OPINION

PRELIMINARY STATEMENT

In this Opinion, the Court holds unconstitutional certain provisions of the Multiemployer Pension Plan Amendment Act, 29 U.S.C. Sec. 1381 ff. (MPPAA), as applied to each Plaintiff. The Plaintiffs are employers who withdrew from a pension plan covered by the Employees Retirement Income Security Act (ERISA), 29 U.S.C. Sec. 1001 ff., prior to the enactment of MPPAA. Under the retroactive provisions of the new statute, they were assessed a substantial liability, and they sued for declaratory and injunctive relief. The Plaintiffs sought a declaration that the statute and the assessments made thereunder were unconstitutional. They also sought an injunction prohibiting collection of the assessments.

The two cases, by agreement of all parties, were consolidated for discovery and pre-trial purposes. Preliminary injunctions, enjoining further efforts to collect the assessments, were entered in both cases on January 14,1982. Thereafter, both sides in both cases filed simultaneous mutual summary judgment motions on the issue of MPPAA’s constitutionality. Argument on the motions in the two cases was consolidated.

Plaintiffs challenge the validity of the statute on a number of separate grounds. As will be seen, those challenges are rejected on certain of the grounds but are sustained on the issue of retroactivity. The Court does not reach the issue of taking without just compensation.

The Opinion which follows was delivered piecemeal, orally from the bench, on March 22, 23 and 24, 1982. As an oral opinion, it does not purport to be a full and comprehensive discussion of all of the applicable case law bearing on all of the issues adjudicated. The Court’s oral remarks have been somewhat edited for publication and, in the interest of clarity, a few minor additions have been made.

FACTS

Counsel for all of the parties have stipulated in open court that there are no disputed issues of material fact and that the only questions presented by the mutual summa ry judgment motions are questions of law. They have also agreed that the matter is ripe for decision now as to the constitutional issues, and no trial on those issues is necessary or appropriate. The lawyers have filed an extensive set of stipulated facts. Other facts not covered in that stipulation emerged during the argument as being uncontroverted and they, too, have been considered by the Court. The following is a summary of the relevant uncontested facts.

Both Plaintiffs are contractors in the construction business. Plaintiff Shelter was party to a collective bargaining agreement with a local of the Carpenters Union from 1976 to 1980. As required by the agreement, Shelter contributed to the Defendant, Carpenters Pension Trust (Trust), throughout that period. Shelter’s union contract expired July 1, 1980, and negotiations for a new agreement reached an impasse on July 16, 1980. Shelter had no obligation to continue its contributions to the Trust after July 16, 1980, but in fact made payments until August 12, 1980. These dates place Shelter’s withdrawal from the plan well before September 26, 1980, the enactment date of MPPAA, but after the retroactive application date of April 29, 1980, set by the statute.

After this withdrawal, the Trust assessed against Shelter a lump sum withdrawal liability of $797,648. This is more than twice Shelter’s net worth. As authorized by the statute, the Trust established a monthly payment schedule of $22,737 per month for 40 months, totalling’ $922,489, which includes interest. These monthly payments amount to $272,850 per year, a sum which exceeds 60% of Shelter’s net worth. The first monthly payment was due June 24, 1981. When Shelter did not make that payment, the Trust declared Shelter in default and thereafter accelerated the total liability. Thus, the monthly payment plan is no longer available to Shelter, and Shelter is currently obligated to pay the original lump sum of $797,648.

Plaintiff G & R was party to a collective bargaining agreement with the same local of the Carpenters Union from 1972 to 1980 and contributed to the Defendant Trust throughout that period. The collective bargaining agreement expired July 1,1980, and renewal negotiations reached an impasse on July 18, 1980. G & R was not obligated to contribute to the Trust after July 18, 1980, but did in fact continue its contributions until August 12, 1980. The date of G & R’s withdrawal from the plan is thus also well before the enactment date of MPPAA, September 26, 1980, and after its retroactive application date of April 29, 1980.

The Trust assessed a lump sum withdrawal liability against G & R in the amount of $687,387. That sum is equivalent to 40% of G & R’s net worth. The Trust, acting under the statute, established a monthly payment schedule of $17,397 per month for 45 months commencing November 5, 1981. The monthly payments include interest and total $787,824. For one year, the monthly payments amount to $208,773, which would equal 94% of G & R’s net income for the year 1980. Because of the preliminary injunction issued by this Court, the Trust has taken no steps to accelerate the liability and has not yet declared the full original sum of $687,387 due and payable.

THE STATUTE

I think I should begin with some brief remarks about the background of the statute and the statutory objectives.

MPPAA is an amendment to a statute passed some years earlier, the Employees Retirement Income Security Act, 29 U.S.C. Sec. 1001 ff. (ERISA). ERISA was an effort by Congress to regulate and govern the conduct and administration of employee pension plans. The Act set up duties and obligations to the beneficiaries of such plans, and gave recourse to the federal courts to participants and beneficiaries who claimed abuses and maladministration.

Along with ERISA, Congress established the Pension Benefit Guaranty Corporation (PBGC) to help insure that vested rights of beneficiaries in these plans would be vindicated, and that vested obligations would be met in cases of financial difficulty.

ERISA included an initial limited approach to the problems that might be encountered when employers in multiemployer pension plans withdrew from such plans. It established a contingent obligation on the part of withdrawing employers to pay a proportionate share of the sums needed to cover vested benefits if the plan were to become insolvent during the 5-year period following withdrawal. 29 U.S.C. Sec. 1365 . “Substantial” employers who withdrew were required to post a bond to meet this contingent liability. If the plan remained solvent for five years, the withdrawing employer paid nothing. Employers could purchase insurance from PBGC to protect themselves from the contingent liability which could ensue if the plan failed within 5 years. 29 U.S.C. Sec. 1323 .

As to MPPAA, the parties have furnished a great deal of material from its legislative history. That material is not controverted and is contained in the stipulation of undisputed facts. The material is also a public record, so I could take judicial notice of it. From that material, the following things can be extrapolated.

The principal purpose of MPPAA, as enunciated to Congress in a report from the PBGC, was to reduce the incentive of employers to terminate their affiliation with multi-employer pension plans by making it more onerous and costly for them to withdraw.

PBGC was concerned with the drain on its own finances which might result if it, as an insurer, had to make good on promised benefits in many plans following withdrawal by numerous employers. PBGC was also concerned with the increased financial burden created by such withdrawals upon those employers who remained, as this increased burden in turn created a greater incentive for the remaining employers to withdraw.

The statute that Congress passed under this impetus from PBGC is an extremely lengthy and complex one. The statutory scheme may be summarized in a bare-bones fashion this way:

Whenever an employer has made a complete withdrawal from a multi-employer pension fund, the fund is required to compute the withdrawing employer’s proportion of the unfunded vested benefits of that fund. Each employer’s share is based on the proportion of its contributions to the fund over the past five years relative to all of the contributions received by the fund during those five years.

The sum so computed becomes an obligation of the withdrawing employer to the fund. That sum bears interest from the time it is assessed until it is paid. If the employer goes 60 days without paying the amount assessed, following written notice of non-payment, the Trust may declare the full amount of the liability due and payable.

The withdrawing employer is given an option of paying the liability assessed to him in monthly installments as prescribed by the pension plan trustees; but to have this option, the first installment must be paid within 60 days of notice of the amount due.

If the employer is unhappy with the amount assessed against him, he may, under certain conditions, have the matter reviewed in arbitration proceedings. He must, however, pay the first monthly in stallment and continue paying monthly installments during the arbitration process.

One of the aspects and facets of MPPAA which may well be central to the issue of its constitutionality, at least for the purposes of this case, is that it was enacted on September 26, 1980. But by the statute’s own wording and terms, it purports to be retroactively applicable to employers who withdrew from multiemployer funds prior to that date. 29 U.S.C. Sec. 1461 (e)(2)(A). It purports to impose the liability that I have described on any employer who withdrew after April 29, 1980, up to and including the enactment date of September 26, 1980. Of course, it also imposes that liability on any employer who withdrew after September 26, 1980. Both of the Plaintiffs in our cases withdrew after April 29, 1980, but before the enactment date of the statute. See footnote 2, supra.

Both Plaintiffs were operating under collective bargaining agreements which required them to make contributions to the fund. The collective bargaining agreements explicitly stated that the employer would not incur any further or future liability whatsoever if it should cease making contributions to the fund when it ceased to be bound by the collective bargaining agreement. Before MPPAA’s enactment date, both Plaintiffs properly ended their adherence to their collective bargaining agreements and thus terminated their obligations to continue making further contributions to the fund. Both Plaintiffs concede that under the law as it read at the time of their withdrawal, they incurred the contingent “five year” liability provided in ERISA and discussed above.

This general description is sufficient to set the stage. I will summarize in more detail other provisions of the statute which may be directly applicable to specific constitutional attacks. I now proceed to a discussion of the separate specific constitutional challenges made by Plaintiffs.

EXHAUSTION OF REMEDIES

Plaintiffs have come directly to this Court, seeking declaratory relief in the form of a holding that MPPAA is unconstitutional, and seeking an injunction against the enforcement of the withdrawal liability asserted by the Trust. PBGC argues that this Court should require exhaustion of the administrative remedy of arbitration before considering the constitutional challenges and that this Court lacks jurisdiction to decide the constitutional issues now. PBGC’s request that constitutional adjudication be deferred pending arbitration is denied.

Whether PBGC’s exhaustion argument is viewed as jurisdictional or discretionary, it is rejected. I hold that the Court has jurisdiction to adjudicate the constitutionality. And, if the matter involves the exercise of discretion, the Court, in its discretion, declines to defer adjudication of constitutionality pending recourse by Plaintiffs to arbitration.

Let me first discuss the claim that the Court lacks jurisdiction at this point to proceed to a constitutional adjudication. PBGC relies on decisions of our Circuit in Montgomery v. Rumsfeld, 572 F.2d 250 (9th Cir. 1978), and Eluska v. Andrus, 587 F.2d 996 (9th Cir. 1978). These cases contain language that could be interpreted as saying that failure to resort to administrative remedies deprives the District Court of jurisdiction. But on the facts of each case, such language seems too broad. The principle for which those cases actually stand is that if one approaches the Court for a form of relief which is within the scope of a statutorily mandated administrative procedure, the Court should not, ought not, and cannot grant such relief until that administrative remedy is availed of. That rule of law is simply inapplicable to the present cases and the present facts.

Plaintiffs here challenge the constitutionality of the statute. All of the parties agree that the statutory arbitration proceeding cannot adjudicate constitutional challenges.-' As I understand its position, PBGC also seems to agree that the arbitrators could not adjudicate the constitutional challenges.

If Plaintiffs were in this court trying to challenge the manner in which their obligation to the Fund was computed, or whether they had in fact withdrawn, or certain other disputes within the general framework of the statute and thus arbitrable under the statutory scheme, we might have a different case, and the argument that this court has no jurisdiction might have some force. But that is not the type of challenge that these Plaintiffs have made. They challenge the basic structure and the basic constitutionality of the statute.

PBGC also makes what I regard as a different kind of argument, perhaps not even a jurisdictional one. PBGC says that under general principles of law, the Court should require the arbitration, because to do so could have many beneficial results of which the Court should avail itself before embarking on constitutional adjudication.

PBGC says that arbitration could possibly eliminate the dispute entirely, or permit resolution of it on non-constitutional grounds, or present to the Court a better record on which constitutional adjudication could be made. Arbitration, says the agency, would permit the Court to have the benefit of the agency’s expertise as to what the statute means.

In certain cases such considerations might well be worthy of serious thought and notice, and might conceivably result in a court declining to make a constitutional adjudication. But those arguments and those considerations are simply not applicable to the present cases. We are dealing here with a straight legal constitutional challenge. Agency expertise is meaningless and of no relevance as applied to the type of challenge these Plaintiffs pose. The only expertise an agency like PBGC might possess is expertise on how the statute is to be applied and administered. That could not aid the Court in deciding a frontal challenge of this type.

Moreover, since the challenges made here are not within the scope of the arbitration, there could be no help in the adjudication of these challenges from any record that an arbitration procedure would create. As far as I can see (and I note that the parties agree with me), there is no possibility that factual findings within an arbitration could moot these constitutional challenges. As the parties agree, there is zero possibility that the assessed liability could be eliminated in an arbitration. And the possibility that the liability of each Plaintiff could even be very materially reduced is remote indeed, if not entirely non-existent. Challenges such as these Plaintiffs made, based on retroactivity, denial of right to jury trial, access to a court, and others, simply could not be mooted by an arbitration.

I am not impressed with the assertion that arbitration would develop a better record for adjudication of these constitutional issues. We have a clear factual record here now. As I see it, there would be no benefit in developing further facts through the arbitration. And I cannot see the possibility of getting any additional facts through the arbitration that would affect the Court’s duties as a constitutional adjudicator.

There is ample case law to the effect that when one is challenging the basic statute, the basic administrative framework, there is no duty to exhaust administrative remedies. Among those cases in our own Circuit are Downen v. Warner, 481 F.2d 642 (9th Cir. 1973), and Montana Chapter of Association of Civil Technicians, Inc. v. Young, 514 F.2d 1165 (9th Cir. 1975).

The doctrine of exhaustion is not an absolute rule, and not a sine qua non. There are well-known exceptions. If exhaustion would be futile, one need not exhaust. Where administrative remedies would be inadequate or inefficacious, one need not exhaust. These exceptions and the case law supporting them are well set out in Aleknagik Natives Ltd. v. Andrus, 648 F.2d 496 (9th Cir. 1980), and in Winterberger v. General Teamsters Auto Truck Drivers, Local Union 162, 558 F.2d 923 (9th Cir. 1977).

As in Winterberger , the challenge here goes to the whole statute. The basic fairness of the entire administrative structure is challenged on due process, equal protection, and other fundamental grounds. I think the Winterberger rationale applies here.

There is no statutorily mandated exhaustion requirement in MPPAA for the type of challenges made to this Court. Even if, arguendo, I construed the statute as containing an exhaustion requirement, I read the cases just mentioned as indicating that a statutorily mandated exhaustion requirement is subject to the same exceptions I have just mentioned, and this case comes within those exceptions.

One other factor is important. To make these Plaintiffs go through the complex, time-consuming, and very expensive process of arbitration to determine the actuarial bases, the accounting bases, and other questions they might be impelled to raise as to the way their liability was calculated, would be unfair and wasteful when arbitration could not adjudicate their principal grievance, which is that the basic statutory scheme is unconstitutional. There is simply no public interest to be served by imposing this barrier before addressing Plaintiffs’ assertions of unconstitutionality. To impose such a requirement under the circumstances of these cases would, in my view, be totally unreasonable.

CONSTITUTIONAL ISSUES

Before I commence a discussion of the separate types of challenges which Plaintiffs have made, I think it is worthwhile to restate some general principles which every court must observe in adjudicating challenges to the constitutionality of legislation.

Legislation is, of course, presumed to be constitutional, and the challenger has the burden of demonstrating its unconstitutionality. Statutes should be construed wherever possible so as to uphold their constitutionality. Since we deal with legislation in the economic area, and not with criminal or First Amendment legislation, special deference must be paid to Congress’ judgment. Generally speaking, economic legislation should be upheld if it bears a rational relationship to a legitimate governmental interest.

Plaintiffs present their constitutional challenges on many grounds, and I have grouped them in the following way for purposes of convenience: where a civil statute imposes penalties for prohibited conduct, the general rule is that one is entitled to reasonable notice and reasonable definition of what the prohibited conduct is. Papachristou v. City of Jacksonville, 405 U.S. 156 , 92 S.Ct. 889 , 31 L.Ed.2d 110 (1972). The entire vagueness concept is based on fairness and notice.

In legislation involving economic regulation, less precision is required. The courts recognize that legislatures, including Congress, cannot dot every “i” and cross every “t”, and that they need only define the scope of their edict so far as is practicable and feasible. The courts have said many times that legislatures must deal with numerous and unforeseen variations in factual situations and for that reason are permitted, in economic areas, to use general and imprecise language.

With respect to economic legislation, many extremely vague phrases have withstood constitutional challenges. Among the significant cases are: United States v. National Dairy Products Corp., 372 U.S. 29, 29-30 , 83 S.Ct. 594, 596 , 9 L.Ed.2d 561 (1963) (“unreasonably low prices”); Boyce Motor Lines v. United States, 342 U.S. 337, 339 , 72 S.Ct. 329, 330 , 96 L.Ed. 367 (1952) (“so far as practicable and where feasible”); Sproles v. Binford, 286 U.S. 374, 393 , 52 S.Ct. 581, 587 , 76 L.Ed. 1167 (1932) (“shortest practicable route”); diLeo v. Greenfield, 541 F.2d 949, 955 (2nd Cir. 1976) (“other due and sufficient cause”), Brennan v. Occupational Safety & Health Review Commission, 505 F.2d 869, 872 (9th Cir. 1974) (“near proximity”).

Among the statutory terms which Plaintiffs challenge as invalidly vague are “building and construction industry,” and the classification .of employers covered by MPPAA as those where “substantially all” of the employees continue to perform work “in the jurisdiction of the collective bargaining agreement”. 29 U.S.C. Sec. 1383 (b). These terms do not seem to me to be constitutionally vague at all. In any event, Plaintiffs do not contend they are vague as applied to them. There is no question that Plaintiffs are construction contractors and are covered by the statute. There is no question that Plaintiffs withdrew from the Trust Fund, and that they continue to work in the regulated field. One to whose conduct a statute clearly applies may not successfully challenge it for vagueness. Parker v. Levy, 417 U.S. 733, 756 , 94 S.Ct. 2547, 2562 , 41 L.Ed.2d 439 (1974).

Plaintiffs also mention alleged vagueness in the parts of the statute defining the actuarial assumptions to be made by the Trust in fixing a withdrawing employer’s obligation. On that subject, the statute contains six pages of instructions, and that can hardly be called vague.

Overall, this Act is about as clear as it can be, and certainly as clear as it needs to be by law. It would be impossible for Congress to predict and codify completely and exactly all of the factual situations which human experience could manufacture. Thus, in my opinion, the challenge for vagueness must be rejected.

(B) Equal Protection

I proceed now to the equal protection challenge. Plaintiffs argue that 29 U.S.C. Secs. 1461 (f)(1) and (f)(2), when read in conjunction with 29 U.S.C. Secs. 1461 (e)(2)(A) and 1363, establish an irrational distinction between substantial and non-substantial withdrawing employers. Plaintiffs claim that they are non-substantial employers and that they are discriminated against under these statutory sections, and are treated so unequally as to amount to a denial of equal protection under the Constitution.

To state my conclusion first, I think that the equal protection challenge must be rejected as moot. It turns out that there is no one in the group which Plaintiffs say was more favorably treated. The undisputed evidence is that there were no substantial employers in any multiemployer plans anywhere in the country who withdrew between April 29 and September 26, 1980. That group is finite and is known to us now, and it is clear that there is no employer in a position to receive more favorable treatment under these sections.

Although I have rejected this challenge as moot, I will review the parties’ arguments. Shelter argues that MPPAA treats substantial employers who terminated before the enactment date, but after April 29, 1980, more favorably than non-substantial employers. To buttress this challenge, Shelter interprets the statute in an unusual way. It asserts that a substantial employer who terminated in that period is permitted to post a bond and escape all further liability if the trust remains in existence for five years. Plaintiff says, in effect, that for substantial employers the old law simply carried over until September 26, 1980.

The Trust and PBGC dispute that interpretation. Defendant says that the statute should be read otherwise, that substantial employers in this time window have the same amount and degree of liability as non-substantial employers; that they, too, must pay their pro rata share of what the Trust believes will be necessary to meet the promises and assurances of the Trust in the future. In fact, Defendant says that the statute imposes a heavier burden on substantial employers; namely, the burden of posting a bond to assure they will make the necessary payments. This bond, they point out, is a requirement imposed on substantial employers by the previous legislation, and is simply carried over, to be superimposed on top of the new legislation until September 26, 1980.

PBGC seems to argue that the Defendant’s interpretation is correct.

If, arguendo, the equal protection challenge is not moot, I would reject it as based on an incorrect reading of the statute. Although I admit that the statute is murky and convoluted in the extreme, the only logical way to read it is the way Defendant reads it. No sensible reading could result in the conclusion which Plaintiffs assert; namely, that substantial employers have entirely escaped their MPPAA liability-

According to principles declared by the Supreme Court, where a statute is unclear, it should be read in the light of its stated purpose and in a manner that does not render it unconstitutional. Certainly, it was part of the statute’s purpose and scheme that substantial employers who dropped out during this period would have to pay their share of the actuarially-determined liability. I also adopt such a construction because it permits the statute to be read constitutionally. So I reject the challenge based on equal protection.

The next three subjects I will be discussing — impartial tribunal, denial of access to the courts, and denial of jury trial — are quite closely related and could be grouped together. But I will discuss them separately. However, much of my discussion under the “Impartial Tribunal” heading is equally applicable to the other two challenges.

(G) Impartial Tribunal

I pass now to the matter of impartial tribunal. In this area, Plaintiffs challenge the law because of (1) the way their liability is fixed, (2) who fixes their liability, and (3) the manner in which the fixing is thereafter reviewed.

The statutory scheme is such that the liability is fixed by the Trust alone. Plaintiffs claim that the trustees are not an impartial tribunal. I do not agree. The tribunal here, the trustees of the Carpenters Pension Trust, could be called bipartite. Half are named by employers and half by the union. They are required by law and by the Trust instrument to give their loyalty to the Trust, and thus they are not to be automatically responsive or responsible to the group or interest which named them.

The employers’ trustees are selected by the Contractors Association. Neither Plaintiff before me was a member of the Association; thus, neither Plaintiff had any voice in selecting the employers’ trustees. Also, neither Plaintiff has or had any of its officers or employees among the employers’ trustees.

Plaintiffs complain that the task of fixing the liability should not be delegated to the trustees, since both the employers’ trustees and the union’s trustees have interests which conflict with those of withdrawing employers. Plaintiffs’ argument runs as follows: all of the trustees are biased against them in fixing the amount of their liability. The union-nominated trustees are anxious to maximize the assets of the Trust so as to guarantee beyond any risk the payment of future benefits as promised. The employer-nominated trustees are really representing the employers who are still contributing; and those trustees are interested in getting as much money from the withdrawing employers as they can, thereby minimizing the amounts that the employers still contributing will be required to contribute in the future.

In my view, this challenge must be rejected. As stated, all of these trustees are required to serve the interests of the Trust impartially and without reference to who appointed them. They are commanded by the statute to act in a fair manner. General provisions of ERISA impose on them fiduciary duties and impose potential liability if they breach those duties. Also, it is my view that the trustees’ function in computing the liability of a withdrawing employer is not judicial or quasi-judicial. The final decision as to the existence and amount of liability belongs to the arbitrator, if the employer is dissatisfied with the trustees’ computation. In the arbitration, new evidence and a new record are permitted.

In computing the obligation, the trustees are not free to do as they wish. As a practical matter, and in recognition of their fiduciary obligations, they are apparently required to get professional actuarial advice. The trustees are also not free because there are six pages of detailed instructions in the statute itself. 29 U.S.C. Sec. 1391 .

Finally, the trustees’ calculation of the unfunded liability will, as a practical matter, govern them in more than the one situation. Trustees could not defend a different ad hoc determination for each employer. If they tried to do so, it would very likely be set aside in the arbitration as arbitrary and capricious. Any employer representative who is a trustee would also necessarily recognize that any precedent created might well apply to his own company if it at a later time also withdrew. A final factor to consider is that setting arbitrary or excessively high withdrawal liability may have the effect of discouraging employers from ever beginning participation in trust funds in the future. All of these factors, it seems to me, limit the effect of any bias that any trustee may have against any withdrawing employer.

The cases relied on by the Plaintiffs are totally distinguishable. The municipal mayor who collects his own salary and a large portion of his town’s funds out of traffic fines he imposes as judge is obviously biased in performing his judicial function. Ward v. City of Monroeville, 409 U.S. 57 , 93 S.Ct. 80 , 34 L.Ed.2d 267 (1972). In Berryhill, the tribunal of optometrists which made the final decision on charges of unprofessional conduct were obviously biased because they were the entrenched competitors, interested in keeping out any new competition. Gibson v. Berryhill, 411 U.S. 564 , 93 S.Ct. 1689 , 36 L.Ed.2d 488 (1973). Carter Coal is in the same mold as Berryhill , the final decision being made by business competitors. Carter v. Carter Coal Co., 298 U.S. 238 , 56 S.Ct. 855 , 80 L.Ed. 1160 (1936). And the Brotherhood of Railway and Steamship Clerks case seems to be simply off point. Brotherhood of Railway and Steamship Clerks v. Allen, 373 U.S. 113 , 83 S.Ct. 1158 , 10 L.Ed.2d 235 (1963).

Any bias that any trustee might have in computing this liability simply does not rise to the level of constitutional infirmity.

The method prescribed in this statute is a rational way to approach the problem of setting and fixing the liability. If there is a liability, someone has to fix it. And this method, it seems to me, is within the parameters of what Congress could rationally have decided was a reasonable way of fixing it. I cannot presume that the trustees are so biased as to make the statute unconstitutional on that ground.

Plaintiffs contend that if the tribunal’s bias does not itself make the statutory scheme unconstitutional, the effect of the bias is increased to the point of unconstitutionality by the fact that the decision of the Trust has many presumptions in its favor after the decision is made. Plaintiffs point out that the determination by the Trust as to what is owed is made unilaterally, without participation of, or input from, the withdrawing employer. Once the Trust’s determination is made, the withdrawing employer may have it reviewed in an arbitration, and, if dissatisfied with the results of the arbitration, may seek a review in court. In the arbitration, the Trust’s determination is “presumed correct” and to obtain relief the employer must show, by a preponderance of the evidence, that it was “unreasonable” or “clearly erroneous”. 29 U.S.C. Sec. 1401 (a)(3). Thereafter, either party may come to the District Court to ask that the arbitration award be enforced, vacated or modified. 29 U.S.C. Sec. 1401 (b)(2).

In the court proceeding, there is a presumption of correctness that the arbitrator’s findings of fact aré correct, which is rebuttable “only by a clear preponderance of the evidence.” 29 U.S.C. Sec. 1401 (c). Although the statute is silent on the subject, I must assume that the District Court has the same power as it has in the usual cases involving arbitration awards, i.e., it can set aside the award for bad faith, arbitrariness, capriciousness or lack of substantial evidence.

It seems to me that the presumptions created by the statute are reasonable. They are the traditional advantages given to administrative decisions and arbitration proceedings which later come before a court for review. If Congress has the power to set up a statutory scheme of this type, it certainly has the power, rationally exercised, to create these presumptions. Usery v. Turner Elkhorn Mining Co., 428 U.S. 1 , 96 S.Ct. 2882 , 49 L.Ed.2d 752 (1976). Whether you or I, as members of Congress, would have done the same thing is irrelevant. The question is whether Congress has acted in accordance with the rationality test. I hold that it has so acted and I reject the “impartial tribunal” challenge.

(D) Denial of Access to the Courts

I pass now to the claim that the statute unconstitutionally denies access to the courts. Much of what I have just said under the heading of impartial tribunal applies equally to this claim. The statute, of course, does not totally deny access to the courts. The courts have the right and jurisdiction not only to judge the constitutionality of the statute, a process in which I am now engaged, but also to review the arbitration, and to have the last word on whether the arbitration which settled the liability will be enforced or set aside.

Essentially, Plaintiffs are arguing that requiring an administrative fixing of liability and then an arbitration, as necessary preconditions to coming to court, is so unreasonably restrictive that the statutory scheme must be stricken down.

The major premise of that argument is one that all lawyers and judges have heard before. It is that arbitration does not protect a litigant’s rights as well as court proceedings, because hearsay is allowed, rules of evidence are not observed, discovery is not as broad, and so on. These complaints are often voiced about arbitration. But the question here is whether requiring arbitration before coming to court is within Congress’ power. I have no difficulty in reaching the conclusion that there is nothing unconstitutional about so preconditioning the right to eventual court review. Andrews v. Louisville & Nashville R. R. Co., 406 U.S. 320 , 92 S.Ct. 1562 , 32 L.Ed.2d 95 (1972).

This argument is coupled with the argument that the presumptions discussed earlier are a further effective restriction on the access to the courts. Plaintiffs argue that judicial review is just an “empty gesture”. As I have said, I find no constitutional invalidity in those presumptions, and I find that the requirement to arbitrate before coming to court, even in combination with the presumptions, is constitutionally valid.

(E) Jury Trial

I move now to the claim that the statute is invalid as infringing a right to jury trial. Again, much of what I have previously said under the last two subject headings applies to this issue as well.

Plaintiffs argue that since the statute results in a money obligation being imposed on them, the Seventh Amendment gives them a right to have that money obligation fixed in a court of law and by a jury. They claim that since the fixing of the obligation is done administratively by the Trust and thereafter in arbitration, they are unconstitutionally denied the right of jury trial.

The mere imposition of a money obligation under a statute does not, in my view, guarantee the right to have the obligation fixed by a court in the first instance or the right to have it fixed by a jury. Just because money is involved, that does not make the procedure for determining the amount of the monetary obligation the equivalent of a common law action for money damages. Several leading cases discuss this principle and indicate clearly that when Congress legislates in the economic field and sets up an administrative scheme for determining a financial liability, there is no right to trial by jury under the Seventh Amendment in the common law sense. Among these cases are Katchen v. Landy, 382 U.S. 323 , 86 S.Ct. 467 , 15 L.Ed.2d 391 (1966); Andrews v. Louisville & Nashville R. R. Co., 406 U.S. 320 , 92 S.Ct. 1562 , 32 L.Ed.2d 95 (1972); and Wardle v. Central States Southeast and Southwest Areas Pension Fund, 627 F.2d 820 (7th Cir. 1980). I note that Wardle was an ERISA case.

Plaintiffs rely on a group of Seventh Amendment cases including Curtis v. Loether, 415 U.S. 189 , 94 S.Ct. 1005 , 39 L.Ed.2d 260 (1974); Rogers v. Loether, 467 F.2d 1110 (7th Cir. 1972); and Pons v. Lorillard, 549 F.2d 950 (4th Cir. 1977). This reliance is misplaced. Those were civil rights cases and they hold that when Congress encourages private plaintiffs to sue in civil courts to enforce statutorily created rights (which may include the right to money damages), the parties are entitled to a jury trial. Those cases point out that Congress was silent in the civil rights statutes as to the right to jury trial, and the opinions made the point that there is no particular purpose to be served by denying the right to jury trial.

Our situation is distinguishable. Here, at the same time Congress created the statutory liability, Congress also established a method for fixing the amount due, first administratively and then by arbitration. Congress has provided that court proceedings occur only at the end of the process and that relief is limited to the usual alternatives available in dealing with arbitration awards — enforcing, setting aside or modifying the award. This statutory scheme must be read as a clear indication that Congress decided against any jury trial right. That decision was, in my view, well within Congressional discretion as part of an economic statutory scheme. The statutory method for establishing the amount of the liability does not violate the Seventh Amendment.

(F) Taking Without a Hearing

I pass now to the last of the issues which I have characterized as less substantial, the claim that the statute is void under due process concepts because it constitutes a taking of Plaintiffs’ property without a hearing. The essence of Plaintiffs’ argument is that there can be no deprivation of property without a • hearing, and that the hearing must be a meaningful one, occurring before the deprivation. Plaintiffs rely on cases such as Fuentes v. Shevin, 407 U.S. 67 , 92 S.Ct. 1983 , 32 L.Ed.2d 556 (1972), which involved repossession by a seller, and Sniadach v. Family Finance Corporation of Bay View, 395 U.S. 337 , 89 S.Ct. 1820 , 23 L.Ed.2d 349 (1969), which involved a garnishment.

As background for this claim, I will review the procedures under MPPAA for assessing the liability, collecting the money, and accelerating the liability. Plaintiffs claim that these procedures are, in toto, the equivalent of a taking.

When the Trust determines the amount due from the withdrawing employer, it is also required to establish an installment payment schedule which the employer may avail itself of. The installments may be payable quarterly or at other intervals established by the Trust. The Trust then is required to notify the employer of the total amount due and of the installment schedule. The lump sum or the first installment (whichever the employer may elect to pay) is due 60 days after that notice to the employer is served. If the employer fails to pay either the lump sum or the first installment payment within that 60 days, the Trust must then notify the employer that he is in default. If the default is not cured within 60 days after notice of default is served, the Trust “may” revoke the employer’s option to pay in installments and require immediate payment of the full amount due plus interest. 29 U.S.C. Sec. 1399 (c)(3). Even if the employer is seeking arbitration as to the amount due, he must start paying the installments and continue paying them while the arbitration is pending. After the arbitration decision is made, the employer must continue to make timely installment payments in accordance with that decision, even if he wishes to contest the arbitration award in court. If those payments do not continue during that time, the employer risks being treated as delinquent, and risks losing the right to pay in installments if the arbitration award is later enforced. 29 U.S.C. Sec. 1401 (d).

Plaintiffs argue that these statutory provisions put irresistible pressure on them to begin paying their obligation before any hearing. Even if the employer preserves the right to an installment schedule by paying the installments while he is seeking arbitration and during the arbitration process, and thereafter while he is seeking court relief, Plaintiffs describe this option as extremely onerous, since, as the facts of our case show, even the monthly installments can be crippling.

If the employer decides not to pay in installments while contesting the liability in arbitration and in court, he is faced with the loss of the installment option and a declaration that the entire liability is due and payable, plus a liability for interest on

the full amount back to the date of the original assessment. The statute provides for additional liquidated damages of up to 20% of the full amount to be assessed against a delinquent employer, plus an obligation to pay the Trust’s attorneys fees in a collection action. 29 U.S.C. Secs. 1399 (c)(5), 1451(e), 1451(b), 1145 and 1132(g)(2). The net practical effect of all this, say Plaintiffs, is that they cannot challenge the liability, or even utilize the statutory arbitration process, before starting to make payments.

I reject this constitutional challenge. Although the statute may be onerous to the withdrawing employer who is presented with a difficult alternative, it does not violate the Constitution because there is no seizure of property. There is no deprivation of property without a hearing. The constitutional doctrine invoked by Plaintiffs has thus far been extended only to cases of actual seizure. Plaintiff Shelter claims that the doctrine is not limited to instances of actual seizure but cites no authority to support that claim. I decline to extend the rule of Fuentes, supra, and Sniadach, supra, as requested by Plaintiffs. I cannot hold that a mere assertion of liability, by itself, constitutes a deprivation of a significant property interest that demands a prior hearing, even if the statute requires that Plaintiffs give up a valuable installment payment option if they determine to pay nothing before the hearing takes place.

I should comment in a little more detail on some of the claims made by both sides under this issue. Plaintiffs claim that if they start to pay the installments during the arbitration proceeding (in order to keep the installment option open to them if they lose) and if later they win the arbitration or a court decides that they owe nothing, there is no provision for their interim payments to be returned. They also say that there is no provision in the statute for payment of either interest or damages for harm to their business that may occur from having to make these “compulsory” installment payments that turn out not to have been owed.

These assertions postulate a chamber of horrors, but do not seem to pose actual problems. Surely a quasi-contractual obligation would exist to return any money paid in error and the obligation to repay would certainly include interest. Whether there would be any consequential damages for harm done to a business from having to pay money improperly demanded might well be governed by local law. The right to such damages cannot be ruled out entirely, but I need not decide that question here.

Defendant argues that we should assume that any arbitration demanded by the employer would be concluded in the 120-day grace period provided by the statute before the employer can be declared in default. Thus, says Defendant, there is no realistic possibility of the whole amount being accelerated before the arbitration ends. I am not persuaded by this argument. As a practical matter, it is hard to conceive of an arbitration being conducted with such promptness now or in the near future. PBGC has not yet gotten around to prescribing any rules, regulations or procedures for these arbitrations. The arbitrations, when they do occur, are obviously going to be very complex affairs, depending a great deal on expert testimony as to sophisticated actuarial questions. As a practical matter, it is very doubtful that such an arbitration could be concluded often, if ever, within the 120-day time frame set up by the statute.

To summarize, I sympathize with Plaintiffs when they describe the coercive effect of the entire statutory scheme. They face an onerous burden if they start installment payments before arbitration, and an even more severe burden if they do not start installment payments before a hearing is afforded to them. I do not think, however, that those hardships and disadvantages rise to the point of making the statute unconstitutional on the ground that it is an invalid taking of property without a hearing.

(G) Retroactivity

The two final issues, retroactivity and taking without just compensation, seem to me to overlap analytically to some extent. But counsel for both sides have approached, briefed and argued these two issues separately. The appellate courts also seem to have separated the two concepts. There is a line of retroactivity cases which delineate what type of retroactivity violates the Constitution and what does not, and there is a separate line of cases involving taking without just compensation. There appears to have been little, if any, cross-citation between these two lines of cases. For all of those reasons, I will treat the two issues separately.

I have determined that MPPAA must be stricken down as applied to employers like those before me, who terminated their collective bargaining agreements and their pension plan membership before the enactment date of the statute, September 26, 1980, but after its application date of April 29, 1980.

In making that holding, I emphasize that Plaintiffs were entitled to terminate their collective bargaining agreements when they did and that those agreements, by their own terms, promised each employer that he had no obligation to make any further payments to the trust fund after he terminated his obligations under the collective bargaining agreement.

I emphasize also that the law as it existed when Plaintiffs withdrew from the plan (ERISA) imposed only a contingent liability, limited to certain circumstances and a certain time frame. I am speaking, of course, of the liability of the withdrawing employer to pay its pro rata share of the unfunded vested benefits if, and only if, the entire plan terminated or became insolvent within five years after an employer’s withdrawal.

Each employer’s withdrawal was a closed transaction when the withdrawal took place, with a liability that was clearly defined and limited. By a later legislative enactment, which purported to reach backwards to a date before these transactions were consummated, Congress imposed on each Plaintiff a different, large, unanticipated liability, without any significant moderating or mitigating factors.

That is my conclusion and I will document it in some detail.

In judging retroactive legislation under the Constitution, I must recognize certain general principles that have been enunciated by the Supreme Court. First, if the legislation is in the economic sphere, great deference must be paid to legislative judgments. Second, retroactive legislation is always to be examined with greater scrutiny than legislation which has only a prospective effect.

Counsel for both sides agree that the best test for judging the constitutional validity of retroactive legislation is found in the Seventh Circuit’s opinion in Nachman Corp. v. Pension Benefit Guaranty Corporation, 592 F.2d 947 (7th Cir. 1979). I agree with counsel. Nachman was affirmed by the Supreme Court on different grounds. Nachman Corp. v. Pension Benefit Guaranty Corporation, 446 U.S. 359 , 100 S.Ct. 1723 , 64 L.Ed.2d 354 (1980). In its affirmance, however, the Court in a footnote cited with apparent approval part of the analysis used by the lower court, [n. 12, 446 U.S. at 367 , 100 S.Ct. at 1729 .] The Seventh Circuit’s opinion listed four factors as relevant to retroactive analysis and I adopt those factors here.

The four factors listed in Nachman are:

(1) “The reliance interests of the parties affected”;

(2) Whether the private interest which is impaired is “in an area previously subjected to regulatory control”;

(3) “The equities of imposing the legislative burdens”; and

(4) The inclusion of “statutory provisions designed to limit and moderate the impact of the burden.”

Nachman also emphasizes that the validity of the legislation does not rest on the way a court balances these factors one against the other. The factors are used as guidelines or measuring devices for the final and ultimate judicial determination, which is “whether the legislation imposes a rational means to a legitimate end.”

I will now consider these factors individually and discuss how MPPAA measures up • under them.

(a) The first factor is reliance. In this connection, I think it is appropriate to examine the extent to which the parties relied on existing law continuing, and to examine the types of interest involved in their reliance. I want to start with examining the reliance of the employers, the Plaintiffs here.

Defendants argue first of all that employers, including these Plaintiffs, could not have relied on existing law continuing because they knew or should have known that (1) the coming legislation was going to impose a substantial new burden on withdrawing employers, and (2) that the new legislation would have a drastic retroactive effect. I reject that argument. No one should be held reasonably to anticipate a given type of Congressional action on any subject. Outside of the tax area, there is no authority that imposes on anyone the burden of predicting Congressional action.

It is an undisputed fact that these Plaintiffs had no knowledge of the forthcoming legislation. The mere fact that one or more trade journals speculated about Congressional interest in the field does not in my view outweigh that uncontroverted evidence.

The cases also tell us that in measuring reliance of a party, one must look at what that party would have done if he had known that the legislation was coming. See Welch v. Henry, 305 U.S. 134 , 59 S.Ct. 121 , 83 L.Ed. 87 (1938); Adams Nursing Home of Williamstown, Inc. v. Mathews, 548 F.2d 1077 (1st Cir. 1977). The Nachman opinion follows this rule.

Here, the Plaintiffs offered convincing evidence that they would have changed their conduct and would have taken one of several alternative actions to avoid this catastrophic liability from being imposed, had they known MPPAA was coming. One op tion was to renew their collective bargaining agreements and continue making the usual contributions to the Trust fund. They could also have gone completely out of business. Or, they could have sold their assets to a company which remained in business and was a participant in the pension plan. Any of those courses of action would have allowed the Plaintiffs to avoid the liability. All would have been open to Plaintiffs if they had had notice of the statute’s potential application to them.

I should note that in Nachman the court likewise found a high degree of reliance by the employer, but held it to be counterbalanced by other factors. Nachman upheld ERISA, a much more moderate retroactive statute.

There is one other important factor. In Nachman, the law that was being discussed was already enacted and in effect when the plaintiff closed up his business. He knew or should have known what the law was when he took the action in question. He knew what he was subjecting himself to when he terminated his plan. Our Plaintiffs, had they withdrawn on September 27, the day after enactment of MPPAÁ, would be in the same position. They would have been subject to an existing statute which had a retroactive effect. But our Plaintiffs withdrew long before the enactment date and were caught by a later-enacted statute which purported to be retroactive in application. The distinction I am making is between statutes which are retroactive in effect, though prospective in application, and statutes which are retroactive in application. The latter present greater hardship to those persons affected and more serious due process problems.

Generally, the concept of reliance in a retroactivity case has focused on the state of mind and the degree of fairness as viewed from the perspective of the party who bears the new or added burden from the legislation, and that party is the employer in our situation. But the Nachman opinion requires us to consider not only reliance from the employer’s point of view but also reliance by the other affected party, the employees. Therefore, I should and will look at reliance from the employees’ standpoint as well.

From that point of view, it seems to me that Nachman is distinguishable from our cases. Nachman involved a one-employer pension plan. When that employer went out of business, the employees who had been relying on pension benefits for their future had their expectations disappointed. In Nachman, only 35 per cent of the vested benefits were funded when the employer went out of business. Non-vested employees would apparently have wound up with nothing, and vested employees would have had their benefits drastically reduced from the expected level. These employees were looking solely to this one employer for future benefits.

In the present cases, it cannot be claimed that the workers of either Plaintiff were relying on their employer’s continuing in the multi-employer pension plan, nor were they relying on that particular employer to pay the pension benefits. The promises that were made to Plaintiffs’ employees, and upon which they may have relied, were made in our cases by the Trust or by the union, and not by their particular employer.

Further, when our Plaintiffs terminated their participation in the Trust fund, that did not end the Trust fund nor did it render the fund insolvent. There is no clear or immediate harm shown to have resulted to the employees’ reliance interest as a result of our Plaintiffs’ withdrawal. The evidence in our cases is that after the withdrawal of both Plaintiffs, the Defendant Trust remains in sound financial shape. Thus, the reliance interest of the employees in our cases is far more nebulous and far less affected than it was in Nachman. Indeed, it may not be affected at all.

(b) I turn now to the second element, whether the interests impaired by the retroactive features of the legislation are in an area previously subject to regulation. That element, although enunciated in Nachman, was little discussed there. There is only a footnote, which merely discusses some prior IRS regulation of the pension field which had occurred before the passage of ERISA. In my view, this element is the least important of the four, and the Nachman court itself did not treat it very significantly.

As I consider this element, I feel that it is really just another facet of reliance. It goes to the reasonable expectations of the party upon whom the burden of the retroactive legislation falls.

It may be instructive, in analyzing the force and effect of this element, to discuss a couple of the leading cases. In Federal Housing Administration v. The Darlington, Inc., 358 U.S: 84, 79 S.Ct. 141 , 3 L.Ed.2d 132 (1958), the plaintiff was in the business of building and renting apartments, and he built them with federal aid in the form of FHA-insured loans. Among the regulations in existence when he took out the loan for this particular project was a requirement that the project be for residential housing. Any type of residential tenant could apparently qualify. After the project was built, the regulations were changed so that the owner was prohibited from renting the units to transients.

First of all, it seems to me that that change could easily be viewed as a mere clarification or modification of the initial restriction. The Court viewed it as such and said that such a change was the anticipated and normal risk taken by those who do business in a regulated field. The Court then upheld the change.

The other case worth discussing is Veix v. Sixth Ward Building and Loan Association, 310 U.S. 32 , 60 S.Ct. 792 , 84 L.Ed. 1061 (1940). In Veix , an investor bought some shares in a savings and loan association. When he bought them, there was a right to a later turnback or redemption of those shares. There were statutes then in effect governing the conditions of redemption and specifying what funds could be used to pay for redemption. After the plaintiff bought his shares, amendments to the law were passed which served further to restrict the conditions under which redemption could take place, including a more restrictive definition of the funds out of which such redemption could be paid.

The case was brought under the Contracts Clause of the Constitution because state law was involved. However, the Supreme Court’s approach is significant because the Court said that when the plaintiff purchased into an enterprise already regulated, “in the particular to which he now objects”, he took the risk of further legislation on the “same topic.”

I interpret this to mean that for retroactive legislation in a previously regulated area to be valid, the new and more burdensome legislation must be closely akin in type and scope to the prior regulation.

It should be noted also that in neither Darlington nor Veix was the legislation retroactive in application. The legislation in each case applied to events that occurred after its enactment date.

It should further be noted that in each case the matter involved was a statutorily-created privilege. In Darlington, it was a government-back loan. In Veix , it was a statutorily-created right of redemption, a right not generally afforded shareholders in an ordinary corporation.

In reviewing the significance of prior regulation, I believe this element tends to favor the Plaintiffs here. It is evident that there was some prior regulation of multiemployer funds and of contributing employers prior to the passage of MPPAA. But MPPAA imposed much heavier and more drastic burdens than had been imposed by the prior law. Our Plaintiffs, although they were in a field that the government had already regulated somewhat, could not reasonably have foreseen a change of this scope and nature in the regulation of the field, especially when applied on a totally retroactive basis.

The essence of the concept of reliance, as I view it, is that a party should be able confidently to anticipate the consequences of a transaction, knowing what the law provides when he concludes that transaction. If he is in a regulated field, he can anticipate some clarification and modification of the scope of regulation, but cannot foresee change of a most drastic type, espe cially when applied to a transaction already completed before the new legislation is enacted. The sole exception to these principles may well be in the field of taxation, which I regard as sui generis because of the taxable-year concept.

(c) I turn now to the third element, the equities of imposing the burdens of retroactive regulation on the parties that must bear the burdens.

The Nachman opinion points out that we look at equities for the essential purpose of determining whether the retroactive law is rational or irrational. The Nachman court regarded several cases as leading authorities on this point, including Allied Structural Steel Company v. Spannaus, 438 U.S. 234 , 98 S.Ct. 2716 , 57 L.Ed.2d 727 (1978), and Railroad Retirement Board v. Alton R. R. Co., 295 U.S. 330 , 55 S.Ct. 758 , 79 L.Ed. 1468 (1935).

In Spannaus , a Minnesota statute, which was passed before the plaintiff employer closed his Minnesota plant, forced any employer who closed a Minnesota business to make pension payments to his Minnesota employees, even though under his overall pension plan, such employees had no vested rights. The law was struck down partly on the ground, as the Nachman opinion says, that the employer was being forced to pay added compensation for fully compensated services.

In Alton, Congress passed a railroad pension law which required pension benefits to be paid not only to workers then working for the railroad, but also to some who had ceased working for the railroad long before the statute was passed. It was struck down on the same ground as the Minnesota statute, since it forced payment of added compensation for services already fully compensated. There are two other relevant cases, which should be compared with Spannaus and Alton. They are Usery v. Turner Elkhorn Mining Company, 428 U.S. 1 , 96 S.Ct. 2882 , 49 L.Ed.2d 752 (1976), and Todd Shipyards Corporation v. Witthuhn, 596 F.2d 899 (9th Cir. 1979).

In Usery , Congress had passed a statute requiring employers to compensate mining employees who contracted black lung disease from their mining activities. The law applied even if the miners had terminated their relationship with the employer long before the law was passed, if the disease was caused by the employment. The law was held valid.

In Todd Shipyards, the Longshoremen and Harbor Workers Compensation Act was amended to provide death benefits for workers injured on the job who later died of such injuries. The amendment was prospective, as it applied to all workers who died after its effective date as a result of on-the-job injuries. In that particular case, it was applied to a worker who had been injured before the effective date of the statute but who died after the effective date. The statute was upheld. It is worth noting that the court determined that this act was prospective both in effect and application.

Assume, arguendo, that both Todd and Usery are true retroactivity cases, as they make the employer liable for consequences of acts which occurred before the legislation was enacted. In my view, both cases can still be distinguished from our cases, in that in both Todd and Usery the employer controlled the conditions which caused the harm to the workers. The retroactive legislation called on the employer to compensate the workers for actual harm caused by the now-burdened employer. But in our cases, it cannot be said that these Plaintiffs, these employers, caused any harm or failure of expectation. These Plaintiffs fulfilled their obligations under the collective bargaining agreement to the letter, and cannot be blamed in any sense for the debts or the failed promises of the Trust fund.

The same distinction I have just made can also be made between our cases and Nachman. Nachman involved a single employer pension plan. The employer himself, alone and directly, caused the harm and the failure of expectations when he closed his business. That cannot be said of our Plaintiffs.

To summarize the discussion under this third factor, I think our cases are much closer in fact and rationale to the Supreme Court holdings in Spannaus and Alton than to the holdings in Usery and Todd Shipyards. Moreover, it should not be forgotten that in all of this we are dealing with equities. This factor, as defined by the Nachman court, emphasizes the equities of imposing or not imposing certain burdens. Our cases have substantial equity factors in favor of the employer and against imposing such a drastic and clearly retroactive burden upon the employer.

It is undeniably true that the harshest burden of MPPAA is placed on the withdrawing employers who withdrew between April 29 and September 29. Employers who withdrew earlier have no burden except the minor contingent liability of the old ERISA statute. Those who stayed in after September 26, 1980 have many options available to them which our Plaintiffs do not have. For example, as discussed earlier, those employers who stayed in the fund can sell their assets to a buyer who will continue the business and continue participating in the fund. If the employees stay in business, they can choose to remain affiliated with the Trust and pay their regular contributions, with no catastrophic lump-sum obligation such as that imposed on the Plaintiffs in our cases. So our Plaintiffs, and those similarly situated, are being forced to bear a very disproportionate burden. They must bear the costs of a large problem which, so far as I can see, they had little if any role in creating.

(d) I turn now to the fourth factor. This final factor delineated in Nachman is whether the statute includes “provisions designed to limit and moderate the impact of the burdens,” and its importance is highlighted by the extensive treatment given to it in the Supreme Court’s Nachman opinion. Nachman Corp. v. Pension Benefit Guaranty Corp., 446 U.S. 359 , 367 n.12; 100 S.Ct. 1723, 1729, n.12 , 64 L.Ed.2d 354 (1980).

To put this another way, the Court should examine whether the burdens imposed by the retroactive legislation are overly drastic or whether they are substantially modified, mitigated, or lessened by other provisions of the statute. On this test, and under this factor, the Plaintiffs in my view have by far the best of it.

Contrast MPPAA, and how harshly it deals with these Plaintiffs, with the statutes dealt with in Nachman, Usery and Spannaus . In Nachman, which upheld ER-ISA in its pre-MPPAA form, the Court leaned very heavily on the various protective limits of the statute. For example, the Court mentioned that ERISA, as it then was, limited possible obligations to 30 per cent of the employer’s net worth. Similarly, withdrawing employers became liable only for the amount which was guaranteed by PBGC, which was less than the total amount of vested benefits. Moreover, there was a phase-in period of 12 months, a kind of grace and transition period, during which the impact of the statute, initially mild, gradually increased. Contrast that grace period with the 5-month retroactive period of MPPAA. The Court also noted that under ERISA, employers had an opportunity to obtain low-cost insurance from PBGC to cover the contingent future liability. Finally, there was statutory authority for PBGC to waive the whole liability or to moderate or defer it. None of these limiting or moderating factors are present here. MPPAA is far more drastic in effect than the statute addressed in Nachman, with far fewer mitigating factors.

In Usery , it should be remembered that one major mitigating factor was that the government paid a substantial portion of the liability for black lung disease.

In Spannaus , where the law was stricken down, it was very important to the Court that there was no grace period and that the law was immediately effective. MPPAA not only has an immediate effect, but even has a retroactive effect, covering transactions already completed.

Defendants have argued, not convincingly, that MPPAA has moderating provisions. They point out that under 29 U.S.C. Sec. 1383 (b)(2), withdrawing employers may avoid the liability of MPPAA by going out of the construction business immediately and remaining out of it for five years. In the context of this case, that cannot be considered a moderating provision because that option was not available to the Plaintiffs. They had no idea that the law would be passed and could thus have made no intelligent decision to go out of the construction business upon withdrawal from the plan, being unaware of both the liability they incurred by staying in business and the option of avoiding it by leaving the business.

The Defendant has also argued how easy it is for Plaintiffs to pay in monthly installments. But that does not impress me either, especially in the context of a case where the monthly payments would take 94 per cent of G & R’s annual income. The severity in our cases is in the total amount of the liability, and it is not much mitigated by the right to pay monthly with accumulating interest.

The Defendant also claims that the Plaintiffs could now rejoin the plan, and that this should be viewed as a moderating factor. But an employer cannot unilaterally rejoin the plan, because he must be party to a collective bargaining agreement to participate in the plan. Moreover, it seems to me that it is not absolutely clear that the imposed liability would be abated by rejoining. At this point there are no regulations to tell us when, under what conditions, and to what degree the liability would be abated if an employer who has withdrawn does rejoin the plan. I would not call the ability to undo the transaction a mitigating factor. Moreover, that ability does not exist realistically, as I see the facts.

To summarize, in evaluating the four factors of Nachman, giving to each the appropriate weight in the light of the facts of our cases and the statute in question, I have come to the conclusion that this retroactive law is invalid as applied to these two Plaintiffs who withdrew before its enactment date.

The Defendant continually argues in our cases that it was absolutely necessary for Congress to apply this statute retroactively. They say that if Congress had not done so, there would have been a large group of people who withdrew before the enactment date, thus endangering the pension plans even further. But there was almost no probative evidence on this point and I do not find this argument very convincing as a practical matter. No employer was free to withdraw just because he wanted to. Every employer was bound by a collective bargaining agreement and bound by various kinds of notice provisions that had to precede any withdrawal.

(H) Taking Without Just Compensation

My holding on retroactivity makes it unnecessary for me to deal with the claim that the MPPAA is unconstitutional on the separate challenge of taking without just compensation. I do not believe in unnecessary dicta, and I will make no ruling on that challenge. However, I must say that there are some serious questions raised against the statute on this ground.

(I) Conclusion

My ruling is that the MPPAA is unconstitutional as applied to these two Plaintiffs, and all others similarly situated, who withdrew before the enactment of the statute. As to employers who withdrew after the statute was enacted, I believe that although there may be some problems with the statute on the issue of retroactive effect, I make no holding with respect to those employers.

I want to make it very clear that I am expressing no opinion on the statute’s validity on either retroactivity or taking grounds as applied to employers who remained in pension plans on or after MPPAA’s enactment date. No such case is before me, and any such case should be decided on its own facts.

Having granted judgment for Plaintiffs, a corollary judgment must go against Defendant on its counterclaims which seek to collect the assessed liability from each Plaintiff. This opinion will constitute the Court’s findings of fact and conclusions of law.

. Prior to the argument of the mutual summary judgment motions, the Court recognized Pension Benefit Guaranty Corporation, a government agency created under ERISA, as an amicus curiae. Its representative appeared and asserted in briefs and argument that the Court had no jurisdiction to adjudicate the constitutional validity of MPPAA before the Plaintiff-employers had availed themselves of the statutorily-provided arbitration procedure. MPPAA provides for arbitration to resolve disputes or objections by withdrawing employers over the existence or amount of withdrawal liability. This threshold issue was argued, and was rejected by the Court, before commencement of the hearing on the constitutional issues. Thereafter, the amicus orally argued the constitutional issues. . An employer is deemed to make a “complete withdrawal” from a multi-employer pension plan when the employer permanently ceased to have an obligation to contribute to the plan; 29 U.S.C. Sec. 1383 . Thus, Shelter completely withdrew from the plan on July 16, 1980, or at the very latest on August 12, 1980. Similarly, G & R completely withdrew on July 18, 1980, or, at the very latest, on August 12, 1980. . I note that PBGC claims to have adopted an “interpretation” of MPPAA which prohibits a declaration of accelerated liability under certain circumstances. But there is no probative evidence of the existence of such a regulation before the Court. I have not seen a PBGC document or any indication that such a regulation has been validly and properly issued, or that its issuance would be within PBGC’s powers under the statute. I note it only as an unsupported claim made by PBGC. . One court has required a plaintiff-employer to exhaust the arbitration remedy before seeking judicial relief and a declaration of unconstitutionalitv. Republic Industries v. Central Pennsylvania Teamsters Pension Fund, 534 F.Supp. 1340 , 3 E.C.B. 1299 (E.D.Pa. 1982) (order granting motion to dismiss). . The withdrawing employer may, after the determination is made, ask the Trust to review the determination, but the Trust is apparently not obligated to grant such review. . See note 3, supra. . In Peick v. Pension Benefit Guaranty Corporation, 539 F.Supp. 1025 (N.D.Ill.1982), the court held the statute valid on its face. The court said: “This is an extremely close call. Congress has probably gone to the very limits of its constitutional power.” 539 F.Supp. at 1056 . The Illinois court also took note of this court’s “tentative view” that the statute was unconstitutional, and stated that because of the facts developed in the Shelter case, such a view was “not necessarily inconsistent with [the] holding that MPPAA survives facial attack.” 539 F.Supp.’at 1056 n.79. . Two courts have held that the statute is valid when applied prospectively. See Peick v. Pension Benefit Guaranty Corporation, supra, n.7; and S & M Paving, Inc. v. The Construction Laborers Pension Trust of Southern California, 539 F.Supp. 867 , 81-5929 TJH (C.D.Cal. April 12, 1982).