Anthony M. Feeherry, Boston, Mass., with whom Joseph L. Cotter, Susan K. Hoffman, Henry C. Dinger, and Goodwin, Procter & Hoar, Boston, Mass., were on brief for plaintiff, appellant.

James T. Grady, Boston, Mass., with whom Gabriel O. Dumont, Jr. and Grady, Dumont & Dwyer, Boston, Mass., were on brief for defendants, appellees.

Mitchell L. Strickler, Deputy Gen. Counsel, Washington, D.C., with whom Henry Rose, General Counsel, Baruch A. Fellner, Associate Gen. Counsel, J. Stephen Caflisch, Sp. Counsel, Peter H. Gould, Deputy Asst. Gen. Counsel, Terence G. Craig, David F. Power, Washington, D.C., and Louise E. Cayne, Nathan Lewin, Martin D. Minsker and Miller, Cassidy, Larroca & Lewin, were on brief for Intervenor Pension Ben. Guar. Corp.

Before CAMPBELL, Chief Judge, ALDRICH, COFFIN, BOWNES, BREYER and TORRUELLA, Circuit Judges.

COFFIN, Circuit Judge.

Appellant Keith Fulton & Sons, Inc. (Fulton) originally challenged the constitutionality of the Multiemployer Pension Plan Amendments Act of 1980 (MPPAA), 29 U.S.C. Secs . 1381, et seq., Pub.L. No. 96-364, 94 Stat. 1208 (1980), on a number of grounds. A panel of this court upheld most of the act, but found that certain presumptions in the statute, 29 U.S.C. Sec . 1401(a)(3), deprived employers of procedural due process. Because of the importance of the issue, the unanimous authority of other circuits against the panel's view,1 and the questions raised by the parties following the panel decision, we decided to reconsider whether the prescribed method of assessing withdrawal liability, with its statutory presumptions, is constitutional. After a careful review of the statute, its legislative history, and the opposing arguments, we have concluded that the statute does meet the requirements of procedural due process.

I. BACKGROUND

The facts surrounding Fulton's withdrawal from the New England Teamsters and Trucking Industry Pension Fund (the Fund) are set out in the opinion of the panel at 762 F.2d 1124 (1st Cir.1984). Although that opinion also describes the history of the MPPAA and some of its provisions, we believe it necessary to review certain aspects of that discussion as background for this decision.

The MPPAA was enacted as an amendment to the Employee Retirement Income Security Act of 1974 (ERISA), 29 U.S.C. Secs . 1001, et seq., and was designed, in part, to strengthen multiemployer pension plans financially by discouraging employers from withdrawing and leaving a plan with unfunded liabilities. H.R.Rep. No. 869, 96th Cong., 2d Sess. 67, reprinted in 1980 U.S.Code Cong. & Ad.News 2918, 2935 [hereinafter House Report]. It does so by requiring a withdrawing employer to pay its share of the shortfall; this payment is known as the "withdrawal liability". The statute provides that the amount of withdrawal liability is calculated by the trustees of the pension plan. If an employer disputes either the amount or the fact of liability, it can negotiate with the pension plan and, if there is no resolution, the dispute must be arbitrated. 29 U.S.C. Sec . 1401(a)(1). Either party to the arbitration may appeal that decision to a district court. 29 U.S.C. Sec . 1401(b)(2).

The constitutional objections which we consider at this time focus on the method of calculating the withdrawal liability and the deference accorded that calculation. Fulton argues that it is a violation of due process for the trustees to determine an employer's withdrawal liability in the first instance, since their goal always would be to maximize the amount in order to swell the pension fund's coffers. The trustees' natural bias is compounded, it is argued, by the statutory presumptions of correctness given to that initial determination. Subsection A of 29 U.S.C. Sec . 1401(a)(3) provides that, for purposes of arbitration proceedings under the MPPAA, "any determination made by a plan sponsor under sections 1381 through 1399 of this title and section 1405 of this title [all relating to calculation of withdrawal liability] is presumed correct unless the party contesting the determination shows by a preponderance of the evidence that the determination was unreasonable or clearly erroneous." Subsection B of Sec. 1401(a)(3) provides that:

"In the case of the determination of a plan's unfunded vested benefits for a plan year, the determination is presumed correct unless a party contesting the determination shows by a preponderance of evidence that--

  (i) the actuarial assumptions and methods used in the determination were, in the aggregate, unreasonable (taking into account the experience of the plan and reasonable expectations), or

  (ii) the plan's actuary made a significant error in applying the actuarial assumptions or methods."

A district court reviewing the arbitrator's award must then presume the arbitrator's findings of fact to be correct unless they are rebutted by "a clear preponderance of the evidence." 29 U.S.C. Sec . 1401(c). Fulton contends that these presumptions make the trustees' determinations "virtually unassailable", and they argue that the inability to meaningfully challenge the initial calculation is a denial of due process.

II. DISCUSSION

We begin our analysis with the recognition that:

" '[i]t is by now well established that legislative Acts adjusting the burdens and benefits of economic life come to the Court with a presumption of constitutionality, and that the burden is on one complaining of a due process violation to establish that the legislature has acted in an arbitrary and irrational way. See, e.g., Ferguson v. Skrupa, 372 U.S. 726 [83 S.Ct. 1028, 10 L.Ed.2d 1347] (1963); Williamson v. Lee Optical Co., 348 U.S. 483, 487-488 [75 S.Ct. 461, 464, 99 L.Ed. 563] (1955).' " Usery v. Turner Elkhorn Mining Co., 428 U.S. 1, 15, 96 S.Ct. 2882, 2892, 49 L.Ed.2d 752 (1976).

Fulton's argument that 29 U.S.C. Sec . 1401(a)(3) is unconstitutional must, therefore, receive rigorous scrutiny. In determining whether due process is satisfied, we are required to balance the private interest that will be affected, the risk of error inherent in the challenged procedure, and the government's interest in using the procedure, Mathews v. Eldridge, 424 U.S. 319, 335, 96 S.Ct. 893, 903, 47 L.Ed.2d 18 (1976), but we also "are mindful that 'due process is flexible and calls for such procedural protections as the particular situation demands,' " Republic Industries v. Teamsters Joint Council, 718 F.2d 628, 640, (quoting Morrissey v. Brewer, 408 U.S. 471, 481, 92 S.Ct. 2593, 2600, 33 L.Ed.2d 484 (1972)). Thus, our inquiry is not whether there is a fairer method for assessing withdrawal liability, but only whether the method Congress chose is fair enough. The statute can not be arbitrary, but it need not be perfect.

In a sense, Fulton takes issue with two separate aspects of the MPPAA arbitration procedures: the use of a non-neutral party, the fund trustees, to calculate the liability in the first instance, and the use of presumptions to support that initial calculation. While statutory presumptions are not uncommon and have been upheld frequently,2 the original panel in this case was concerned about the due process implications of granting the presumptions on behalf of apparently non-neutral trustees. Although we recognize that the trustees come to their task of calculating withdrawal liability with a bias, we do not believe that their lack of neutrality, even when aided with the statutory presumptions, deprives Fulton and other employers of due process. We do not minimize the hardship to Fulton, a sole stockholder corporation assessed a withdrawal liability of $468,637 when it withdrew from the Fund after the City of Cambridge acquired its land in a federally subsidized taking for a public transportation project. We simply rule that the action Congress took was within its discretion.

We emphasize first that we do not perceive the trustees to be performing an adjudicatory role when they calculate the withdrawal liability, and so we do not believe we in any way mar the purity of the judicial process in deciding that their bias is not of constitutional dimension. Rather than serving as judicial decisionmakers, the trustees are simply part of an administrative procedure set up by Congress for reaching an initial determination of withdrawal liability which is then challengeable in arbitration and in court. "If there is a liability, someone has to fix it", Shelter Framing Corp. v. Carpenters Pension Trust, 543 F.Supp. 1234, 1244 (C.D.Cal.1982), aff'd in part and rev'd in part on other grounds, 705 F.2d 1502 (9th Cir.1983), rev'd sub nom. Pension Benefit Guaranty Corp. v. R.A. Gray & Co., --- U.S. ----, 104 S.Ct. 2709, 81 L.Ed.2d 601 (1984), and Congress simply assigned the duty to the persons with the most information. See Republic Industries, 718 F.2d at 640 n. 13. The trustees, moreover, do not have unbridled discretion in fixing the amount of liability. They are required to compute the employer's debt according to "detailed and explicit statutory guides", Textile Workers Pension v. Standard Dye & Finishing, 725 F.2d at 855.

A brief description of how the trustees determine withdrawal liability underscores the difference between their work and that of a judge. The trustees must use an accepted actuarial method for computing the unfunded vested liability of a plan, 29 U.S.C. Sec . 1393(a), which is the total amount owed to participating employees at the time of the withdrawal.3 The Act also sets out four methods which may be used for allocating this amount among employers. 29 U.S.C. Sec . 1391(c). Although the trustees unquestionably are allowed to make initial choices for allocating the fund shortfall, and the choices they make unquestionably can have the effect of increasing or decreasing the amount of withdrawal liability, their discretion is primarily that of selecting one from a number of procedures prescribed by Congress the first time a withdrawal issue arises. The Act specifically requires that a plan's rules for determining withdrawal liability "operate and be applied uniformly with respect to each employer." 29 U.S.C. Sec . 1394(b). Uniformity would seem to require that a rule applying to all employers (should they withdraw) in Year One also would apply to all employers in subsequent years, in the absence of significant changes of circumstance.4 Thus, in future years, the trustee's function is likely to become that of simply applying a designated method of computation to all later withdrawals. In addition, their knowledge that they must act uniformly over time, treating like cases alike, significantly cabins their discretion, for it tends to create a uniform employer interest in favor of fair rules and, given the need to maintain employer membership and satisfaction, it entails a trustee obligation of fair treatment to withdrawing employers as part of their obligation to the fund.

This case, therefore, is unlike Ward v. Village of Monroeville, 409 U.S. 57, 93 S.Ct. 80, 34 L.Ed.2d 267 (1972), in which the town mayor also acted as the initial judge for certain minor violations. The Supreme Court in Ward found a due process violation because the mayor had an obvious interest in increasing town revenues and the "[p]etitioner [was] entitled to a neutral and detached judge in the first instance." Id. at 61-62, 93 S.Ct. at 84. Unlike the mayor in Ward, whose role was to adjudicate traffic offenses and impose fines on a case-by-case basis, "the trustees play a mixed role since much, but admittedly not all, of their task is ministerial in nature," Republic Industries, 718 F.2d at 640 n. 13.

On this issue, then, we are in accord with the court in Dorn's Transp., Inc. v. I.A.M. Nat. Pension Fund, 578 F.Supp. 1222, 1237-38 (D.D.C.1984):

"[T]he role they perform in the statutory scheme is one of administration or enforcement rather than of adjudication. Both the fact and the degree of liability are determined by the trustees according to statutorily mandated standards and methods which otherwise satisfy the requirements of due process.... In this respect, the precision of the Congressionally-prescribed standards cures the process of the innate taint of partiality." See also Textile Workers Pension, 725 F.2d at 855; Republic Industries, 718 F.2d at 640-41 and 640 n. 13; Shelter Framing Corp., 543 F.Supp. at 1244.

Even if the trustees need not be as neutral as judges, their decisions must nevertheless provide due process to the employers. Fulton believes the trustees can not calculate liabilities in a constitutional manner because of their institutional bias and the presumptions which reinforce their biased decisions. We disagree for two reasons. First, we do not believe the trustees are as biased against the employers as Fulton suggests; and, second, we believe the statutory procedure lawfully envisions a less perfect calculation than Fulton demands. We discuss each of these points separately below.

Trustee Bias

An equal number of trustees are chosen by the participating employers and the participating unions, and so their allegiances are not uniformly against the withdrawing employer. Although it has been argued that even employer trustees would be inclined to inflate the withdrawal liability to lessen the burden on employers who remain in the plan, Republic Industries, 718 F.2d at 640; Shelter Framing, 543 F.Supp. at 1244, such an approach is unlikely since the initial choice of a method for calculating the liability will be used on all subsequent withdrawals from that fund, possibly including withdrawal of the trustees' own employers. Although the trustees are fiduciaries to the fund, and must consider the fund's interest above all else, that does not mean always choosing the highest withdrawal liability:

"Plan fiduciaries are given a great deal of flexibility to strike a balance among the competing considerations of encouraging new entrants, discouraging withdrawals, easing administrative burdens, and protecting the financial soundness of a fund. The committee wishes to make it clear that in choosing the rules that would eliminate or reduce liability, the choice of such a rule is not per se a violation of fiduciary standards; the determination must be made as to whether the fiduciary has acted reasonably and in the interests of plan participants and beneficiaries and otherwise in accordance with the fiduciary standards." H.R.Rep. No. 869, 96th Cong., 2d Sess. 67, reprinted in 1980 U.S.Code Cong. & Ad.News 2935.5

This commentary makes it clear that the trustees' role is not simply to find the highest value for withdrawal liability; they would not meet their fiduciary duty to the fund, for example, if they set excessively high withdrawal liabilities which would have the effect of discouraging future participation. Shelter Framing, 543 F.Supp. at 1244.

It also must be remembered that the trustees' determination is not, in fact, irrebuttable. The employer always has the opportunity to demonstrate that the trustees' calculation is unreasonable or erroneous,6 and presumably the employer could also challenge the trustees' determination by showing how bias in a particular case affected the decisionmaking process and produced a result that was unreasonable.

Thus, we can not say that an institutional bias on the part of the trustees makes unconstitutional the MPPAA's procedure for calculating withdrawal liability.

The Presumption of Reasonableness

Fulton argued in its supplemental brief that we should find the MPPAA's presumptions of correctness unconstitutional because "an employer simply has no financial incentive to go to arbitration solely over the issue of actuarial assumptions unless there is something so utterly aberrant about the case that no person could seriously doubt that the trustee erred." Although we think Fulton overstated the likely impact of the presumptions on employers, we also think it is true that the presumptions discourage litigation. That, however, is precisely their point:

"These rules are necessary in order to ensure the enforcability [sic] of employer liability. In the absence of these presumptions, employers could effectively nullify their obligation by refusing to pay and forcing the plan sponsor to prove every element involved in making an actuarial determination." House Report at 86, U.S.Code Cong. & Admin.News 1980, p. 2954.

We think it is not irrational for Congress to impose presumptions as a disincentive to challenging the trustees' determination precisely because of circumstances which Fulton emphasizes, namely that the uncertainties inherent in making actuarial assumptions mean that there is a range of "reasonable" withdrawal liability amounts. Fulton uses the fact that there are a number of reasonable values for withdrawal liability to argue that one figure must be the best, or the fairest; that the trustees will not be looking for that figure but for the highest figure; that the presumptions operate to sustain that figure; and that that is a denial of due process. What must not be overlooked, however, as Fulton concedes, is that "virtually any of the available methods of valuing the Fund's assets and liabilities may well be 'reasonable' in an actuarial sense", and that the nature of the "actuarial art" enables the trustees "to choose from a wide range of assumptions, all of which are, in some sense, 'reasonable' " (emphasis added).

We think this factually compelled concession gives the game away. Fulton concedes there are several "correct" methods, yet insists it is entitled to a better "correct" method than the Fund chose. We do not think due process requires this much. Congress apparently recognized the imprecision of forecasting a plan's financial future, and created the presumptions to avoid "[l]engthy, futile and costly bickering over the adequacy of the methods chosen." Dorn's, 578 F.Supp. at 1239.7 We think a convincing test of the constitutionality of this approach is to ask if Congress could have enacted a flat requirement that would have been even harsher on employers. We have no doubt that in furtherance of its objective of shoring up financially troubled multiemployer pension plans, see House Report at 54-55, reprinted in 1980 U.S.Code Cong. & Ad.News 2922-23, Congress could have unilaterally imposed the highest reasonable withdrawal liability, since Congress need not have a perfect fit in matching economic legislation with its purpose. Turner Elkhorn, 428 U.S. at 19, 96 S.Ct. at 2894; Ferguson v. Skrupa, 372 U.S. 726, 730-32, 83 S.Ct. 1028, 1031-32, 10 L.Ed.2d 93 (1963); Williamson v. Lee Optical, 348 U.S. 483, 487-88, 75 S.Ct. 461, 464, 99 L.Ed. 563 (1955). If the imposition of withdrawal liability is constitutional, and we have held that it is, 762 F.2d 1124, then Congress' decision to impose any reasonable amount must also be constitutional. Thus, instead of a flexible system with presumptions, Congress could have constitutionally required the trustees or even an independent agency to use the actuarial method which would produce the highest reasonable withdrawal liability. A rigid system imposing the highest reasonable amount would do the job of minimizing disputes over withdrawal liability amounts even better than presumptions since it would avoid arguments, like Fulton's, over what amount is "most reasonable". Since Congress could draft a statute to accomplish precisely what it has accomplished through the presumptions, but with less responsiveness to individual cases, "we do not think that Congress' choice of statutory language can invalidate the enactment when its operation and effect are clearly permissible." Turner Elkhorn, 428 U.S. at 23, 96 S.Ct. at 2896. We think it aids a proper sense of perspective to reflect that, after all, the Supreme Court upheld the retroactive application of the MPPAA, Pension Benefit Guaranty Corp. v. R.A. Gray & Co., --- U.S. ----, 104 S.Ct. 2709, 81 L.Ed.2d 601. Certainly, if due process allows the retroactive imposition of this potentially substantial liability, it must allow deference to a calculation of that liability which even Fulton admits is likely to be reasonable, and which can be struck down if shown to be unreasonable.

Our rejection of Fulton's argument does not mean that we believe Congress picked the best method for ascertaining an employer's withdrawal liability. It is quite likely that an expert arbitrator who would handle all disputes concerning a specific pension fund's withdrawal liability would be "fairer" than a set of trustees who, to one extent or another, have an interest in the calculation. See Note, Trading Fairness for Efficiency: Constitutionality of the Dispute Resolution Procedures of the Multiemployer Pension Plan Amendments Act of 1980, 71 Geo.L.J. 161, 190-91 (1982). Finding the best method, however, is not our function; under the due process precedents which guide us, we must only find that Congress acted rationally in designing the procedure for calculating withdrawal liability.

We acknowledge that the system of presumptions means that, at times, a "reasonable" calculation by the trustees will prevail over the "most reasonable" calculation to which Fulton argues it is entitled. Congress could rationally have decided, however, that the search for the "most reasonable" withdrawal liability figure would undermine its goal of securing financial stability for multiemployer pension plans and so, in the balance, a procedure that in most cases will produce at least a reasonable assessment is good enough.8 Moreover, if an injustice does occur, the procedure provides for relief in arbitration or in federal court.

It is also instructive to consider the impact of a decision for Fulton. Without the presumption in favor of the trustees' determination, results in arbitration are likely to vary for withdrawals even in the same year from a given plan, undermining Congress' evident intent that plan rules be administered uniformly with respect to each employer, 29 U.S.C. Sec . 1394(b). Elimination of the presumption will also lead to an increase in litigation over the withdrawal liability assessments, even in cases when the amount apparently is correct but the employer hopes it can slip a marginal case for another amount past a sympathetic arbitrator. Fulton concedes the presumptions now discourage such wistful litigation. Congress passed this legislation to protect financially shaky pension plans, and we would be stripping a layer of lawful protection from the plans by removing the presumption and opening them up to an increased load of litigation.

We recognize that there are no precise analogies to the type of delegation Congress made to pension fund trustees in the MPPAA. Perhaps this is because it is a further extension of a growing trend of delegating to those closer to a problem the difficult task of unraveling conflicting facts and assumptions, a trend already amply reflected by the work of administrative agencies. Its emergence in the area of multiemployer pension funds is perhaps brought about by the nature and impact of the task; calculating withdrawal liabilities is a highly technical endeavor, affecting many thousands of employees, in which certainty is lacking and litigation expenses are likely to impose precisely the economic harm which Congress is trying to avoid. It is in such circumstances that the detached view of the judge is least helpful.

We are not without some comparisons, however, and those we have found convince us that we have reached the correct resolution of this case. In Turner Elkhorn, 428 U.S. 1, 96 S.Ct. 2882, the Supreme Court upheld a number of evidentiary presumptions in the Black Lung Benefits Act of 1972, including two irrebuttable ones, establishing coal miners' entitlement to death or disability benefits for pneumoconiosis, or black lung disease. In finding the statute constitutional, the Court noted that presumptions arising in civil statutes involving matters of economic regulation do not violate due process as long as there is a " 'rational connection between the fact proved and the ultimate fact presumed' ", id. at 28, 96 S.Ct. at 2898, quoting Mobile, J. & K.C.R. Co. v. Turnipseed, 219 U.S. 35, 43, 31 S.Ct. 136, 138, 55 L.Ed. 78 (1910). The Court also stated:

" 'The process of making the determination of rationality is, by its nature, highly empirical, and in matters not within specialized judicial competence or completely commonplace, significant weight should be accorded the capacity of Congress to amass the stuff of actual experience and cull conclusions from it.' " 428 U.S. at 28, 96 S.Ct. at 2898, quoting United States v. Gainey, 380 U.S. 63, 67, 85 S.Ct. 754, 757, 13 L.Ed.2d 658 (1965).

We recognize the difference between the type of evidentiary presumption at issue in Turner Elkhorn, involving inferences from one set of facts to another, and the presumptions at issue here, which allocate the burden of proof. We do not think the difference is of significance, however, in determining the appropriate deference to pay to Congress' judgment on an issue unquestionably outside of "specialized judicial competence". As we discussed above, Congress' goal of protecting multiemployer pension plans is rationally related to its decision to presume the correctness of a calculation of withdrawal liability which almost certainly will fall within a range of reasonableness. In this context, moreover, the employers have the ability to challenge the trustees' determinations; in Turner Elkhorn, certain of the presumptions were irrebuttable.

Other cases of which we take note include Vance v. Terrazas, 444 U.S. 252, 100 S.Ct. 540, 62 L.Ed.2d 461 (1980), in which the Supreme Court upheld an evidentiary presumption that a U.S. citizen who declares allegiance to a foreign state intended the expatriating conduct. Despite a recognized preference for requiring clear and convincing evidence to prove expatriation, the Court upheld the presumption because the proceeding involved was civil in nature and did not threaten a loss of liberty. Again, while we recognize the distinction between the presumption at issue in this case and the presumption in Vance, we contrast the importance of the matter at issue there, an individual's citizenship, with the economic interest at stake here.

The Supreme Court itself made a similar contrast when it upheld the constitutionality of duration-of-relationship requirements for Social Security benefit eligibility for surviving wives and stepchildren of deceased wage earners:

"The Constitution does not preclude such policy choices as a price for conducting programs for the distribution of social insurance benefits. [Citation omitted.] Unlike criminal prosecutions, or the custody proceedings at issue in Stanley v. Illinois [405 U.S. 645, 92 S.Ct. 1208, 31 L.Ed.2d 551 (1972) ], such programs do not involve affirmative Government action which seriously curtails important liberties cognizable under the Constitution. There is thus no basis for our requiring individualized determinations when Congress can rationally conclude not only that generalized rules are appropriate to its purposes and concerns, but also that the difficulties of individual determinations outweigh the marginal increments in the precise effectuation of congressional concern which they might be expected to produce." Weinberger v. Salfi, 422 U.S. 749, 785, 95 S.Ct. 2457, 2476, 45 L.Ed.2d 522 (1975).

While Salfi, too, provides an imperfect analogy with the issue of Congress' prescribed method for computing withdrawal liability, it, like Vance, suggests to us that the present case falls even more clearly within the bounds of due process. Unlike Salfi, which reflected a generalized, irrebuttable presumption that recent marriages were more likely than not shams designed specifically to secure Social Security benefits, this case involves only a rebuttable presumption that a calculation which probably will be within a range of reasonableness is an appropriate withdrawal liability.

Although we concede that Congress may not have prescribed the fairest plan for assessing withdrawal liability, our review of the statute in light of the standard of due process convinces us that the provisions of 29 U.S.C. Sec . 1401(a)(3) are not constitutionally deficient. "Read together ... these provisions do little more than allocate the burden of proof to the challenger and direct that issues which are close be resolved in favor of the nonjudicial dispute resolver", Republic Industries, 718 F.2d at 640-41.

The district court's denial of Fulton's motion for summary judgment on this issue is affirmed.

BAILEY ALDRICH, Senior Circuit Judge, dissenting.

Possibly it is pointless, and even presumptuous, to dissent, given that the court's opinion accords with the conclusions of all other courts that have spoken on this issue. However, with respect, I am troubled. To put plaintiff's claim baldly, he finds himself, without possibility of relief, in a boxing ring from which he had been told he might be exempt, the prize being his own money, in an uneven match, but assured by the referee that he would see to it that it was a fair fight. Although not put in exactly those terms, the justification offered for this is that it was his own fault for associating with people like that, and his opponent needs the money. I agree that some of plaintiff's contentions are incorrect. I do not feel they all are.

Translating, plaintiff was a contributor to a multiemployer ERISA pension fund pursuant to its collective bargaining agreements with Teamster Local 379. By the Fund's Agreement and Declaration of Trust, Art. VI, Sec. 7, "[t]he financial liability of any Employer shall in no event exceed the obligation to make contributions as set forth in its applicable collective bargaining agreement with the Union or Unions." Under the original version of ERISA, to which this language was obviously subject, if the Fund were to terminate, all employers who had contributed at any time within the preceding five years were collectively liable to the Pension Benefit Guaranty Corporation (PBGC), the government corporation responsible for insuring against insufficiency in the trust funds, in an amount not to exceed 30% of the employer's net worth. In anticipation of this possibility, every withdrawing employer was required to furnish security, in one manner or another. If the Fund did not terminate in five years, the employer would receive its payment back; if the Fund did terminate, the employer would be entitled to a refund of any amount not needed to meet the Fund's liabilities. 29 U.S.C. Sec . 1363 (Pub.L. 93-406, Title IV, Sec. 4063, Sept. 2, 1974, 88 Stat. 1030).

On November 10, 1980, plaintiff's land adjoining the railroad tracks needed to carry on its activities was taken by eminent domain. Since no other usable land was available, plaintiff was promptly out of business. It, accordingly, ceased contributing to the Fund. In the meantime, on September 26, 1980, Congress had passed an amendment to ERISA, the MPPAA, which provided that all employers withdrawing from plans after April 29, 1980 were immediately subject to an assessment, payable in monthly installments over a period of five years. 29 U.S.C. Secs . 1381-1405, 1461(e). Unlike the prior statutory scheme, this assessment was not refundable under any circumstances, regardless of how well the Fund fared. Hence plaintiff's future contingent liability, if any, in an unknown amount, under the prior section 1363, was replaced by an absolute liability, commencing forthwith, under the MPPAA.

Nor was this an assessment required only in case of voluntary withdrawal, and hence one that could be avoided by remaining a Fund participant. While the legislative history shows that the 1980 amendment to ERISA was sparked by reports that employers were withdrawing from plans in such numbers as to threaten their future solvency--as employers then had a right to do, subject to the five year contingency assessments--so that legislation was desirable to put a halt to such conduct, the act as passed applied to all employers alike, even if withdrawal had been forced upon them. Although I mention this particular hardship to plaintiff, my dissent is not based upon it, but goes simply to the assessment procedure. Some might think, however, that it emphasizes the need of fairness, to the extent that fairness was practicably achievable.

While substituting absolute liability for contingent liability would be an obvious violation of contract, were that the test, I agree that is not the test, and the case is governed by the less stringent requirements of due process. However, I cannot share the ease with which other judges are content to find such. To put the case bluntly, could Congress, deciding that ERISA plans were a great idea but, with inflation and all, insufficient, increase the benefits and require retroactive contributions? Obviously not, but to what extent, then, is Congress justified in substituting an absolute liability for a contingent one? That this is an increase in the burden a participating employer originally assumed would seem manifest from the primary reason Congressional leaders advanced for enacting the amendments--to discourage employers from, viz., to put a premium upon, voluntary withdrawal. H.R.Rep. No. 96-869, Part I, 96th Cong., 2d Sess. 54-55, reprinted in 1980 U.S.Code Cong. & Ad.News 2918, 2922-23; H.R.Rep. No. 96-869, Part II, 96th Cong., 2d Sess. 10, 15, reprinted in 1980 U.S.Code Cong. & Ad.News 2992, 3001, 3004. See also 29 U.S.C. Sec . 1001a(a)(4).

However, even though more burdensome in both procedure and substance than the prior statutory scheme, I do not object insofar as the MPPAA merely filled a loophole to assure the employees' ultimate receipt of their promised payments. Employers have had the benefit of offering their employees so-called vested pension rights, and I do not quarrel with Congress's power to call on them to make fully good. Nor, although not without some reluctance, do I cavil against the fact that this obligation is averaged across the board, so that some withdrawing employers may be charged unevenly and not in accordance with their exact deserts. In the interests of practicality, some play may be allowed in "allocat[ing] to the [employer] an actual, measurable cost...." Usery v. Turner Elkhorn Mining Co., 428 U.S. 1, 19, 96 S.Ct. 2882, 2894, 49 L.Ed.2d 752 (1976). But this is far enough. My difficulty stems from a belief that when contracts are set aside and an agreed-upon insurance premium is increased retroactively, Congress cannot add the burden of having the amount of the premium determined by a body that is so seriously lacking in impartiality as to create a serious risk of a substantial overcharge. Further, although in the ordinary case an original factual determination may properly be buttressed on review by a presumption of correctness, according that presumption here has entrenched the unfairness.

As the court points out, the amount to be assessed against a withdrawing employer is determined in the first instance by the trustees of the Fund. If the employer is dissatisfied, he may negotiate with the trustees and, if still dissatisfied, he may obtain arbitration. 29 U.S.C. Sec . 1401(a)(1). Either party may appeal the arbitrator's decision to the district court. Sec. 1401(b)(2). The employer reaches the arbitrator facing a finding by the trustees that "is presumed correct unless the [employer] shows by a preponderance of the evidence that the determination was unreasonable or clearly erroneous." Sec. 1401(a)(3)(A). Specifically, the actuarial calculation underlying the trustees' finding "is presumed correct unless [the employer] shows by a preponderance of the evidence that (i) the actuarial assumptions and methods used ... were, in the aggregate, unreasonable ..., or (ii) the plan's actuary made a significant error in applying the actuarial assumptions or methods." Sec. 1401(a)(3)(B). And, of course, the arbitrator's decision carries similar weight in the district court. Sec. 1401(c). At the base of all this procedure is a decision by trustees who are not disinterested, but who have, as the court concedes, a bias.

This is not a mere appearance of bias; it is a real, indeed multiple, bias, rooted both in the trustees' statutory duty and in their personal circumstances. The trustees, as fiduciaries under 29 U.S.C. Sec . 1002(21)(A), are subject to the duties imposed by 29 U.S.C. Sec . 1104:

"[A] fiduciary shall discharge his duties with respect to a plan solely in the interest of the participants and beneficiaries and--

  (A) for the exclusive purpose of:

  (1) providing benefits to participants and their beneficiaries; and

  (ii) defraying reasonable expenses of administering the plan...."

In spite of language in my brothers' opinion, there appears no equal duty to an employer who has withdrawn from the plan. See NLRB v. Amax Coal Co., 453 U.S. 322, 329-30, 101 S.Ct. 2789, 2794, 69 L.Ed.2d 672 (1981). And, in any event, as between protecting employers who have abandoned the plan and protecting the solvency of a fund within their charge, could it be doubted where the trustees' primary, and important interest lies? The trustees face potential civil liability if they breach these statutorily-imposed fiduciary obligations. 29 U.S.C. Sec . 1109.

In addition, the trustees have personal interests in their decision beyond avoiding being charged for misfeasance. If, as the court suggests, they look ahead to possible future withdrawals on their own part, obviously the more solvent they make the fund today, at the withdrawing employer's expense, the smaller the possible deficit they will be personally responsible for in the future. Even if they do not contemplate future withdrawal, augmenting the fund's assets today could avert possible future pressure on them to increase contributions to meet unfunded liability. I cannot bring myself to believe that assigning a liability determination to decision-makers in this interested position comports with fundamental concepts of due process.

My brethren recognize the general impropriety of an interested decision-maker. See Gibson v. Berryhill, 411 U.S. 564, 578-79, 93 S.Ct. 1689, 1697-98, 36 L.Ed.2d 488 (1973); Ward v. Village of Monroeville, 409 U.S. 57, 93 S.Ct. 80, 34 L.Ed.2d 267 (1972); In re Murchison, 349 U.S. 133 , 75 S.Ct. 623, 99 L.Ed. 942 (1935); Tumey v. Ohio, 273 U.S. 510, 47 S.Ct. 437, 71 L.Ed. 749 (1927); United Church of the Medical Center v. Medical Center Comm'n, 689 F.2d 693 (7th Cir.1982). They also recognize the trustees' position, stating, "[T]he trustees are fiduciaries to the fund, and must consider the fund's interest above all else...." They offer several answers. First, "[W]e do not perceive the trustees to be performing an adjudicatory role when they calculate the withdrawal liability.... Rather than serving as judicial decision-makers, the trustees are simply part of an administrative procedure set up by Congress for reaching an initial determination of withdrawal liability which is then challengeable in arbitration and in court." This is a sequential account chronologically, but, with respect, it is a total non-sequitur so far as meeting the issue of bias is concerned. In what way are the trustees not performing an adjudicatory role? I am at a loss to understand that statement. And why is it any the less adjudicatory because it is subject to review? And particularly I ask this question when it carries a presumption of correctness.

Nor does it seem helpful to say, "Congress simply assigned the duty to the persons with the most information." An administrative law judge who owns stock in a corporate employer might be the best informed as to what was going on in the company, but would his interest in a proceeding involving that company be acceptable because its decision could be reviewed by the NLRB and challenged in court?

The court then says that the trustees do not have "unbridled discretion." The most biased judge does not have unbridled discretion; he is bound by the evidence and the rules of law. Even the need of uniformity, of which the court makes much (e.g., the sentence quoted from Dorn's Transportation, Inc. v. I.A.M. National Pension Fund, 578 F.Supp. 1222, 1238 (D.D.C.1984): "[T]he precision of the Congressionally-prescribed standards cures the process of the innate taint of partiality."), does not adequately cabin the trustees' discretion. "The Congressionally-prescribed standards" certainly do not produce anything like the precision of an automatic computer printout. It is far from it. Great discretion resides in the trustees in calculating the amount of unfunded vested liability, in some cases the actual value of the fund assets, and in all cases the present value of vested accrued benefits. Indeed, the court's objection to what it claims to be the employer's objection to the presumption of correctness (which, standing alone apart from bias, the employer does not object to) is the wide range of possible alternative "reasonable" liability figures that could thereafter be pressed if there were no presumption to dampen litigation. In one breath the court says the trustees are not really deciding anything, and in the next it says there is so much room for disagreement that the trustees' decision must be given the protection of a presumption of correctness. I cannot conclude that the trustees are performing merely "ministerial" calculations, as distinguished from quasi-judicial decision-making. See Marshall v. Jerrico, Inc., 446 U.S. 238, 243, 247, 100 S.Ct. 1610, 1613, 1615, 64 L.Ed.2d 182 (1980); Ward, ante, 409 U.S. at 62 n. 2, 93 S.Ct. at 84 n. 2.

Nor are the trustees' decisions permanently "standardized." Of enormous importance is the choice of interest rates for discounting, and surely that is not standardized for all time. Nor do I understand why standardization means that withdrawing employers get an impartial shake; the trustees' standard assumptions may be uniformly harsh. Is not setting the size of the safety factor against possible future losses precisely the kind of decision as to which the decision-maker should not be engaged in protecting himself?

The court would answer this by saying that while "the trustees ... must consider the fund's interest above all else, that does not mean always choosing the highest withdrawal liability...." It seems an odd principle to determine the permissibility of manifest bias by weighing the decision-maker's conflicting interests in the scales of justice and finding which side predominates. If there could be such a principle, the interests favorable to the party who must pay should outweigh those contrary, or at least there should be some balance. Neither is be the case here. The court suggests that the employer-half of the board's membership would hesitate to set too high a standard, since such a standard ultimately might injure any of them who might later want to withdraw. I cannot believe that such forethought, were it engaged in, could begin to compete with the personal desirability of building up the fund in order to minimize the future deficit that employer-trustees might have to meet. Nor, as the court inferentially concedes, would the employee trustees--the other half of the board's membership--feel any such possible ambivalence.

I agree it is possible that the trustees might give some thought to weighing the desirability of building up the Fund's assets against the danger of scaring off potential future joiners, but, again, for the trustees the latter danger would hardly be likely to meet the immediacy of the calls in the former direction. Indeed, increasing the solvency of the Fund would seem to make joining more rather than less attractive. In sum, I must find the court's small list of alleged contra interests speculative, and in no way comparable to the very real bias in favor of over-assessments.

Next, and this seems both its theme and its coda, the court says, "It also must be remembered that the trustees' determination is not, in fact, irrebuttable.... [I]f an injustice does occur, the procedure provides for relief in arbitration or in federal court." (ital. in orig.) To this I would add an exclamation point. What biased decision by any body that is subject to review cannot be reversed if unreasonable? With an antiseptic gesture the court has put an end to the whole disease of bias, root and branch.

Nor is the court justified in attacking plaintiff by charging it with seeking the "most reasonable" result. This is a totally unfair characterization of plaintiff's objections to a decision made by a biased body that will be affirmed so long as it is within the confines of reasonableness. Plaintiff is not asking for the best; it is only asking for a fair shake. As we said in Marlboro v. Association of Independent Colleges, 556 F.2d 78, 82 (1st Cir.1977), "Decision by an impartial tribunal is an element of due process." The court's charge is but another way of saying that lack of an impartial tribunal does not count if it does not produce an unreasonable result.

Supplementing these arguments which I cannot find persuasive, the court seeks to distinguish on the facts the cases which point out the impropriety of biased tribunals. Thus Ward, ante, 409 U.S. 57, 93 S.Ct. at 80, where the deciding officer, the mayor, was disqualified because he had an interest in increasing town revenues, is found not applicable because "much [of the trustees'] task is ministerial in nature." But, how "ministerial" is it for the trustees to select, as here, a discount rate of 7 1/2 for determining future values, an interest rate unheard of in my memory in a decade, while the Fund's actuary admitted the rate could reasonably have been set at 14 1/2%? In Ward the mayor had no direct, or personal, interest, but only the general interest of adding to the town's assets. Here the trustees had the interest of protecting a fund to which they had an affirmative fiduciary obligation. I add that in Ward a dissatisfied defendant could obtain a trial de novo, without the handicap of a presumption.

The court does not seek to distinguish Ward as being a criminal case, but if one did, I would ask what would one prefer to have decided by a biased decision-maker, a traffic fine or a $468,000 assessment? This is financial life.

The court says, "If there is a liability, someone has to fix it," and accepts the statutory procedure because it discourages litigation. Both are quite true. I must concede that I cannot think of a better way of discouraging litigation than to erect a procedure by which a potential litigant knows it faces a biased tribunal whose decisions are backed by a presumption of correctness. This is not, however, my conception of due process.

Finally, I find peculiarly specious, despite its superficial appeal, the court's argument that any liability determination the trustees may make is acceptable so long as it is within reason because Congress could have made that choice to begin with. If Congress had enacted harsh standards, admittedly they would pass, if reasonable. But in that case a majority of the legislators would have made the selection. It is totally novel to me that possession of a discretionary power should include the power to delegate that discretion to a body not constituted to exercise it impartially, but has direct interests, in the particular case, conspicuously weighted against the payor. Such a principle would seem to have consequences far beyond this case.

What to do? The proper solution, accepting the general constitutionality of the MPPAA, would be for Congress to designate an impartial decision-maker. Until that be accomplished a make-do solution, by no means advocated as the most desirable one, would be to deny to trustees' liability decisions the presumption of correctness presently contained in section 1401(a)(3), a presumption that under due process should not be accorded to the decisions of a body that lacks impartiality. Erasing that presumption would not affect the remainder of the MPPAA and would retain the arbitration proceeding provided for in Sec. 1401(a) as the forum for the resolution of disputes over withdrawal liability in the first instance, subject to judicial review pursuant to Sec. 1401(b). The knowledge and expertise of the trustees would not be lost, since the neutral arbitrator--which could itself be expert--would have the benefit of the trustees' estimate and assumptions, as well as the withdrawing employer's. If the arbitrator were permanent (like the NLRB), it should not produce the wildly inconsistent and divergent results feared by my brethren. Absent this, to return to my beginning, the referee has been fixed. I must respectfully dissent.

Circuit Judge TORRUELLA joins in this opinion.

1 Textile Workers Pension v. Standard Dye & Finishing Co., 725 F.2d 843, 854-55 (2d Cir.1984), cert. denied sub nom. Sibley, Lindsay & Curr Co. v. Bakery, Confectionery & Tobacco Workers Int'l, --- U.S. ----, 104 S.Ct. 3554, 82 L.Ed.2d 856 (1984); Washington Star Co. v. Typographical Union Negotiated Pension Plan, 729 F.2d 1502, 1511 (D.C.Cir.1984); Republic Industries, Inc. v. Teamsters Joint Council No. 83, 718 F.2d 628, 640-41 (4th Cir.1983), cert. denied, --- U.S. ----, 104 S.Ct. 3553, 82 L.Ed.2d 855 (1984). Cf. Peick v. Pension Ben. Guar. Corp., 539 F.Supp. 1025, 1047-49 (N.D.Ill.1982), aff'd, 724 F.2d 1247, 1268 (7th Cir.1983), cert. denied, --- U.S. ----, 104 S.Ct. 3554, 82 L.Ed.2d 855 (1984) (appeals court upholds constitutionality of MPPAA without expressly addressing the specific issues we consider here, the role of the pension plan trustees and the presumptions supporting their actions, although the district court did consider the trustees' role); Terson Co. v. Bakery Drivers & Salesmen Local 194, 739 F.2d 118, 121 (3rd Cir.1984) (concluding generally that MPPAA does not violate due process)

2 See, e.g., Vance v. Terrazas, 444 U.S. 252, 100 S.Ct. 540, 62 L.Ed.2d 461 (1980); Usery v. Turner Elkhorn Mining Co., 428 U.S. 1, 96 S.Ct. 2882, 49 L.Ed.2d 752 (1976); Mathews v. Lucas, 427 U.S. 495, 96 S.Ct. 2755, 49 L.Ed.2d 651 (1976); Weinberger v. Salfi, 422 U.S. 749, 95 S.Ct. 2457, 45 L.Ed.2d 522 (1975); Welch v. Helvering, 290 U.S. 111, 54 S.Ct. 8, 78 L.Ed. 212 (1933)

3 Calculation of this amount requires the use of actuarial assumptions about such things as employee life expectancy and future rates of return on the plan's assets. The MPPAA requires that plans use actuarial assumptions and methods that "in the aggregate, are reasonable" and that "in combination, offer the actuary's best estimate of anticipated experience under the plan...." 29 U.S.C. Sec . 1393(a)(1). It is also worth noting that this calculation is made by an actuary, a professional consultant to the plan, and not by the trustees themselves. Although the trustees have the option of accepting or rejecting a given projection, the trustees' actions with regard to the actuary's estimate can be used in challenging the reasonableness of the withdrawal liability calculations. In Woodward Sand Co. and Operating Engineers Pension Trust, 3 E.B.C. (BNA) (Employee Benefits Cases) (Kaufman, Arb.) 2351 (1982), the plan trustees had rejected the actuary's recommended assumptions and the arbitrator found the plan's substitute assumptions to be unreasonable. The arbitrator reduced the employer's liability by more than 20% and awarded the employer $10,000 in attorney's fees

4 It appears to us that, under 29 U.S.C. Sec . 1394(b), the same figures would be used for all employers withdrawing in a given year. Although we assume the precise numbers on some variables--for example, interest rates--might change from year-to-year to reflect changing economic conditions, it also appears to us that Sec. 1394(b) would require use of the same figures for unchanging variables such as mortality rates. In any case, we do not believe the trustees have any significant flexibility after the first employer withdraws from a fund and the procedures for calculating withdrawal are chosen. Any deviation from the original procedure to reach a result that harms the employer could well be found unreasonable if challenged

5 We agree with the observation of the court in Peick v. Pension Ben. Guaranty Corp., 539 F.Supp. at 1048 n. 48, quoting N.L.R.B. v. Amax Coal Co., 453 U.S. 322, 343, 101 S.Ct. 2789, 2801, 69 L.Ed.2d 672 (1981) (Stevens, J., dissenting), that "[e]mployer-appointed trustees do not violate their fiduciary duties simply because they are sensitive to the management perspective of what is best for the plan:

'[T]he administration of a trust fund often gives rise to questions over which representatives of management and representatives of labor may have legitimate differences of opinion that are entirely consistent with their fiduciary duties.' "

The majority holding in Amax is not to the contrary.

6 This is not pure theory. See, e.g., Classic Coal Corp. v. U.M.W. 1950 and 1974 Pension Plans, 5 E.B.C. (BNA) (Employee Benefits Cases) 1449 (1984) (Nagle, Arb.); Perkins Trucking Co. and Local 807 Pension Fund, 4 E.B.C. (BNA) 1489 (1983) (O'Loughlin, Arb.)

7 A committee report on the MPPAA explained that the presumptions were created "to reduce the likelihood of dispute and delay over technical actuarial matters with respect to which there are often several equally 'correct approaches'. Without such a presumption, a plan would be helpless to resist dilatory tactics by a withdrawing employer--tactics that could, and could be intended to, result in prohibitive collection costs to the plan. The presumptions can be rebutted in those cases where injustice would otherwise result to the withdrawing employer." Senate Committee on Labor and Human Resources, 96th Cong., 2d Sess., Sec. 1076, The Multiemployer Pension Plan Amendments Act of 1980: Summary and Analysis of Consideration, 21 (1980)

8 "While additional procedural safeguards might contribute to the perception of the withdrawn contributors that justice had been equitably served, the extensive cost to the taxpayer and to the fund far outweighs the benefit to be gained." Dorn's, 578 F.Supp. at 1239