Peick v. Pension Ben. Guaranty Corp., 539 F. Supp. 1025 (N.D. Ill. 1982)
May 14, 1982
PENSION BENEFIT GUARANTY CORPORATION, et al., Defendants.
United States District Court, N. D. Illinois, E. D.
*1026 *1027 *1028 *1029 Sherman Carmell, Carmell, Charone & Widmer, Ltd., Peter M. Sfikas, Peterson, Ross, Schloerb & Weidel, Chicago, Ill., for plaintiffs.
Henry Rose, Mitchell L. Strickler, Baruch Fellner, Washington, D. C., for defendants.
Harris Weinstein, Arvid E. Roach, III, Covington & Burling, Washington, D. C., Sheldon W. Witcoff, Daniel A. Boehnen, Allegretti, Newitt, Witcoff & McAndrews, Ltd., Chicago, Ill., for Transport Motor Express, Inc., E. W. Bohren Transport, Inc., and Essex Group, Inc.
Lester M. Bridgeman, Lester M. Bridgeman & Associates, P. C., Washington, D. C., Philip B. Kurland, John J. Coffey III, Christopher G. Walsh, Jr., Rothschild, Barry & Myers, Chicago, Ill., for Johnson Motor Lines, Inc. and Republic Industries, Inc.
Carl L. Taylor, Arthur F. Sampson III, Kirkland & Ellis, Washington, D. C., Kent A. Zigterman, Kirkland & Ellis, Chicago, Ill., for American Trucking Associations, Inc.
MEMORANDUM OPINION AND ORDER
GETZENDANNER, District Judge.
On September 26, 1980, President Carter signed into law the Multiemployer Pension Plan Amendments Act of 1980 ("MPPAA" or "the Act"). This statute sets forth a comprehensive scheme of federal law regulating multiemployer pension plans. Multiemployer plans are those "to which more than one employer is required to contribute" under the terms of "one or more collective bargaining agreements between one or more employee organizations and more than one employer." 29 U.S.C.A. ß 1002(37) (A) (Supp.1981).
Plaintiffs attack the facial constitutionality of MPPAA on various grounds and cross-motions for summary judgment have been filed. These motions raise complex and novel issues which have been very ably briefed by the parties. The ultimate questions are close, but in my opinion, the challenged provisions of MPPAA survive facial attack.
I. THE BACKGROUND OF MPPAA
The 1974 enactment of the Employee Retirement Income Security Act ("ERISA") marks the initial attempt by the federal government to regulate pension plans in a comprehensive manner. This statute contains numerous provisions:
Title I attacks the lack of adequate "vesting" provisions in many plans. Before ERISA, for example, if a plan did not provide for vesting until retirement, an employee with 30 years of service could lose all rights in his pension benefits in the event that his employment was terminated prior to retirement. Title I establishes minimum vesting standards to ensure that after a certain length of service an employee's benefit rights would not be conditioned upon remaining in the service of his employer. Employers were required to amend the terms of their plans to reflect these minimum standards effective January 1, 1976. [29 U.S.C.] ß 1053(a). A second area of difficulty was the inadequacy of the funding cycle used by many plans. To improve the fiscal soundness of these pension funds, Title II amends the Internal Revenue Code to require minimum funding. Title III imposes fiduciary responsibilities on the trustees of the pension funds and provides for greater information and disclosure to employee-participants. The final area of concern addressed by ERISA was the loss of employee benefits which resulted from plan terminations. In order to protect an employee's interest in his accrued benefit rights when a plan failed or terminated with insufficient funds, Title IV establishes a system of *1030 termination insurance, effective September 2, 1974.
Nachman Corp. v. Pension Benefit Guaranty Corp., 592 F.2d 947, 951 (7th Cir. 1979), aff'd, 446 U.S. 359, 100 S. Ct. 1723, 64 L. Ed. 2d 354 (1980) (hereafter cited as Nachman). Most relevant for present purposes is the termination insurance program contained in Title IV. This program is run by the Pension Benefit Guaranty Corporation ("PBGC"), a governmental entity which receives no direct federal appropriations. The PBGC relies instead primarily on premium payments: In 1974, multiemployer plans paid $.50 per participant per year while single employer plans Äî those created, operated and maintained by a single employer acting alone Äî paid $1.00 per participant per year.
Upon enactment of ERISA in 1974, the PBGC immediately insured the receipt of all "nonforfeitable benefits" that had been earned by employees in single employer plans. A single employer that wished to terminate its plan was thus first required to notify the PBGC. 29 U.S.C. ß 1341(a) (1976). If an investigation subsequently revealed that the plan lacked sufficient assets to pay its "nonforfeitable benefits," the PBGC itself became obligated for the shortfall. Id. at ß 1341(c), (b). Any amounts so expended could be recovered from the terminating employer, id. at ß 1362, but the latter's liability could in no event exceed thirty per cent of its net worth. Id. at ß 1362(b) (2).
Multiemployer plan benefits were treated differently. They were not insured absolutely upon enactment, but rather were guaranteed solely at the discretion of the PBGC until January 1, 1978. At that time, the guarantees were to become mandatory. Id. at ß 1381(c) (1). In the interim, the PBGC was authorized to determine on a case-by-case basis whether it would pay a terminating plan's participants the difference between the value of their guaranteed benefits and the value of the plan's assets on the date of termination. Id. at ß 1381(c) (2). As in the single employer context, secondary employer liability was imposed in all cases in which PBGC funds were actually expended. Specifically, all employers that contributed to a terminated multiemployer plan during the five years immediately preceding termination were collectively liable to the PBGC for the amount the latter had disbursed, each employer for its proportionate share of the total. As before, no single employer's termination liability could exceed thirty per cent of its net worth. Id. at ß 1364.
Employers that withdrew from an on-going (i.e., non-terminating) multiemployer plan thus incurred a contingent liability. It was contingent first upon the plan's terminating within the next five years, and second, in the absence of mandatory benefit insurance, upon the PBGC's deciding to insure the plan's benefits. ERISA did not, in general, obligate a withdrawing employer to provide the PBGC with any security for this potential debt. An exception was recognized, however, in the case of a "substantial" employer, one that had contributed at least ten per cent of all contributions received by the plan over a specified period of time. Id. at ß 1301(a) (2). Withdrawing employers meeting this description were required to place in escrow an amount equalling what their termination liability would have been had the plan terminated on the date of withdrawal. Id. at ß 1363(b). Alternatively the employer could furnish a bond. Id. at ß 1363(c) (1). If no termination actually occurred during the next five years, the escrow was refunded or the bond cancelled. Id. at ß 1363(c) (2).
There were several reasons why Congress chose not to insure all multiemployer plan benefits immediately in 1974. Congress viewed multiemployer plans as more stable and secure than single employer plans and thus saw less need to insure the former. *1031 See, e.g., Connolly v. Pension Benefit Guar. Corp., 581 F.2d 729, 734 (9th Cir. 1978); 126 Cong.Rec. H4116 (daily ed. May 22, 1980) (remarks of Rep. Biaggi). Congress, moreover, was worried about the potential costs of such a program. These worries became more prevalent as January 1, 1978 approached. Senator Javits warned his colleagues in late 1977 that he knew of several multiemployer plans which planned to terminate soon after the first of the year. See id. at S10099 (daily ed. July 29, 1980). Recognizing that it needed more time to study the entire problem, Congress delayed the effective date of the mandatory guarantee program and extended the PBGC's discretionary authority through June 30, 1979. Pub.L.No.95-214, 91 Stat. 1501 (1977). At the same time Congress ordered the PBGC to prepare a comprehensive report analyzing the multiemployer situation.
The PBGC submitted its report on July 1, 1978. Its major factual findings were that:
1. There were about two thousand covered multiemployer pension plans with approximately eight million participants. Pension Benefit Guaranty Corporation, Multiemployer Study Required by P.L. 95-214, at 1, 20 (1978) (hereafter cited as PBGC Report).
2. About ten per cent of these plans were experiencing financial difficulties that could result in plan terminations before 1988. These plans had about 1.3 million participants. Id. at 1, 138.
3. If all of these plans were to terminate, it could cost the insurance system about $4.8 billion to fund all plan benefits then covered by Title IV's guarantee. The annual premium needed to fund this liability would be unacceptably high. Id. at 2, 16, 139.
4. Limiting consideration to only those covered multiemployer pension plans which were experiencing sufficiently serious financial difficulties that it was likely they would become insolvent before 1988, the cost to the insurance system to fund all guaranteed plan benefits could be approximately $560 million. The annual per capita premium needed to fund this liability could rise from fifty cents to as much as nine dollars. Id. at 2, 16, 140.
The PBGC derived these figures by using a computer model that analyzed and predicted the projected financial health of a selected sample of plans. The PBGC stressed that it relied solely on economic data and statistical analysis in forecasting the expected number of terminations. It did not attempt to factor in as well any incentives favoring termination which ERISA itself might foster. Id. at 137, Appendix XIV. Nevertheless the PBGC argued that such incentives were both present and troubling:
Under the current statutory provisions, mandatory termination insurance for multiemployer plans would protect virtually all vested benefits in multiemployer plans, since the maximum guaranteeable benefit of $1,000 per month at age 65 is well above the average vested benefit level in multiemployer plans ...
Since all, or nearly all, of the vested benefits of participants would be guaranteed upon termination under the current law, the cost of plan termination to participants would be greatly reduced. This does not necessarily mean that participants will have an incentive to bargain for plan termination merely to take advantage of the insurance program. However, the removal of the threat of benefit losses does make termination a viable option to active employees in situations in which a high proportion of pension contributions is being used for the benefits of retirees.
The principal deterrent to plan termination under the current program is employer liability, which imposes a direct cost upon employers for termination, and an indirect cost on active employees since less money will be available for other labor costs. However, to assure that termination liabilities do not cause undue business hardship and loss of jobs, employer liability is limited to 30 percent of net worth. Because of this net worth limitation, employer liability may very well be less than the cost of maintaining *1032 the plan in some situations. Since the insurance program would cover most, if not all, of participants' vested benefits, it may be to the mutual economic advantage of the employers, the union, and the active employees to terminate the plan.
Other ERISA rules also may weaken a plan and result in eventual termination. The withdrawal rules may discourage large employers from entering multiemployer plans. The restrictions on benefit reductions contained in ERISA may cause a financially troubled plan to terminate, even though the benefits that would be paid if the plan terminated would be less (because of the guarantee limitations) than the benefits that would be paid if the plan were permitted to reduce its obligations to avoid termination. Id. at 23-24 (footnote omitted).
The PBGC analyzed in addition a number of ways ERISA could be amended. It examined proposals that would:
1. Require the PBGC to pay guaranteed benefits only when a multiemployer plan became insolvent, rather than simply terminated. Id. at 56, 57, 69, 70.
2. Reduce the level of benefits which were guaranteed. Id. at 56, 57.
3. Authorize the PBGC to provide financial assistance to multiemployer plans experiencing temporary financial problems. Id. at 56.
4. Permit multiemployer plans experiencing financial difficulties to reduce benefit payments. Id. at 40.
5. Require faster funding of multiemployer plan obligations. Id. at 56.
6. Increase the premiums paid by multiemployer plans. Id. at 18, 137-63.
7. Impose upon a withdrawing employer a fixed liability equal to that employer's share of the plan's unfunded vested liability. Id. at 40, 57.
On February 27, 1979, the PBGC submitted a legislative proposal advocating these ideas. This was followed on May 3, 1979 by the formal introduction in both houses of Congress of the legislation which ultimately became MPPAA. Because of the scope of the bill, Congress once again delayed the effective date of the 1974 mandatory guarantee program, this time until May 1, 1980. Pub.L.No.96-24, 93 Stat. 70 (1979).
The House Education and Labor Committee favorably reported MPPAA on April 3, 1980. The Committee specifically agreed with the PBGC's assessment of the 1974 Act:
Under the existing termination insurance rules, guarantees are provided by the PBGC to participants in a terminated plan. Guarantee levels are high enough to result in coverage of virtually 100 percent of the vested benefits of participants in certain multiemployer plans. Employers who withdraw from a multiemployer plan more than five years before termination have no further obligation to fund the liabilities of the plan, while employers who remain with a plan until it terminates, or withdraw within five years of termination, are liable to the PBGC for unfunded guaranteed benefits up to 30 percent of net worth.
In the case of a financially troubled plan, termination liability creates an additional incentive for employers to withdraw early. In such a plan, contribution increases may be escalating so sharply that termination liability may prove cheaper than continuing the plan. The remaining employers have an incentive to terminate the plan. Where active employees determine that benefits may be provided for them at considerably less cost than current contributions and are satisfied that vested benefits for retirees and others are virtually 100 percent covered by the guarantees, there is an incentive for the union to agree to terminate the plan. The result is to transfer the cost of providing benefits to the insurance system. The current termination insurance provisions of ERISA thus threaten the survival of multiemployer plans by exacerbating the problems of financially weak plans and encouraging employer withdrawals from and termination of plans in financial distress.
*1033 H.R.No.96-869, Part I, 96th Cong. 2d Sess., 54-55, reprinted in  U.S.Code Cong. & Ad.News 2918, 2922-23 (hereafter cited as Education and Labor Report); accord, id. at 60-61, reprinted in id. at 2928-29. The House Ways and Means Committee expressed similar views in its report released April 23, 1980. See H.R.No.96-869, Part II, 96th Cong. 2d Sess., 15, reprinted in  U.S.Code Cong. & Ad.News 3004 (hereafter cited as Ways and Means Report). On April 30, 1980, Congress for a third time delayed the implementation of the 1974 mandatory guarantees. This extension lasted until July 1, 1980. Pub.L.No.96-239, 94 Stat. 341 (1980). Finally, on May 22, 1980, the House approved MPPAA by a vote of 374-0. 126 Cong.Rec. H4170 (daily ed.).
Senate approval followed on July 29, 1980, but only after yet another extension Äî to August 1, 1980 Äî of the PBGC's discretionary authority under the 1974 law. Pub. L.No.96-293, 94 Stat. 610 (1980). The Senate vote in favor of MPPAA was 85-1. 126 Cong.Rec. S10169 (daily ed.). Differences between the House and Senate versions were eventually reconciled in September of 1980 and President Carter's approval followed soon thereafter on the 26th of that month.
For present purposes, what is most significant is that on September 26, 1980, the rules governing an employer's withdrawal from an on-going multiemployer pension plan changed. No longer did such behavior give rise, as it had under ERISA, to a contingent liability payable to the PBGC. Under MPPAA, an employer that withdraws must immediately begin to pay a fixed and certain debt owed to the plan.
The details of this "withdrawal liability" are extremely complex. To obtain a basic grasp, it is important to realize that MPPAA regulates multiemployer plans of the "Taft-Hartley" variety. These plans are in reality trusts created by collective bargaining between a union and several employers. By law, the union appoints half the fund's trustees, and the employers appoint the other half. 29 U.S.C. ß 186(c) (5) (B) (1976). The trust is funded by employer contributions which are made at a rate established by the terms of the collective contract. Id. This rate is usually expressed as an amount per time worked or product produced, e.g., $.75 per hour or $1.50 per item. The trustees collect the contributions and then determine, after considering all restraints imposed by the contract and all necessary actuarial data, the level of benefits which can prudently be offered. All decisions as to benefits are within the sole province of the trustees. As a general rule, once an employer parts with its contribution, it retains no rights thereafter to determine how that money should be spent. But see Borden, Inc. v. United Dairy Workers Pen. Program, 517 F. Supp. 1162 (E.D.Mich.1981), discussed at p. 1049, n. 50, infra.
A plan's vested liability is the actuarial present value of the benefit obligations which have vested. The difference between this figure and the value of the plan's assets is called its unfunded vested *1034 liability. Hansen Aff. at 24. Under MPPAA, a withdrawing employer becomes liable on the date of withdrawal for a proportionate share of the latter figure. 29 U.S.C.A. ß 1381 (Supp.1981). The duty to calculate and collect this liability falls to the plan's trustees.Id. at ß 1382. The trustees have substantial discretion in deciding how much to assess any given employer; the statute lists several different methods of allocating a plan's unfunded vested liability, and it further empowers the trustees to seek PBGC approval of a completely different method of their own design. Id. at ß 1391. Under the "presumptive" method of section 1391(b), the liability is derived basically by multiplying the plan's aggregate unfunded vested liability by a fraction whose numerator is the sum of all contributions required to have been made by the withdrawing employer during the previous five years. The denominator is the sum of all contributions made by all employers during this same period. If disputes between an employer and the trustees arise over an assessment, they are to be resolved, at least initially, in arbitration. Id. at ß 1401.
One final aspect of MPPAA requires comment. Though its provisions take effect in general upon enactment, the withdrawal liability rules are expressly retroactive to April 29, 1980. Id. at ß 1461(e) (2) (A). Any employer that withdraws after this date and before MPPAA's enactment is thus just as liable as those who leave after September 26, 1980.
II. THE PARTIES
This suit was brought in part by the trustees of the Local 705 International Brotherhood of Teamsters Pension Fund (the "Fund"). The Fund was created through collective bargaining in 1954 and is now one of the largest multiemployer plans in the country. Hansen Aff. at 4. As of January 31, 1980, 15,733 workers were employed by companies contributing to the Fund. Nearly 4300 additional workers enjoyed a vested pension right even though they no longer worked for a contributing employer. The Fund thus served roughly 20,000 "participants". Id. at 12.
During the five plan years ending January 31, 1980, over 600 employers withdrew from the Fund, and over 100 other companies joined. The bulk of all contributing employers employ fewer than five employees. Id. at 13. In 1981 each firm contributed $51.00 per week per employee. Id. at 17. This contribution rate has steadily increased since the plan's inception and especially in recent years. As late as 1971, it was only $16 per week. Id. at 22.
The trustees have increased benefits for active workers over 100% since 1971, keeping these workers roughly even with inflation. However, "[d]ue to the prohibitive high cost, it has not been possible to increase retiree benefits to meet this goal." Id.
On January 31, 1980 the Fund possessed assets with a market value of $174.7 million. Id. at 7. During the fiscal year ending on that date, the Fund collected $29.8 million in contributions, earned $13.6 million more on investments and disbursed $17.2 million in benefit payments. Administrative costs consumed $677,000. Id. at 9.
Despite the positive cash flow generated during this year, the Fund's unfunded vested liability exceeded $183 million on January 31, 1981. This translates into a figure of $11,645 per active worker. Id. at 25.
*1035 Joining the trustees as plaintiff is the Truck Drivers, Oil Drivers, Filling Station & Platform Workers Local 705, International Brotherhood of Teamsters, Chauffeurs, Warehousemen and Helpers of America (the "Union"). The Union bargained for the Agreement and Declaration of Trust establishing the Fund as well as all subsequent amendments thereto.
Plaintiffs Illinois Motor Truck Operators Association (IMTOA), Illinois Trucking Association, Inc., (ITA), Cartage Exchange of Chicago (CEC), and Motor Carriers Labor Advisory Council (MCLAC) are employer associations that represent numerous companies that contribute to the Fund. IMTOA, ITA, and CEC negotiated the initial Declaration of Trust establishing the Fund. MCLAC became a party to the Second Amended Trust Agreement in 1975.
The named defendants are the PBGC, the Secretary of Labor and the Secretary of the Treasury.
III. THE PLAINTIFFS' CLAIMS
Plaintiffs base their main constitutional challenge on the due process clause of the fifth amendment. They point out that each contributing employer's sole contractual obligation is to make timely contributions to the Fund in the amounts and under the conditions set forth in the governing collective bargaining agreement; once an employer pays its contributions, it is specifically exempted by the terms of the contract from any future liability to either the Fund or an employee seeking vested benefits. MPPAA thus increases each employer's obligations beyond the level agreed upon. Because of the withdrawal liability provisions, each withdrawn employer must continue to fund the plan following withdrawal even though contributions are not then required under the contract. MPPAA as well increases the obligations of the trustees since it obligates them to calculate and collect a liability that would not otherwise have been due and owing. Plaintiffs conclude from all this that their contractual rights have been impaired to such an extent that due process has been denied.
Plaintiffs argue further that the duties and obligations imposed upon them are arbitrarily more onerous than those imposed upon their single employer counterparts by ERISA. Such invidious discrimination is said to violate the equal protection component of the fifth amendment. Various phrases in MPPAA are also claimed to be impermissibly vague. Plaintiffs finally challenge the arbitration provisions on seventh amendment grounds.
Before addressing the merits of the constitutional attacks on MPPAA, it is first necessary to consider the arguments which have been pressed by various parties appearing as amici curiae. The TMX amici note that though twenty-two employers withdrew from the Fund between April 29, *1036 1980 and the date suit was filed, none was ever assessed a withdrawal liability by the Fund's trustees. The trustees instead came to this court seeking a declaratory judgment that MPPAA is void. This scenario leads TMX to conclude that no Article III "case or controversy" exists. TMX, joined by the American Trucking Association, Inc. (ATA), further contends that even if subject matter jurisdiction is present, the controversy nevertheless remains insufficiently ripe for adjudication.
The constitutional argument is that no plaintiff has standing to challenge MPPAA. TMX contends that only the twenty-two withdrawn employers have actually suffered injury-in-fact within the meaning of Article III. Since there are no allegations that any one of these twenty-two companies is now a member of a plaintiff association, it follows in TMX' view that no plaintiff can assert the needed injury. In addition, TMX argues that in any event, the named defendants have done nothing by themselves to cause any plaintiff harm. Cf. Valley Forge v. Americans United, ___ U.S. ___, ___, 102 S. Ct. 752, 758, 70 L. Ed. 2d 700 (1982) ("at an irreducible minimum, Art. III requires the party who invokes the court's authority to `show that he personally has suffered some actual or threatened injury as a result of the putatively illegal conduct of the defendant,' ...") (quoting Gladstone, Realtors v. Village of Bellwood, 441 U.S. 91, 99, 99 S. Ct. 1601, 1607, 60 L. Ed. 2d 66 (1979) (emphasis added)). Neither argument withstands analysis.
Subsequently filed affidavits establish that at least three of the withdrawn employers currently remain members of the plaintiff associations. Supplemental Exhibits 4-6. Moreover, TMX is mistaken in believing that the requisite injury is felt only by those employers who have actually withdrawn from the Fund. Injury-in-fact is also suffered by all employers who are members of the plaintiff associations and who are seriously contemplating withdrawal in the near future. These employers must disclose their potential liabilities to the financial community, and this likely diminishes their access to credit. See Supplemental Exhibits 7-9. Furthermore, in that the trustees must now discharge extensive unbargained-for responsibilities and obligations, they too have been injured in a concrete way. On the other hand, I agree with TMX that the Union's assertion of injury is too speculative to support federal jurisdiction. The possibility that the withdrawal liability provisions might create long-run incentives harmful to both the Fund and its Union beneficiaries is simply too remote and nebulous.
TMX further errs when it asserts that the PBGC is an inappropriate defendant for the trustees and employers to sue for redress of their injuries. Hardly an innocent bystander, the PBGC itself possesses ultimate responsibility for administering and enforcing the very law which causes the noted injuries. The PBGC, for example, can bring civil actions to enforce all provisions of Title IV. 29 U.S.C.A. ß 1303 (Supp.1981); see also id. at ß 1451 (authorizing PBGC intervention in private suits relating to withdrawal liability). The PBGC also promulgates all regulations needed to implement the statute. Id. at *1037 ß 1302(b) (3) (Supp.1981). In short, there is a sufficiently strong nexus between the acts of the PBGC, on the one hand, and the harm felt by the trustees and the employers, on the other, "that the[se] injur[ies] `fairly can be traced to the challenged action' and `[are] likely to be redressed by a favorable decision.'" Valley Forge v. Americans United, supra, 102 S. Ct. at 758 (quoting Simon v. Eastern Kentucky Welfare Rights Org., 426 U.S. 26, 38, 41, 96 S. Ct. 1917, 1924, 1925, 48 L. Ed. 2d 450 (1976)); cf. Hodel v. Virginia Surface Mining & Reclamation Association, 452 U.S. 264, 295, 101 S. Ct. 2352, 2370, 69 L. Ed. 2d 1 (1981) (claim that a federal statute was unconstitutional on its face held to be "properly before" the court in a suit for declaratory and injunctive relief brought against the federal official responsible for administering the law); Duke Power Co. v. Carolina Env. Study Group, 438 U.S. 59, 68-81, 98 S. Ct. 2620, 2627-2634, 57 L. Ed. 2d 595 (1978) (same).
A constitutional "case or controversy" thus pits the PBGC against both the trustees and the employers that either withdrew from the Fund or seriously contemplate doing so. Since the plaintiff associations properly represent these employer interests, both the associations and the trustees possess standing to litigate this action.
The question of ripeness implicates different issues. The ATA argues here that constitutional adjudication should be stayed until it can be determined exactly how MPPAA affects the various plaintiffs. It claims first that the Fund may be eligible to enjoy the relatively lenient treatment accorded "trucking industry" plans. An employer that ceases to contribute to such a plan does not "withdraw," and does not incur withdrawal liability, unless it "continue[s] to perform work within the jurisdiction of the plan." 29 U.S.C.A. ß 1383(d) (1) (Supp.1981). Even then, no liability is imposed if the PBGC determines that the employer's acts have caused no "substantial damage to [the plan's] contribution base." Id. at ß 1383(d) (3) (B) (i). The ATA thus argues that the trustees could obviate the need for constitutional adjudication by seeking and obtaining a PBGC determination that no withdrawal from the Fund has thus far caused "substantial damage."
Alternatively, the ATA suggests that the trustees could ask the PBGC for permission to treat the Fund as a "construction" or "entertainment" industry plan. In such plans an employer that ceases to contribute does not "withdraw" unless it "continues to perform work in the jurisdiction of the collective bargaining agreement of the type for which contributions were previously required," or resumes such work within five years. 29 U.S.C.A. ß 1383(b) (2) (B) (Supp. 1981) (construction industry); id. at ß 1383(c) (1) (entertainment industry). Moreover, even if a plan (like the Fund) cannot take advantage of these rules because its contributing employers are not a part of the construction or entertainment industry, its trustees can nevertheless petition the PBGC for the right to adopt a similar definition of withdrawal. Id. at *1038 ß 1383(f); see also 47 Fed.Reg. 12622 (1982) (to be codified in 29 C.F.R. Part 2645) (establishing procedure through which plans can petition). Since a positive response to such a request might result in a determination that no employer has yet "withdrawn" from the Fund, the ATA argues as before that the trustees should be forced to exhaust this possibility before their claims are heard.
A similar argument arose in Hodel v. Virginia Surface Mining & Reclamation Association, supra. Plaintiffs there included an association of surface coal miners and 63 of its member coal companies. They filed suit for declaratory and injunctive relief arguing that certain provisions of the Surface Mining Control and Reclamation Act were unconstitutional. One argument asserted was that the Act, by restricting the ability of the plaintiffs to use their land as they pleased, effected a taking without compensation. In analyzing this challenge, the Court distinguished between a facial challenge to the statute and a challenge to the statute as applied in specific circumstances. The Court determined that the plaintiffs could not challenge the application of the act because they presented no record as to the effect of the act on particular surface mining operations or on the specific parcels of land. Additionally, the court noted that the plaintiffs could have sought either a variance or a waiver from the statute's requirements, and that the "potential for such administrative solutions" confirmed the lack of ripeness of any challenge to the act as applied. 452 U.S. at 297, 101 S. Ct. at 2371.
Nevertheless, the Court reached and decided the merits of plaintiffs' facial challenge to the act. Neither the lack of a factual record nor the possibility of administrative relief in particular cases precluded the Court from determining that on its face the challenged statute was constitutional. Id. at 295-97, 101 S. Ct. at 2370. Similarly, in the instant case, the lack of a factual record and the possibility of an exemption might indicate a lack of ripeness if plaintiffs were challenging MPPAA as applied. However, they are bringing a facial challenge to the statute and, as in Hodel, these factors present no bar to their action. See generally Village of Euclid v. Ambler Realty Co., 272 U.S. 365, 386, 47 S. Ct. 114, 117, 71 L. Ed. 303 (1926).
V. DUE PROCESS: POST-ENACTMENT WITHDRAWALS
The focus here is on employers that withdraw from a multiemployer plan after September 26, 1980. The question to be resolved is whether it violates due process to subject such an employer to withdrawal liability, if, prior to enactment, an enforceable contract protected it from this form of liability.
The governing framework for analysis is set forth in Nachman, supra, a decision in which the Seventh Circuit rejected a similar due process attack on the single employer insurance system created by ERISA. This case arose when Nachman terminated its single employer plan in 1975. At that time, the plan's assets were insufficient to satisfy *1039 all vested claims for benefits, and the contract establishing the plan exempted Nachman from any obligation to make up this difference:
Benefits provided for herein shall be only such benefits as can be provided by the assets of the Fund. In the event of termination of th[e] Plan, there shall be no liability or obligation on the part of the company to make any further contribution to the Trustee except such contributions, if any, as on the effective date of such termination, may then be accrued but unpaid.
Nachman, supra, 592 F.2d at 950 (quoting Article V, section 3 of the Nachman plan). Nachman nevertheless became concerned that it could be held liable for the shortfall under ERISA. It brought suit for declaratory relief to resolve this question.
Nachman argued first that its employees' vested benefits were not insurable at all since they were not "nonforfeitable" within the meaning of Title IV. Thus, since no PBGC primary liability existed, no secondary liability could be imposed on Nachman. Alternatively, Nachman contended that to the extent ERISA did indeed subject it to liability, the statute was void for working an unconstitutional impairment of Nachman's contractual disclaimer of liability.
The Seventh Circuit rejected both arguments. The Court construed "nonforfeitable" to encompass the benefits which had vested under the terms of the Nachman plan, and it rejected Nachman's constitutional claim. Judge Sprecher began the discussion of the latter point by noting that the retroactivity of ERISA was at issue:
Title IV of ERISA does affect Nachman retroactively. The defendants argue that since ERISA only requires employers to assume liability for pension benefits which become due upon termination after the effective date of the Act, it assesses liability prospectively. This argument, however, relates only to the degree of retroactive impact. Although it is true that the statute applies only to prospective terminations, it also applies retrospectively to invalidate exclusion of liability clauses in pension plans agreed upon prior to ERISA. Thus to the extent that ERISA invalidates Nachman's otherwise valid acts which occurred prior to enactment, it is retroactive.
Id. at 958 (citation and footnote omitted). The Court next explored the proper standard against which to test ERISA, a federal statute that retroactively impaired contractual rights. On the one hand, only due process review was implicated; the generally stricter restraints imposed by the contract clause were facially irrelevant since the latter provision applies by its terms only when state legislation is challenged. See U.S.Const. Art. I, ß 10. The Court recognized, though, that "several authorities have suggested that the analysis employed in contract clause cases is also relevant to judicial scrutiny of Congressional enactments under the Due Process Clause. Both employ a means-end rationality test. Nachman, supra, 592 F.2d at 959 (citations omitted). Ultimately, however, the court found it unnecessary to resolve this point: "Since we are convinced that ERISA withstands the scrutiny employed under the Contract Clause cases, we need not decide whether the two clauses in fact impose identical restraints on legislative impairment of contracts." Id. (emphasis added).
The Court thus sustained ERISA after subjecting it to the essentially two-step process of review mandated by prior contract clause cases. The Court inquired initially whether Nachman's contractual rights had been altered only minimally, noting that a positive response would end the inquiry then and there. When the Court found instead that the impairment was substantial, id. at 961 & n.29, it moved to the next stage of analysis: an assessment of the "rationality" of Congress' legislating such an impairment. The Court defined "rationality" in the following manner:
Rationality must be determined by a comparison of the problem to be remedied with the nature and scope of the burden imposed to remedy that problem. In evaluating the nature and scope of the burden, it is appropriate to consider the *1040 reliance interests of the parties affected, whether the impairment of the private interest is effected in an area previously subjected to regulatory control, the equities of imposing the legislative burdens, and the inclusion of statutory provisions designed to limit and moderate the impact of the burdens. It must be emphasized that although these factors might improperly be used to express merely judicial approval or disapproval of the balance struck by Congress, they must only be used to determine whether the legislation represents a rational means to a legitimate end.
Id. at 960 (citations and footnote omitted). After examining these factors, the Court determined that, on balance, the single employer liability provisions were "rational." Id. at 961-63. In so doing, Judge Sprecher distinguished Allied Structural Steel Co. v. Spannaus, 438 U.S. 234, 98 S. Ct. 2716, 57 L. Ed. 2d 727 (1978) (hereafter cited as Allied Steel), a decision in which the Supreme Court had invalidated a Minnesota pension reform law on identical contract clause grounds.
The Supreme Court granted Nachman's petition for certiorari, but limited its review to the nonconstitutional questions presented, ultimately affirming the Court of Appeals. Significantly, however, the majority quoted extensively in one footnote from the Seventh Circuit's constitutional analysis. 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). Moreover, in that the decision to affirm exposed Nachman to a substantial liability, it seems doubtful that the Court would have so acted had it truly felt that Nachman's constitutional argument had merit. For these reasons, at least one court has read the Supreme Court's decision as a sub silentio affirmance of the constitutional aspect of Nachman. See A-T-O, Inc. v. Pension Benefit Guaranty Corp., 634 F.2d 1013, 1024 (6th Cir. 1980); see also Pension Benefit Guaranty Corp. v. Ouimet Corp., 630 F.2d 4, 12 (1st Cir. 1980), cert. denied, 450 U.S. 914, 101 S. Ct. 1356, 67 L. Ed. 2d 339 (1981).
Clearly, the task at hand is to apply the Seventh Circuit's approach in Nachman. Like ERISA, MPPAA retroactively impairs contractual rights. It disrupts the expectancies of all employers that withdraw from a multiemployer plan which is governed by a pre-MPPAA document disclaiming postwithdrawal liability. This is true even as to withdrawals occurring after enactment. See p. 1039, supra. Because of this impact, the question arises, as in Nachman, whether such a law must survive contract clause scrutiny. The court in Nachman left this question unresolved, and I will do likewise, *1041 since I believe that MPPAA survives such heightened review in any event.
This is not to say that MPPAA works but a mere minimal disruption of contractual expectancies. As one commentator has explained, contractual limitations upon employer liability are important aspects of the multiemployer mechanism:
The same contribution rate is required of each employer without reference to the cost factors of his own employee group. As a result, some employers may pay more and others less than their share of the cost of benefits for their own employees ...
[T]he Union is the cohesive force demanding that employers accept the plan's average experience in lieu of their own costs and offering a limitation of contribution liability as a quid pro quo.
J. Melone, Collectively Bargained Multi-Employer Pension Plans 95-96 (1963); see also Peick Aff. at ∂∂ 5, 11; Tr. at 7. A law which upsets an arrangement of such centrality cannot be deemed de minimis.
The Act nevertheless withstands facial due process attack because it is "rational" within the meaning of Nachman. This conclusion follows from an examination of the factors deemed relevant in that decision.
A. The Reliance Interests
The basic justification for withdrawal liability is not hard to discern. By obligating a withdrawn employer to amortize its share of the unfunded vested liability which remains in its former plan, withdrawal liability increases the likelihood that sufficient funds will actually be accumulated to satisfy the debt. The possibility of default is reduced, diminishing in turn the threat that employee expectations of payment will be dashed. Withdrawal liability thus protects employee reliance upon the promise of vested pension benefits. To this extent, it clearly furthers a legitimate and commendable objective. Nachman, supra, 592 F.2d at 962; accord, Pension Benefit Guaranty Corp. v. Ouimet Corp., 470 F. Supp. 945, 957 n.28 (D.Mass.1979), aff'd, 630 F.2d 4 (1st Cir. 1980), cert. denied, 450 U.S. 914, 101 S. Ct. 1356, 69 L. Ed. 2d 339 (1981); Lear Siegler, Inc. v. Pension Benefit Guaranty Corporation, 238 Pens.Rep. D-3, D-5 (E.D.Mich.1979).
Yet there is a second side to this equation. In deciding to join and remain a part of a multiemployer plan, an employer relies heavily on its contractual protection from post-withdrawal liability. See p. 1041, supra. MPPAA undermines this expectancy and accordingly cuts both ways in terms of reliance. It protects the interests of the employees, but does so at the expense of the employers.
The issue is whether it was rational for Congress to adopt this hierarchy of interests. When faced with the identical question in Nachman, Judge Sprecher answered in the affirmative. Nachman, supra, 592 F.2d at 961-62. A similar result must follow here in the absence of some basis for distinction.
In upholding ERISA, the Nachman court relied heavily on the strength of the record before Congress in 1974. This record proved beyond doubt that employee expectations were often frustrated when a single employer terminated its plan; concrete evidence established "that each year somewhere in the vicinity of 20,000 workers lost vested pension benefits due to causes beyond their control when a pension plan terminated." Nachman, supra, 592 F.2d at 960-61 (footnote omitted). By contrast, plaintiffs contend, the record underlying MPPAA pales in comparison. They stress *1042 that the PBGC cited in its entire 1978 Report to Congress only four actual instances in which a multiemployer plan failed to pay vested benefits during the years following the enactment of ERISA. See PBGC Report at Appendix XV. They further argue that Congress could not have reasonably relied upon the PBGC's computer-generated predictions of future plan terminations. Even the PBGC realized that "the number of terminations ... cannot be projected with any degree of certainty," and that its figures consequently "[c]ould not be viewed as precise projections." PBGC Report at 4, 138 (emphasis in original). Plaintiffs thus conclude that Nachman has no relevance for this case; that since there was no evidence upon which Congress could rationally conclude that multiemployer plan benefits were in fact threatened in 1980, the record does not support the "Draconian penalty" of withdrawal liability; and the more relevant precedent is Allied Steel: "[T]here is no showing in the record before [the Court] that this severe disruption of contractual expectations was necessary to meet an important general social problem." Allied Steel, supra, 438 U.S. at 247, 98 S. Ct. at 2723 (emphasis added).
I am not persuaded. That few multiemployer plans actually terminated prior to the enactment of MPPAA does not by itself prove that Congress had no grounds for concern. Congress need not wait for an actual disaster to strike before attempting a cure. Congress, moreover, was well aware of limitations inherent in the PBGC study. It knew full well that the report did not "establish  with any degree of exactitude ... the probable incidence of [plan] terminations." Education and Labor Report, supra, at 57, reprinted in  U.S.Code Cong. & Ad.News 2925. Congress nevertheless believed that the report did establish one basic fact: that the "magnitude of risk" Äî whatever its exact size Äî was "intolerably high." Id. Congress decided, in short, to take the report seriously. It chose not to gamble on the hope that the PBGC had predicted only an improbable, worst-case possibility. After examining the report in its entirety, I cannot say that this decision was either irrational or unreasonable. Much more would have to be shown to prove otherwise:
Legislation adjusting the rights and responsibilities of contracting parties must be upon reasonable conditions and of a character appropriate to the public purpose justifying its adoption. As is customary in reviewing economic and social regulation, however, courts properly defer to legislative judgment as to the necessity and reasonableness of a particular measure.
United States Trust Co. of New York v. New Jersey, 431 U.S. 1, 22, 97 S. Ct. 1505, 1517, 52 L. Ed. 2d 92 (1977) (footnote and citations omitted) (emphasis added); accord, Norman v. B&O R. Co., 294 U.S. 240, 311, 55 S. Ct. 407, 417, 79 L. Ed. 885 (1935); Retirement Board v. Alton R. Co., 295 U.S. 330, 379, 55 S. Ct. 758, 775, 79 L. Ed. 1468 (1935) (Hughes, C. J., dissenting); Slawson, Constitutional *1043 and Legislative Considerations in Retroactive Lawmaking, 48 Calif.L.Rev. 216, 248 (1960); The Supreme Court, 1977 Term, 92 Harv.L.Rev. 57, 97 (1978); see generally Duke Power Co. v. Carolina Env. Study Group, 438 U.S. 59, 83-84, 98 S. Ct. 2620, 2635, 57 L. Ed. 2d 595 (1978); Note, A Process-Oriented Approach to the Contract Clause, 89 Yale L.J. 1623, 1636-37 (1980) (hereafter cited as Yale Note). In 1980 as in 1974, there was sufficient evidence before Congress to justify a law impairing employer reliance.
Plaintiffs next argue that Nachman is distinguishable because MPPAA impairs employer reliance interests of a more formidable nature than those disrupted by ERISA. Further background is needed to explore this contention.
In Allied Steel the Supreme Court voided a Minnesota statute that obligated employers to pay a "pension funding charge" upon the termination of their Minnesota operations. In essence, employers were required to purchase annuities sufficient to fund pensions for all Minnesota employees with more than ten years experience. Under the terms of the actual Allied Steel plan, by comparison, no pension rights vested in less than fifteen years, and pensions were payable upon termination only to the extent that funds remained in the plan. The Minnesota law thus disrupted two of Allied's expectancies: (1) that it would never have to pay anything at all to its employees with more than ten, but less than fifteen, years experience; and (2) that it would never have to pay its vested employees anything more than the amounts it had put into its plan prior to termination, regardless of how much these employees had actually earned. In condemning the Minnesota law, the Supreme Court stressed only the former impairment. Allied Steel, supra, 438 U.S. at 246, 98 S. Ct. at 2723. This reasoning prompted the court in Nachman to conclude that the latter type of disruption is relatively insignificant compared to the gain in employee security that thereby results. Nachman, supra, 592 F.2d at 961 & n.31; accord, A-T-O, Inc. v. Pension Benefit Guaranty Corp., supra, 634 F.2d at 1026; Pension Benefit Guaranty Corp. v. Ouimet Corp., supra, 470 F. Supp. at 956. Implicit in Nachman is a determination that a single employer cannot reasonably rely upon a putative contractual right to terminate without liability a plan whose vested obligations are not fully funded.
One could argue that MPPAA merely places similar restraints upon employers that belong to a multiemployer plan. They too are simply denied the right to abandon a plan that cannot meet its current vested debt. However, as plaintiffs point out, this analogy is flawed. The essence of a single employer plan is a promise to pay benefits running directly from one employer to its employees. This promise, when stripped to its "true nature," is contingent only upon the employer's receiving a specified "length of service." Nachman, supra, 592 F.2d at 962 (quoting Alabama Power Co. v. Davis, 431 U.S. 581, 593, 97 S. Ct. 2002, 2009, 52 L. Ed. 2d 595 (1977)). Since employees rely on this promise and agree to accept a wage package in which the non-pension component is smaller than it would otherwise have been, the issue in Nachman ultimately boiled down to "who should bear the costs of a plan termination: a solvent employer who has received the full benefit he bargained for or the employee with vested benefit rights." Id. Judge Sprecher's answer was clear: Nachman would be held liable for the shortfall and would not be allowed to break its promise. In relying upon its liability disclaimer clause, Nachman had relied upon nothing more than an asserted right to break the "true" deal it had struck with its employees. Cf. A-T-O, Inc. v. Pension Benefit Guaranty Corp., supra, 634 F.2d at 1026 n.14 (under state law, an employer's obligation to pay vested benefits *1044 may be absolute, notwithstanding a liability disclaimer clause); Pension Benefit Guaranty Corp. v. Ouimet Corp., supra, 470 F. Supp. at 957 & n.27 (same); Lear Siegler, Inc. v. Pension Benefit Guaranty Corp., supra, 238 Pens.Rep. at D-5 (employer liability disclaimers cannot be given effect because their enforcement frustrates valid employee expectations). It was thus hardly surprising that the court found Nachman's reliance to be worthy of little protection. Courts have often sustained retroactive laws which merely enjoin contracting parties to observe the spirit and terms of their initial contract. See, e.g., City of El Paso v. Simmons, 379 U.S. 497, 85 S. Ct. 577, 13 L. Ed. 2d 446 (1965); Hochman, The Supreme Court and the Constitutionality of Retroactive Legislation, 73 Harv.L.Rev. 692, 720-21 (1960).
The same conclusions cannot be drawn in the multiemployer filed. Here, the promise to pay benefits runs not from one employer, but from a trust financed by an entire group of firms acting collectively. It follows that the withdrawal of a single employer does not automatically lead to a breach of the payment duty. For even when the plan lacks sufficient assets to pay all claims which have vested as of the date of withdrawal, the employers that remain may pick up the slack to such an extent that the harm caused by the withdrawal substantially abates. An employer that relies upon a liability disclaimer clause contained in a multiemployer plan is therefore not relying upon an asserted right to frustrate employee expectations at will. The employer's reliance is in this sense more reasonable than anything present in Nachman.
However, too much cannot be made of this argument. As discussed more fully at p. 1046, infra, withdrawals may very well increase the possibility of default. An employer that relies upon a contractual exemption from post-withdrawal liability is thus relying upon a right to unleash forces that are potentially destructive of employee benefit security. The difference between this case and Nachman is not as great as plaintiffs assert.
Moreover, in a second and perhaps more important sense, Nachman's reliance interest was more reasonable than those impaired by MPPAA. It has long been a tenet of constitutional law that "[t]hose who do business in the regulated field cannot object if the legislative scheme is buttressed by subsequent amendments to achieve the legitimate end." F. H. A. v. The Darlington, Inc., 358 U.S. 84, 91, 79 S. Ct. 141, 146, 3 L. Ed. 2d 132 (1958) (citations omitted); accord, Allied Steel, supra, 438 U.S. at 249, 98 S. Ct. at 2724; Veix v. Sixth Ward Building & Loan Assn., 310 U.S. 32, 38, 60 S. Ct. 792, 794, 84 L. Ed. 1061 (1940). Reliance upon existing rights Äî even those which are reasonable in the abstract Äî is itself unreasonable when the relying party has notice that future legislation may alter these rights. In Nachman, the Court drew upon this doctrine in finding an additional reason to denigrate Nachman's reliance: Since pension plan terminations had previously been subject to federal regulation, Nachman had received warning that its contractual rights might one day be impaired. Nachman, supra, 592 F.2d at 962 & n.33.
The same reasoning applies with even greater force here. As plaintiffs acknowledge, multiemployer plans have been subject to decades of pre-MPPAA federal regulation under Section 165 (now section 401) of the Internal Revenue Code, section 302 of the Labor-Management Relations Act of 1947, and the Welfare and Pension Plans Disclosure Act of 1958. Moreover, and most significant, the 1974 enactment of ERISA demonstrated in the clearest possible way that contractual limitations on withdrawal liability were themselves susceptible to federal displacement. By its terms, ERISA voided all absolute exemptions, and installed in their place a regime of contingent liability. See p. 1030, supra. This development alone afforded clear warning that the federal government might one day act again and further buttress the legislative *1045 scheme it had created. Nachman never received such explicit warning. Compare Nachman, supra, 592 F.2d at 962 n.33.
Indeed, by September 26, 1980, the date MPPAA was enacted, Congress acted a second time. It introduced mandatory multiemployer insurance and thus rendered employer liability less contingent and more certain. See p. 1030, supra. In addition, constitutional challenges to the termination insurance system had by then been rejected by the Seventh Circuit in Nachman and the First Circuit in Ouimet. Two similar district court opinions (Ouimet and Lear Siegler) were also on the books as was the Supreme Court's decision in Nachman, a decision which can be read to approve of the Seventh Circuit's constitutional analysis. See p. 1040, supra. None of these decisions, to be sure, dealt specifically with the multiemployer aspects of ERISA. They did nevertheless indicate that a constitutional attack on these provisions was unlikely to succeed. See generally, Duke Note, supra, at 661-71 (arguing that the 1974 multiemployer provisions of ERISA were constitutional). Thus, to the extent that any employer still contributed to a multiemployer plan on September 26, 1980, and did so solely because it believed that its contractual exemption from post-withdrawal liability was constitutionally immune from all legislative modification, it was not acting in the most reasonable of fashions. Cf. Yale Note, supra, at 1629 n.34 (a court "might safely assume that any legislation closely resembling a law previously upheld against a contract clause challenge would violate no legitimate expectations.")
In sum, it can be argued that in one sense MPPAA impairs employer interests which are more reasonable than those displaced by ERISA. Yet this contention is largely, if not completely, counterbalanced by the *1046 more extensive history of prior federal regulation found in this case. MPPAA, moreover, furthers employee rights that are at least on a par with those shielded by ERISA. In light of these considerations, I find that the "reliance" aspect of Nachman has not been distinguished and that the conclusion reached there applies equally here as well: Congress acted rationally in making its basic decision to subordinate the reliance interests of the employers to those of the employees. MPPAA survives this level of analysis.
B. The Equities
In the absence of withdrawal liability, a withdrawn employer ceases to fund the plan it has abandoned. The plan's unfunded vested debt remains, however, and must be financed by the employers that contribute in the years that follow. Since these employers must therefore furnish the funds which the withdrawn employer would have contributed had it not departed, Congress concluded that "withdrawals ... unfairly burden remaining contributors with unfunded benefit obligations left behind by the withdrawn employer." Education and Labor Report, supra, at 60, reprinted in  U.S.Code Cong. & Ad.News 2928.
In economic terms, Congress believed that withdrawals place upward pressure on plan contribution rates. Following a withdrawal, the contribution base of the affected plan is smaller than it would otherwise have been; in order to finance the plan's vested liability, it may become necessary to increase the contribution rate in order to extract more income from each remaining contribution unit. If this occurs, plan participation becomes less desirable. New employers will be less inclined to join, see id. at 77, reprinted in id. at 2945; PBGC Report at 13, and the remaining contributors may find withdrawal to be economic for them as well. See Education and Labor Report, supra, at 54, reprinted in  U.S.Code Cong. & Ad.News 2922. The possibility of plan insolvency increases to some extent, and with it the chance that the PBGC ultimately will become obligated to pay guaranteed benefits. Employee interests are, of course, also thereby threatened. Withdrawal liability responds to these concerns by deterring withdrawals and by shoring up the contribution base of an abandoned plan when withdrawals nevertheless occur. MPPAA thus protects the interlocking interests of the PBGC, its *1047 premium payers, the non-withdrawing employers and the vested employees. It does so at the expense of the withdrawn employers, the parties whose conduct threatens the harm. This seems equitable. See generally L. Tribe, American Constitutional Law, ß 9-4 (1978); id. at 43 (Supp. 1979).
But as with the reliance analysis, there is a second side to the ledger. It was assumed above that an unfunded vested liability remained to be amortized at the time of the withdrawal. But whether this is in fact the case Äî and whether a withdrawal liability must as a result be paid Äî is a contingency the withdrawing employer cannot control. The extent of a plan's unfunded vested liability depends upon the value of the plan's assets and vested liability, two factors a withdrawing employer does not regulate. Current asset levels, for example, depend in part upon the economy generally, the trustees' investment acumen, and the promptness with which third-party employers pay contributions and withdrawal assessments. Direct employer control over these contingencies is minimal at best. Moreover, the size of the vested liability depends most crucially upon the level of benefits that the trustees have promised. Yet in setting these amounts, as in everything else the trustees occupy a fiduciary relationship with the plan's beneficiaries. They consequently owe their allegiance to the interests of the employees and to no other party; even the management-appointed trustees must be guided solely by what is best for the participants: "[T]he duty of the management-appointed trustee ... is directly antithetical to that of an agent of the appointed party." N. L. R. B. v. Amax Coal Co., 453 U.S. 322, 331-32, 101 S. Ct. 2789, 2794, 69 L. Ed. 2d 672 (1981) (footnote omitted). MPPAA thus creates, in plaintiffs' view, a liability which is determined in large part by parties who cannot legally consider the impact their actions have on the employers who foot the bill.See 126 Cong.Rec. H4163 (daily ed. May 21, 1980) (colloquy between Representatives Erlenborn and Peyser); Duke Note, supra, at 662, 668. Not even the statute condemned in Allied Steel shared this defect.
The problem of trustee independence cannot be ignored, as the recent Bay Area arbitration decision evidences. See Bay Area Painters Pension Trust Fund, 2 Employee Benefit Cases (BNA) 1724 (1981). The Bay Area case arose when the "labor" trustees of the Bay Area Painters Pension *1048 Trust Fund proposed a benefit increase which had the effect of increasing the plan's unfunded vested liability by $625,000. The "management" trustees objected, and arbitration followed. The arbitrator held for the labor trustees, finding that the increase in the unfunded vested liability was relatively "[in]significant and would not put the Trust into jeopardy of insolvency or other problems." Id. at 1734. The arbitrator specifically reasoned that the proposal could not be rejected solely because it increased each employer's potential withdrawal liability. The trustees could take into account only the interests of "the participants of the Trust and their beneficiaries. Trustees cannot act for the interest of some third party not a part of the trust concern." Id. at 1733.
However, there are limits to this principle. Bay Area does not establish a per se rule barring trustee consideration of withdrawal liability:
While the withdrawal liability of the employers, on its face, is a concern of the employers and not of the participants in the trust, it is possible that such withdrawal liability could have an impact on the overall health of Trust.... If in the opinion of the trustees or their advisors, the withdrawal liability of employers has some impact on the ability of the Trust to meet its obligations to the participants and the beneficiaries, then withdrawal liability ought to be considered.
Id. at 1734. The legislative history of MPPAA confirms that plan trustees do not violate their fiduciary duties simply because they take steps which minimize the impact of the withdrawal liability rules. If these steps are also in the best interests of the plan, they should be taken. See Education and Labor Report, supra, at 67, reprinted in  U.S.Code Cong. & Ad.News 2935; 126 Cong.Rec. H3949 (daily ed. May 21, 1980) (remarks of Rep. Thompson).
A second arbitrator has in fact ruled on the basis of this logic. In Texas Carpenters Pension Plan, 9 Pens.Rep. (BNA) 99 (1982), the arbitrator was asked to choose from between two proposals designed to increase benefits. Under the proposal put forth by the union trustees, the plan's unfunded vested liability increased $5.3 million or roughly 95 percent. Under the management proposal, there was no increase at all. Both proposals were economically viable, the real difference between them lying in their relative impact upon the employer's potential withdrawal liability. Id. at 103. The arbitrator voted with management. Id. at 104.
It is not true, therefore, that the trustees are completely barred from considering the impact of their actions on the employers' withdrawal liability. Actual consideration of these interests is, moreover, likely, given management's statutory right to select half of the trustees. See generally Note, Withdrawal Liability for Double-Breasted Construction Employers Under the Multiemployer Pension Plan Amendments Act of 1980, 54 So.Cal.L.Rev. 1359, 1373 (1981). The perceived harshness of MPPAA is clearly mitigated by these considerations.
*1049 The problem may in fact be one of only limited duration. At a subsequent round of collective bargaining, employers can insist that restrictions upon trustee behavior be written into the actual contract and declaration of trust establishing their plan. They can bargain for language forbidding future benefit improvements which increase the plan's unfunded vested liability by more than a set amount or percentage. See generally Borden, Inc. v. United Dairy Workers Pension Program, 517 F. Supp. 1162 (E.D. Mich.1981); 126 Cong.Rec. S10103 (daily ed. July 29, 1980) (remarks of Sen. Dole); Bay Area Painters Pension Trust Fund, supra, 2 Employee Benefit Cases at 1735. If these demands are accepted, the employers will clearly regain substantial control over their level of exposure.
MPPAA itself ameliorates the additional concerns generated by the employers' inability to control current asset levels. In the first place, the statute places heavy penalties upon employers that are delinquent in paying either contributions or withdrawal assessments. This fact alone should decrease the incidence of tardy payments. MPPAA also allows plans to insure against the possibility that withdrawal liability debts may ultimately prove to be uncollectible. Plans can do this by joining a private "withdrawal liability payment fund," 29 U.S.C.A. ß 1403(c) (1) (C) (Supp.1981), or by enrolling in the "supplemental program" the PBGC is to create. Id. at ß 1402. By enacting each of these provisions, Congress made it more certain that multiemployer plans will actually receive the funds they are owed. Congress lowered the risk that a withdrawing employer will be forced to pay for amounts which others should have supplied in the past.
However, as a general rule, the considerations which have been canvassed thus far offer protection only against assessments which are based on post-enactment increases in the level of a plan's unfunded vested liability. These factors do not mitigate the employers' burden of amortizing the liabilities *1050 which already existed on September 26, 1980. Plaintiffs in fact claim that the statute contains no moderating features whatsoever. They draw comparisons with Nachman, wherein the Court stressed four aspects of ERISA which tempered the burdens it imposed:
(1) Liability could not exceed thirty percent of net worth.
(2) Employers were not liable for the full value of the employees' vested benefits, but only for the amount guaranteed by the PBGC.
(3) Employers could guard against their liability through optional PBGC insurance.
(4) The PBGC could in its discretion allow for reasonable terms of payment.
Nachman, supra, 592 F.2d at 963. MPPAA contains the essence of the latter two features. To at least some extent, withdrawal liability can be insured against through the mechanism of a withdrawal liability payment fund. See, p. 1049 & n. 53, supra. The statute further allows the liability to be paid over time; it is not due in one lump sum, but is rather amortized in accordance with the payment schedule established by the trustees. 29 U.S.C.A. ß 1399(c) (Supp. 1981). The trustees develop this schedule in two steps. They first calculate the amount the employer must pay each year. In doing this, they rely exclusively on the employer's past contribution history; the payment amount is set at a figure designed roughly to approximate what the employer would have contributed had it not withdrawn. Id. at ß 1399(c) (1) (C). The trustees then determine how many payments are needed to amortize the withdrawal liability in full. Id. at ß 1399(c) (1) (A) (ii).
By contrast, no provision of MPPAA embodies the first two moderating features deemed significant in Nachman. This is an important basis for distinction, particularly the absence of a net worth limitation. MPPAA, however, lessens its impact in several distinct ways of its own. For example, the mandatory "De Minimis" exemption excuses in general all assessments under $50,000; it further reduces to some extent all others under $150,000. Id. at ß 1389(a). More sweeping relief Äî a $100,000 absolute exemption and a $250,000 limit on sliding scale reductions Äî is also available if the trustees opt for it in their discretion. Id. at ß 1389(b). Plaintiffs belittle this provision, noting that it affords no relief at all to large employers that incur high levels of withdrawal liability. While this is true, it is equally true that the "De Minimis" rule protects a substantial number of employers. Indeed, when a $100,000 exemption was first proposed, Senator Williams objected that "[a]s much as 40 per cent of all unfunded vested benefits would be excluded under the proposed amendment. In some plans, the proposed amendment would exclude 100 per cent of the unfunded vested benefits." 126 Cong.Rec. S10144 (daily ed. July 29, 1980). Plaintiffs' own estimates support these remarks. They show that because of the "De Minimis" rule, nearly seventy percent of the firms contributing to the Local 705 Fund could withdraw without incurring any liability whatsoever. They further indicate that if the $100,000 exemption option were adopted, over eighty percent could withdraw and face a maximum penalty of only $3,500. Hansen Aff. at 13, 33. The "De Minimis" rule is a significant moderating feature.
Other provisions further reduce employer liability. If more than twenty years are needed to amortize an employer's withdrawal *1051 assessment, the "liability shall be limited to the first 20 annual payments." 29 U.S.C.A. ß 1399(c) (1) (B) (Supp.1981). Liability is reduced as well if a withdrawal results from the liquidation or dissolution of an employer's business. Id. at ß 1405. Plan trustees have broad discretion to "adopt rules providing for other terms and conditions for the satisfaction of an employer's withdrawal liability." Id. at ß 1404.
Also significant are the provisions of MPPAA which modify the basic definition of "withdrawal" itself. Three of these sections Äî those pertaining to the construction, entertainment and trucking industries Äî have already been discussed. See pp. 1037-1038, supra. Each reflects Congress' recognition that the contribution base of such plans is not harmed by many events that are considered "withdrawals" under the basic provisions of the Act. See, e.g., Ways & Means Report, supra, at 15, reprinted in  U.S.Code Cong. & Ad. News 3004 (construction and entertainment industries); 126 Cong.Rec. S11671 (daily ed. August 26, 1980) (remarks of Sen. Durenberger) (trucking industry). Congress further recognized that the same might be true of other plans not covered by the technical language of these sections. Congress thus authorized all plans to petition the PBGC for permission to adopt "special withdrawal liability rules similar to the rules described in" the entertainment and construction industry sections. 29 U.S.C.A. ß 1383(f) (Supp.1981); see pp. 1037-1038, supra. Moreover, a withdrawal does not generally occur when an employer ceases covered operations or ceases to contribute "as a result of a bona fide, arm's-length sale of assets to an unrelated party," if the latter entity is itself obligated to contribute to the plan in such amounts that the plan's contribution base substantially retains its pre-sale health.Id. at ß 1384. Each of these provisions evidences a Congressional attempt "to impose liability only to the extent necessary to achieve the legislative purpose." Nachman, supra, 592 F.2d at 962.
Moreover, and perhaps most significant, Congress did not saddle withdrawing employers with the entire burden of stabilizing underfunded multiemployer plans. Rather, as the PBGC argues,
withdrawal liability was only one feature of a comprehensive revision of the law governing multiemployer plans. Funding rules were tightened so that employers contributing to ongoing plans have to make minimum payments higher than under the 1974 law. Premiums were increased. PBGC funds were made available to supplement employer-derived assets so that insolvent plans can continue *1052 operating, though with reduced benefit payments. Participants' vested benefits were made subject to limited reductions in plans suffering substantial financial difficulties, and the ultimate protection of benefit payments, PBGC's guarantee, was reduced. PBGC Brief at 34.
Congress thus spread the pain around. It chose a mix of options which it thought best responded to the problems it perceived. To be sure, its solution may not have been the wisest. Plaintiffs have so argued, and not without force. Moreover, sentiment already exists on Capitol Hill to amend MPPAA in significant ways. See The Wall Street Journal, March 5, 1982, at 40, col. 1. Yet all this is not my concern. My concern is only with the basic fairness of the choices Congress actually made in 1980. Plaintiffs have convinced me that withdrawal liability can be harsh and onerous in certain instances. However, after examining the statute as a whole and the context in which it operates, I cannot say that the "mere enactment" of MPPAA was unconstitutional. The basic structure of the statute is not so inequitable that a facial attack on its validity can succeed. At least as to withdrawals occurring after enactment, no due process violation has been shown.
VI. DUE PROCESS: PRE-ENACTMENT WITHDRAWALS
Though the President signed MPPAA on September 26, 1980, the withdrawal liability provisions were rendered effective as of April 29, 1980. 29 U.S.C.A. ß 1461(e) (2) (A) (Supp.1981). Employers that withdrew during the intervening 150 day period thus found that the governing law changed after-the-fact. When they withdrew, their withdrawal liability was contingent upon plan termination within five years. Yet, on September 26, it became fixed and due, even if no termination had occurred in the interim. The issue is whether it was permissible for Congress to alter the legal effect of a closed transaction in this manner.
The starting point for analysis is my previous conclusion that withdrawal liability is constitutional in the context of post-enactment withdrawals. This finding dictates the relevant question for present purposes: *1053 Does the shift to the pre-enactment context introduce added considerations of a sufficiently compelling nature that the result must change? In other words, I will not repeat the analysis found in Part V. I simply note that the factors discussed there balance out to a finding in the statute's favor. In this section, I consider only the unique and additional issues posed by Congress' decision to utilize a retrospective date of effectiveness.
The most obvious new factor is the presence of a closed transaction. This was not a concern in the post-enactment context because the relevant employers were all plan contributors on the date of enactment. They were thus subject to liability only to the extent they later opted to withdraw in full awareness of the consequences. The same cannot be said of the employers that withdrew during the retrospective period. Their liability was fixed from the moment MPPAA was signed since they had already withdrawn, and at a time when the governing law had failed to inform them of the relevant consequences. The difference in position seems apparent. See Concord Control, Inc. v. International Union, 647 F.2d 701, 705 (6th Cir.), cert. denied, ___ U.S. ___, 102 S. Ct. 599, 70 L. Ed. 2d 590 (1981); A-T-O, Inc. v. Pension Benefit Guaranty Corp., supra, 634 F.2d at 1026.
In Nachman's terms, this difference translates into an added element of reliance. The employers that withdrew prior to enactment relied on more than a belief that their contracts would remain valid in the future. They relied as well on a belief that their agreements were enforceable at the actual moment they withdrew. Whether this is significant depends upon the reasonableness of the latter belief.
The legislative development of MPPAA becomes relevant at this juncture. When the bill was originally introduced on May 3, 1979, the withdrawal liability rules were effective as of February 27, 1979, the date the PBGC transmitted its initial legislative proposal to Congress. See 126 Cong.Rec. S10156 (daily ed. July 29, 1980) (remarks of Sen. Matsunaga). The February 27, 1979 date remained a part of the bill until June 1980, when the Senate Finance Committee determined that it was an unnecessarily harsh starting point. The Committee shortened the retrospective period by more than a year, advancing the effective date to April 29, 1980, where it now stands. Id.
By the latter date, the concept of a noncontingent withdrawal liability was hardly novel. It had been the subject of serious discussion for nearly two years, ever since the PBGC had submitted its July 1978 report to Congress. As a result of this legislative scrutiny, MPPAA was far along the road to enactment by April 29, 1980. It had by then already received the approval of the House Education and Labor Committee, the House Ways and Means Committee, and the Senate Labor and Human Resources Committee. Each approved version, moreover, contained the original February 27, 1979 date of effectiveness. These facts, taken together, clearly temper the employers' claims of surprise. They indicate that even the pre-enactment withdrawals were made under circumstances that afforded some notice of the consequences. During the entire retrospective period, there was strong evidence to suggest that MPPAA was likely to pass and that it would be retrospective when it did.
*1054 Similar reasoning has been employed by the Supreme Court in decisions sustaining retrospective tax legislation. The Silver Purchase Act of 1934, for example, was upheld even though it taxed transactions consummated prior to enactment:
The period of retroactivity prescribed for this taxing provision reaches backward from June 19, 1934, the date of the act, to and including May 15, 1934 Äî 35 days. For some months prior to this period there was strong pressure for legislation requiring increased acquisition and use of silver by the government, and several bills providing therefor were presented in the Senate and House of Representatives. On May 22 the President sent to Congress a message recommending legislation for increasing the amount of silver in our monetary stocks and further recommending the imposition of a tax of at least 50 per cent. on profits accruing from private dealing in silver. The bill which became the Silver Purchase Act was introduced May 23 in response to this message. In these circumstances we think the period of retroactivity fixed in the act is not unreasonable, but consistent with the practice sustained by this Court in the cases already cited.
United States v. Hudson, 299 U.S. 498, 501, 57 S. Ct. 309, 310, 81 L. Ed. 370 (1937) (footnote omitted). Under Hudson, it is thus very relevant that "there was strong pressure for legislation" imposing withdrawal liability in the "months prior to [the retrospective] period." Indeed, the present facts are more sympathetic to a finding of constitutionality than those in Hudson. In this case, significant legislative activity preceded April 29, 1980. In Hudson, the legislation became effective eight days before it was even introduced.
Hudson, moreover, is not an obsolete oddity. Just last year, the Supreme Court utilized similar logic while sustaining a retrospective application of the Tax Reform Act of 1976:
The proposed increase in rate had been under public discussion for almost a year before its enactment. The Tax Reform Act of 1976 reflected a compromise between the House and Senate proposals. Both bills, however, provided that changes in the minimum tax were to be effective for taxable years beginning after 1975. Appellee, therefore, had ample notice of the increase in the effective minimum rate.
United States v. Darusmont, 449 U.S. 292, 299, 101 S. Ct. 549, 553, 66 L. Ed. 2d 513 (1981) (per curiam) (citations omitted); see also Cooper v. United States, 280 U.S. 409, 411, 50 S. Ct. 164, 165, 74 L. Ed. 516 (1930). To be sure, the analogy between the present case, on the one hand, and Darusmont and Hudson, on the other, can be drawn only so far. Tax cases are to an extent sui generis. See generally Welch v. Henry, 305 U.S. 134, 146-47, 59 S. Ct. 121, *1055 125, 83 L. Ed. 87 (1938). Nevertheless, the quoted passages contain a message that has relevance beyond the narrow realm of the tax code. They indicate that fair warning of a statute's terms can conceivably exist even before it is enacted. The substantiality of Congress' pre-April 29, 1980 activity is thus a factor that militates against the reasonableness of the employers' second reliance interest.
Moreover, rational theories justify a pre-enactment date of effectiveness. Congress was concerned that if the statute became effective only upon enactment, "opportunistic" employers would be encouraged to withdraw while Congress was still debating. See 126 Cong.Rec. S10156 (daily ed. July 29, 1980) (remarks of Sen. Matsunaga); id. at S10101 (remarks of Sen. Javits). One of the very flaws Congress perceived in ERISA Äî its encouragement of early withdrawals Äî would thus have been perpetuated had MPPAA not included a retrospective date of effectiveness. Congress further realized that pre-enactment withdrawals affect equity and plan stability to the same extent as withdrawals occurring after enactment. Congress was particularly concerned with the relative equities between withdrawing and remaining employers: If pre-enactment withdrawals were excused, the firms still contributing on the date of enactment would face greater potential liabilities. See id. at S10101; id. at S10158 (remarks of Sen. Matsunaga).
Plaintiffs seemingly do not challenge these theories in the abstract. Rather, they argue, somewhat as they did in part V., that these contentions are not supported by the factual record before Congress: Since "there is no recorded instance of employers withdrawing during congressional debate," it was irrational for Congress to attack a problem that did not exist. Once again, I am unpersuaded. To begin with, plaintiffs' premise is wrong. Some employers clearly withdrew during the retrospective phase; thirteen did from the Local 705 Fund alone. See Hansen Aff. at 34. I assume, therefore, that plaintiffs are really arguing that not enough employers left to warrant concern, that since a "stampede" of pre-enactment withdrawals never materialized, Congress lacked the authority to impose liability on those which in fact occurred. This logic fails. First of all, Congress' policies were threatened by all pre-enactment withdrawals, not simply those which occurred in conjunction with others. It is also not clear why a withdrawn employer should escape liability simply because others "behaved" and did not withdraw themselves. In any event, Congress showed through the course of the legislative debate that it was sensitive to the plaintiffs' concerns. Congress advanced MPPAA's effective date from February 27, 1979 to April 29, 1980 in part precisely because employers had shown "restraint" and had not withdrawn in droves during the interim. See 126 Cong.Rec. S10101 (daily ed. July 29, 1980) (remarks of Sen. Javits). In A-T-O similar legislative behavior was found to be indicative of rational law-making:
Congress gave careful attention to the retroactivity problem.... Congress not only passed Title IV, it also advanced the effective date from that originally proposed in the House and Senate bills.
A-T-O, Inc. v. Pension Benefit Guaranty Corp., supra, 634 F.2d at 1026. The very care Congress exercised in picking *1056 MPPAA's date of effectiveness bolsters the case in favor of the statute.
One final factor bears mentioning. Conceivably, many of the employers that withdrew during the retrospective period can still avoid an assessment. Section 4207 of the Act requires the PBGC to promulgate regulations prescribing the conditions under which a withdrawn employer can vitiate its liability by reentering its plan. If employers make use of this provision, they will be subject to further liability only if they engage in an additional post-enactment withdrawal.
There are thus several arguments in MPPAA's favor even in the pre-enactment context. While none is sufficient standing alone, in combination they carry the day. They offset the added burdens of the statute's retrospective effect to such a degree that the balance of considerations continues to tip, as it did in the post-enactment context, in the statute's favor. I must emphasize that this is an extremely close call. Congress has most probably gone to the very limits of its constitutional power. Nevertheless, I cannot say that it went too far.
VII. EQUAL PROTECTION
Plaintiffs further argue that MPPAA imposes burdens which are irrationally more onerous than those levied on single employer trustees, contributors and participants. Plaintiffs point to four distinctions:
(1) Though multiemployer plan trustees are obligated to calculate and collect each withdrawn employer's withdrawal liability, single employer trustees share no similar responsibility. The PBGC calculates and collects a single employer's termination liability itself.
(2) Employers that contribute to a multiemployer plan can incur liability even when the plan continues in existence. Single employers become liable only upon plan termination.
*1057 (3) Single employers can in no event become liable for more than thirty percent of their net worth. Multiemployer liability has no upper limit.
(4) Multiemployer benefits are allegedly protected to a much lesser degree than those earned by single employer plan participants. For example, the PBGC's guarantees become operative upon the termination of a single employer plan, but not a multiemployer plan. In the latter context, only plan insolvency is an insurable event. Moreover, guaranteed benefit levels are generally lower for multiemployer benefits.
Before discussing these claims in detail, a few prefatory comments are in order. Plaintiffs have never argued that these distinctions run along "suspect" lines or that they impinge upon "fundamental rights" as that term is understood in the equal protection context. This being the case, plaintiffs can succeed only upon a showing that Congress has acted in a "patently arbitrary or irrational way." U. S. Railroad Retirement Bd. v. Fritz, 449 U.S. 166, 177, 101 S. Ct. 453, 460, 66 L. Ed. 2d 368 (1980); see The Supreme Court, 1980 Term, 95 Harv.L.Rev. 91, 155 (1981). As the Supreme Court recently explained:
Social and economic legislation ... that does not employ suspect classifications or impinge on fundamental rights must be upheld against equal protection attack when the legislative means are rationally related to a legitimate governmental purpose. Moreover, such legislation carries with it a presumption of rationality that can only be overcome by a clear showing of arbitrariness and irrationality. As the Court explained in Vance v. Bradley, 440 U.S. 93, 97 [99 S. Ct. 939, 942, 59 L. Ed. 2d 171] (1979), social and economic legislation is valid unless "the varying treatment of different groups or persons is so unrelated to the achievement of any combination of legitimate purposes that [a court] can only conclude that the legislature's actions were irrational." This is a heavy burden ....
Hodel v. Indiana, 452 U.S. 314, 331-32, 101 S. Ct. 2376, 2386, 69 L. Ed. 2d 40 (1981) (citations omitted). In light of these considerations, plaintiff's arguments plainly lack merit.
Withdrawal liability is a debt owed to the plan which has been abandoned. See 29 U.S.C.A. ß 1383(a) (Supp.1981). Single employer termination liability is payable to the PBGC. See 29 U.S.C. ß 1362(b) (1976). In both contexts, therefore, Congress imposed the duty of calculation and collection upon the party to whom the debt runs. Congress rationally treated similarly situated entities, and did not violate the norm of equal protection.
Plaintiffs' second argument also fails. It was not "patently irrational" for Congress to conclude that both withdrawals and terminations pose threats to employee benefit security. See p. 1044, supra.
The third and fourth arguments are no more substantial. Congress eliminated the thirty percent limitation on multiemployer liability for basically two reasons:
(1) Congress wished to deter withdrawals by making them more expensive. See, e.g., Education and Labor Report, supra, at 61, reprinted in  U.S.Code Cong. & Ad.News 2929; 126 Cong.Rec. S10115 (daily ed. July 29, 1980) (Joint Explanation of S.1076 by Senators Williams and Long).
(2) Congress further wished to insulate plans from the deleterious effects of withdrawals that nevertheless occur. By making more employer assets available for this purpose, Congress lowered the PBGC's potential exposure and, thus, the required level of premiums needed to finance the termination insurance system. See, e.g., Education and Labor Report, supra, at 63, reprinted in  U.S.Code Cong. & Ad.News 2931.
Similar considerations also motivated Congress' decision to dilute the guarantees afforded multiemployer plan participants:
The purpose of the reduced guarantees is to create an incentive to avoid plan insolvency and mass withdrawal. Current *1058 high guarantees offer no disincentives to the parties to let a plan fall into insolvency, or to abandon the plan. But high guarantees offer illusory protection since they result in potentially intolerable costs for the insurance system. After much consideration, the committee has determined that plan continuation, which alone gives participants the opportunity for benefit improvements and future accruals, should be encouraged. At the same time guarantee levels must be high enough to protect participants against loss of the modest level of benefits provided in the plans in declining industries which have been projected to become insolvent because required contributions will rise to uneconomic levels. The committee believes that the level of guarantees in the bill is appropriate to meet these objectives.
In addition, lower guarantees in the early years of the program reduce the risk of excessive costs which might destroy the program altogether. They provide a trial period during which the prophylactic effects of other aspects of the program can be assessed and the guarantee levels then re-evaluated if adequate benefit security is not provided. Id. at 69, reprinted in id. at 2937.
Plaintiffs concede that the challenged aspects of MPPAA rationally further these legitimate objectives. As a result, their argument rests at bottom on a mere charge that Congress acted in an underinclusive fashion: Since the same logic would have justified an elimination of the thirty percent limit on single employer termination liability, and a reduction of single employer benefit guarantees, it was irrational for Congress to single out multiemployer plans for harsher treatment. The law, however, is clearly to the contrary:
The problem of legislative classification is a perennial one, admitting of no doctrinaire definition. Evils in the same field may be of different dimensions and proportions, requiring different remedies. Or so the legislature may think. Or the reform may take one step at a time, addressing itself to the phase of the problem which seems most acute to the legislative mind. The legislature may select one phase of one field and apply a remedy there, neglecting the others.
Williamson v. Lee Optical Co., 348 U.S. 483, 489, 75 S. Ct. 461, 465, 99 L. Ed. 563 (citations omitted) (emphasis added); accord, e.g., Michael M. v. Superior Court of Sonoma Cty., 450 U.S. 464, 481 n. 13, 101 S. Ct. 1200, 1210 n.13, 67 L. Ed. 2d 437 (1981) (Stewart, J., concurring); Minnesota v. Clover Leaf Creamery Co., 449 U.S. 456, 466, 101 S. Ct. 715, 725, 66 L. Ed. 2d 659 (1981); Fullilove v. Klutznick, 448 U.S. 448, 485-86, 100 S. Ct. 2758, 2778, 65 L. Ed. 2d 902 (1980) (plurality opinion); City of New Orleans v. Dukes, 427 U.S. 297, 303-06, 96 S. Ct. 2513, 2516, 49 L. Ed. 2d 511 (1976) (per curiam); Katzenbach v. Morgan, 384 U.S. 641, 657, 86 S. Ct. 1717, 1727, 16 L. Ed. 2d 828 (1966); Semler v. Dental Examiners, 294 U.S. 608, 610, 55 S. Ct. 570, 571, 79 L. Ed. 1086 (1935); L. Tribe, American Constitutional Law ß 16-4 (1978).
Congress was not obligated to "choose between attacking every aspect of [the] problem or not attacking the problem at all." Dandridge v. Williams, 397 U.S. 471, 487, 90 S. Ct. 1153, 1162, 25 L. Ed. 2d 491 (1970). Congress had a third option Äî to remedy only those evils which it reasonably believed to require immediate attention. Indeed, this is exactly what Congress did. Relying on the evidence found in the PBGC's Report, Congress concluded that the potential exposure of the multiemployer insurance system was unacceptably large. *1059 See, e.g., Education and Labor Report, supra, at 57, reprinted in  U.S.Code Cong. & Ad.News 2925. It adopted measures designed to neutralize this threat. Yet lacking similar data implicating the single employer program, Congress left the rules operative in that sphere unchanged. Such behavior was completely rational.
Plaintiffs next allege that MPPAA is unconstitutionally vague. In assessing the statute, the court is guided by three circumstances. First, the challenged statute is federal rather than state, and as one commentator has observed:
Where a contention of vagueness is advanced with regard to federal legislation, ... the Court may narrowly interpret the act ... rather than void it, and there has been a significant tendency to adopt this narrowing, rather than annihilating course.
Note, The Void-for-Vagueness Doctrine in the Supreme Court, 109 U.Pa.L.Rev. 67, 86 (1960) (hereafter cited as Pennsylvania Note), and cases cited therein. The federal judiciary's ability to interpret federal legislation distinguishes such cases from vagueness challenges to state legislation. See, e.g., Dennis v. United States, 341 U.S. 494, 501-02, 71 S. Ct. 857, 863, 95 L. Ed. 1137 (1951); Pennsylvania Note, supra, at 83 n.80.
Second, the challenge here is to the statute on its face and not as applied to a particular fact situation. The Supreme Court recently had occasion to define the standard to be used in a pre-enforcement facial challenge. See Village of Hoffman Estates v. Flipside, ___ U.S. ___, 102 S. Ct. 1186, 71 L. Ed. 2d 362 (1982), where the Court upheld a municipal "drug paraphernalia" ordinance.
The Court began its analysis by explaining that a "`facial' challenge, in this context, means a claim that the law is `invalid in toto Äî and therefore incapable of any valid application.' Steffel v. Thompson, 415 U.S. 452, 474, 94 S. Ct. 1209, 1223, 39 L. Ed. 2d 505 (1974)." Id. at 1191 n.5. A facial vagueness challenge can thus succeed only if "the complainant ... demonstrate[s] that the law is impermissibly vague in all of its applications." Id. at 1191. This is a heavy burden. The plaintiff must do more than establish that the statutory standard of conduct is imprecise; the plaintiff must show that the statute specifies no standard at all. Id. at 1191 n.7.
Finally, the Flipside Court reasoned that the degree of vagueness tolerated under the constitution varies according to the nature of the challenged statute. Civil, as opposed to criminal, statutes are afforded greater tolerance "because the consequences of imprecision are qualitatively less severe." Id. at 1193. Statutes regulating economic conduct are also given greater deference:
Thus, economic regulation is subject to a less strict vagueness test because its subject-matter is often more narrow, and because businesses, which face economic demands to plan behavior carefully, can be expected to consult relevant legislation in advance of action. Indeed, the regulated enterprise may have the ability to clarify the meaning of the regulation *1060 by its own inquiry, or by resort to an administrative process. Id.
In the present case, plaintiffs challenge on its face a federal civil statute regulating business activity. Accordingly, the constitution tolerates a fair amount of imprecision in the statutory terms, and in weighing such imprecision, this court may properly consider the availability of judicial and administrative interpretation to clarify the statute's language. With this in mind, the court now turns to an examination of the specific terms plaintiffs challenge.
Plaintiffs challenge three sections of MPPAA. The first is the "trucking exemption." See 29 U.S.C.A. ß 1383(d) (Supp. 1981). As we have seen, this provision creates a special withdrawal rule for employers contributing to certain plans, those in which "substantially all of the contributions required under the plan are made by employers primarily engaged in the long and short haul trucking industry, the household goods moving industry, or the public warehousing industry." Id. at ß 1383(d) (2). An employer covered by this special rule who ceases contributions or operations "withdraws" within the meaning of the statute only if the PBGC determines that the cessation causes "substantial damage" to the plan's contribution base or the employer fails to furnish a bond or set up an escrow account equal to half of its withdrawal liability.
Plaintiffs contend that the statutory language fails to give "fair warning" as to when the exemption applies. Specifically, they assert that the terms "substantially all," "primarily engaged" and "substantial damage" are unconstitutionally vague. The short answer to this argument is that, in a case involving federal civil economic regulation, unconstitutional vagueness exists only where the statute provides no standard, rather than merely an imprecise standard. Terms such as "substantially all," "primarily engaged" and "substantial damage" provide courts and administrative agencies some guidance, and the boundaries of such terms may be refined through later interpretation.
In Joseph E. Seagram & Sons v. Hostetter, 384 U.S. 35, 86 S. Ct. 1254, 16 L. Ed. 2d 336 (1966), the Supreme Court considered a vagueness challenge to a statute regulating the price of liquor sold by brand owners and "related persons." The statute defined a "related person" as one whose "exclusive, principal or substantial business" was the sale of liquor purchased from the brand owner. The challengers contended that the term "principal or substantial" was unconstitutionally vague because it permitted the state administering agency to make arbitrary decisions. The Court disagreed, stating:
[W]e think it plain under our decisions that if substantiality is the statutory guide, the limits of administrative action are sufficiently definite or ascertainable so as to survive challenge on the grounds of unconstitutionality.
Id. at 49, 86 S. Ct. at 1263 (quoting Board of Governors v. Agnew, 329 U.S. 441, 449, 67 S. Ct. 411, 415, 91 L. Ed. 408 (1947)). This reasoning applies here.
Clearly, if less than 50 percent of the contributions to a plan come from employers in the covered industries, the exemption does not apply. In such a case, the application of the statute is not at all vague; therefore, the statute is not vague in all of its possible applications.
Moreover, the term "substantially all" need not be construed in a vacuum; at least one sponsor of MPPAA indicated that the phrase means "at least 85 percent." 126 Cong.Rec. H7900 (daily ed. Aug. 26, 1980) (remarks of Rep. Thompson); accord, Ford, McNamara and Stanger, Withdrawal Liability under ERISA After the Multiemployer Pension Plan Amendments Act of 1980, NYU Thirty-Ninth Annual Institute on Federal Taxation (1981 ERISA Supp.), at *1061 10-20 & n.30; but see 126 Cong.Rec. S11671-72 (daily ed. Aug. 26, 1980) (remarks of Sen. Durenberger). Plaintiffs concede that the Local 705 plan does not meet the "85% rule." See Plaintiffs' Brief at 23 n.18. Nevertheless, they argue that they "cannot be completely certain" that the exemption does not apply to them. Id. at 36. Such uncertainty, however, in view of the available opportunities for clarification, does not exceed that tolerated under the constitution in cases such as the present one.
The same reasoning disposes of plaintiffs' argument that the phrases "primarily engaged" and "substantial damage" are impermissibly vague. Again, the terms provide some guidance, which precludes plaintiffs from arguing that the statute is vague in all of its applications.
The second section of MPPAA challenged by plaintiffs is 29 U.S.C.A. ß 1384 (Supp. 1981), the "sale of assets" provision. It provides that a withdrawal may "not occur solely because, as a result of a bona fide, arm's length sale of assets to an unrelated party," the selling employer ceases operations or contributions. See p. 1051, supra. Plaintiffs contend that ß 1384 gives absolutely no guidance as to what constitutes a "bona fide, arm's length sale of assets." This contention is without merit.
Plaintiffs' final challenge is to 29 U.S. C.A. ß 1389 (Supp.1981), the section of MPPAA containing the De Minimis rules. See pp. 1050-1051, supra. These rules do not apply when an employer withdraws in the same year that "substantially all" employers withdraw. 29 U.S.C.A. ß 1389(c) (1) (Supp.1981). It is similarly inapplicable when an employer withdraws pursuant to an agreement by which "substantially all" employers withdraw. Id. at ß 1389(c) (2). Plaintiffs challenge the term "substantially all," but this challenge fails for the same reasons discussed above with reference to the trucking exemption.
In sum, plaintiffs have failed to sustain their burden of showing that the statute provides no guidance and is impermissibly vague in all of its applications. Plaintiffs' motion for summary judgment on the ground of vagueness is denied.
IX. RIGHT TO A JURY TRIAL
MPPAA provides that disputes between an employer and the plan sponsor concerning withdrawal liability shall be resolved through arbitration. 29 U.S.C.A. ß 1401 (Supp.1981). Plaintiffs assert that this mandatory arbitration requirement violates the seventh amendment right to a jury trial. They seek to distinguish the leading case in this area, Atlas Roofing Co. v. OSHA, 430 U.S. 442, 97 S. Ct. 1261, 51 L. Ed. 2d 464 (1977), by arguing that MPPAA did not create a new "public rights."
In Atlas Roofing, the Supreme Court held that the seventh amendment did not prevent Congress from assigning to an administrative agency the task of adjudicating violations of the Occupational Safety and Health Act. The court reasoned that such congressional delegation is permissible in cases involving "public rights":
Our prior cases support administrative fact-finding in only those situations involving "public rights," e.g., where the Government is involved in its sovereign capacity under an otherwise valid statute creating enforceable public rights. Wholly private tort, contract, and property cases, as well as a vast range of other cases as well [sic] are not at all implicated.
Id. at 458, 97 S. Ct. at 1270. Plaintiffs argue that MPPAA does not create "public rights" because the liabilities that it imposes are enforceable not by the Government but by the trustees of the private multiemployer plans.
Plaintiffs give too narrow an interpretation to the term "public rights." In N.L. R.B. v. Jones & Laughlin Steel Corp., 301 U.S. 1, 57 S. Ct. 615, 81 L. Ed. 893 (1937), the *1062 Supreme Court held that delegation of the fact finding function to an administrative tribunal under the National Labor Relations Act did not violate the seventh amendment, even though the subject matter of the dispute to be adjudicated was the discharge of certain employees allegedly for union activity Äî arguably a private dispute Äî and even though the administrative proceedings were initiated by the union Äî a private as opposed to governmental entity. In finding no right to a jury trial in that case, the Supreme Court reasoned:
The instant case is not a suit at common law or in the nature of such a suit. The proceeding is one unknown to the common law. It is a statutory proceeding. Reinstatement of the employee and payment for time lost are requirements imposed for violation of the statute and are remedies appropriate to its enforcement. The contention under the Seventh Amendment is without merit. Id. at 48-49, 57 S. Ct. at 629.
In Crowell v. Benson, 285 U.S. 22, 52 S. Ct. 285, 76 L. Ed. 598 (1932), cited in Atlas Roofing with reference to the private-public dichotomy, the Court stated:
[T]he distinction is at once apparent between cases of private rights and those which arise between the Government and persons subject to its authority in connection with the performance of the constitutional functions of the executive or legislative departments.... Familiar illustrations of administrative agencies created for the determination of [public rights] are found in connection with the exercise of the congressional power as to interstate and foreign commerce, taxation, immigration, the public lands, public health, the facilities of the post office, pensions and payments to veterans.
The present case [a seaman's injury case] does not fall within the categories just described but is one of private right, that is, of the liability of one individual to another under the law as defined." Id. at 50-51, 52 S. Ct. at 292.
In the instant case, the liability at issue is noncontractual; it is imposed by law, and though the law obligates the trustees to initiate enforcement of the liability, the benefits do not run from one private individual to another. The liability imposed incurs in the first instance to the numerous beneficiaries of the plan and ultimately to the fiscal integrity of the PBGC, a government corporation. Given these circumstances, the court concludes that MPPAA seeks to protect public interests and therefore the seventh amendment is inapplicable. Cf. Republic Industries, Inc. v. Central Pennsylvania Teamsters Pension Fund, No. 82-0146, slip op. at 21 (E.D.Pa. March 22, 1982) (seventh amendment extends only to suits for claims that existed at common law and initial resort to nonjudicial form for adjudication of statutory rights under MPPAA does not offend the amendment).
For the reasons stated in this opinion, plaintiffs' motion for summary judgment is denied, and the PBGC's cross-motion for summary judgment is granted in its entirety. The case is dismissed.NOTES
 This is not to say the federal regulation was nonexistent prior to ERISA. See pp. 1044-1045, infra.
 Actually, these guarantees were made effective as of June 30, 1974, a date prior to ERISA's enactment. 29 U.S.C. ß 1381(b) (1976).
 Employers were in no event liable to the PBGC on account of terminations occurring prior to the date of ERISA's enactment. 29 U.S.C. ß 1381(b) (1976).
 The House repassed its version 363-0 on August 25, 1980. 126 Cong.Rec. H7909 (daily ed.). The Senate approved a slightly different bill the following day. Id. at S11676 (daily ed.). A conference followed, with the Senate and House agreeing to the conference report on September 18 and September 19 respectively. Id. at S12901 (daily ed.); id. at H9180 (daily ed.) (vote of 324-1).
 A proposal to extend the PBGC's discretionary authority a fifth time failed on August 1. 126 Cong.Rec. H6984 (daily ed.). The 1974 mandatory program was thus ultimately in effect for less than two months, from August 1, 1980 to September 26, 1980, the date President Carter approved MPPAA.
 MPPAA also provides for liability in certain cases of "partial" withdrawal. See 29 U.S.C. ßß 1381, 1385. These provisions are not at issue here.
 An employee's right to collect a pension benefit is "vested" when it survives a pre-retirement termination of employment. Nachman v. Pension Benefit Guaranty Corporation, 446 U.S. 359, 363-64, 100 S. Ct. 1723, 1727, 64 L. Ed. 2d 354 (1980).
 The Hansen Affidavit was submitted by the A. S. Hansen Company, an actuarial consulting firm.
 In general, a complete withdrawal occurs when an employer
(1) permanently ceases to have an obligation to contribute under the plan, or
(2) permanently ceases all covered operations under the plan.
29 U.S.C.A. ß 1383(a) (Supp.1981).
 Trustees that fail to collect these payments are subject to suit for breach of fiduciary duty. 29 U.S.C. ß 1105 (1976).
 This is a gross oversimplification. The full flavor of section 1391(b) can be obtained only by studying it.
 ITA was formerly known as the Central Motor Freight Association.
 Because of my rejection of the plaintiffs' claims, I need not address the issues raised by the Secretaries in their separate motion to dismiss.
 Plaintiffs press this argument with respect to MPPAA's treatment of withdrawals occurring after enactment as well as those occurring in the "retrospective" period of April 29, 1980 to September 26, 1980. See p. 1034 supra. With respect to the latter category of withdrawals, the trustees additionally argue that MPPAA constitutes a forbidden ex post facto law since it penalizes them for failing to collect a liability that is fixed by events preceding enactment. See U.S.Const. Art. 1, ß 9.
 In the complaint, the Union makes the additional argument that MPPAA "interfere[s] with rights granted by the National Labor Relations Act." There was little, if any, discussion of this point in the briefing, and I assume that it has been dropped. In any event, a federal statute such as MPPAA is not void simply because it disrupts previous rights also conferred by statute. What Congress gives, Congress can take.
 Transport Motor Express, Inc.; E. W. Bohren Transport, Inc.; Essex Group, Inc.; Johnson Motor Lines, Inc.; and Republic Industries, Inc.
 See Hansen Aff. at 34.
 One withdrew between April 29, 1980 and September 26, 1980. The two others withdrew after enactment.
 The existence of such member-employers has been verified by affidavits filed under seal. Supplemental Exhibits 1-3.
 Similarly, the Union does not suffer injury simply because it is a party to a collective contract that has been abrogated in part by MPPAA. At least in the short run, the only entities that are hurt by the elimination of the employer liability disclaimer are the employers and the trustees. While in the long run this might not be true, this contingency remains as speculative as before.
 At oral argument counsel for the PBGC stated that it was "very likely" that the PBGC would take action if needed. Transcript ("Tr.") at 59. This case is not one in which the threat of enforcement is insubstantial.
 The PBGC has already promulgated certain rules. See 47 Fed.Reg. 12622 (1982) (to be codified in 29 C.F.R. Part 2645).
 See Hunt v. Washington State Apple Advertising Commission, 432 U.S. 333, 343, 97 S. Ct. 2434, 2441, 53 L. Ed. 2d 383 (1977); p. 1036 & n. 19 supra.
 None of the "prudential" limitations on standing has relevance here. See generally Valley Forge v. Americans United, supra, 102 S. Ct. at 759-60.
 Until the PBGC makes such a determination, the employer must post a bond or place in escrow an amount equal to fifty percent of its potential withdrawal liability. 29 U.S.C.A. ß 1383(d) (3) (B) (ii). If it does not, a "withdrawal" within the meaning of the statute occurs. The bond is cancelled or the escrow refunded if the necessary PBGC finding is made. Id. at ß 1383(d) (5) (A).
On the other hand, if the PBGC determines within 60 months of the date the employer ceases to contribute that "substantial damage" has in fact occurred, the full withdrawal liability becomes due and owing as of the initial date. Id. at ß 1383(d) (4). If no determination is made either way within this period, the bond is once again cancelled or the escrow refunded. Id. at ß 1383(d) (5) (B).
 TMX argues that the present suit is unripe for this very reason. However, the Hansen Affidavit contains much pertinent information. I will draw upon this data from time to time in order to illustrate various general points.
 One other jurisdictional objection has been asserted, this time by the PBGC. The PBGC believes that no employer can sue if it has actually withdrawn, since it must instead pursue arbitration. However, no employer can demand arbitration unless the trustees first demand payment. 29 U.S.C.A. ßß 1399(b), 1401(a) (Supp.1981). As no demands have here been made, it is hard to understand how the employers could pursue their arbitration "remedy." Compare R. A. Gray & Co. v. Oregon-Washington Carpenters-Employers Pension Trust Fund, et al., No. 81-912 (D.Ore. Dec. 1, 1980) in which an employer was ordered to arbitrate only after it had been served with a demand for payment by the trustees.
 Because I believe that Congress could constitutionally impose this liability upon employers, I think it follows a fortiori that it could impose upon the trustees the subsidiary duties of calculation and collection. I will not discuss their separate claims any further.
 Several other courts have followed Nachman in upholding the single employer provisions of ERISA. See Concord Control, Inc. v. International Union, 647 F.2d 701 (6th Cir.), cert. denied, ___ U.S. ___, 102 S. Ct. 599, 70 L. Ed. 2d 590 (1981); A-T-O, Inc. v. Pension Benefit Guaranty Corp., 634 F.2d 1013 (6th Cir. 1980); Pension Benefit Guaranty Corp. v. Ouimet Corp., 470 F. Supp. 945 (D.Mass.1979), aff'd, 630 F.2d 4 (1st Cir. 1980), cert. denied, 450 U.S. 914, 101 S. Ct. 1356, 67 L. Ed. 2d 339 (1981); Lear Siegler, Inc. v. Pension Benefit Guaranty Corp., 238 Pens.Rep. D-3 (E.D.Mich. 1979).
 MPPAA does not, however, deprive employers of any rights in specific tangible property. The "takings" analysis set forth recently in Matter of Gifford, 669 F.2d 468 (7th Cir. 1982), is hence inapposite. See also In re Ashe, 669 F.2d 105 (3rd Cir. 1982).
 Besides the authorities cited by Judge Sprecher in Nachman, a comparison of two decisions written by Justice Holmes further supports the notion that federal and state laws should be tested against an identical standard when they impair private contracts. Marcus Brown Co. v. Feldman, 256 U.S. 170, 41 S. Ct. 465, 65 L. Ed. 877 (1921); Block v. Hirsch, 256 U.S. 135, 41 S. Ct. 458, 65 L. Ed. 865 (1921). Contrary inferences, however, do appear in the cases. See, e.g., Pension Benefit Guaranty Corp. v. Ouimet Corp., 470 F. Supp. 945, 956 (D.Mass.1979); see also Note, ERISA's Title IV and the Multiemployer Pension Plan, 1979 Duke L.J. 644, 663-65 (hereinafter cited as Duke Note).
There seems to be more agreement that the same standards govern when the law in question impairs a contract made by the legislating body itself, rather than two private parties. See Rivera v. Patimo, 524 F. Supp. 136, 143 (N.D.Cal.1981); cf. L. Tribe, American Constitutional Law ß 9-5, at 465 n.1 (making the broad assertion that "the due process clause of the fifth amendment has essentially the same effect" as the contract clause, but citing in support only cases dealing with congressional attempts to abrogate government contracts); but cf. United States Trust Co. of New York v. New Jersey, 431 U.S. 1, 17 n.13, 97 S. Ct. 1505, 1515 n.13, 52 L. Ed. 2d 92 (1977).
 This discussion includes an examination of prior federal regulation of multiemployer plans, the second factor identified in Nachman.
 The desire to protect employee reliance interests clearly motivated Congress. Other provisions of MPPAA further this goal as well. See generally pp. 1051-1052, infra.
 All sides agree that the PBGC's Report was the predicate for MPPAA.
 At oral argument counsel made the additional charge that the PBGC's computer model was biased since it "was based upon certain premises and the premises were that ERISA was in place, ERISA as we knew it in '74 [with] all the problems that were attendant with ERISA." Tr. at 9. This statement was incorrect to the extent counsel meant to imply that the PBGC's model incorporated ERISA's incentives favoring plan termination. See pp. 1031-1032, supra.
Plaintiffs also argue that the PBGC's 1978 report should be discounted because "as the act began to take shape with new funding schedules and the concept of reorganization, the PBGC's January 2, 1980 report to Congress dramatically scaled down [the] estimates of multiemployer plans which might fail to pay promised benefits." Plaintiffs' Brief at 21. The claim seems to be that the 1980 Report establishes that withdrawal liability is unnecessary, that the non-withdrawal liability provisions of MPPAA independently protect employee reliance interests to a sufficient degree. See Tr. at 9-10. This is incorrect. The PBGC's relatively low 1980 estimates were based on the assumption that all of MPPAA's provisions were in force, including those calling for withdrawal liability. See Education and Labor Report, supra, at 219, reprinted in  U.S. Code Cong. & Ad.News 2985.
 In Allied Steel, the Supreme Court did not defer to the judgments of the Minnesota legislature simply because they were insufficiently articulated. Allied Steel, supra, 438 U.S. at 247-48, 98 S. Ct. at 2723. The decision thus affords "no reason to believe that the Court will not be deferential to legislative policy judgments when the grounds for such deference appear on the record." The Supreme Court, 1977 Term, 92 Harv.L.Rev. 57, 97 (1978).
 Similar reasoning appears in Central States v. Alco Express Company, 522 F. Supp. 919 (E.D.Mich.1981). At issue there was a provision of MPPAA which affords prevailing trustees an absolute right to attorneys fees in suits brought to recover delinquent plan contributions. The court held that it was constitutional to apply this provision to suits already pending on the date MPPAA was enacted. The court rejected the argument that this was unfair to the defendant since it could not have possibly known at the time it failed to contribute that its acts would inevitably lead to this added liability:
[The] Court already had discretion to make such an award under ERISA. Consequently, Alco was certainly on notice that it might be subject to an attorney fee award should it not prevail.
Id. at 930. Central States was recently followed in San Pedro Fishermen's Welfare v. DiBernardo, 664 F.2d 1344, 1346 (9th Cir. 1982).
 Arguably the analysis should disregard both the creation of mandatory multiemployer insurance and the decisions upholding ERISA. The reason would be that each of these events (with the exception of the Seventh Circuit's decision in Nachman) occurred after February 27, 1979, and from this date forward, a pending version of MPPAA threatened to impose full liability upon all subsequent withdrawals. See p. 1053, infra. No employer could thus have reacted to these later incidents by withdrawing in the certainty that MPPAA would not apply. After this date, employers may have remained in their plans as much out of a fear of MPPAA as out of a continuing belief in the validity of their contractual protection. See generally p. 1053, infra.
The gist of this argument is that employers "really" relied upon their contracts only prior to February 27, 1979, and that the reasonableness of this reliance should be tested as of then, when it was actually utilized. Several responses can be made. First, the flipside of this argument is that no contract clause claim can be made at all on behalf of those employers that entered a plan after February 27, 1979. By definition, these employers never contributed at a time when they could have "really" relied upon their contracts. When they joined their plans, they "took with notice" and acted at their peril in the clearest sense.
Second, it is not at all obvious that the relevant date for assessing MPPAA is the date the employer's expectations were utilized. An alternative hypothesis seems equally, if not more, plausible: The relevant date should be the date of enactment since the reasonableness of a legislative impairment should be judged solely with reference to the strength remaining in the impaired interest on the date the legislature acts.
Finally, even if the post-February 27, 1979, events are irrelevant, the ultimate conclusion Äî that the scope of prior federal regulation is greater here than in Nachman Äî remains unchanged. The enactment of ERISA alone provides sufficient grounds for holding that the Nachman logic applies a fortiori in the MPPAA context.
 This discussion includes an analysis of MPPAA's moderating features, the fourth and final factor deemed relevant in Nachman.
 A plan's contribution base is the aggregate number of units upon which contributions are assessed. For example, if under the terms of a collective bargaining agreement, all employers contribute $2.00 per employee hour worked, the contribution base is the sum total of all hours worked by all employees in all covered firms.
 Congress was most concerned about withdrawals that occur in conjunction with a general economic decline in the industry covered by the plan. See, e.g., Education and Labor Report, supra, at 55, reprinted in  U.S.Code Cong. & Ad.News 2923. However, Congress also believed that prophylactic measures were needed:
[H]ealthy multiemployer plans and the participants in those plans ... need the provisions in this bill that will ensure that they continue to accumulate sufficient assets to pay benefits.
126 Cong.Rec. H4162 (daily ed. May 22, 1980) (remarks of Rep. Thompson). Counsel illustrated this point at oral argument with reference to the plaintiffs' Fund. He conceded that the Fund is at present in sound financial shape. Yet, drawing upon the information found in the Hansen affidavit, he argued:
Now ... in a decade you have moved from a ratio [of active participants to retirees and terminated vested participants] of 5.97 to 2.12 and I would only suggest that if unfortunately that were to continue another 10 or 20 years, this plan could be in deep trouble....
The purpose or one of the purposes of the withdrawal liability besides discouraging withdrawals is [to insure] that pension plans will not get in the kind of trouble that Congress was having to deal with. It was a preventive measure. It wasn't just remedial. Tr. at 41.
 To the extent that vested rights are worth more than the maximum amount guaranteed by MPPAA, employee interests are threatened even if PBGC funds are actually paid. Employee interests are in any event implicated if withdrawals prevent the granting of benefit increases.
 The need to keep premiums low can itself justify some form of employer liability. A-T-O, Inc., v. Pension Benefit Guaranty Corp., supra, 634 F.2d at 1025; Nachman, supra, 592 F.2d at 962-63.
 Plaintiffs evidently disagree with much of Congress' economic reasoning. They believe that withdrawal liability will exacerbate rather than cure the problems Congress identified: "[T]he additional employer financial obligations resulting under the 1980 Act may discourage potential new employers from joining the plan." Hansen Aff. at 30. However, as plaintiffs themselves recognize, economic legislation does not violate due process simply because it is unwise. Courts do not sit as "super legislatures" in these matters. See, e.g., Exxon v. Governor of Maryland, 437 U.S. 117, 124, 98 S. Ct. 2207, 2213, 57 L. Ed. 2d 91 (1979). Plaintiffs' arguments have therefore been construed as attacks on the fairness, rather than the wisdom, of MPPAA.
 This problem can be exacerbated when the employees of a withdrawn employer are subsequently rehired by a firm that continues to contribute to the plan. Under these circumstances, the plan obtains a windfall. It receives both withdrawal liability payments and contributions made on behalf of the rehired employees, though it would have received only the latter had there been no withdrawal. What makes this significant is that MPPAA does not obligate the trustees to escrow or in any way set aside the withdrawal liability payments for the purpose of reducing the plan's unfunded vested liability. Conceivably, the trustees could instead use the money to raise benefit levels even further. This in fact is what plaintiffs claim is likely to occur given the trustees' fiduciary responsibilities. By paying its withdrawal liability, one firm may thus set in motion a chain of events ultimately leading to an increased level of unfunded vested liability.
 Allied's liability was based on the benefit levels it itself had set. Cf. Allied Steel, supra, 438 U.S. at 237, 98 S. Ct. at 2718 ("The company ... retained a virtually unrestricted right to amend the plan in whole or in part....") The same was true of Nachman's liability.
 The plaintiff trustees are apparently of this view. See p. 1047 n.44, supra.
 Employer-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 trustees of employee benefit funds often exercise broad discretion on policy matters with respect to which management and labor representatives may reasonably have different views.
N.L.R.B. v. Amax Coal Co., 453 U.S. 322, 343-44, 101 S. Ct. 2789, 2801, 69 L. Ed. 2d 672 (Stevens, J., dissenting). The majority holding in Amax is not to the contrary.
 The problem of trustee discretion is also mitigated by MPPAA's higher funding requirements. Because of these provisions, fewer benefits can legally be offered per given amount of income.
 Under one decision, employers may be able to exercise direct control over the extent of their liability even prior to negotiating a new contract. In Borden, Inc. v. United Dairy Workers Pension Program, 517 F. Supp. 1162 (E.D.Mich.1981), Borden sought and obtained a preliminary injunction against a trustee-approved benefit improvement which increased Borden's potential withdrawal liability by $800,000. The court refused to enforce the provision in the contract granting the trustees plenary power over benefit levels, reasoning "that the grant of power to the Fund Committee to set the benefit factor was granted at a time when [Borden's] withdrawal liability was contingent instead of absolute, and that the change in the law thus expands upon the Committee's powers in a manner not contemplated by Borden." Id. at 1165.
The Borden holding obviously supports my ultimate conclusion that MPPAA is constitutional. However, I have doubts as to the decision's ultimate correctness. It rests at bottom on a belief that "this change in the law [MPPAA] was not one which could have been foreseen by the parties when entering into the collective bargaining agreements." Id. at 1166. I have already explained why the extensive nature of pre-MPPAA federal regulation militates against this conclusion. See pp. 1044-1045, supra.
 See generally Central States v. Alco Express Company, 522 F. Supp. 919 (E.D.Mich.1981).
 See p. 1050 n. 56, infra.
 Indeed, by joining a withdrawal liability payment fund ("fund"), a plan can guarantee that no employer ever pays a withdrawal liability that is inflated by a third party contributor's past default. This is because the fund can pay for the employer's "unattributable liability." 29 U.S.C.A. ß 1403(c) (1) (A) (Supp.1981). The employer's "attributable liability" is the "value of vested benefits accrued as a result of service with the employer" less "the value of plan assets attributed to the employer." Id. at ß 1403(c) (3) (A). Its "unattributable liability" is "the excess of withdrawal liability over attributable liability." Id. at ß 1403(c) (3) (B). If the fund pays for the employer's "unattributable liability," it cannot recover this amount from the employer. Id. at ß 1403(d) (1) (A), (e) (1). Payments of "attributable liability" can, however, be recovered. Id. at ß 1403(c) (2). The long and short of all this is that the employer is ultimately liable only for its "attributable liability." The amount it pays is therefore not influenced by whether asset contributions have been made by others.
 By selecting half the trustees, the employers also exercise some control over the plan's investment performance.
 Prior to enactment, for example, even employer-appointed trustees regularly approved benefit improvements that increased unfunded vested liability. See generally, Texas Carpenters Pension Plan, supra, 9 Pens.Rep. at 103; Bay Area Painters Pension Trust Fund, supra, 2 Employee Benefit Cases at 1732.
 However, if the employer defaults on its payments, the entire liability (plus accrued interest) becomes immediately due and owing. 29 U.S.C.A. ß 1399(c) (5) (Supp.1981).
 Plaintiffs further argue that to the extent any amount is deducted from one employer's bill, it simply increases the potential liabilities of the firms that remain. Withdrawal liability payment funds offer a solution to this problem. See 29 U.S.C.A. ß 1403(c) (1) (B) (Supp.1981).
 29 U.S.C.A. ßß 1383(b), (c), (d) (Supp.1981).
 The PBGC has indicated that it will carefully consider all
information establishing industry characteristics which would indicate that withdrawals in the industry do not typically have an adverse effect on the plan's contribution base. Such industry characteristics include the mobility of employees, the intermittent nature of employment, the project-by-project nature of the work, extreme fluctuations in the level of an employer's covered work under the plan, the existence of a consistent pattern of entry and withdrawal by employers, and the local nature of the work performed.
47 Fed.Reg. 12625 (1982) (to be codified in 29 C.F.R. Part 2645).
 Further, but more limited, relief is afforded by 29 U.S.C.A. ß 1398 (Supp.1981) (withdrawals do not occur if an employer changes business form or suspends contributions during a labor dispute). The PBGC's reliance upon the "free look" rule of 29 U.S.C.A. ß 1390 (Supp. 1981) is, however, misplaced. This provision protects only those employers that first "had an obligation to contribute to the plan after September 26, 1980." Id. at ß 1390(a) (1). It therefore does not assist the relevant class of employers, those already contributing to a plan on the date MPPAA was enacted.
 Representative Ashbrook even hinted that Congress had gone too far in tempering the burden of withdrawal liability:
The existence of special exceptions to the employer withdrawal liability rules; for example, for small employers; the construction, entertainment, and trucking industries; and employers selling their businesses, may restrict their applicability to only a small fraction of the contributing employers in any given plan Äî thus diminishing the desired objective of discouraging early employer withdrawals.
126 Cong.Rec. H7906 (daily ed. Aug. 26, 1980).
 It is perhaps also relevant that Congress did not legislate in a vacuum, but rather adopted MPPAA only after consulting with the affected parties for "3 long and torturous years." 126 Cong.Rec. H6933 (daily ed. July 31, 1980) (remarks of Rep. Thompson); see generally Yale Note, supra, at 1638-39. Indeed, it was often remarked upon in the legislative debate that MPPAA enjoyed "extremely broad support" outside of Congress. See, e.g., 126 Cong.Rec. S10098 (daily ed. July 29, 1980) (remarks of Sen. Williams). Senator Javits listed its backers:
Groups which have had a significant impact on and generally support the plan termination insurance amendments include the National Coordinating Committee for Multiemployer Plans, the National Construction Employers Council, the United Food and Commercial Workers International Union, the Western Conference of Teamsters Pension Trust Fund, the United Mine Workers of America, the Food Marketing Institute, the Bituminous Coal Operators Association, and the AFL-CIO.
In addition to the foregoing groups, other organizations which have had an impact on S. 1076 include the American Association of Retired Persons, the Graphic Arts Supplemental Retirement and Disabilities Fund, and the Associated General Contractors of America. The general plan termination insurance principles of the bill are supported by the ERISA Industry Committee, the American Bankers Association, the American Council on Life Insurance, the National Association of Manufacturers, and the U.S. Chamber of Commerce.
Id. at S10100. Too much should not be made of this point. Employer enthusiasm for MPPAA has clearly waned. See, e.g., The Wall Street Journal, March 5, 1982, at 40, col. 1.
 I express no opinion as to whether any plaintiff can successfully challenge MPPAA as applied to a particular situation. See p. 1038, supra; see generally Hodel v. Virginia Surface Mining & Reclamation Association, supra, 452 U.S. at 297 n.40, 101 S. Ct. at 2371 n.40; id. at 306, 101 S. Ct. at 2375 (Powell, J., concurring); Nectow v. Cambridge, 277 U.S. 183, 48 S. Ct. 447, 72 L. Ed. 842 (1928).
 I will refer to this period as the "retrospective period."
 Of course, prior to August 1, 1980, it was also contingent upon the PBGC's deciding to pay guaranteed benefits in the event of such termination. See p. 1030, supra.
 Of course, some of these factors are not relevant in the pre-enactment context. An example would be the possibility of future negotiated limits on trustee behavior.
 All employers presumably relied on this expectation when they decided to join, or remain in, their plans prior to enactment.
 By referring to an employer's belief in the validity of its contract, I am using shorthand. The most any employer could have reasonably believed was that its contract had been modified only to the extent contemplated by ERISA.
 Senator Matsunaga believed that employers had "ample notice" by May 3, 1979, nearly a year before MPPAA became effective. 126 Cong.Rec. S10158 (daily ed. July 29, 1980).
 The evidence grew, of course, as September 26, 1980 approached. Significant milestones were reached on May 22, 1980 and July 29, 1980, the dates of House and Senate passage.
 The Hudson court's willingness to consider legislative activity that occurred subsequent to the start of the retrospective period is perhaps explained by the fact that the taxpayer challenged the Silver Act only as applied to two sales postdating this activity. The language of the opinion, however, was broader: "[T]he period of retroactivity fixed in the act is not unreasonable." United States v. Hudson, supra, 299 U.S. at 501, 57 S. Ct. at 310. This statement seemingly sanctions the entire retrospective period, not simply that part of it which followed the beginning of the formal legislative process.
Moreover, even if legislative behavior is relevant only when it precedes a challenged transaction, the present facts remain more conducive to a finding of constitutionality than those found in Hudson. In that case, both taxed transactions occurred less than a week after the statute had merely been introduced. Here, three Congressional committees approved MPPAA before any retrospective withdrawal took place. Still more significant legislative action preceded many others. See p. 1053, n.70, supra.
 The taxed transaction occurred on July 15, 1976. The statute was signed October 4, 1976.
 Compare Untermyer v. Anderson, 276 U.S. 440, 445-46, 48 S. Ct. 353, 354, 72 L. Ed. 645 (1928): "The taxpayer ... cannot foresee and ought not to be required to guess the outcome of pending measures." This categorical assertion cannot be reconciled with the subsequent approaches of Hudson and Darusmont. See generally Fleming v. Rhodes, 331 U.S. 100, 102 n.3, 67 S. Ct. 1140, 1141 n.3, 91 L. Ed. 1368 (1947).
 The tax cases also indicate that a retrospective law is more likely to be upheld if it merely increases a tax rate rather than creates a whole "new tax." See United States v. Darusmont, supra, 449 U.S. at 298, 300, 101 S. Ct. at 552, 553; Milliken v. United States, 283 U.S. 15, 23, 51 S. Ct. 324, 327, 75 L. Ed. 809 (1931). One could argue that MPPAA is more analogous to the former type of law; it doesn't create withdrawal liability; it simply amends ERISA so that the debt is fixed instead of contingent.
 See, e.g., 126 Cong.Rec. H4170 (daily ed. May 22, 1980) (remarks of Rep. Miller). Under ERISA, the longer an employer remained in its plan, the greater was its risk that its eventual withdrawal would occur within five years of termination.
 See pp. 1046 1047, supra.
 See Plaintiffs' Reply Brief at 14.
 Plaintiffs also cite various comments made late in the legislative debate by Representative Ashbrook to the effect that, despite dire predictions, "the worst has not happened." 126 Cong.Rec. H7906 (daily ed. August 26, 1980); accord, id. at H7865. Congressman Ashbrook was referring to the fact that no plan had yet terminated under the mandatory 1974 guarantees, though they had been in effect for nearly a month. He was not discussing withdrawals from ongoing plans.
 Indeed, one court has expressed a "tentative view" that withdrawal liability cannot constitutionally be imposed upon pre-enactment withdrawals. Shelter Framing Corp. v. Carpenters Pension Trust for Southern California, 9 Pens. Rep. (BNA) 467, 472-76 (C.D.Cal.1982). In reaching this determination, the court appears to have relied heavily on facts peculiar to the parties before it. See id. at 473, 475. The views expressed in Shelter Framing are thus not necessarily inconsistent with my holding that MPPAA survives facial attack.
 This assumes that contract clause principles are relevant in this suit. See pp. 1040-1041, & n. 31, supra. If they are not, the case for Congress is much clearer. See Usery v. Turner Elkhorn Mining Company, 428 U.S. 1, 14-21, 96 S. Ct. 2882, 2891 2895, 49 L. Ed. 2d 752 (1976).
 As before, the failure of the employers' claims mandates dismissal of the trustees' due process contentions as well. See p. 1038, n. 28, supra.
The trustees' ex post facto argument, see p. 1035, n. 14, supra, is also without merit. In the first place, MPPAA is not retrospective with respect to these parties. They are subject to suit for breach of fiduciary duty only if they fail to collect a withdrawal liability in the period following enactment. This fact alone is fatal to their argument. See Weaver v. Graham, 450 U.S. 24, 29, 101 S. Ct. 960, 964, 67 L. Ed. 2d 17 (1981). Moreover, "ex post facto analysis ... is concerned solely with whether a statute assigns more disadvantageous criminal or penal consequences to an act than did the law in place when the act occurred." Id. at 29 n.13, 101 S. Ct. at 964 n. 13. Civil fiduciary liability is not a "criminal or penal consequence." See generally Agustin v. Quern, 611 F.2d 206, 211 (7th Cir. 1979); United States v. Nasser, 476 F.2d 1111, 1117 (7th Cir. 1973). The one decision cited by plaintiffs, Hiss v. Hampton, 338 F. Supp. 1141 (D.D.C.) (1972) (three judge court), establishes no contrary rule. In Hiss, the court merely invalidated a statute which increased the penalties for certain crimes committed before enactment.
 The fifth amendment does not tolerate irrational distinctions since the due process clause embodies an equal protection "component." Bolling v. Sharpe, 347 U.S. 497, 74 S. Ct. 693, 98 L. Ed. 884 (1954).
 The bulk of these decisions involved state legislation and were thus decided under the equal protection clause of the fourteenth amendment. They nevertheless remain pertinent to this case: "[T]he Fifth Amendment imposes on the Federal Government the same standard required of state legislation by the Equal Protection Clause of the Fourteenth Amendment." Schweiker v. Wilson, 450 U.S. 221, 226 n.6, 101 S. Ct. 1074, 1078 n.6, 67 L. Ed. 2d 186 (1981).
 See pp. 1042 1043, supra.
 In any event, plaintiffs' final argument may rest on an erroneous assumption. It is not clear that fewer PBGC funds are actually available to protect multiemployer, as opposed to single employer, benefits. MPPAA authorizes the PBGC to extend loans to ongoing multiemployer plans which have not terminated. See 29 U.S.C.A. ß 1431 (Supp.1981). Ongoing single employer plans are not eligible for such aid.
 Certain language in the Court's opinion suggests that a failure to show statutory vagueness in all circumstances deprives a facial challenger of standing: "A plaintiff who engages in some conduct that is clearly proscribed cannot complain of the vagueness of the law as applied to the conduct of others." Village of Hoffman Estates v. Flipside, supra, 102 S. Ct. at 1191. But, on the whole, the Court seems to consider it as a matter of failing on the merits. See, e.g., id. at 1191 n.7, where the Court quotes from Parker v. Levy, 417 U.S. 733, 756, 94 S. Ct. 2547, 2561, 41 L. Ed. 2d 439 (1974): "One to whose conduct a statute clearly applies may not successfully challenge it for vagueness."
 Much less vagueness is tolerated when first amendment rights are implicated. See, e.g., Smith v. California, 361 U.S. 147, 150-51, 80 S. Ct. 215, 217, 4 L. Ed. 2d 205 (1959).
 The trucking industry rules are described more completely at p. 1037 & n.25, supra.
 The PBGC has promised that "[w]here clarification of congressional language is vital it will be forthcoming." PBGC Brief at 60; see 29 U.S.C.A. ß 1302(b) (3) (Supp.1981).
 See Atlas Roofing Co. v. OSHA, supra, 430 U.S. at 450 n.7, 97 S. Ct. at 1266 n.7.