.

After taking evidence on the issue of Unisys's prudence, the District Court held that the actions taken by the Fund's trustees satisfied the prudence standard: "Based on the evidence at trial I find that the Unisys fiduciaries undertook adequate and reasonable steps before purchasing the three Executive Life contracts." CL p 3. The District Court concluded, "Measured by any standard, the Unisys fiduciaries' actions are consistent with the prudence requirements of ERISA." CL p 13. We hold that the District Court's findings of fact support its conclusions of law that Unisys was prudent in investing in Executive Life GICs, and thus Meinhardt failed to prove an essential element of his ERISA claim.

As an alternate theory for holding that Unisys was not imprudent, the District Court considered the objective prudence of Unisys investments in Executive Life GICs by applying the "hypothetical prudent investor" test. Unisys I, 74 F.3d at 436 (citing Fink v. National Savings & Trust Co., 772 F.2d 951, 962 (D.C.Cir.1985) (Scalia, J., concurring in part and dissenting in part) and Roth v. Sawyer-Cleator Lumber Co., 16 F.3d 915, 919 (8th Cir.1994)).

The District Court held that a hypothetical prudent fiduciary would have invested in Executive Life GICs because (1) Executive Life was qualified under federal regulations as an insurance company authorized to provide annuities to facilitate the termination of benefit pension plans because it was state licensed, 46 Fed.Reg. 9532, 9534 (1981); (2) other judicial decisions endorsed the purchase of Executive Life annuities, e.g., Riley v. Murdock, 890 F. Supp. 444, 458-60 (E.D.N.C.1995);12 (3) Becker, the adviser Unisys retained for the first bid, included Executive Life on his approved list of GIC bidders until six months after Unisys made its third and final purchase of Executive Life, App. 559; and (4) other well-known pension plans purchased Executive Life GICs.App. 559, 1742. The District Court did not err in concluding that a hypothetical prudent fiduciary would have made the same investments in Executive Life GICs as the investments made by Unisys. FF p 24 n. 3; CL pp 8-10.

In sum, we are satisfied that the District Court's holdings that Unisys was prudent, and in the alternative, that a hypothetical prudent fiduciary would have made the same investments, are supported by the evidence.

In addition to holding that the trustees' actions satisfied the standard of prudence required by ERISA, the District Court went on to discuss why the trustees' actions were not arbitrary and capricious. The District Court need not have discussed application of an arbitrary and capricious standard in this case. In light of the District Court's holding of prudence and our affirmance of that holding, the District Court's discussion of an arbitrary and capricious standard cannot affect the judgment in favor of Unisys.

In Unisys I, we stated that fiduciaries of investment plans had to satisfy the "prudent" legal standard specified in ERISA, cited supra. We stated that one of ERISA's underlying purposes was "to enforce strict fiduciary standards." Unisys I, 74 F.3d at 434 (citing 29 U.S.C. § 1001 and H.R.Rep. No. 93-533 (1974), reprinted in 1974 U.S.C.C.A.N. 4639, 4639-43).

In Struble v. New Jersey Brewery Employees' Welfare Trust Fund, 732 F.2d 325 (3d Cir.1984), we held specifically that the duties of loyalty and prudence demanded by ERISA should not be reviewed through an "arbitrary and capricious" lens. We held that the "standards set forth explicitly in ERISA" should be used to evaluate the trustees' conduct. Id. at 333-34 (finding support in cases from the Second, Fifth, and Eleventh Circuits). Struble governs the question of Unisys's duty of prudence under ERISA.

The District Court should have applied only the standard of "prudence under the circumstances" as required by the statute. Only the standard found in ERISA, 29 U.S.C. § 1104(a)(1)(B); Unisys I, 74 F.3d at 434, should have been applied to determine whether Unisys's investment methods were prudent.

As authority for an arbitrary and capricious standard, the District Court cited Varity Corp. v. Howe, 516 U.S. 489, 116 S. Ct. 1065, 134 L. Ed. 2d 130 (1996), Firestone Tire & Rubber Co. v. Bruch, 489 U.S. 101, 109 S. Ct. 948, 103 L. Ed. 2d 80 (1989), and Moench v. Robertson, 62 F.3d 553 (3d Cir.1995), cert. denied, 516 U.S. 1115, 116 S. Ct. 917, 133 L. Ed. 2d 847 (1996). These cases, however, involved situations not relevant to the present claims asserted against Unisys.

In the context of a case challenging a denial of benefits under 29 U.S.C. § 1132(a)(1)(B)--and not challenging the prudence of investment decisions--the Supreme Court held that the "inherently discretionary" nature of fiduciary functions does not necessarily require a deferential standard of review. Firestone Tire & Rubber Co. v. Bruch, 489 U.S. 101, 112, 109 S. Ct. 948, 103 L. Ed. 2d 80 (1989). Under ERISA, the standard of review over a trustee's decision to deny benefits or the interpretation of the plan is de novo as a general rule; only when the plan gives the trustee discretion to deny benefits or construe the terms of the plan should a court employ the arbitrary and capricious standard. As the instant case does not concern a denial of benefits under 29 U.S.C. § 1132(a)(1)(B) or an interpretation of Unisys's Plan, Firestone 's standard is inapplicable. Nor did Varity Corp. v. Howe, 516 U.S. 489, 116 S. Ct. 1065, 134 L. Ed. 2d 130 (1996), involve a claim that a trustee's investments were imprudent--the claim that Meinhardt makes here.

In Moench v. Robertson, 62 F.3d 553 (3d Cir.1995), cert. denied, 516 U.S. 1115, 116 S. Ct. 917, 133 L. Ed. 2d 847 (1996), we did apply a deferent arbitrary and capricious standard of review to a claim by a plan's participants that the fiduciary was imprudent. We were careful to point out in Moench, however, that our holding was limited to the specific type of plan involved in that case, an Employee Stock Ownership Plan ("ESOP"). Here, of course, the Unisys Plan was not an Employee Stock Ownership Plan. Furthermore, Moench specifically held its analysis was in "complete harmony with the prudent man standard of care obligations imposed by 29 U.S.C. § 1104 on fiduciaries, as our holding implicates only the standard of review of the conduct of a fiduciary and not the standards governing that conduct," Moench, 62 F.3d at 566 n. 3 (emphasis added), as is the case here.

In sum, Unisys's methods of making that decision must be evaluated using the ERISA standard mandated by Unisys I and in accordance with Struble. The District Court did not err in concluding that Unisys did not breach its duty of prudence.

For these reasons, the District Court's discussion of an arbitrary and capricious standard was superfluous and constitutes no more than harmless error. Fed.R.Civ.P. 61. The District Court's statements that Unisys's actions were not arbitrary and capricious do not negate its otherwise correct holding that Unisys satisfied the ERISA prudence standard.

Meinhardt also argues that the District Court abused its discretion when it decided to exclude the testimony of Dr. George M. Gottheimer, one of Meinhardt's two proposed expert witnesses.13 Meinhardt offered Dr. Gottheimer to testify on the subject of "the customary methods of investigating the financial condition and creditworthiness of insurance companies."

To qualify as an expert under Fed.R.Evid. 702, a witness must have sufficient qualifications in the form of knowledge, skills, and training. Additionally, the court must find that the testimony of the expert will be reliable and that the testimony will "fit," that is, assist the trier of fact. In re Paoli R.R. Yard Litig., 35 F.3d 717, 741-43 (3d Cir.1994). While it may be arguable that the "fit" prescribed in In re Paoli is a "fit" reflecting on the substance of the witness' testimony, we think it just as relevant to the "fit" reflecting on the witness' credibility. Indeed, in the recent Supreme Court decision discussing the standard of review applicable to the admission and exclusion of expert evidence, the Supreme Court refers to the District Court's "gatekeeper" role of screening such evidence to ensure that it is not only relevant but reliable. The Court goes on to write, "A court of appeals applying 'abuse of discretion' review to [expert testimony] rulings may not categorically distinguish between rulings allowing expert testimony and rulings which disallow it." General Elec. Co. v. Joiner, 522 U.S. 136, 118 S. Ct. 512, 517, 139 L. Ed. 2d 508 (1997). Thus in our view, the Court's emphasis on reliability as well as on relevancy embraces within its standard the credibility of the witness proffering expert opinion. This is particularly true where, as here, it is the district court judge sitting as a finder of fact who must rule on issues of evidence. See Goodman v. Highlands Ins. Co., 607 F.2d 665, 668 (5th Cir.1979) ("[A] trial judge sitting without a jury is entitled to even greater latitude concerning the admission or exclusion of evidence.").

While we could understand issue being taken with the Goodman precept, it appears to us after Joiner that the Goodman standard has been given increased viability. Although Joiner was a summary judgment decision, it nevertheless emphasized that Daubert did not alter the general rule announced in Spring Co. v. Edgar, 99 U.S. 645, 25 L. Ed. 487 (1878). In Edgar, the Court stated, " 'cases arise where it is very much a matter of discretion with the court whether to receive or exclude the evidence; but the appellate court will not reverse in such a case, unless the ruling is manifestly erroneous.' " Joiner, 118 S. Ct. at 517 (quoting Edgar, 99 U.S. at 658). After so stating, the Court rejected any alteration of this rule and in the context of expert testimony said, "But Daubert [v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S. Ct. 2786, 125 L. Ed. 2d 469 (1993) ] did not address the standard of appellate review for evidentiary rulings at all...." Id. The Court then repeated that under Daubert " 'the trial judge must ensure that any and all scientific testimony or evidence admitted is not only relevant, but reliable.' " Id. (quoting Daubert, 509 U.S. at 589, 113 S. Ct. 2786).

The Court went on to hold in Joiner that the Eleventh Circuit had erred in reviewing the exclusion of Joiner's experts' testimony because it failed to give the trial court the deference that is the hallmark of abuse of discretion review. Id. (citing Koon v. United States, 518 U.S. 81, 97-99, 116 S. Ct. 2035, 135 L. Ed. 2d 392 (1996)). The Court further held that the studies upon which Joiner's experts relied were not sufficient whether individually or in combination, to support their conclusions that Joiner's exposure to PCBs contributed to Joiner's cancer, and that the District Court therefore did not abuse its discretion in excluding their testimony.14

In the instant case, the District Court ruled that Dr. Gottheimer's testimony would not be admissible for three reasons. First, the court found that Dr. Gottheimer's educational credentials were not of the highest caliber.15 Second, during Dr. Gottheimer's voir dire, the District Court found Dr. Gottheimer not to be credible because he had made statements about his credentials that were inconsistent with his deposition testimony. Finally, the District Court found that Dr. Gottheimer's expertise was in property casualty insurance, not life insurance, and that Gottheimer had admitted in his deposition that there are "fundamental" differences in evaluating the two types of insurance. FF pp 101-05; CL pp 17-19.

These three reasons coincide with the three requirements articulated in Paoli II: qualifications, reliability, and fit. The record discloses that Dr. Gottheimer's qualifications were less than stellar. Because he had testified untruthfully at voir dire, his testimony could well have been held unreliable. Finally, Dr. Gottheimer's alleged expertise, limited in any event to methods of investing with respect to property casualty insurance, did not fit with or meet the need of the District Court for expert testimony in life insurance investing. See Surace v. Caterpillar, Inc., 111 F.3d 1039, 1055-56 (3d Cir.1997). The District Court determined that Dr. Gottheimer could not be a credible witness based on admissions secured through the use of his prior sworn testimony. See FF pp 102-104. Because the District Court, as fact-finder, listened to the testimony and assessed the credibility of the witness, these findings are entitled to great deference by this Court. See Anderson v. City of Bessemer City, N.C., 470 U.S. 564, 105 S. Ct. 1504, 84 L. Ed. 2d 518 (1985).

Judge Becker, in his thoughtful and comprehensive dissent, while focusing almost exclusively on the standards for admissibility of scientific expert testimony, nevertheless acknowledges that "the factfinder is ordinarily the arbiter of general credibility," Dissent at 167, and that "the power to evaluate the credibility of witnesses and give testimony its proper weight primarily resides with the trier of fact." Dissent at 167. He also properly acknowledges that the decision to admit or to exclude expert testimony is reviewed for abuse of discretion.

Judge Becker, however, fails to recognize or credit in his dissent two significant and controlling issues. First, Dr. Gottheimer's testimony does not fall within the scope of scientific testimony, and accordingly, it should not be tested by the particular standards required for testimony based on a particular scientific ethic.16 See Carmichael v. Samyang Tire, Inc., 131 F.3d 1433 (11th Cir.1997), cert. granted sub nom. Kumho Tire Co. v. Carmichael, --- U.S. ----, 118 S. Ct. 2339, 141 L. Ed. 2d 711 (1998). It is true of course that we have referred to the Paoli requirements of qualifications, reliability and fit--but we have done so only to emphasize that, measured by any standard, scientific or non-scientific, the District Court did not abuse its discretion in excluding Dr. Gottheimer's testimony.

Second, and more important, the dissent, while acknowledging that it is the fact finder that determines weight and credibility of an expert's testimony, apparently overlooks the distinguishing circumstance here: the fact finder was the District Court judge himself, and not a jury. Judge Hutton made the critical credibility determination that he could not believe the testimony of Dr. Gottheimer, and that being so, no reason has been put forward as to why we should not credit and defer to the District Court's finding. Even if we might have made a different determination regarding Dr. Gottheimer's credibility, we are not at liberty to impose our opinion on the District Court.17

We would be hard pressed to require a District Court judge sitting in a non-jury case who credibly and with reason found that he could not believe a witness to nevertheless hear the witness's direct examination, cross-examination, and rebuttal examination in an extended trial when he knew that he would only reject it as unbelievable. All the instances and cases cited by the dissent, see, e.g., Dissent at 162 n. 1, 170 n. 13, are those in which it was for the jury as fact finder to determine credibility and weight of the expert testimony. Thus, those cases are wholly inapplicable to the facts of this case. When the role of the gatekeeper to admit or exclude evidence (the judge) and the role of the factfinder to assess and weigh the evidence that was admitted (the jury) are one and the same, the judge who becomes the factfinder as well as the gatekeeper must be given great deference by this Court, and, as we note below, should not be required to waste judicial time. See Fed.R.Evid. 403. Therefore, we cannot say on this record that it was an abuse of the District Court's discretion to exclude Dr. Gottheimer's testimony.

Even if the District Court had abused its discretion, and we hold that it did not, the error must be deemed harmless in light of the District Court's finding that Dr. Gottheimer was not credible. The District Court concluded, "If given the chance to testify, I could not find [Dr. Gottheimer] to be a credible witness given his evasiveness, if not his propensity to state falsehoods." CL p 18. Obviously, hearing testimony from a witness who was given no credence at all by the District Court judge presiding at a bench trial would have resulted in the "waste of time" proscribed by Federal Rule of Evidence 403, particularly where the false statements identified by the judge were material to the purported expert's qualifications.

ERISA requires a fiduciary to "diversify[ ] the investments of the plan so as to minimize the risk of large losses, unless under the circumstances it is clearly not prudent to do so." 29 U.S.C. § 1104(a)(1)(C). Meinhardt complains that Unisys invested an excessive amount of Fund investments in Executive Life GICs. As discussed in Unisys I, Congress did not try to quantify diversity with percentages, leaving that determination to the facts and circumstances as found by the courts. Unisys I, 74 F.3d at 438-40.

The District Court determined that Meinhardt had not introduced evidence by way of expert testimony as to what would have constituted a properly diversified fund. At the time Executive Life went into receivership, the Fund had 20% of its assets in Executive Life investments, and the court held this level of diversification to be proper.18 CL pp 32-33.

Nor did Meinhardt prove his case with respect to the duty to diversity because he did not show that the Fund suffered "large losses" as a result of any failure to diversify. This was an issue that the Court of Appeals specifically held was not known at the summary judgment stage. See Unisys I, 74 F.3d at 440. Meinhardt presented no evidence of investments that would have constituted proper diversification in order to prove that Unisys did not properly diversify the investments in the Fund or to enable the District Court to assess losses, if any, to Meinhardt for the difference. FF p 90.

The duty to diversify and resulting damages from a breach of that duty could not be determined from the summary judgment record in Unisys I. Now that the trial has concluded, it is evident to us, as it was to the District Court, that Meinhardt's failure of proof did not lead to any determination that Unisys breached its duty to diversify or that damages resulted.

Meinhardt claims that Unisys did not fulfill its obligation of making necessary disclosures under ERISA.

In Unisys I, we stated that "a fiduciary may not materially mislead those to which section 1104(a)'s duties of loyalty and prudence are owed." Unisys I, 74 F.3d at 440-41. A fiduciary must make disclosures if silence would be harmful. Bixler v. Central Pa. Teamsters Health & Welfare Fund, 12 F.3d 1292, 1300 (3d Cir.1993). ERISA also requires plaintiffs to prove losses for any breach of fiduciary duty claim:

Any person who is a fiduciary with respect to a plan who breaches any of the responsibilities, obligations, or duties imposed upon fiduciaries by this subchapter shall be personally liable to make good to such plan any losses to the plan resulting from each such breach.

29 U.S.C. § 1109(a) (emphasis added). As Meinhardt and the other class plaintiffs were seeking individual relief under 29 U.S.C. § 1132(a)(3) (in contrast to § 1132(a)(2), which only allows relief on behalf of the Plan), Meinhardt was required to prove individual losses. Varity Corp. v. Howe, 516 U.S. 489, 507-15, 116 S. Ct. 1065, 134 L. Ed. 2d 130 (1996).

Although the District Court's discussion of Meinhardt's claims that Unisys breached its duty to disclose information about Executive Life spans 18 pages, we can dispose of this issue on appeal briefly. Essentially, Meinhardt complains that Unisys received reports about Executive Life's financial troubles in early 1990. Meinhardt claims that Unisys (1) did not warn the Fund participants, (2) drafted misleading letters in order to dispel participants' concern, (3) did not disclose the fact that a high-level Unisys executive had purged his own portfolio of Executive Life securities, and (4) entered into an agreement with Executive Life not to disclose information to participants that would cause participants to change their investments.19 The District Court held that Meinhardt did not prove that the alleged failures to disclose were material. CL pp 22-23.

We need not address the question of whether the alleged nondisclosures were material, however, because it is clear that Meinhardt did not prove that any alleged failures to disclose caused the participants to suffer damages. CL pp 29-30. The District Court found that Meinhardt and the other class plaintiffs (1) already had actual knowledge of much of the information it is claimed that Unisys failed to disclose, (2) did not read the Plan documents, and (3) testified that they would not have withdrawn or transferred their money from the Fund even if they had known about Executive Life's problems. FF pp 70-71; CL p 28; App. 1681-82, 1688-89.

Moreover, Meinhardt's expert, Tsetsekos, offered testimony on losses suffered as a result of the alleged failures to disclose, but referred only to those losses incurred by the Fund and not to any losses incurred by individual participants named as plaintiffs. FF pp 76-77; CL pp 29-31. Meinhardt also failed to prove individual damages suffered by each participant as ERISA requires. 29 U.S.C. § 1132(a)(3).

We hold that these factual findings of the District Court are not clearly erroneous and that they support the conclusion reached by the District Court that Meinhardt failed to prove his claim that Unisys breached its duty of disclosure.

A final word should be said about Meinhardt's claim for damages. First and foremost, in the absence of proof of a breach of fiduciary duty, no relief in the way of damages or losses could accrue to Meinhardt. We have held that Unisys satisfied ERISA's prudence standard and that accordingly, no breach of fiduciary duty occurred. This being so, no claim for increased interest, i.e., the promised interest minus the actual amount of interest received, or any other damages, can be sustained. As we have noted earlier, Meinhardt had received the return of the entire principal invested as well as some interest.20

It is therefore irrelevant as to whether any losses sustained by the Fund from GIC investments could have been offset by any gains derived from other Fund investments. While we acknowledge that a trustee, when he is imprudent and breaches his trust, is liable for all gains and may not offset losses against them,21 in the instant case, the Unisys fiduciaries were neither imprudent nor did they breach their fiduciary duties.22

Additionally, Meinhardt and the Department of Labor as amicus also argued that the burden of proving causation of damages shifts to the defendant after the plaintiff has proved that the defendant breached a fiduciary duty. Here, the District Court assigned the burden of proof to the plaintiff Meinhardt. Because we have held that Unisys did not breach its fiduciary duties, we have no need to address the issue of which party bears the burden of proving causation of damages resulting from a breach of fiduciary duty. While we recognize that our sister circuits have divided in deciding this question,23 we have yet to express ourselves on this issue.24

(1) The District Court's findings of fact issued after hearing evidence on Meinhardt's ERISA claims will be affirmed as not clearly erroneous.

(2) The District Court's conclusions of law that the Unisys fiduciaries were prudent, and in the alternative, that a hypothetical prudent fiduciary would have made the same investments in Executive Life GICs, also will be affirmed.

(3) Although we do not agree with the District Court's additional discussion of an arbitrary and capricious standard in reviewing the investments made by the ERISA fiduciaries, this discussion was superfluous, constitutes no more than harmless error, and does not affect our judgment.

(4) The District Court did not abuse its discretion in excluding the testimony of Meinhardt's proposed expert witness on the issue of Unisys's prudence, and in any event, that action was harmless in light of the District Court's finding that the expert was not credible.

(5) The District Court did not err in denying relief to Meinhardt inasmuch as no losses or damages could be sustained. As a consequence, we have found it unnecessary to address the disputed issue of which party bears the burden of proving causation of damages that result from a breach of fiduciary duty.

(6) The District Court did not err in holding that Meinhardt's other ERISA claims, failure to diversify and failure to disclose information, were not proved.

Thus, we will affirm the District Court's orders of November 24, 1997, and January 9, 1998, in favor of Unisys and against Meinhardt.

Each party shall bear its own costs.

BECKER, Chief Judge, dissenting.

Although I join in Parts I through V and VII through IX of the majority's opinion, I believe that the majority has made a significant error regarding expert testimony, and I dissent on this point. Because I believe that this error is not harmless, I believe we should vacate the judgment and remand this case for a new trial.

In concluding that the District Court properly excluded the testimony of Dr. Gottheimer, the majority seriously misconceives the proper approach to the admission of expert opinion testimony under Federal Rule of Evidence 702. As an initial point, the majority overstates the degree to which we owe deference to the District Court's decision in a case, such as this, tried to the court. The District Court's decision deserves no additional deference simply because the court sat as both evidentiary gate-keeper and fact-finder. If anything, the practicalities of the matter suggest that the trial court in such a situation should be more reluctant than usual to exclude evidence, although I do not suggest that we should apply a correspondingly more stringent standard of review.

The District Court and the majority also make three important errors in analyzing the substantive requirements of Rule 702. First, the majority confuses the reliability of an expert witness--a matter for the jury--with the reliability of his or her methodology--a matter initially for the trial judge--and therefore erroneously concludes that questions about an expert witness's general credibility are a proper basis for excluding his or her testimony. The question for the judge under Rule 702 is not whether the witness is reliable but whether the methodology the expert uses in reaching his conclusions is reliable. As to this latter point, the witness's general credibility is simply irrelevant; the relevant issues in determining the reliability of an expert's principles and methods are of the sort set forth with respect to scientific testimony in Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 593-95, 113 S. Ct. 2786, 125 L. Ed. 2d 469 (1993).

Second, the majority misconstrues the nature of the requirement of "fit" between the expert's testimony and the facts at issue. The majority concludes that the District Court correctly found that no fit exists in this case because Dr. Gottheimer's experience, in the area of property-casualty insurance, on which his testimony is based, is not in the specific area--life insurance--with which the facts in this case deal. But the majority's focus on the connection between Dr. Gottheimer's claimed basis for being an expert and the facts at issue, as opposed to the connection between the substance of his testimony and the facts, is irrelevant to the question of fit. Their concern about his background should more appropriately be directed at Dr. Gottheimer's qualifications, not the fit between his testimony and the facts.

Third, the majority permits the District Court to set the qualifications bar for expert testimony too high. It approves the District Court's rejection of the expert's testimony simply because his qualifications are not of the "highest caliber." This conclusion is inconsistent with our longstanding liberal approach to the matter of expert witness qualifications. Also, the connection between Dr. Gottheimer's expertise and the issues in this case--which the majority discusses in the context of fit--are not too remote for him to qualify as an expert under Rule 702.

Finally, I think that the majority wrongly concludes that any error in the exclusion of Dr. Gottheimer's testimony was harmless. While improper admission of evidence is usually harmless error in a bench trial, the improper exclusion of an expert witness who would have offered a party's sole expert testimony on an element of its case ordinarily is not harmless. The fact that the District Court found some inconsistencies in Dr. Gottheimer's voir dire testimony was not a sufficient basis for changing the ordinary rule. Such inconsistencies are not enough to convince me that it is highly probable that Dr. Gottheimer's testimony, if admitted, would not have changed the outcome. This is especially true in light of the fact that the inconsistencies the District Court discusses were at worst minor.

Under Rule 702, "[i]f scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise." Fed.R.Evid. 702. The Supreme Court has recognized that Rule 702, although it limits the scope of permissible evidence, is part of "the 'liberal thrust' of the Federal Rules and their 'general approach of relaxing the traditional barriers to "opinion" testimony.' " Daubert, 509 U.S. at 588, 113 S. Ct. 2786 (quoting Beech Aircraft Corp. v. Rainey, 488 U.S. 153, 169, 109 S. Ct. 439, 102 L. Ed. 2d 445 (1988)). We have recognized that this "principle of liberal admission of expert testimony is found in Rule 702 itself, in the advisory committee note to the rule, and in our case law."1

To these ends, Rule 702 embodies three distinct substantive restrictions on the admission of expert testimony: qualifications, reliability and fit. See In re Paoli R.R. Yard PCB Litig. ("Paoli II"), 35 F.3d 717, 741-43 (3d Cir.1994). First, an expert witness must be qualified by virtue of specialized expertise. See Fed.R.Evid. 702 (permitting expert testimony of a witness "qualified as an expert by knowledge, skill, experience, training, or education"); Paoli II, 35 F.3d at 741. Second, "an expert's testimony is admissible so long as the process or technique the expert used in formulating the opinion is reliable." Paoli II, 35 F.3d at 742 (citing Daubert, 509 U.S. at 589-90, 113 S. Ct. 2786). Third, "[e]xpert testimony which does not relate to any issue in the case is not relevant and, ergo, non-helpful." Daubert, 509 U.S. at 591, 113 S. Ct. 2786 (citations omitted).2II. Standard of Review

As an initial matter, I think the majority errs in its analysis of the applicable standard of review and the extent to which we owe deference to the District Court's decision. Of course, the decision whether to admit or exclude expert testimony is largely within the hands of the trial judge. We review such a decision for abuse of discretion. See General Elec. Co. v. Joiner, 522 U.S. 136, 118 S. Ct. 512, 517, 139 L. Ed. 2d 508 (1997) ("We have held that abuse of discretion is the proper standard of review of a district court's evidentiary rulings." (citations omitted)); In re Paoli R.R. Yard PCB Litig. ("Paoli I"), 916 F.2d 829, 856 & n. 33 (3d Cir.1990). The standard does not change when we are reviewing a court's decision to exclude, as opposed to admit, expert testimony, in spite of the liberal standard for the admission of such testimony. See Joiner, 118 S. Ct. at 517.

Even though we apply an abuse of discretion standard of review, however, "to the extent the district court's ruling turns on an interpretation of a Federal Rule of Evidence our review is plenary."3 Furthermore, although our review is highly deferential, it is not a complete bar to reversing a district court's decision even where the court does not commit purely legal error. See, e.g., Paoli I, 916 F.2d at 855-56 (rejecting as an abuse of discretion the trial court's insistence on certain credentials as expert qualifications); 4 Weinstein's Federal Evidence § 702.02(2) n. 9 (2d ed. Nov.1998) (collecting cases).

The majority concludes that the fact that this case involved a bench trial requires additional deference to the District Court's evidentiary exclusion decision. In reaching this conclusion, the majority relies on Goodman v. Highlands Insurance Co., 607 F.2d 665 (5th Cir.1979), which states that "a trial judge sitting without a jury is entitled to even greater latitude concerning the admission or exclusion of evidence." Goodman, 607 F.2d at 668. The majority asserts that, after Joiner--in which the Supreme Court held that the same abuse of discretion standard of review applies to an evidentiary ruling on expert testimony regardless of whether the trial court admits or excludes the evidence--the "Goodman " rule "has been given increased viability." Slip Op. at 20. For a variety of reasons, I cannot agree with the majority's conclusion that our review is affected by the fact that this case was tried to the court.

As an initial matter, Goodman does not support the meaning the majority draws from it. The statement in Goodman upon which the majority relies refers not to the trial court's decision on admissibility but to whether the trial court's decision, if erroneous, was a harmless error. This reading is made manifest by an examination of the sentence following that quoted above:

In a non-jury case, the admission of incompetent evidence will not warrant reversal unless all of the competent evidence is insufficient to support the judgment, or unless it affirmatively appears that the incompetent evidence induced the court to make an essential finding which would otherwise not have been made.

Goodman, 607 F.2d at 668 (citations omitted). This is simply a restatement of the familiar harmless error test for review of decisions admitting evidence in bench trials.4 It is, however, irrelevant to our determination of whether the district court's decision to exclude expert testimony from evidence is reversible error.5

Furthermore, even assuming the majority correctly states the Fifth Circuit's Goodman rule, I think that rule is not a proper one and would not follow it. The Federal Rules of Evidence apply with full force to bench trials. See Fed.R.Evid. 1101(b); 9 Charles Alan Wright & Arthur R. Miller, Federal Practice & Procedure § 2411, at 587 (2d ed. 1995) ("In theory, the Federal Rules of Evidence apply equally in court trials and jury trials."). After all, a trial is a trial. To me, the proposition that we owe more deference to trial court decisions excluding evidence in bench trials is inconsistent with the Federal Rules and encourages sloppy district court decisionmaking.6

If anything, trial courts should be more chary of excluding evidence in bench trials than in jury trials. See Builders Steel Co. v. Commissioner, 179 F.2d 377, 379 (8th Cir.1950) ("[A] trial judge who, in the trial of a nonjury case, attempts to make strict rulings on the admissibility of evidence, can easily get his decision reversed by excluding evidence which is objected to, but which, on review, the appellate court believes should have been admitted."), quoted in 9 Wright & Miller, supra, § 2411, at 587. The better course is to admit the evidence and then take factors that otherwise might affect its admissibility into consideration in determining its weight, rather than waste time debating the propriety of admitting the evidence.7 I believe the majority's approach grants undue deference to trial court's decisions excluding evidence in bench trials.

I now turn to the substantive requirements of Rule 702. Under Rule 702, expert testimony is admissible only if it meets the requirements of Rule 702: qualifications, reliability and fit. The majority concludes that the District Court properly found that Dr. Gottheimer's testimony met none of these requirements. I disagree, and discuss each of these factors, although not in the usual order.

The majority begins with the indisputable premise that reliability is a key factor in determining the admissibility of expert testimony. "[U]nder the Rules the trial judge must ensure that any and all scientific testimony or evidence admitted is not only relevant, but reliable." Joiner, 118 S. Ct. at 517 (quoting Daubert, 509 U.S. at 589, 113 S. Ct. 2786). From this, the majority concludes that, as part of the reliability analysis, a trial court may consider the general credibility of a witness in determining whether his or her testimony is appropriately admitted as expert witness testimony. See Slip Op. at 20 ("Thus in our view, the Court's emphasis on reliability as well as on relevancy embraces within its standard the credibility of the witness proffering expert opinion."). Accordingly, the majority holds that, since the District Court found that Dr. Gottheimer was not a credible witness it could properly exclude his expert testimony under Rule 702. The majority misconceives the fundamental nature of the reliability inquiry under Daubert. The reliability inquiry focuses not on the witness's reliability in an evidentiary sense, but on the reliability of the methodology that the expert applies in arriving at an opinion.

[I]n order to qualify as "scientific knowledge," an inference or assertion must be derived by the scientific method. Proposed testimony must be supported by appropriate validation--i.e., "good grounds," based on what is known. In short, the requirement that an expert's testimony pertain to "scientific knowledge" establishes a standard of evidentiary reliability.

Daubert, 509 U.S. at 590, 113 S. Ct. 2786 (footnote omitted). This inquiry focuses on the expert's principles and methodology, not his results.

The inquiry envisioned by Rule 702 is, we emphasize, a flexible one. Its overarching subject is the scientific validity--and thus the evidentiary relevance and reliability--of the principles that underlie the proposed submission. The focus, of course, must be solely on principles and methodology, not on the conclusions that they generate.

Daubert, 509 U.S. at 594-95, 113 S. Ct. 2786. Under Rule 702, the trial court's preliminary reliability analysis in making an admissibility determination must focus on the witness's methods, not his or her testimony as a whole.8 A current proposed amendment to Rule 702 reemphasizes this focus on the reliability of the methodology, as opposed to the witness.9

Credibility plays no appropriate part in the analysis of the reliability of a proposed expert's methodology. The Court in Daubert recognized a number of factors pertinent to the reliability inquiry, including testability, peer review or publication, potential rate of error, existence of standards and controls, and general acceptance. See 509 U.S. at 593-94, 113 S. Ct. 2786. We, as well as others, have suggested numerous additional factors relevant to reliability as set forth in the margin.10 The key point is that none of these factors requires consideration of the proffered expert's credibility in general. "[E]valuating the reliability of scientific methodologies and data does not generally involve assessing the truthfulness of the expert witnesses...." Paoli II, 35 F.3d at 749.11 Of course, a particular witness can lie about whether one of the factors mentioned above is present. But the witness's general credibility--i.e., credibility based on matters not directly related to these factors, such as bias or unrelated prior inconsistent statements--is not relevant to a trial court's preliminary determination that these factors are present. These factors are relatively objective matters that the court can generally analyze independent of the witness's testimony about them. For example, a court is fully capable of determining whether a theory or method is testable.

Furthermore, permitting the trial court to consider the general credibility of an expert witness in its analysis of the reliability of proposed expert opinion testimony improperly trenches on the province of the fact-finder. The fact-finder is ordinarily the arbiter of general credibility. Rule 702 is intended not to impinge on the authority of the fact-finder in making credibility determinations, but rather to ensure that the fact-finders' ability to find facts independently is not overwhelmed by complex and authoritative-seeming expert testimony. See 29 Wright & Gold, supra, § 6262, at 179 ("If the trier of fact is unable or disinclined to question the expert's opinion, it surrenders its central function to an expert whose testimony may be unreliable."). In order to avoid this eventuality, Rule 702 provides for a preliminary inquiry into whether proposed expert testimony is the kind that is appropriately admitted into evidence, i.e., that the method is reliable and based on expertise. See id. at 183-84. But this inquiry cannot extend into matters that are the proper province of the fact-finder, such as general credibility. See id. at 184 ("Importantly, however, Rule 702 on its face creates no general power in the trial judge to exclude expert testimony on the grounds that it is unreliable.").

This limitation on the power of the trial court to admit and exclude expert testimony rests on sound policy considerations. First, it is emblematic of the Rules' generally liberal approach to the admissibility of evidence, discussed above. See supra Part I. Second, it reflects the fact that the power to evaluate witness's credibility lies traditionally with the fact-finder, not the trial court making evidentiary rulings. See 29 Wright & Gold, supra, § 6262, at 185 ("[T]he power to evaluate the credibility of witnesses and give testimony its proper weight primarily resides with the trier of fact.").

Finally, it rests on the general assumption underlying the Federal Rules of Evidence as a whole: the fact-finder is best situated to determine the credibility of witnesses. See id. at 185 ("[J]uries generally have the ability to accurately weight and evaluate witness credibility."). "Accordingly, the most common judicial response to attacks on the reliability of expert testimony is that such matters go to weight, not admissibility." Id. at 185-86. The Court in Daubert recognized the importance of leaving such matters to the fact-finder, and not determining them on evidentiary grounds:

Respondent expresses apprehension that abandonment of "general acceptance" as the exclusive requirement for admission will result in a "free-for-all" in which befuddled juries are confounded by absurd and irrational pseudoscientific assertions. In this regard respondent seems to us to be overly pessimistic about the capabilities of the jury and of the adversary system generally. Vigorous cross-examination, presentation of contrary evidence, and careful instruction on the burden of proof are the traditional and appropriate means of attacking shaky but admissible evidence.

Daubert, 509 U.S. at 595-96, 113 S. Ct. 2786 (citation omitted; emphasis added).

Accordingly, I conclude that the District Court erred to the extent that it considered Dr. Gottheimer's general credibility as a factor in determining whether the principles and methodology underlying his proposed expert testimony were reliable under Rule 702. This result is unaffected by the fact that the District Court--post hoc--went ahead and made credibility findings that arguably would have resulted in his rejection of Dr. Gottheimer's testimony in its role as fact-finder. I deal with this aspect of the matter in my discussion of harmless error. See infra Part IV. Questions about Dr. Gottheimer's credibility should have been left for substantive fact-finding, not evidentiary rulings. Since the District Court and Unisys suggest no other concerns about the reliability requirement, I would conclude that Dr. Gottheimer's proposed testimony satisfies it.

The majority also concludes that the District Court properly found that Dr. Gottheimer's testimony did not fit with the question in issue at the trial. In particular, the majority concludes that Dr. Gottheimer's experience and knowledge lay in a field not sufficiently connected with the question at issue to meet the requirement of fit. "Dr. Gottheimer's alleged expertise, limited in any event to methods of investing with respect to property casualty insurance, did not fit with or meet the need of the District Court for expert testimony in life insurance investing." Slip Op. at 21. The majority misconceives the requirement of fit under Rule 702.

The requirement of fit is essentially a relevance requirement. Under Rule 702, expert testimony is admissible if it "will assist the trier of fact." Fed.R.Evid. 702. "This condition goes primarily to relevance." Daubert, 509 U.S. at 591, 113 S. Ct. 2786. "[A]dmission depends upon the 'fit,' i.e., upon a specific proffer showing that scientific research has established" some point relevant to the facts of the case. United States v. Downing, 753 F.2d 1224, 1226 (3d Cir.1985). We have further clarified this point:

An additional consideration under Rule 702--and another aspect of relevancy--is whether expert testimony proffered in the case is sufficiently tied to the facts of the case that it will aid the jury in resolving a factual dispute. In this regard, we hold that a defendant who seeks the admission of expert testimony must make an on-the-record detailed proffer to the court, including an explanation of precisely how the expert's testimony is relevant to the [factual issue] under consideration. The offer of proof should establish the presence of factors ... which have been found by researchers to [provide a basis for the proffered opinion]. Failure to make such a detailed proffer is sufficient grounds to exclude the expert's testimony.

753 F.2d at 1242 (citations omitted); see also Lauria v. National R.R. Passenger Corp., 145 F.3d 593, 600 (3d Cir.1998); United States v. Velasquez, 64 F.3d 844, 850 (3d Cir.1995).

The majority errs in concluding that Dr. Gottheimer's field of expertise has any relevance to the fit inquiry. As the foregoing discussion of the fit requirement shows, it is satisfied if the proffered expert opinion is relevant to a factual issue before the fact-finder. The expert's field of expertise is irrelevant to an inquiry into the connection between the opinion itself and the issues in the case. A simple example shows this. In Lauria, a railroad worker slipped and was injured when he stepped on a loose railroad tie that was sitting between some tracks. "The primary issue [was] whether Amtrak was negligent in failing to remove a [railroad] tie from Lauria's workplace." Lauria, 145 F.3d at 600. An expert opinion "that the tie was a dangerous obstruction that should have been discovered and removed and, ... that Amtrak's negligence made the workplace unsafe" clearly satisfies the requirement of fit. 145 F.3d at 600.

But the expertise of the proponent of the opinion is irrelevant to this inquiry. This is true even if the witness proposing to testify to the above opinion is a medical doctor who has no experience with train tracks. Under the majority's reasoning, however, the doctor's testimony would not meet the fit requirement. Clearly, a medical doctor with no experience working with train tracks has no expertise with regard to the placement of loose railroad ties. If Dr. Gottheimer's testimony would not meet the fit requirement because his expertise lay in a different field of insurance than that in issue in the case, clearly the hypothetical doctor cannot provide testimony that meets the fit requirement. But the doctor's testimony should be excluded because of his lack of qualifications, not because of a supposed lack of fit. The proposed expert's expertise is simply irrelevant to determining whether the proffered expert opinion is relevant to issues in the case.

As this example shows, the majority's and the District Court's concerns about the connection between Dr. Gottheimer's expertise and the issues in the case are actually relevant to the step of the Rule 702 inquiry which I consider infra: qualifications. Under Rule 702, a witness can offer an expert opinion if he or she is "qualified as an expert by knowledge, skill, experience, training, or education." Fed.R.Evid. 702. The nature of a witness's specific field of expertise is part of the expert's background that is considered in determining whether a witness is qualified. See Paoli II, 35 F.3d at 741.12

I think a proper fit inquiry would show that Dr. Gottheimer's testimony meets the fit requirement. Plaintiffs' counsel said that Dr. Gottheimer would have testified that certain standard tests exist in the insurance industry to analyze the financial condition of insurance companies, and that he was familiar with those tests. He would have further testified that he performed these tests on Executive Life data, and concluded "that the tests set up certain red flags that should have caused a person familiar with the tests and performing the tests, to ask further questions about the solvency and the credit worthiness of Executive Life." I think this is the kind of "on-the-record detailed proffer" that we required in Downing, "including an explanation of precisely how the expert's testimony is relevant to" determining whether Unisys acted prudently. Downing, 753 F.2d at 1242. Accordingly, I would conclude that Dr. Gottheimer's proposed testimony meets the fit requirement. The nature of Dr. Gottheimer's expertise, and its connection with the factual issues in this case, are relevant only to the Rule 702 inquiry into qualifications, to which I now turn.

The majority also concludes that the District Court properly excluded Dr. Gottheimer's testimony because he was not qualified as an expert with respect to the issues in this case. Specifically, the majority adopts the District Court's conclusion that Dr. Gottheimer was not qualified because his "qualifications were less than stellar." Slip Op. at 21. The District Court noted that "Dr. Gottheimer claims a doctoral degree from a correspondence school, an additional ground for my refusal to qualify him as an expert." In re Unisys Sav. Plan Litig., No. 91-3067, 1997 WL 732473, at * 26 (E.D.Pa. Nov. 24, 1997). In addition, as discussed above, see supra section III.B, the majority concludes that the District Court properly rejected Dr. Gottheimer's testimony because his field of expertise was not sufficiently connected to the issues in the case. "Dr. Gottheimer's experience in the insurance area was limited to property casualty insurance and not life insurance." Unisys, 1997 WL 732473, at * 22. Neither of these are appropriate grounds for excluding an expert witness for lack of qualifications.

The requirement that an expert witness be qualified is well established. A witness may only provide expert testimony to the extent that he or she is "qualified as an expert by knowledge, skill, experience, training, or education." Fed.R.Evid. 702. The bases for concluding that a witness is qualified are flexible. "[I]nsistence on a certain kind of degree or background is inconsistent with our jurisprudence in this area." Paoli I, 916 F.2d at 855; accord Waldorf v. Shuta, 142 F.3d 601, 626 (3d Cir.1998) ("[I]n considering the qualification of witnesses as experts, we stress that ordinarily an otherwise qualified witness is not disqualified merely because of a lack of academic training."); Paoli II, 35 F.3d at 741 ("We have eschewed imposing overly rigorous requirements of expertise and have been satisfied with more generalized qualifications."). A particular educational background is unnecessary; practical experience is sufficient to conclude that a witness is qualified as an expert. See Lauria, 145 F.3d at 599 (citing American Tech. Resources v. United States, 893 F.2d 651, 656 (3d Cir.1990)). "Following this logic, it is an abuse of discretion to exclude testimony simply because the trial court does not deem the proposed expert to be the best qualified or because the proposed expert does not have the specialization that the court considers most appropriate."13

Although, as always with evidentiary questions, we apply a deferential standard of review to a trial court's determination of whether a proposed expert is qualified, we have on numerous occasions found that a district court abused its discretion in excluding a proffered expert because of his or her qualifications. Some of these cases are summarized in the margin.14

Supported by these cases, I believe that Dr. Gottheimer was in fact qualified to offer the expert testimony proffered, and that the District Court's conclusion to the contrary was an abuse of discretion. The District Court found that Dr. Gottheimer was not qualified on two grounds: the nature of his educational credentials and the distinction between his experience in property-casualty insurance companies and the issues in the case involving life insurance companies. In light of the case law discussed in the margin above, both of these findings are inconsistent with the exercise of sound discretion.

The District Court concluded that Dr. Gottheimer could not be qualified because his doctorate was awarded by a correspondence school. If this were plaintiffs' sole basis for claiming that Dr. Gottheimer was qualified, I would probably agree with the District Court. It was not, however. Rather, the record discloses numerous grounds on which to conclude that Dr. Gottheimer was qualified as an expert. Dr. Gottheimer's resume demonstrates his expertise through three distinct areas: experience, education and teaching. First, he has worked for a dozen years as a consultant in the insurance industry, following thirty years of employment by various insurance companies. His consulting work has included analyses of both property-casualty and life insurance companies.15 Second, he has bachelor's and master's degrees in insurance-related fields, as well as a doctorate from a correspondence school.16 He also possesses several professional affiliations in insurance professionals' organizations. Finally, Dr. Gottheimer has taught for over twenty-five years at the College of Insurance. The College of Insurance is an accredited, industry-sponsored school that offers classes in all aspects of insurance business. He is now on the fulltime faculty there, and has taught courses in a variety of fields, including insurance company management.

In light of these extensive qualifications, I have no doubt that Dr. Gottheimer was qualified and should have been permitted to testify as an expert under Rule 702. The District Court's decision to the contrary was an abuse of discretion. As the majority points out, the District Court refused to qualify Dr. Gottheimer because his qualifications "were not of the highest caliber." Slip Op. at 21. But in light of our longstanding jurisprudence, this is not an appropriate basis for excluding a proffered expert witness. See Kannankeril v. Terminix Intl., Inc., 128 F.3d 802, 809 (3d Cir.1997); Holbrook, 80 F.3d at 782; Paoli II, 35 F.3d at 741. Accordingly, I must conclude that the District Court abused its discretion in excluding his testimony on this basis.

For all these reasons, I think that Dr. Gottheimer should have been permitted to testify as an expert witness under Rule 702. As discussed above, Dr. Gottheimer fully met all three of the Rule 702 requirements: qualifications, reliability and fit. The only remaining question is whether that error was reversible or harmless error.IV. Harmless Error

In the closing paragraph of its discussion of the Rule 702 evidentiary issue, the majority concludes that, even if the District Court did commit an error in excluding Dr. Gottheimer's testimony, the error was harmless. In reaching this conclusion, it relies on the District Court's statement that it "could not find [Dr. Gottheimer] to be a credible witness given his evasiveness if not his propensity to state falsehoods." 1997 WL 732473, at * 26. The District Court pointed to a few alleged inconsistencies in Dr. Gottheimer's deposition and trial testimony in support of this conclusion. The majority concludes that, since the District Court would not have believed Dr. Gottheimer's testimony, his testimony could not have been given any weight if admitted. Thus its exclusion did not have a substantial effect on the outcome and any error in excluding it was harmless. Once again, I must disagree.

Under the Federal Rules of Evidence, an evidentiary error to which a party has raised a proper objection is not a grounds for reversal "unless a substantial right of the party is affected." Fed.R.Evid. 103(a); see also 28 U.S.C. § 2111; Fed.R.Civ.P. 61. An error is harmless, i.e., it does not affect a substantial right, only if "it is highly probable that the error did not contribute to the judgment." Murray v. United of Omaha Life Ins. Co., 145 F.3d 143, 156 (3d Cir.1998) (citing McQueeney v. Wilmington Trust Co., 779 F.2d 916, 923-27 (3d Cir.1985)). Although, as discussed above, the improper admission of evidence is usually harmless error in a bench trial, see supra Part II, the erroneous exclusion of evidence in a bench trial can be reversible error just as in a jury trial. See 11 Charles Alan Wright, Arthur R. Miller & Mary Kay Kane, Federal Practice & Procedure § 2885, at 454 (2d ed. 1995) ("In nonjury cases the district court can commit reversible error by excluding evidence but it is almost impossible for it to do so by admitting evidence." (footnote omitted)). Error is especially likely not to be harmless where the excluded expert was the only one a party offered to prove an essential element of its case.17

In this case, Dr. Gottheimer was the only expert witness plaintiffs offered to prove that Unisys acted imprudently. His proffered testimony, set forth in the margin, was strong.18 By excluding Dr. Gottheimer's testimony, the District Court deprived plaintiffs of their best evidence that Unisys breached its duty of prudence, a key element of their case. In light of our conclusion in Lauria and Holbrook, I cannot say that it is highly probable that the exclusion of Dr. Gottheimer did not affect the outcome of the trial. This is especially true in light of the other evidence admitted at trial, set forth in the margin.19 The District Court essentially decided all questions the evidence raised in favor of Unisys and concluded that Unisys acted prudently. Although I agree with the majority that, based on the admitted evidence, this conclusion was not clearly erroneous, this is to me an exceedingly close question. Accordingly, Dr. Gottheimer's testimony, if admitted, stood a good chance of changing this balance and consequently changing the decision of the District Court.

That the District Court had questions about Dr. Gottheimer's credibility should not affect our harmless error analysis. The District Court's conclusion that it would not have found Dr. Gottheimer's testimony credible, based only on his voir dire testimony, is not sufficient grounds for concluding that exclusion of his testimony was harmless. I believe that there is a reasonable chance that, if the District Court had given Dr. Gottheimer the opportunity to present his testimony in full, it would have found him to be a credible witness.

Our decisions in Lauria and Holbrook at least implicitly support this conclusion. In each of those cases, we reversed a district court decision excluding expert testimony on the grounds that the expert witness lacked the necessary qualifications. Such evidence goes to the weight, not the admissibility, of the expert testimony. See Kannankeril, 128 F.3d at 809. In neither Lauria nor Holbrook did the effect of the expert's particular qualifications on the weight properly accorded to his testimony play any part in our harmless error analysis. See Lauria, 145 F.3d at 600 (discussing harmless error without mentioning the quality of the improperly excluded expert's qualifications); Holbrook, 80 F.3d at 787 (same). Similarly, the impact of questions about Dr. Gottheimer's credibility on the weight due his testimony should not play a part in our harmless error analysis in this case.

I find further support for this conclusion in an examination of what the District Court identified as "impeachment" of Dr. Gottheimer. The District Court stated that "Dr. Gottheimer was impeached no fewer than four times on the relatively straight forward questions on his qualifications." 1997 WL 732473, at * 26; see also 1997 WL 732473, at * 21-* 22. But this so-called "impeachment" involved at most minor inconsistencies. First, Dr. Gottheimer testified at trial that he could not recall having testified in court in a case involving a life insurance company, although he stated at his deposition three years earlier that twenty-four out of the twenty-five times he had testified in court involved property-casualty insurance companies. See 1997 WL 732473, at * 21. Second, although at his deposition he could not recall having done consulting work involving life insurance company solvency, he testified that he had been retained once before his deposition and several times after his deposition to do such consulting. See 1997 WL 732473, at * 21-* 22. Third, while he testified at trial that there were "some differences and there are also some similarities in the way" one analyzes life insurance as opposed to property-casualty insurance company solvency, in his deposition he agreed with Unisys's attorney that there were "fundamental differences." See 1997 WL 732473, at * 22.

I cannot see how these answers, under any reasonable reading, suggest that Dr. Gottheimer is a completely incredible witness. At most, they suggest that he could remember some things at his deposition that he could not at trial, and vice versa. Certainly, Dr. Gottheimer's statements at trial were not directly inconsistent with those in his deposition. A comparison of the statements does not raise an inference that Dr. Gottheimer was being evasive, let alone lying. Thus, I cannot agree with the majority that the at worst minor inconsistencies in Dr. Gottheimer's testimony make it highly probable that his testimony would not have affected the District Court's judgment.

Accordingly, I do not think that the District Court's erroneous exclusion of Dr. Gottheimer's expert witness testimony was harmless. Therefore, I would remand this case for a new trial in which Dr. Gottheimer's testimony could be presented; hence, I dissent.

Present: BECKER, Chief Judge, SLOVITER, STAPLETON, MANSMANN, GREENBERG, SCIRICA, NYGAARD, ALITO, ROTH, LEWIS, McKEE, RENDELL, WEIS,* and GARTH,* Circuit Judges.

SUR PETITION FOR PANEL REHEARING WITH SUGGESTION FOR

The petition for rehearing filed by Appellants having been submitted to the judges who participated in the decision of this Court, and to all the other available circuit judges in active service, and no judge who concurred in the decision having asked for rehearing, and a majority of the circuit judges of the circuit in regular active service not having voted for rehearing by the court en banc, the petition for rehearing is DENIED. Chief Judge Becker would grant rehearing en banc.

One day after we filed the opinion in this appeal, on March 22, 1999, the Supreme Court rendered its decision in Kumho Tire Co. v. Carmichael, 119 S. Ct. 1167 (1999). Kumho directed that the Daubert factors be applied equally to non-scientific expert testimony as well as to scientific expert testimony. Although the Petition for Rehearing urges that Kumho now requires a change in this court's panel majority opinion, we are satisfied that if anything, Kumho strengthens our analysis.1

It should be noted that this court's majority opinion specifically emphasized that "measured by any standard, scientific or non-scientific, the District Court did not abuse its discretion in excluding Dr. Gottheimer's testimony." Slip Op. at ----. In so stating, we equated non-scientific expert testimony with scientific expert testimony insofar as the Daubert factors are concerned. Thus, with the emphasis on a district court's discretion to exclude or to admit expert testimony under General Electric Co. v. Joiner, 522 U.S. 136, 118 S. Ct. 512, 139 L. Ed. 2d 508 (1997), and Kumho, we are not persuaded that the Supreme Court's Kumho opinion, which reversed the Eleventh Circuit, requires any modification of our disposition as reflected in Part VI of the filed majority opinion. Judge Weis joins in this opinion.