Summary
affirming arbitration award upholding a fund's use of the Segal Blend rate to calculate withdrawal liability where the fund used different rates in other contexts
Summary of this case from Sofco Erectors, Inc. v. Trs. of Ohio Operating Eng'rs Pension FundOpinion
Civ. No. 17-5076 (KM)(MAH)
07-03-2018
MANHATTAN FORD LINCOLN, INC., Plaintiff, v. UAW LOCAL 259 PENSION FUND, Defendant.
Jennifer Lynn Del Medico, Jones Day, New York, NY, for Plaintiff. William T. Josem, Cleary & Josem, LLP, Philadelphia, PA, for Defendant.
Jennifer Lynn Del Medico, Jones Day, New York, NY, for Plaintiff.
William T. Josem, Cleary & Josem, LLP, Philadelphia, PA, for Defendant.
Kevin McNulty, United States District Judge
The plaintiff, Manhattan Ford Lincoln, Inc. ("Manhattan Ford") brings this action against UAW Local 259 Pension Fund ("Pension Fund") pursuant to the Employee Retirement Income Security Act of 1974 ("ERISA"), 29 U.S.C. § 1001, et seq. as amended by the Multi-Employer Pension Plan Amendment Act of 1980 ("MPPAA"), 29 U.S.C. §§ 1381 - 1461. This case arises from Manhattan Ford's withdrawal from the Pension Fund, a multiemployer pension plan. The Arbitrator upheld the Pension Fund's calculation of about $2.55 million in withdrawal liability. Manhattan Ford now challenges that decision.
The issues here turn largely on amendments contained in the MPPAA. For simplicity, from time to time I will refer to the statute generically as ERISA.
ERISA Section 1002(37)(A) defines a multiemployer plan as a plan "maintained pursuant to one or more collective bargaining agreements" between a union or unions and employers, "to which more than one employer is required to contribute." 29 U.S.C. § 1002(37)(A).
Two essential questions are raised:
(1) As a matter of ERISA law, must a pension plan's actuary use identical actuarial assumptions to calculate the plan's satisfaction of minimum funding requirements and its unfunded vested benefits ("UVB") for withdrawal liability?
2) Assuming the answer to question 1 is "no," did the Arbitrator err in this case when he found that the discount rate applied by the Pension Fund's actuary to determine Manhattan Ford's withdrawal liability, the Segal Blend, did not render the actuarial assumptions "in the aggregate, unreasonable (taking into account the experience of the plan and reasonable expectations)"? See 29 U.S.C. § 1401(a)(3)(B)(i).
Now before this Court are Manhattan Ford's motion for summary judgment and the Pension Fund's cross-motion for summary judgment. For the reasons discussed below, the two questions raised in this case are answered in the negative. Accordingly, Manhattan Ford's motion for summary judgment is denied, and the Pension Fund's cross-motion for summary judgment is granted. I will therefore affirm the Arbitrator's Interim and Final Awards. I. Background
In this Opinion, certain record items will be abbreviated as follows:
"Compl."= Complaint (ECF no. 1)
"W. L. Report"= The Pension Fund's Report on Manhattan Ford's Withdrawal Liability for withdrawal date of December 31, 2014, prepared by the Segal Consulting, submitted as Exh. 7 in support of Manhattan Ford's motion for summary judgment before the Arbitrator and authenticity stipulated in Joint Stip. of Facts ¶ 14 (ECF no. 6, Exh. A to Compl.)
"Interim Op."= The Arbitrator's July 25, 2016 Interim Award Opinion (ECF no. 6, Exh. B to Compl.)
"Final Op."= The Arbitrator's June 14, 2017 Final Award Opinion (ECF no. 6, Exh. C to Compl.)
"Ans."= Answer (ECF no. 17)
"Pl. Br."= Plaintiff Manhattan Ford's Memorandum in Support of its Motion for Summary Judgment (ECF no. 22-1)
"Arb. Hrg. Tr."= Transcript of December 6, 2016 Arbitration Hearing (ECF no. 22-2, Exh. A to Declaration of Jacob ("Yaakov") M. Roth, Esq.)
"Def. Opp."= Defendant Pension Fund's Memorandum in Opposition to Plaintiff Manhattan Ford's Motion for Summary Judgment, and in support of its Cross-Motion for Summary Judgment (ECF no. 23-3)
"ASOP No. 27"= Actuarial Standard of Practice No. 27, Selection of Economic Assumptions for Measuring Pension Obligations, Revised Edition, Adopted by the Actuarial Standards Board in September 2007, updated for deviation language effective May 1, 2011, admitted into evidence at the Arbitration Hearing as Exh. 2 (ECF no. 23-5, Exh. 1)
"Levy Report"- Report of Thomas Levy, considered part of the record at the Arbitration Hearing (ECF no. 23-6, Exh. 2 to Declaration of William T. Josem, Esq.) See (Arb. Hrg. Tr. at 8:19-9:9, 11:14-12:9) (stating that documents submitted with the parties' summary judgment motions would be considered part of the record).
"Levy Dep."= Transcript of Deposition of Thomas Levy, the Pension Fund's actuarial expert, considered part of the record at the Arbitration Hearing (ECF no. 23-7, Exh. 3 to Declaration of William T. Josem, Esq.) See (Arb. Hrg. Tr. at 8:19-9:9, 11:14-12:9) (stating that documents submitted with the parties' summary judgment motions would be considered part of the record).
"Gleave Dep."= Transcript of Deposition of Diane Gleave, the Pension Fund's actuary, admitted into evidence at the Arbitration Hearing as Exh. 4 (ECF no. 23-8, Exh. 4 to Declaration of William T. Josem, Esq.)
"2014 Actuarial Valuation"= Pension Fund Actuarial Valuation and Review as of January 1, 2014 by Segal Consulting, admitted into evidence at the Arbitration Hearing as Exh. 3 (ECF no. 23-9, Exh. 5)
"Joint Stip. of Facts"= Joint Stipulation of Facts, admitted into evidence at the Arbitration Hearing as Exh. 1 and incorporated into the Arbitrator's Final Opinion (ECF no. 23-10, Exh. 6 to Declaration of William T. Josem, Esq.)
"Pl. Reply"= Plaintiff Manhattan Ford's Reply Brief in Further Support of its Motion for Summary Judgment and in Opposition to Pension Fund's Cross-Motion for Summary Judgment (ECF no. 24)
"Def. Reply"= Defendant Pension Fund's Reply Brief in Further Support of its Cross-Motion for Summary Judgment (ECF no. 25)
"Pl. Letter"= Plaintiff Manhattan Ford's Letter regarding N.Y. Times Co. v. Newspaper & Mail Deliverers'-Publishers' Pension Fund , 303 F.Supp.3d 236 (S.D.N.Y. 2018) (ECF no. 26)
"Def. Letter"= Defendant Pension Fund's responsive letter (ECF no. 27)
For purposes of this motion, I consider Plaintiff Manhattan Ford's Statement of Undisputed Material Facts in support of its motion for summary judgment ("PSMF") (ECF no. 22-3), Defendant Pension Fund's Responsive Statement of Undisputed Material Facts ("DRSMF") (ECF no. 23), Pension Fund's Statement of Undisputed Material Facts in support of its cross-motion for summary judgment ("DSMF") (ECF no. 23-1), and Manhattan Ford's Responsive Statement of Undisputed Material Facts ("PRSMF") (ECF no. 24-1) pursuant to Local Rule 56.1, as well as the deposition testimony, arbitration hearing testimony, and documentary evidence. Facts not contested are assumed to be true.
Manhattan Ford was a contributing employer to the Pension Fund, a multiemployer defined benefit pension plan. (DSMF ¶ 1.) As such, it was required to make contributions to fund the Pension Fund. In 2014, Manhattan Ford's contributions to the Pension Fund ceased. (Id. ) This, everyone agrees, constituted a complete withdrawal from the Pension Fund. That withdrawal triggered Manhattan Ford's withdrawal liability-i.e., its obligation to pay in to the Fund to ensure that any unfunded pension liabilities were covered, and that the employers who remained in the plan would not be unfairly burdened.
See 29 U.S.C. § 1002(37)(A) (defining "multiemployer plan"); id. at § 1002(2)(A) (defining "pension plan").
Pension funds, through their actuaries, necessarily make assumptions or predictions. These include estimates of future contributions, investment return, and liabilities, all of which depend on a number of actuarial factors. To simplify a bit, the Plan's estimated future liabilities are reduced to a present value using a percentage discount rate (which is itself an actuarial assumption). The resulting figure is used to determine whether the Plan's assets are sufficient to meet its obligations.
A. The 7.5% funding rate and the Segal Blend withdrawal rate
Diane Gleave of the Segal Company, the Pension Fund's actuary, calculated the minimum funding level of the plan and Manhattan Ford's withdrawal liability using the following discount rates:
The Funding Rate (used to calculate minimum required funding). To calculate the minimum required funding of the Pension Fund, Ms. Gleave used a funding discount rate of 7.5%. (Id. at ¶ 4.) That funding rate was developed by "employing the ‘building block’ method, looking to the asset mix of the [Pension] Fund's investment portfolio and analyzing the likely return for each asset class." (Final Op. at 7. See Gleave Dep. 10:18-11:7, 11:14-:29.) Manhattan Ford does not dispute, and indeed embraces, that 7.5% rate. Based on the 7.5% funding rate, the Segal firm reported that the Pension Fund was "fully funded- i.e. that, ‘assuming experience is consistent’ with its assumptions, ‘the current value of [the Pension Fund's] assets plus future investment earnings and contribution income is projected to exceed benefit payments and administrative expenses." (2014 Actuarial Valuation at 8.) Indeed, the Pension Fund's funded percentage for 2014 was 111.7%. (Final Op. at 7. See 2014 Actuarial Valuation at 7,10.)
Paragraph 13 of the Joint Stipulation of Facts incorporated paragraphs 1 through 16 of the Interim Opinion (as corrected) as stipulated facts. (See Joint Stip. of Facts ¶ 13.) Accordingly, the Arbitrator incorporated those paragraphs into his Final Opinion. (See Final Op. at 6-8.)
See also (Gleave Dep. at 20:23-21:3) (recognizing that the 111.7% indicates that the value of the assets is greater than the value of the liabilities under that measurement); 28:3-:5 (explaining that 111.7% is "the ratio of the actuarial value of assets to the PPA mandated calculation of the Plan's liabilities."). The actuarial value of assets was $87,890,792 and the liability was $78,672,878. (2014 Actuarial Valuation at 10.)
Pursuant to the Pension Protection Act of 2006, the Pension Fund provided employers with an "Annual Funding Notice for Plan Year Beginning January 1, 2014 and Ending December 31, 2014" which advised participants of that percentage. (Id. at Section 3, Exh. E.) The 2014 Actuarial Valuation and Review Report further noted: "[t]he funded percentage is one measure of a plan's funded status. It is not indicative of how well funded a plan may be in the future, especially in the event of plan termination." (Id. at 36. See 2014 Actuarial Valuation at 10 (describing the "PPA '06 Liability and Annual Funding Notice" by stating "[m]easures the present value of accrued benefits using the current participant census and financial data. As defined by the Pension Protection Act of 2006, based on long-term funding investment return assumption of 7.50% and the actuarial value of assets.").)
The Segal Blend (used to calculate withdrawal liability). To calculate the Pension Fund's UVB at the time of Manhattan Ford's withdrawal, Ms. Gleave used a different discount rate, the Segal Blend. (DSMF ¶ 3.) The Segal Blend has been used by the Pension Fund for purposes of calculating withdrawal liability for more than 25 years. (Id. ¶ 7; Joint Stip. of Facts ¶ 11.) It represents a blend of interest rates prescribed by the Pension Benefit Guaranty Corporation ("PBGC"), and the 7.5% funding rate. More specifically, the Segal Blend "values vested benefit liabilities based on: 1) PBGC, or risk-free rates, to the extent that there are assets on hand that are attributable to the withdrawing employer; and 2) long-term funding assumptions used for minimum funding purposes to the extent such assets are not on hand." (DSMF ¶ 10.) In setting the Segal Blend rate, Ms. Gleave considered two sets of liabilities, and then blended the rates as to those liabilities. (Arb. Hrg. Tr. 92:10-:19.) Gleave first valued a portion of the Pension Fund's liabilities using the long-term funding assumption of 7.5% (i.e. , the same rate as the "funding rate" that was used to value the minimum funding). (DSMF ¶¶ 3, 10.) Gleave then valued another portion of the Pension Fund's liabilities using risk-free interest rates published by the PBGC. (Id. ) "Mathematically, the Segal Blend is the equivalent of using an ‘effective’ discount rate that falls somewhere between the PBGC rates and the investment return assumption used for funding purposes." (Final Op. at 8.) Using the Segal Blend rate, Ms. Gleave calculated the present value of vested plan benefits for withdrawal liability at about $117.8 million. (2014 Actuarial Valuation at 10). That figure minus the market value of current assets ($86,105,701) yielded a $31,737,875 figure for UVB. (Id. at 8). Of that UVB total, about $2.55 million was allocated to Manhattan Ford. See infra .
As explained further herein, the actuary calculates the plan's UVB by subtracting the value of the plan's assets from the present value of vested benefits under the plan. 29 U.S.C. § 1393(c). After calculating the plan's UVB, the actuary must determine the withdrawn employer's allocable share of the UVB. See Dan M. McGill et al., Fundamentals of Private Pensions 582 (9th ed. 2010) (stating that "[d]eriving the actuarial present value of future benefit payments ... is referred to as the valuation of the liabilities of the plan ").
See Section III.C, infra , for a more detailed discussion of discount rates and investment-return assumptions in actuarial calculations, and the relationship between them.
Congress created the PBGC, "a wholly owned Government corporation, to administer an insurance program for participants in both single-employer and multiemployer pension plans." Connolly v. Pension Ben. Guar. Corp. , 475 U.S. 211, 214, 106 S.Ct. 1018, 89 L.Ed.2d 166 (1986) (citing 29 U.S.C. § 1302 ). In particular, "in 1974, ERISA created the Pension Benefit Guarantee Corporation (PBGC) to administer and enforce a pension plan termination insurance program, to which contributors to both single-member and multiemployer plans were required to pay insurance premiums." Concrete Pipe & Prod. of Cal., Inc. v. Constr. Laborers Pension Tr. for S. Cal. , 508 U.S. 602, 607, 113 S.Ct. 2264, 124 L.Ed.2d 539 (1993).
See Gleave Dep. 22:19-:21 (stating "[w]e blended the liabilities at 7.5 with the liabilities on another Interest Rate."); 57:19-:22 (explaining that "[t]o the extent that there is an effective Interest Rate, it is developed by blending the liabilities under the two interest rates that are used.").
See Arb. Hrg. Tr. 95:19-:23 (agreeing with the description of PBGC rates as being "essentially a proxy for rates on a portfolio of investment-grade bonds or treasury instruments."); see also Gleave Dep. 49:23 to 50:4 (explaining that the PBGC rates are "promulgated by the PBGC for single-employer plan terminations and for multiemployer mass withdrawals."); id. at 51:13-:16 (recognizing that PBGC rates are "intended to approximate the cost of buying annuities to settle pension obligations."); (Levy Dep. 48:5-:21) (acknowledging that the PBGC rates are "a proxy for annuitization rates in the commercial marketplace."); id. at 49:7-:11 (explaining that "an annuity rate is a transfer of the risk to an insurance company and therefore settling that particular risk and not taking that risk anymore.").
In particular, the Actuarial Valuation Report explained that "[f]or liabilities up to market value of assets," the following PBGC interest rates were used: a 3% rate was used for liabilities up to 20 years, and a 3.31% rate was used for liabilities beyond 20 years. (2014 Actuarial Valuation at 37.) "For liabilities in excess of market value of assets," the funding rate of 7.5% was used. (Id. )
At the arbitration hearing, Ms. Gleave explained how the Segal Blend works:
We look at the market value of assets and we compare that to the present value of vested benefits at the PBGC interest rates. So to the extent that there are assets on hand to cover those liabilities, that's the portion that's valued using the PBGC rates. To the extent that there are not, the unfunded portion of the liabilities are valued using the 7-and-a-half percent.
So, in effect, you're using the market value of assets as the rating between the PBGC interest rates and the long-term funding interest rates.
(Arb. Hrg. Tr. 94:5-:17.)
The Withdrawal Liability Report explains the valuation assumption as follows:
The actuarial assumptions to be used [for withdrawal liability valuations] are the valuation assumptions used for plan funding, except for the investment return rate and the expense charge . To the extent the assets, valued at market, cover the vested benefits, benefits will be valued at an investment return rate consistent with current annuity rates; the portion of the benefit that is not yet funded will be valued on the interest assumptions used for plan funding.
Specifically, the withdrawal liability valuation assumptions and methods are:
1. Investment Return
a) To the extent the present value of vested benefits is matched by the market value of plan assets on hand: interest assumptions prescribed by the Pension Benefit Guaranty Corporation under 29 C.F.R., Part 4044, which are in effect for the applicable withdrawal liability valuation date.
b) The portion of the vested benefits that is not matched by plan assets (at market) will be valued on the interest assumption used for plan funding, as of the applicable withdrawal liability valuation date.
c) The portion of the vested benefits that is matched by assets will be determined by comparing the total present value of benefits - at the PBGC rates - with the total market value of assets; each vested benefit will be treated as covered by assets to the same extent as all other vested benefits ...
(W. L. Report at 2) (emphasis added). The "Plan Funding investment return assumption" is listed as 7½ %. (Id. at 7. See also 2014 Actuarial Valuation at 37-38.)
Vested benefits are "benefits that are currently being paid to retirees and that will be paid in the future to covered employees who have already completed some specified period of service, 29 U.S.C. § 1053." Concrete Pipe , 508 U.S. at 609, 113 S.Ct. 2264.
The market value of the assets for purposes of withdrawal liability was about $86.1 million. The Pension Fund's funded percentage for withdrawal liability as of January 1, 2014 was 73.1%. (See 2014 Actuarial Valuation at 10.) In other words, the assets represented about 73% of the liabilities.
To determine the portion of the Pension Fund's UVB allocable to Manhattan Ford, Ms. Gleave used the "presumptive method" under 29 U.S.C. § 1391(b). (W. L. Report at 1; Joint Stip. of Facts ¶ 10.) That allocation does not seem to be at issue.
B. The Pension Fund's Assessment of $2.55 Million Withdrawal Liability
Based on Manhattan Ford's cessation of contributions in 2014, the Pension Fund found that Manhattan Ford had completely withdrawn from the Fund. On December 10, 2014, the Pension Fund issued a Notice and Assessment of Withdrawal Liability. (Compl. ¶ 25; Ans. ¶ 25.) Two months later, on February 20, 2015, the Pension Fund sent an Amended Notice to Manhattan Ford, assessing a withdrawal liability of $2,553,692 as of the December 31, 2014 withdrawal date. (Compl. ¶ 25, Ans. ¶ 25; DSMF ¶ 2.) Under that amended assessment, the $2,553,692 withdrawal liability was to be paid in quarterly payments of $99,640.50 over eight years, with a final payment of $78,662.04. (DSMF ¶ 2. See W. L. Report at 7.)
All seem to agree that if Gleave had used the 7.5% funding rate (rather than the Segal Blend) to value the Pension Fund's liability, Manhattan Ford's withdrawal liability would have been $0 (rather than $2.55 million). (Final Op. at 7.) On April 13, 2015, Manhattan Ford challenged the revised assessment on just that basis. The Pension Fund, said Manhattan Ford, was required to use the 7.5% funding rate, not the Segal Blend, to compute Manhattan Ford's withdrawal liability. (Compl. ¶ 26; Ans. ¶ 26.) See 29 U.S.C. § 1399(b)(2)(A). On May 26, 2015, the Pension Fund rejected Manhattan Ford's challenge. (Compl. ¶ 27; Ans. ¶ 27.) See 29 U.S.C. § 1399(b)(2)(B).
C. Arbitration
On July 20, 2015, Manhattan Ford timely initiated arbitration proceedings pursuant to 29 U.S.C. § 1401(a)(1). (Compl. ¶ 28; Ans. ¶ 28.)
Before the Arbitrator, Michael D. McDowell, Esq., Manhattan Ford disputed the Pension Fund's computation of withdrawal liability. On July 25, 2016, the Arbitrator issued an Interim Award denying both Manhattan Ford's motion for summary judgment and the Pension Fund's cross-motion for summary judgment. (See Interim Op. at 18; Compl. ¶ 29; Ans. ¶ 29.) The matter went to a hearing on December 6, 2016. At that hearing, the Arbitrator heard testimony from three witnesses: 1) Darren French, Manhattan Ford's actuarial expert; 2) Ms. Gleave, the Pension Plan's actuary from The Segal Company; and 3) Thomas Levy, the Pension Plan's actuarial expert from The Segal Company. See (Arb. Hr. Tr.) The parties also presented some seven exhibits. (See Final Op. at 4-5.) On June 14, 2017, the Arbitrator issued a Final Award, rejecting Manhattan Ford's challenge to the Pension Fund's withdrawal liability determination. (Compl. ¶ 30; Ans. ¶ 30; Final Op.)
Additionally, by agreement of the parties, all of the documents submitted by the parties in support of their summary judgment motions were also deemed admitted without a question of authenticity. (Arb. Hrg. Tr. 8-9.) (See Joint Stip. of Facts ¶ 14) (stating that "[t]he parties stipulate[d] to the authenticity of each of the Exhibits submitted with their respective Motions for Summary Judgment in this case.").)
The parties submitted additional post-hearing briefing in lieu of oral argument. (Arb. Hrg. Tr. 186-87; Final Op. at 5). Copies of the briefing related to the summary judgment motions and post-hearing briefing were not provided to this Court.
D. This Action and the Cross-Motions for Summary Judgment
On July 12, 2017, pursuant to 29 U.S.C. § 1401(b)(2), Manhattan Ford filed the present action, asking this Court to vacate the arbitrator's interim and final awards in their entirety. (Compl. 13.) On September 6, 2017, the Pension Fund filed its Answer, Affirmatives Defenses, and Counterclaim. (ECF no. 17.) On September 12, 2017, Manhattan Ford filed an Answer to the Pension Fund's Counterclaim. (ECF no. 19.)
On October 16, 2017, Manhattan Ford filed a motion for summary judgment requesting that this Court vacate the Pension Fund's assessment of withdrawal liability and "the arbitration award that sustained it." (Pl. Brf. at 25.) On November 20, 2017, the Pension Fund filed its opposition and a cross-motion for summary judgment (ECF nos. 23-2, 23-3) asking this Court to affirm the assessment of withdrawal liability against Manhattan Ford and the arbitration award sustaining it. (Def. Brf. at 22.) On December 11, 2017, Manhattan Ford filed a reply in support of its own motion and its opposition to the Pension Fund's cross-motion. (Pl. Reply.) On December 29, 2017, the Pension Fund filed a reply in support of its cross-motion. (Def. Reply.)
Manhattan Ford's summary judgment motion and the Pension Fund's cross-motion are equivalent to motions to vacate or affirm the Arbitration Award, which is understood to encompass both the Interim and Final Awards.
On March 27, 2018, Manhattan Ford submitted a letter addressing a recent decision from the Southern District of New York, N.Y. Times Co. v. Newspaper & Mail Deliverers'-Publishers' Pension Fund , 303 F.Supp.3d 236 (S.D.N.Y. 2018). (Pl. Letter.) On March 29, 2018, the Pension filed a responding letter. (Def. Letter.) Briefing is complete, and the motions are poised for decision.
That case is now on appeal to the United States Court of Appeals for the Second Circuit. As luck would have it, the plaintiff-employer in N.Y. Times was represented by two of the attorneys who now represent Manhattan Ford here.
II. Legal Standards Governing Review of Arbitration Award
ERISA "provides the procedure for calculating and assessing withdrawal liability." Concrete Pipe , 508 U.S. at 609, 113 S.Ct. 2264. Disputes regarding withdrawal liability from multiemployer pension plans follow a four-step process:
[1] The plan sponsor has the responsibility of determining this withdrawal liability, notifying the employer, and collecting payment. 29 U.S.C. § 1382.
[2] If the employer disputes the amount set, it may ask the plan sponsor to conduct a reasonable review of the computed liability. 29 U.S.C. § 1399(b)(2)(A).
[3] In the event the dispute is unresolved, either party may request arbitration. 29 U.S.C. § 1401(a)(1).
[4] The arbitrator's award, in turn, may be challenged in federal court. 29 U.S.C. § 1401(b)(2).
Galgay v. Beaverbrook Coal Co. , 105 F.3d 137, 138-39 (3d Cir. 1997) ( [bracketed] numbers and line breaks added). See also Steelworkers Pension Tr. by Bosh v. Renco Grp., Inc. , 694 F. App'x 69, 73 (3d Cir. 2017). The dispute now before the Court followed that pattern.
ERISA mandates arbitration for disputes "concerning a determination made under sections 1381 through 1399." 29 U.S.C. § 1401(a)(1).
A. Presumptions and Burdens of Proof in Arbitration
In reviewing an arbitration award, the court must be mindful of two presumptions that govern an arbitration proceeding under ERISA. See 29 U.S.C. §§ 1401(a)(3)(A)-(B).
The first presumption is that any factual determinations made by a plan sponsor under Sections 1381 through 1399 and 1405 are "presumed correct unless the party contesting the determination shows by a preponderance of the evidence that the determination was unreasonable or clearly erroneous." Id. § 1401(a)(3)(A) ; Concrete Pipe , 508 U.S. at 620-21, 113 S.Ct. 2264. "The Supreme Court has interpreted this language to place the burden of persuasion on the employer during arbitration to ‘disprove a challenged factual determination by a preponderance.’ " Bd. of Trustees of Trucking Empls. of N. Jersey Welfare Fund, Inc.-Pension Fund v. Kero Leasing Corp. , 377 F.3d 288, 307 (3d Cir. 2004) (quoting Concrete Pipe , 508 U.S. at 629, 113 S.Ct. 2264 ). The presumption's "purpose is to prevent the employer from ‘forcing the plan sponsor to prove every element involved in making an actuarial determination.’ " Concrete Pipe , 508 U.S. at 628, 113 S.Ct. 2264 (quoting H.R. Rep. No. 96-869, pt. 1, p. 86 (1980) ).
See Concrete Pipe , 508 U.S. at 626, 113 S.Ct. 2264 (noting that "it is indeed entirely sensible to burden the party more likely to have information relevant to the facts about its withdrawal from the Plan with the obligation to demonstrate that facts treated by the Plan as amounting to a withdrawal did not occur as alleged").
The second presumption applies specifically to challenges to determinations of UVB:
In the case of the determination of a plan's unfunded vested benefits for a plan year, the determination is presumed correct unless a party contesting the determination shows by a preponderance of evidence that-
(i) the actuarial assumptions and methods used in the determination were, in the aggregate, unreasonable (taking into account the experience of the plan and reasonable expectations), or
Section 1401(a)(3)(B)(i)'s language regarding the employer's burden at arbitration parallels that of Section 1393. Compare 29 U.S.C. § 1401(a)(3)(B)(i) (requiring employer to prove by a preponderance of the evidence that the actuarial assumptions and methods used to determine a plan's UVB "were, in the aggregate, unreasonable (taking into account the experience of the plan and reasonable expectations"), with id. § 1393(a)(1)(emphasis added) (requiring a plan to determine UVB on the basis of actuarial assumptions and methods "which, in the aggregate, are reasonable (taking into account the experience of the plan and reasonable expectations ) and which, in combination, offer the actuary's best estimate of anticipated experience under the plan").
(ii) the plan's actuary made a significant error in applying the actuarial assumptions or methods.
29 U.S.C. § 1401(a)(3)(B)(i)-(ii).
The essential distinction between the two presumptions is that the first is directed to the Plan itself, whereas the second is directed to "the assumptions and methods used in calculating withdrawal liability!, which] are selected in the first instance not by the trustees, but by the plan actuary." Concrete Pipe , 508 U.S. at 632, 113 S.Ct. 2264.
In addition to applying those two presumptions, an Arbitrator must of course "follow applicable law, as embodied in statutes, regulations, court decisions, interpretations of the agencies charged with the enforcement of ERISA, and other pertinent authorities" in reaching his or her decision. 29 C.F.R. § 4221.5(a)(1).
B. Standard of Review of ERISA Arbitration Awards
Where, as here, an arbitrator has issued a final decision, ERISA provides that a party may bring suit in federal district court to "enforce, vacate, or modify an arbitrator's award." 29 U.S.C. § 1401(b)(2). A district court reviewing such an award must apply distinct standards of review to legal conclusions, factual findings, or mixed questions of law and fact.
See Chicago Truck Drivers, Helpers & Warehouse Workers Union (Indep.) Pension Fund v. CPC Logistics, Inc. , 698 F.3d 346, 349-50 (7th Cir. 2012) (citing Central States, Se. & Sw. Areas Pension Fund v. Midwest Motor Express, Inc. , 181 F.3d 799, 804-05 (7th Cir. 1999) ); Board of Trustees, Sheet Metal Workers' National Pension Fund v. BES Services, Inc. , 469 F.3d 369, 375 (4th Cir. 2006) (stating that ERISA arbitration decisions are subject to "judicial review similar in scope to appellate review of district court decisions").
As to an arbitrator's factual findings, ERISA mandates a deferential standard of review. An arbitrator's findings of fact are presumed correct, and that presumption is "rebuttable only by a clear preponderance of the evidence." 29 U.S.C. § 1401(c). Thus, factual findings will be reversed only if "clearly erroneous." Crown Cork & Seal Co. v. Cent. States Se. & Sw. Areas Pension Fund , 982 F.2d 857, 860 (3d Cir. 1992). "A factual finding is clearly erroneous ‘when although there is evidence to support it, the reviewing court on the entire evidence is left with the definite and firm conviction that a mistake has been committed.’ " See United States v. Murray , 821 F.3d 386, 391 (3d Cir.), cert. denied , ––– U.S. ––––, 137 S.Ct. 244, 196 L.Ed.2d 185 (2016) (quoting United States v. U.S. Gypsum Co. , 333 U.S. 364, 395, 68 S.Ct. 525, 92 L.Ed. 746 (1948) ). "Where there are two permissible views of the evidence, the factfinder's choice between them cannot be clearly erroneous." Anderson v. City of Bessemer , 470 U.S. 564, 574, 105 S.Ct. 1504, 84 L.Ed.2d 518 (1985).
As to an arbitrator's legal conclusions, ERISA does not specify the applicable standard of review. The United States Court of Appeals for the Third Circuit, however, has held that legal conclusions must be reviewed de novo. Crown Cork & Seal Co. , 982 F.2d at 860.
As to mixed questions of fact and law, ERISA does not expressly state a standard of review. Under Third Circuit precedent, however, a district court must apply "a clearly erroneous standard to findings of fact and conduct plenary review of conclusions of law, applying the appropriate standard to each component." Crown Cork & Seal Co. , 982 F.2d at 861 (citing Crown Cork & Seal Co. Inc. v. Central States Se. & Sw. Areas Pension Fund , 881 F.2d 11, 18 n. 19 (3d Cir. 1989) ; In re Sharon Steel Corp. , 871 F.2d 1217, 1222 (3d Cir. 1989) ).
Cf. Concrete Pipe , 508 U.S. at 630, 113 S.Ct. 2264 (describing the date of "complete withdrawal" as a mixed question of fact and law, and explaining that "[t]he relevant facts are about the closure of the Shafter plant (such as the intent of Concrete Pipe with respect to the plant, its expression of that intent, its activities while the plant was not operating, and the circumstances of the plant's reopening), while the question whether these facts amount to a ‘complete withdrawal’ is one of law").
[Mixed questions of law and fact are] questions in which the historical facts are admitted or established, the rule of law is undisputed, and the issue is whether the facts satisfy the statutory standard, or to put it another way, whether the rule of law as applied to the established facts is or is not violated.
Pullman-Standard v. Swint , 456 U.S. 273, 289 n.19, 102 S.Ct. 1781, 72 L.Ed.2d 66 (1982).
The Arbitrator's determination that ERISA and Concrete Pipe do not always require the use of the same discount rate for funding and withdrawal liability calculations presents a pure conclusion of law. I will therefore review that determination de novo.
The Arbitrator's determination that the use of the Segal Blend rate was reasonable, when considered "in the aggregate" with the other actuarial methods and assumptions, presents a mixed question of law and fact. I will review the Arbitrator's interpretations of the law embedded within that determination de novo, but will apply a "clearly erroneous" standard to the Arbitrator's application of that legal standard to reach his findings of fact. See Crown Cork & Seal Co. , 982 F.2d at 861 ; N.Y. Times , 303 F.Supp.3d at 247-48, 255 (in the context of mixed question review, applying clear error standard to the sub-issue of "[t]he Arbitrator's decision that the Segal Blend's use was reasonable in the aggregate").
C. Summary Judgment Standard
The parties' contentions are presented in the form of cross-motions for summary judgment. Federal Rule of Civil Procedure 56(a) provides that the court should grant summary judgment "if the movant shows that there is no genuine dispute as to any material fact and that the movant is entitled to judgment as a matter of law." Fed. R. Civ. P. 56(a) ; see also Anderson v. Liberty Lobby, Inc. , 477 U.S. 242, 248, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986) ; Kreschollek v. S. Stevedoring Co. , 223 F.3d 202, 204 (3d Cir. 2000). In deciding a motion for summary judgment, a court must construe all facts and inferences in the light most favorable to the nonmoving party. See Hayes v. Harvey , 874 F.3d 98, 103 (3d Cir. 2017). The moving party bears the burden of establishing that no genuine issue of material fact remains. See Celotex Corp. v. Catrett , 477 U.S. 317, 322-23, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986). "[W]ith respect to an issue on which the nonmoving party bears the burden of proof ... the burden on the moving party may be discharged by ‘showing’-that is, pointing out to the district court-that there is an absence of evidence supporting the non-moving party's case." Id. at 325, 106 S.Ct. 2548. Once the moving party has met that threshold burden, the non-moving party "must do more than simply show that there is some metaphysical doubt as to the material facts." Matsushita Elec. Indus. Co., Ltd. v. Zenith Radio Corp. , 475 U.S. 574, 586, 106 S.Ct. 1348, 89 L.Ed.2d 538 (1986). The opposing party must present actual evidence that creates a genuine issue as to a material fact for trial. Anderson , 477 U.S. at 248, 106 S.Ct. 2505 ; see also Fed. R. Civ. P. 56(c) (setting forth types of evidence on which nonmoving party must rely to support its assertion that genuine issues of material fact exist). "[U]nsupported allegations ... and pleadings are insufficient to repel summary judgment." Schoch v. First Fid. Bancorp. , 912 F.2d 654, 657 (3d Cir. 1990) ; see also Gleason v. Norwest Mortg., Inc. , 243 F.3d 130, 138 (3d Cir. 2001) ("A nonmoving party has created a genuine issue of material fact if it has provided sufficient evidence to allow a jury to find in its favor at trial."). If the nonmoving party has failed "to make a showing sufficient to establish the existence of an element essential to that party's cases, and on which that party will bear the burden of proof at trial, ... there can be ‘no genuine issue of material fact,’ since a complete failure of proof concerning an essential element of the nonmoving party's case necessarily renders all other facts immaterial." Katz v. Aetna Cas. & Sur. Co. , 972 F.2d 53, 55 (3d Cir. 1992) (quoting Celotex , 477 U.S. at 322-23, 106 S.Ct. 2548 ).
In deciding a motion for summary judgment, the court's role is not to evaluate the evidence and decide the truth of the matter but to determine whether there is a genuine issue for trial. Anderson , 477 U.S. at 249, 106 S.Ct. 2505. Credibility determinations are the province of the fact finder. Big Apple BMW, Inc. v. BMW of N. Am., Inc. , 974 F.2d 1358, 1363 (3d Cir. 1992).
When the parties file cross-motions for summary judgment, the governing standard "does not change." Auto-Owners Ins. Co. v. Stevens & Ricci, Inc. , 835 F.3d 388, 401 (3d Cir. 2016) (citing Appelmans v. City of Phila. , 826 F.2d 214, 216 (3d Cir. 1987) ). The court must consider the motions independently, in accordance with the principles outlined above. Goldwell of N.J., Inc. v. KPSS, Inc. , 622 F.Supp.2d 168, 184 (D.N.J. 2009) ; Williams v. Phila. Housing Auth. , 834 F.Supp. 794, 797 (E.D. Pa. 1993), aff'd , 27 F.3d 560 (3d Cir. 1994). That one of the cross-motions is denied does not necessarily imply that the other must be granted. For each motion, "the court construes the facts and draws inferences in favor of the party against whom the motion under consideration is made" but does not "weigh the evidence or make credibility determination" because "these tasks are left for the fact-Tinder." Pichler v. UNITE , 542 F.3d 380, 386 (3d Cir. 2008) (internal quotation and citations omitted).
III. Withdrawal Liability and the ERISA/MPPAA Scheme
A. The Statutory Rationale
Under ERISA, as amended by the MPPAA, "if an employer withdraws from a multiemployer pension plan, it incurs ‘withdrawal liability’ in the form of ‘a fixed and certain debt to the pension plan.’ " Concrete Pipe , 508 U.S. at 609, 113 S.Ct. 2264 (quoting Pension Ben. Guar. Corp. v. R.A. Gray & Co. , 467 U.S. 717, 725, 104 S.Ct. 2709, 81 L.Ed.2d 601 (1984) ). The MPPAA was enacted "out of a concern that ERISA did not adequately protect multiemployer pension plans from the adverse consequences that result when individual employers terminate their participation or withdraw." SUPERVALU, Inc. v. Bd. of Trustees of Sw. Pa. & W. Md. Area Teamsters & Emp'rs. Pension Fund , 500 F.3d 334, 336 (3d Cir. 2007) (quoting Warner-Lambert Co. v. United Retail & Wholesale Emp's. Teamster Local No. 115 Pension Plan , 791 F.2d 283, 284 (3d Cir. 1986) (internal citation omitted) ). It was " ‘designed to prevent employers from withdrawing from a multiemployer pension plan without paying their share of unfunded, vested benefit liability, thereby threatening the solvency of such plans.’ " Id. (quoting Mfrs. Indus. Relations Ass'n v. E. Akron Casting Co. , 58 F.3d 204, 205-06 (6th Cir. 1995) ). Accordingly, to "insure[ ] that the financial burden will not be shifted to the remaining employers," a withdrawing employer must pay its fair share of the plan's unfunded liability. Id. at 337.
See C & S Wholesale Grocers , 802 F.3d 534, 536 (3d Cir. 2015) (footnote omitted) (recognizing that "[t]he MPPAA was enacted to mitigate the incentives that employers would otherwise have to withdraw from multiemployer pension plans mired in financial difficulty"). See also Bruce Perlin & Ralph L. Landy, "Special Rules for Multiemployer Plans," in ERISA Litigation at 1394 (Jayne E. Zanglein et al. eds., 6th ed. 2017) (footnote omitted) (describing the unique characteristics of multiemployer plans and stating that those characteristics "may lead employers in financially troubled multiemployer plans, such as those suffering from industry-wide declines, to withdraw and attempt to leave the funding problem to the remaining employers and ultimately to participants and the Pension Benefit Guaranty Corporation (PBGC). Congress addressed this problem in the Multiemployer Pension Plan Amendments Act of 1980 (MPPAA). MPPAA, which added subtitle E to Title IV of ERISA, established withdrawal liability, an immediate and noncontingent liability that the employer owes to the plan when it withdraws from the plan").
B. Calculation of Liability for UVB Upon Withdrawal
"When an employer withdraws from a multiemployer pension plan, ... it incurs withdrawal liability corresponding to its pro rata share of the [UVB] due to the pension fund at the time of its withdrawal." Renco Grp., Inc. , 694 F. App'x at 73. When an employer has completely withdrawn from a multiemployer pension plan, the employer "incurs withdrawal liability that corresponds to the value of the benefits in the plan that have vested and are attributable to its employees." C & S Wholesale Grocers , 802 F.3d at 537 (footnote omitted).
As relevant to this case, a "complete withdrawal" from a multiemployer plan occurs when an employer "permanently ceases to have an obligation to contribute under the plan" or "permanently ceases all covered operations under the plan." 29 U.S.C. § 1383(a). A withdrawal may also be partial, but "partial withdrawal" is not implicated by this case. See id. § 1385(a) (defining "partial withdrawal").
In particular, a withdrawing employer is liable to the plan for its allocable share of the plan's UVB, which is " ‘calculated as the difference between the present value of vested benefits and the current value of the plan's assets,’ " as of a specific date. In re Marcal Paper Mills, Inc. , 650 F.3d 311, 316 (3d Cir. 2011) (quoting Gray , 467 U.S. at 725, 104 S.Ct. 2709 ). That means that the plan actuary must first determine the present value of the plan's future liability for vested benefits. 29 U.S.C. § 1393(c)(A). From that present value, the actuary must subtract the actual value of the plan's assets. 29 U.S.C. § 1393(c). The resulting figure is the UVB.
In the context of a single-employer plan, withdrawal liability is owed to the PBGC. In the context of a multiemployer plan, withdrawal liability is owed to the plan. See 29 U.S.C. §§ 1362(b), 1381(a).
That determination of UVB for purposes of withdrawal liability is a forward-looking one. Future investment returns, life expectancies of employees, and so forth, cannot be known for certain. Thus in calculating the plan's UVB, the actuary must choose appropriate actuarial assumptions and methods. The statute requires selection of "actuarial assumptions and methods, which, in the aggregate, are reasonable (taking into account the experience of the plan and reasonable expectations) and which, in combination, offer the actuary's best estimate of anticipated experience under the plan." 29 U.S.C. § 1393(a)(1). Those actuarial assumptions "must cover such matters as mortality of covered employees, likelihood of benefits vesting, and, importantly, future interest rates." Concrete Pipe , 508 U.S. at 610, 113 S.Ct. 2264. See also CPC Logistics , 698 F.3d at 353 ("[E]stimating the interest rate at which the pension fund's assets are likely to grow is required for determining withdrawal liability.").
Actuarial cost method is the process of assigning the cost of the promised benefits (and expenses) to individual plan years as an annual cost. The cost method determines a normal cost component and an accrued liability component.
Actuarial assumptions are the assumptions used by the actuary to estimate the costs of the promised benefits under the cost method. The true cost of the plan is not known until the last participant or retiree dies; in the interim, actuaries use assumptions such as interest, mortality rates, turnovers, disability rates, and so forth to estimate the cost of the plan.
Kathryn J. Kennedy, Pension Funding Reform: It's Time to Get the Rules Right (Part 1) , 108 Tax Notes 907, Appendix A (Aug. 22, 2005). See also id , at Section II ("Actuarial assumptions are used by actuaries to approximate the cost of the promised level of benefits under the plan. In contrast, actuarial cost methods are used by actuaries to allocate those costs over different periods of time, affording employers with discretion over the incidence of pension expenses.").
Subsection (a)(2) provides that, as an alternative, the actuary may use assumptions provided for in PBGC regulations. To date, however, no such regulations have been promulgated.
Here, Manhattan Ford only challenges the Pension Fund actuary's use of the Segal Blend, the discount rate used by the Pension Fund's actuary to determine the present value of plan liabilities for purposes of calculating withdrawal liability. (Joint Stip. of Facts ¶ 12.) It does not challenge the actuary's other assumptions, including assumptions as to administrative expenses, mortality, and retirement rates. (Id. See also Final Op. at 5-6 (incorporating the Joint Stipulation of Facts into the "Factual Allegations" Section of the Opinion).)
Judge Richard Posner (ret.) of the United States Court of Appeals for the Seventh Circuit has defined a plan's UVB in terms of a "shortfall," i.e., "the difference between the present value of the pension fund's assets and the present value of its future obligations to employees covered by the pension plan." It follows that "[i]f the present value of the assets exceeds the present value of the plan's future obligations, there is no shortfall." CPC Logistics , 698 F.3d at 347-48 (citing 29 U.S.C. §§ 1381, 1391 ).
Such actuarial issues, of course, do not arise only at the stage of a withdrawal from the plan. Multiemployer plans must annually value their future liabilities to ensure compliance with minimum funding rules under ERISA and the Internal Revenue Code. See 29 U.S.C. § 1084 ; 26 U.S.C. § 431. To preview, one issue in dispute here is whether the actuary must use the identical actuarial assumptions, particularly the discount rate, for purposes of minimum funding and withdrawal liability.
Once the plan actuary has estimated the plan's UVB, the actuary must calculate the employer's withdrawal liability. "In essence, the withdrawal liability imposes on the withdrawing employer a share of the unfunded vested liability proportional to the employer's share of contributions to the plan during the years of its participation." Concrete Pipe , 508 U.S. at 610, 113 S.Ct. 2264. "The employer's allocable share depends on the value of the plan's assets and benefits, which, in turn, depends on the date the assets and benefits are valued, the actuarial assumptions and methods used to value the assets and benefits, and the withdrawal liability allocation formula chosen by the plan." Employee Benefits Law at 17-4.
"The date of valuation is the last day of the plan year preceding the date of withdrawal." ABA Section of Labor and Employment Law, Employee Benefits Law 17-4 (4th ed. 2017) (footnote omitted). In the case of a complete withdrawal, the relevant date is that of "the cessation of the obligation to contribute or the cessation of covered operations." 29 U.S.C. § 1383(e).
ERISA Section 1391 allows a plan to select one of four allocation formulas. 29 U.S.C. § 1391. "A plan can also develop its own allocation method, subject to PBGC approval." Employee Benefits Law at 17-4 (citing 29 U.S.C. § 1391 ). Here, the Pension Fund used the "presumptive method," which is one of the specified formulas in Section 1391. (See W. L. Report at 1.) This pro rata method allocates liability based on the employer's proportionate share of contributions. Employee Benefits Law at 17-4 to 17-5.
The MPPAA also provides several adjustments to the employer's allocable share of UVB, including a de minimis reduction, 29 U.S.C. § 1389(a), an adjustment for partial withdrawal, id. § 1386, and special forgiveness rules for employers that have sold their assets or are insolvent and liquidating, id. §§ 1405(a) -(b). See id. § 1381(b)(1). Here, the de minimis rule was applied in calculating Manhattan Ford's withdrawal liability but seemingly did not affect the withdrawal liability allocable to Manhattan Ford. (See W. L. Report at 6.)
In any case, however, the employer's withdrawal liability is not eternal. ERISA provides that any remaining balance is forgiven after the employer has made required annual payments for 20 years. See 29 U.S.C. § 1399(c)(1)(B) ("In any case in which the amortization period ... exceeds 20 years, the employer's liability shall be limited to the first 20 annual payments....").
C. The Role of Discount Rates
The parties agree that a key actuarial assumption, whether for withdrawal liability or minimum funding levels, is the interest rate or discount rate. That rate is used to reduce the fund's future obligations to a present dollar value. I therefore review some basic concepts as developed before the arbitrator and in the case law.
The long-term investment-return assumption is the interest rate that the pension plan uses to predict how much its current assets will grow through investments. (See Levy Dep. 7:8-:12) (describing the assumption as an estimate of "the expected long-term return on the invested assets, both present and future invested assets of the fund"). Based on such assumptions, an actuary will apply a rate to discount the projected benefits stream to a present value of plan liabilities as of the valuation date. (See Arb. Hrg. Tr. 80:19-:22, Gleave Dep. 14:22-15:7 (explaining that the terms "discount rate" and "investment-return assumption" are used interchangeably); Levy Dep. 9:9-:25 (explaining that a discount rate is a rate used "to take expected future events and bring them back to a present value" and is "necessary for an actuary to determine the value of plan benefits today that will be paid in the future").) See also Kathryn L. Moore, Understanding Employee Benefits Law (2015) at 36 (explaining that "a dollar promised in the future must be discounted to its ‘present value’ today" and "[t]he present value is the amount that would have to be invested today at a specified interest rate in order to have a specified amount at some specified future date."); id. at n.42 (noting that the discounted present value is calculated by "discounting the future payment by the interest rate over the relevant period.").
"Actuarial cost methods generally divide the present value of future benefits into two parts, the part attributable to the past and the part attributable to the future. The part attributable to the past is the actuarial liability." Fundamentals of Private Pensions 599. (See Arb. Hr. Tr. 26:10-:22 (describing how an actuary measures the present value of a plan's liabilities as a two-step process, and stating that after using assumptions to determine what the pension fund is likely to pay out in the future, an actuary must "discount" those calculations "back to today, and that's what makes it a present value.").)
The actuarial present value of future benefits is the amount that, together with future interest earnings, is expected to be sufficient to pay those benefits. The term actuarial present value connotes that the derivation of such a value involves population decremental factors, salary scales, and other functions, in addition to an interest discount for the time value of money.
Fundamentals of Private Pensions at 599. See 29 U.S.C. § 1002(27) (defining "present value" "with respect to a liability" as "the value adjusted to reflect anticipated events").
At his deposition, Mr. Levy also testified that a component of calculating minimum funding is calculating the present value of vested benefits. "That calculation is done by projecting for all current participants the expected future benefit payments each year into the future and discount[ing] those back to a present value." (Levy Dep. 14:3-:6.) That is typically done by using a discount rate that is equal to the investment-return assumption. (Id. at 14:7-: 11.)
The most important assumption underlying such a discounting calculation is usually the interest rate:
‘The present value of a series of future contingent payments is a function of the rate of investment return or of interest at which the payments are discounted-the higher the interest assumption, the smaller the present value. Pension plan costs and liabilities are extremely sensitive to the interest assumption in the valuation formula because of the long time-lapse between the accrual of a benefit credit and its payment...’
Langbein et al., supra , at 167 (quoting Dan M. McGill, Kyle N. Brown, John J. Haley, & Sylvester J. Scheiber, Fundamentals of Private Pensions 611-12 (8th ed. 2005) ).
It is elementary that a higher interest rate assumption requires a lower present investment to produce the necessary return, and a lower interest rate requires a higher present investment:
In computing an employer's withdrawal liability, a plan sponsor must calculate the present value of the vested benefits that are to be paid out in the future. An interest rate, or rate of return, is applied in order to determine what present amount of investment will yield the future amounts required to satisfy those vested benefits. In other words, the future benefits are ‘discounted’ to present value. The higher the assumed rate of return, then the less present investment is needed to pay out the future benefits. After the present value of the vested benefits is determined, the asset value of the plan is deducted to come up with the amount of unfunded vested benefits, of which a withdrawing employer is assessed a share. Thus, a higher interest rate applied to discount future benefits will result in a lesser amount of withdrawal liability for the employer.
As discussed above, withdrawal liability is based on the employer's proportionate share of the plan's UVB, "calculated as the difference between the present value of vested benefits and the current value of the plan's assets." C & S Wholesale Grocers , 802 F.3d at 537 (quoting Gray , 467 U.S. at 725, 104 S.Ct. 2709 ) (emphasis added). See 29 U.S.C. § 1393(c).
Masters, Mates & Pilots Pension Plan v. USX Corp. , 900 F.2d 727, 733 (4th Cir. 1990). Those principles, as applied to withdrawal liability, mean that "[i]ncreasing the interest rate assumption decreases the employer's withdrawal liability," and vice versa. Bd. of Trustees, Mich. United Food & Commercial Workers Union v. Eberhard Foods, Inc. , 831 F.2d 1258, 1260 (6th Cir. 1987). Because of compounding over time, even "[a] small adjustment in the interest rate assumption can lead to a major change in the withdrawal liability calculation." Id. See Employee Benefits Law at 17-8 (stating that "[a] small change in the interest rate used to discount the plan's future benefit obligations can have a significant impact on the value of vested benefits, especially because the UVB are based on the difference between liabilities and assets.").
See also Eberhard Foods , 831 F.2d at 1259 (stating that the calculation of UVB "requires a determination of the present value of the vested payments as they will come due over time" and in making that determination, "t]he plan actuary must select an appropriate interest rate to apply in making that determination"). See Pl. Br. at 4 (explaining that the "investment-return assumption is alternatively known as the ‘discount rate,’ because it allows for future obligations to be ‘discounted’ to present value").
Manhattan Ford argues that the court must construe withdrawal liability in parallel with minimum funding requirements, which I will touch on briefly here. Discount rates and other actuarial assumptions, of course, do not become relevant only at the stage of withdrawal: "Estimating the growth of the fund's assets is required not only for determining withdrawal liability but also for determining whether employers are contributing to the fund the minimum amount required by ERISA in order to reduce the probability that the [PBGC] may have to make up for the fund's not being able to pay vested benefits ...." CPC Logistics , 698 F.3d at 353. Just as in the case of withdrawal liability, then, "[t]he minimum required contributions to multiemployer defined benefit pension plans are based on long-term [actuarial] assumptions." Fundamentals of Private Pensions 638. To set funding levels, actuaries for multiemployer plans will select "an interest rate that reflects the long-term expectation of investment earnings given the plan's investment structure." Id. at 638-39. Actuaries perform such calculations to ensure that the employers' contributions are sufficient to fund the future promised level of benefits to employees, and to satisfy the minimum funding standards of ERISA and the Internal Revenue Code.
ERISA established minimum funding requirements to ensure that pension plans " ‘will accumulate sufficient assets within a reasonable time to pay benefits to covered employees when they retire.’ " Employee Benefits Law at 2-29 (citing H.R. Rep. No. 93-1280, 2d Sess., at 283 (1974)(Conf. Rep.), reprinted in 1974 U.S.C.C.A.N. 5038) (footnote omitted). "The contributing employers are responsible for the contributions to be made to the plan to satisfy the minimum funding requirements, unless the plan is in critical status and the trustees have adopted a rehabilitation plan." Id. at 16-65 (footnotes omitted).
See also John H. Langbein et al., Pension and Employee Benefit Law 165 (6th ed. 2015) ("A defined benefit plan envisions pension promises that may extend across four or five decades... An employee may begin to accrue benefits thirty or more years before benefit payments begin, which can be fifty or sixty years before the final benefit check is written. The plan sponsor's challenge is to devise a savings program appropriate to liabilities that are so distant and contingent").
IV. Discussion
This case presents two interrelated issues regarding withdrawal liability.
The first issue, which I review de novo, is whether, as a matter of law, ERISA requires actuaries to use identical actuarial assumptions for purposes of calculating minimum funding and withdrawal liability. If so, then the actuary's use of a 7.5% rate for purposes of minimum funding and the lower Segal Blend rate for purposes of withdrawal liability would be erroneous on its face. If not, however, I must address a second issue.
The second issue is whether application of the Segal Blend in this case satisfied ERISA's requirement that the rate reflect the actuary's "best estimate of anticipated experience under the plan." That requires review of the Arbitrator's decision that Manhattan Ford did not meet its burden of proving by a preponderance of the evidence that the actuarial assumptions and methods used in the Pension Fund actuary's determination of UVB "were, in the aggregate, unreasonable (taking into account the experience of the plan and reasonable expectations)." 29 U.S.C. § 1401(a)(3)(B)(i). The second issue has both a legal and a factual component.
I agree with the Arbitrator that an actuary's use of distinct rates to calculate minimum contribution and withdrawal liability is not prohibited as a matter of law. Additionally, after reviewing the Arbitrator's Final Award and the record as a whole, I uphold the Arbitrator's finding that Manhattan Ford failed to discharge its burden of demonstrating that the actuary's selection of the Segal Blend rate for purposes of withdrawal liability was unreasonable.
A. Must the Actuarial Assumptions Underlying Minimum Funding and Withdrawal Liability be Identical?
In the arbitration, Manhattan Ford challenged the Pension Fund actuary's use of the Segal Blend rate to calculate the present value of the Pension Fund's UVB. (Interim Op. at 14.) According to Manhattan Ford, a pension plan, through its actuary, must use the same assumptions for calculating minimum funding requirements and withdrawal liability. (Id. ) The minimum funding requirements had been calculated on the basis of a 7.5% rate. The use of the lower Segal Blend rate solely for the purpose of calculating withdrawal liability, it claimed, was therefore facially impermissible. (Id. ) The Arbitrator rejected Manhattan Ford's arguments and determined that an actuary is not required to use the same assumptions for minimum funding and withdrawal liability calculations. (Id. at 17.)
I conclude that the Arbitrator was correct as to that narrow issue. The use of disparate rates is not precluded as a matter of statutory law.
1. Comparison of statutory language
By statute, actuarial assumptions that are used to determine adequacy of funding and withdrawal liability must satisfy certain requirements. Those requirements are phrased fairly generically in terms of reasonableness and the actuary's best estimate. They are similar, but not identical.
Section 1084(c)(3) addresses actuarial assumptions in the context of minimum funding, and Section 1393(a) addresses actuarial assumptions in the context of withdrawal liability. See 29 U.S.C. §§ 1084(c)(3), 1393(a).
Regarding actuarial assumptions in the context of minimum funding, Section 1084(c)(3) of ERISA, entitled "[a]ctuarial assumptions must be reasonable," provides as follows:
[f]or purposes of this section, all costs, liabilities, rates of interest, and other factors under the plan shall be determined on the basis of actuarial assumptions and methods-
(A) each of which is reasonable (taking into account the experience of the plan and reasonable expectations), and
(B) which, in combination, offer the actuary's best estimate of anticipated experience under the plan.
Id. § 1084(c)(3). See also 26 U.S.C. §§ 412(a)(2)(C), 431(c)(3) (using similar language regarding actuarial assumptions and minimum funding standards for multiemployer plans under the Internal Revenue Code).
Regarding actuarial assumptions in the context of withdrawal liability, Section 1393(a) of ERISA, entitled "[u]se by plan actuary in determining unfunded vested benefits of a plan for computing withdrawal liability of employer," provides as follows:
Withdrawal liability under this part shall be determined by each plan on the basis of-
(1) actuarial assumptions and methods which, in the aggregate, are reasonable (taking into account the experience of the plan and reasonable expectations) and which, in combination, offer the actuary's best estimate of anticipated experience under the plan ....
The statute provides in the alternative for a UVB determination based on "(2) actuarial assumptions and methods set forth in [PBGC]'s regulations for purposes of determining an employer's withdrawal liability." 29 U.S.C. § 1393(a)(2). No such regulations have been promulgated, however.
In making the UVB determination, the actuary may also permissibly "rely on the most recent complete actuarial valuation used for purposes of Section 412 of Title 26," the Section of the Internal Revenue Code governing calculation of a pension plan's compliance with minimum funding requirements. 29 U.S.C. § 1393(b)(1).
Id. § 1393(a)(1).
Manhattan Ford urges that under "ordinary principles of statutory interpretation," Congress's use of the same language in both Sections leads to the conclusion that a plan actuary must use identical assumptions and interest rates in the context of withdrawal liability and minimum funding. (Pl. Br. at 17 (citing Smith v. City of Jackson, Miss. , 544 U.S. 228, 233, 125 S.Ct. 1536, 161 L.Ed.2d 410 (2005) (citing the canon of construction that "identical words used in different parts of the same act are intended to have the same meaning.") ). To be sure, a comparison of the language in the withdrawal liability Section ( § 1393 ) and the minimum funding Section ( § 1084 ) reveals significant commonalities. Specifically, both Sections require actuaries to use assumptions that are "reasonable (taking into account the experience of the plan and reasonable expectations)" and which, "in combination, offer the actuary's best estimate of anticipated experience under the plan." 29 U.S.C. §§ 1393(a)(1), 1084(c)(3). However, there is also a difference: While the withdrawal liability Section requires that the assumptions be reasonable "in the aggregate," § 1393(a)(1), the minimum funding Section requires that "each" assumption be reasonable, § 1084(c)(3)(A).
The minimum funding Section, then, requires actuarial assumptions and methods each of which is reasonable. The withdrawal liability Section differs in requiring that actuarial assumptions and methods be found reasonable in the aggregate . The former standard is tighter, more granular; it invites item-by-item comparison in a way the latter does not. Requiring that all assumptions, taken in the aggregate, be reasonable would seem to grant the actuary (and the Arbitrator) more latitude to craft a solution that is reasonable and fair overall , in light of the Plan's experience and expectations. And that is what the Arbitrator found that the actuary did. By focusing solely on the (admittedly important) discount rate number, to the exclusion of the differing contexts of the two calculations, Manhattan Ford fails to give sufficient scope to that standard.
Manhattan Ford recognizes of course that the statute has been amended. See Section IV.A.3 (discussing the amendment). It argues, however, that at least one aspect of its argument is untouched by the amendment. The reasonableness requirements, it says, are "on top of the mandate that both sets of assumptions must ‘offer the actuary's best estimate of anticipated experience.’ " (Pl. Reply at 13)(emphasis in original). It argues that the actuary's calculation here flunks that discrete test: "Two divergent assumptions may both be ‘reasonable,’ but they cannot both be the same actuary 's ‘best estimate’ of the same thing ." (Id. ) (emphasis in original).
As discussed below in Sections IV.A.4 and IV.B.1, these two statutory sections operate in distinct arenas: one in the context of ongoing funding and the other in the context of withdrawal. To adopt Manhattan Ford's terminology, those distinct contexts imply that these actuarial projections are not necessarily estimates of precisely "the same thing." The context of withdrawal liability differs from that of funding in ways that do justify, or at least may permissibly justify, a different approach. Whether that disparity is justified in a particular case may raise a factual question; for present purposes, however, it is sufficient that the statute does not rule out such a disparity in all cases, as a matter of law.
2. Legislative history
Manhattan Ford also offers arguments that are to some degree anachronistic. One is its citation of the legislative history of Section 1082(c)(3), the predecessor of Section 1084(c)(3), the minimum funding provision.
In 1974, Section 1082(c)(3) addressed the actuarial assumptions to be used in determining whether minimum funding standards had been met. It required the use of assumptions that "in the aggregate, are reasonable," and that reflect the actuary's "best estimate" of anticipated plan experience. 29 U.S.C. § 1082(c)(3) (1974). As a result of the Pension Protection Act of 2006 ("PPA"), Pub. L. No. 109-280, that language was moved to Section 1084(c)(3) and was also substantively amended in that the reasonableness-related language was changed from "in the aggregate" to "each of which." Section 1082(c)(3) still exists, but it no longer addresses actuarial assumptions. See n.41, infra (discussing the relevant legislative history in more detail).
A relevant legislative report states, as to the "best estimate" language, that " ‘[t]he conferees intend that under this provision a single set of actuarial assumptions will be required for all purposes (e.g., for the minimum funding standard, reporting to the Department of Labor and to participants and beneficiaries, financial reporting to stockholders, etc.) ." (Def. Br. at 17) (quoting H.R. Rep. No. 93-1280, at 284-85 (1974), reprinted in 1974 U.S.C.C.A.N. 5038, 5065) (emphasis added) ).
That funding Section of ERISA, however, was enacted in 1974, six years prior to the enactment of the withdrawal liability section of MPPAA in 1980. "Congress enacted the MPPAA in particular because it found that existing legislation ‘did not adequately protect plans from the adverse consequences that resulted when individual employers terminate[d] their participation in, or withdr[e]w from, multiemployer plans.’ " Flying Tiger Line v. Teamsters Pension Tr. Fund of Phil. , 830 F.2d 1241, 1243 (3d Cir. 1987) (footnote omitted) (quoting Gray , 467 U.S. at 722, 104 S.Ct. 2709 ). The legislature, stating in 1974 that actuarial assumptions should be uniform for all purposes, could not have been contemplating a withdrawal liability scheme whose enactment lay years in the future. In short, the Section 1082 funding formulation might just as plausibly be regarded as part of the problem that MPPAA was attempting to remedy.
I note also that the list of "purposes" requiring a single set of actuarial assumptions, though not presented as exclusive, does not include withdrawal liability. See H.R. Rep. No. 93-1280, quoted supra ("the minimum funding standard, reporting to the Department of Labor and to participants and beneficiaries, financial reporting to stockholders, etc."). This, too, may reflect nothing more than Congress's failure to comprehensively address withdrawal liability until the passage of the MPPAA in 1980.
3. Concrete Pipe
For the proposition that actuarial standards must be uniform, Manhattan Ford relies primarily on the 1993 case of Concrete Pipe & Prod. of Cal., Inc. v. Constr. Laborers Pension Tr. for S. Cal. , 508 U.S. 602, 113 S.Ct. 2264. Indeed, Manhattan Ford asserts that in Concrete Pipe , the Supreme Court "squarely" adopted the proposition that actuaries must employ the same actuarial assumptions in calculating withdrawal liability and in calculating minimum funding. Not quite. While this argument has anachronism problems of its own, I also find that the Concrete Pipe case does not go as far as Manhattan Ford would wish.
The anachronism is this. The language that Manhattan Ford has quoted from Concrete Pipe rested on the proposition that ERISA used entirely "identical language" to describe the actuarial assumptions and methods that must be used in the "different contexts" of withdrawal liability and minimum funding:
The statutory requirement (of ‘actuarial assumptions and methods-which, in the aggregate, are reasonable ...’) is not unique to the withdrawal liability context, for the statute employs identical language in 29 U.S.C. § 1082(c)(3) to describe the actuarial assumptions and methods to be used in determining whether a plan has satisfied the minimum funding requirements contained in the statute.
Concrete Pipe , 508 U.S. at 632-33, 113 S.Ct. 2264 (emphasis added). In 1993, when Concrete Pipe was decided, that was true. It was only thirteen years later, in 2006, that the old funding provision was replaced by one which required that "each" actuarial assumption (not the assumptions "in the aggregate") be reasonable. It was only then, in 2006, that the formerly identical language of these provisions first diverged as described in Section IV.A.1, supra . So Concrete Pipe's discussion of the "identical" statutes must now, post-2006, be taken with a grain of salt.
In discussing the actuarial assumptions that must be used for minimum funding purposes, the Court in Concrete Pipe cited ERISA Section 1082(c)(3). Concrete Pipe , 508 U.S. at 632-33, 113 S.Ct. 2264. At the time of the Court's 1993 decision, Section 1082, entitled "minimum funding standards," addressed actuarial assumptions. Subsection (c)(3) provided:
(3) For purposes of this section, all costs, liabilities, rates of interest, and other factors under the plan shall be determined on the basis of actuarial assumptions and methods -
(A) in the case of-
(i) a plan other than a multiemployer plan, each of which is reasonable (taking into account the experience of the plan and reasonable expectations) or which, in the aggregate, result in a total contribution equivalent to that which would be determined if each such assumption and method were reasonable, or
(ii) a multiemployer plan, which, in the aggregate, are reasonable (taking into account the experiences of the plan and reasonable expectations), and
(B) which, in combination, offer the actuary's best estimate of anticipated experience under the plan.
29 U.S.C. § 1082(c)(3) (1993) (emphasis added).
In 2006, however, Section 1082 was amended by the PPA. See C & S Wholesale Grocers , 802 F.3d at 538 (quoting Trs. of the Local 138 Pension Tr. Fund v. F.W. Honerkamp Co., Inc. , 692 F.3d 127, 130 (2d Cir. 2012) ) (explaining that Congress "enacted the PPA ‘to protect and restore multiemployer pension plans in danger of being unable to meet their pension distribution obligations in the near future.").
Effective August 17, 2006, the PPA added Section 1084, "Minimum funding standards for multiemployer plans." Under the PPA, each actuarial assumption must be individually reasonable, thereby replacing the previous requirement that the assumptions be reasonable in the aggregate. Section 1084 states, in relevant part:
(3) Actuarial assumptions must be reasonable
For purposes of this section, all costs, liabilities, rates of interest, and other factors under the plan shall be determined on the basis of actuarial assumptions and methods-
(A) each of which is reasonable (taking into account the experience of the plan and reasonable expectations), and
(B) which, in combination, offer the actuary's best estimate of anticipated experience under the plan.
29 U.S.C. § 1084(c)(3) (2006)(emphasis added). Section 1082 still exists; however, it no longer addresses actuarial assumptions.
The PPA also amended and altered related sections of the Internal Revenue Code. Inter alia , it removed the actuarial assumption-related language from Section 412(c)(3) and placed it in the newly-created Section 431(c)(3). See 26 U.S.C. § 431(c)(3) (2006) (including the same language as ERISA Section 1084(c)(3) ).
More recently, "[o]n December 16, 2014, Congress passed the Multiemployer Pension Reform Act of 2014 ("MPRA"). Pub. L. No. 113-235, Div. O, 128 Stat. 2130, 2773-2822 (amending the PPA, 29 U.S.C. §§ 1084 -1085 and 26 U.S.C. §§ 431 -432, among other things)." C & S Wholesale Grocers , 802 F.3d at 538. Section 1084(c)(3) was not amended.
The main upshot, for our purposes, is that under current law, "each" actuarial assumption must be reasonable for the purpose of minimum funding, whereas they must be reasonable "in the aggregate" for purposes of withdrawal liability.
That is not to say, however, that the case should be disregarded, and I do analyze it here. I conclude, however, that Concrete Pipe does not impose a statutory bar; it leaves open the possibility of separate actuarial assumptions being permissibly applied to funding and withdrawal liability.
In Concrete Pipe , the Court was not considering the issue presented here. Rather it was hearing a due process challenge to the presumption of correctness that an arbitrator must apply to the plan sponsor's calculation of a plan's UVB. 508 U.S. at 631, 113 S.Ct. 2264. Then, as now, a plan sponsor's calculation of a plan's UVB for a plan year was "presumed correct" unless proven to be unreasonable by a preponderance of evidence. 29 U.S.C. § 1401(a)(3)(B)(i). (See pp. 376–78, supra .) The withdrawing employer claimed, however, that the presumption of reasonableness violated its procedural due process rights because the plan trustees were institutionally biased to impose the greatest possible withdrawal liability. The Court, however, upheld the presumption.
Concrete Pipe also discussed the presumption in Section 1401(a)(3)(A), which is not implicated here.
One key to the Court's analysis was the moderating effect of professional actuarial standards. Under the statute, the Court noted, "the assumptions and methods used in calculating withdrawal liability are selected in the first instance not by the trustees, but by the plan actuary." 508 U.S. at 632, 113 S.Ct. 2264.
For a variety of reasons, this actuary is not, like the trustees, vulnerable to suggestions of bias or its appearance. Although plan sponsors employ them, actuaries
are trained professionals subject to regulatory standards. The technical nature of an actuary's assumptions and methods, and the necessity for applying the same assumptions and methods in more than one context, as a practical matter limit the opportunity an actuary might otherwise have to act unfairly toward the withdrawing employer.
Id. (internal citations omitted).
In support of its conclusion, the Court made a second observation, heavily relied on by Manhattan Ford here. Pointing out the (then) "identical language" in ERISA regarding actuarial assumptions for withdrawal liability and minimum funding, the Court found that these matched provisions would "tend[ ]to check the actuary's discretion in each of them":
Using different assumptions [for different purposes] could very well be attacked as presumptively unreasonable both in arbitration and on judicial review.
[This] view that the trustees are required to act in a reasonably consistent manner greatly limits their discretion, because the use of assumptions overly favorable to the fund in one context will tend to have offsetting unfavorable consequences in other contexts. For example, the use of assumptions (such as low interest rates) that would tend to increase the fund's unfunded vested liability for withdrawal liability purposes would also make it more difficult for the plan to meet the minimum funding requirements of § 1082.
508 U.S. at 632-33, 113 S.Ct. 2264 (alterations in original; internal quotation marks omitted)
The Court concluded that actuarial standards and consistency were sufficient to stave off a due process challenge to the statutory presumption:
[The withdrawing employer] has not shown that any method or assumption unique to the calculation of withdrawal liability is so manipulable as to create a significant opportunity for bias to operate, and arguably the most important assumption (in fact, the only actuarial assumption or method that [the employer] attacks in terms ...) is the critical interest rate assumption that must be used for other purposes as well.
Id. at 633, 113 S.Ct. 2264 (footnote omitted). The Court did not lay down a requirement of strict uniformity, or deny that certain actuarial considerations might have particular application to withdrawal liability alone. Indeed, the Court pointed out that its point was "not significantly blunted by the fact that the assumptions used by the Plan in its other calculations may be ‘supplemented by several actuarial assumptions unique to withdrawal liability .’ " 508 U.S. at 633, 113 S.Ct. 2264 (emphasis added).
The Court also discussed the employer's burden of proof under Section 1401(a)(3)(B)(i). I will discuss that aspect of the opinion in Section IV.B.1, infra .
There are, then, two answers to Manhattan Ford's argument that Concrete Pipe held that there was a statutory ban on divergent discount rates.
The more straightforward (if not quite dispositive) answer is that, as pointed out above, the two sections are no longer "identical," as they were in 1993. Intervening amendments have weakened the notion that today, the funding rate must be used for all "other purposes," including withdrawal liability.
The second, more complex answer is that the Supreme Court's comments were general, were not made in relation to the issue now before this Court, and did not in fact rule out non-identical actuarial assumptions. At most, the Court suggested that, under unspecified circumstances, the use of inconsistent assumptions "could very well be attacked as presumptively unreasonable," id. at 633, 113 S.Ct. 2264 -not that it was prohibited per se.
On that point, the sparse interpretive case law is in accord. In CPC Logistics , Judge Posner, writing for the Seventh Circuit, seemingly endorsed the Fund's view of Concrete Pipe , and rejected that of Manhattan Ford:
[ Concrete Pipe ] could be read to suggest that having two different interest-rate assumptions-one for withdrawal liability and one for avoiding the tax penalty-might make a plan vulnerable to claims that either or both were ‘unreasonable’ within the meaning of 29 U.S.C. § 1393(a)(1). The danger was remote; the Court had indicated that "supplemental" assumptions that might cause the rates to diverge were permissible. 508 U.S. at 633, 113 S.Ct. 2264.
698 F.3d at 354-55. "Vulnerability" to a charge of "unreasonableness" is a far cry from "prohibited." And indeed, CPC upheld the arbitrator's disallowance of the plan's "questionable" decision to adopt the funding rate for withdrawal liability purposes, because it appeared that in fact the Segal Blend, not the funding rate, embodied the actuary's "best estimate." 698 F.3d at 356. That, of course, is the opposite of the result sought by Manhattan Ford here.
In N.Y. Times, supra , District Judge Robert W. Sweet of the Southern District of New York acknowledged Concrete Pipe's description of the funding rate assumption as one that "must be used for other purposes as well." 303 F.Supp.3d at 254-55 (citing 508 U.S. at 633, 113 S.Ct. 2264 ). "In the same breath, however," he wrote, "the [ Concrete Pipe ] Court stated that the ‘assumptions used by [a] Plan in its other calculations may be supplemented by several actuarial assumptions unique to withdrawal liability.’ " Id. (quoting 508 U.S. at 633, 113 S.Ct. 2264 ) (emphasis added, internal quotation marks omitted). Thus Judge Sweet, like Judge Posner, rejected the interpretation that Concrete Pipe had imposed a per se ban on using different discount rates for purposes of funding and withdrawal liability:
The expectation is that a standard, uniform interest rate is applied in all contexts, and any deviation ‘could very well be attacked as presumptively unreasonable both in arbitration and on judicial review.’ ... That does not mean, however, that deviation is, at all times, impermissible by law-were that the case, the Court would not have included the open-ended ‘could very well be’ language rather than something more definitive.
Id. (internal citations omitted).
Concrete Pipe , the main authority cited by Manhattan Ford, therefore does not require adoption of the per se rule for which Manhattan Ford advocates. And the case law interpreting Concrete Pipe is in accord.
4. Miscellaneous differences in context
Finally, the differing contexts of funding and withdrawal support the idea that Congress did not mean to categorically rule out any divergence between them as to the actuarial assumptions used.
Both funding and withdrawal require actuarial predictions as to the fund's future; to that extent their concerns are parallel. And to be sure, Manhattan Ford is not wrong to cite the canon of construction that "identical words used in different parts of the same act are intended to have the same meaning." It is perhaps a bit facile to refer to "different parts of the same act" when we are actually talking about ERISA and MPAAA, statutory enactments separated by years, the latter of which was intended to remedy gaps and deficiencies in the former. In any event, however, the presumption may be overcome when "there is such variation in the connection in which the words are used as reasonably to warrant the conclusion that they were employed in different parts of the act with different intent." Official Comm. of Unsecured Creditors of Cybergenics Corp. ex rel. Cybergenics Corp. v. Chinery , 330 F.3d 548, 559 (3d Cir. 2003) (en banc) (quoting Atl. Cleaners & Dyers v. United States , 286 U.S. 427, 433, 52 S.Ct. 607, 76 L.Ed. 1204 (1932) ).
Minimum funding and withdrawal liability are different concepts under ERISA with different, although related, policy concerns. Moreover, one cannot ignore the limiting language in each Section. The withdrawal liability Section is prefaced with the language "[w]ithdrawal liability... under this part," 29 U.S.C. § 1393(a) (emphasis added), while the minimum funding Section is prefaced with the phrase "[f]or purposes of this section," id. § 1084(c)(3) (emphasis added). This suggests that Congress viewed minimum funding and withdrawal liability as distinct calculations warranting the use of different assumptions by actuaries. If total parallelism were mandated, a cross-reference or a common definition would have been a more natural way to draft the statutes.
Furthermore, the Pension Fund correctly points out that in Section 1393(a)(2), the PBGC was given the authority to provide actuarial assumptions to be used in the calculation of withdrawal liability only. Such a delegation would have been unwarranted if Section 1393 wholly precluded the use of assumptions that were not the minimum funding assumptions.
I consider also that funding and withdrawal invoke some disparate concerns, which I discuss further in section IV.B.2, infra . Funding is an ongoing process, subject to adjustment for an employer that is remaining in the plan. The Plan's needs and actuarial projections are reassessed annually, and a participating employer may be required to make additional contributions to make up for any shortfall. Withdrawal liability, however, is calculated once, as of the time of withdrawal. Should the unexpected occur after that employer's departure, the burden may unfairly fall on other plan employers (or ultimately the taxpayer, through PBGC). As discussed in more detail below, a responsible actuary might therefore opt to calculate that withdrawal liability on a more risk-averse or conservative basis. And it must always be remembered, of course, that ERISA is a remedial statute, which must be liberally construed to ensure that workers who were promised pension benefits will actually receive them.
For all of these reasons, I conclude that ERISA imposes no per se ban on the use of different actuarial assumptions for purposes of funding and withdrawal liability. I therefore move to the question of whether the Arbitrator erred in permitting application of the Segal Blend under the circumstances of this case.
B. Did the Arbitrator Err in Upholding the Application of the Segal Blend?
Manhattan Ford argues that the Pension Fund's actuary violated Section 1393(a)(1) of ERISA by not using the 7.5% funding rate, which represented her "best estimate of anticipated experience under the plan," and instead using the Segal Blend, which did not. (Pl. Br. at 9-13.) The Segal Blend, recall, represents a combination of the 7.5% funding rate and the PBGC risk-free rate. The PBGC component of the Segal Blend does not satisfy the applicable standard, says Manhattan Ford, because "the Fund is not principally invested in risk-free assets" and "does not intend to use Manhattan's withdrawal liability payments to buy such assets." (Pl. Br. at 11.) Rather, an average annual investment return of 7.5% would be the "best estimate of anticipated experience under the plan." (Id. at 12.) As a result, says Manhattan Ford, the actuarial assumptions used in the withdrawal liability determination were unreasonable "in the aggregate." (Id. at 24.) See 29 U.S.C. § 1401(a)(3)(B)(i).
As noted above, Manhattan Ford does not challenge any actuarial assumption aside from the discount rate. (Joint Stip. of Facts ¶ 12.)
Manhattan Ford seeks to present this issue as a pure question of ERISA law. As I have already held, there is no legal mandate that funding and withdrawal be calculated on identical actuarial assumptions, which disposes of much of the legal issue. What remains is a mixed question of the application of the law to the facts. Because, as I find, the Arbitrator did not err as a matter of law, the issue here is not what another actuary would have done, or whether the court would have favored another approach, but rather whether the Arbitrator committed clear error.
In Section IV.B.1, I find that the Arbitrator did not err in his interpretation of the governing legal standard. In Section IV.B.2, I summarize the evidentiary record, and in Section IV.B.3, I conclude that the Arbitrator did not commit clear error in finding that Manhattan Ford failed to overcome the presumption of correctness of the actuary's best estimate.
1. "Best estimate" as a procedural standard that defers to actuarial expertise
ERISA does not define the phrase "best estimate of anticipated experience under the plan." See 29 U.S.C. § 1393(a)(1) ; § 1084(c)(3)(B). However, case law provides some guidance. Even in the context of minimum funding, "courts have generally accepted the view that the ‘best estimate’ standard is procedural only and that actuarial assumptions may properly err on the side of conservatism and overfunding." Jay Conison, Employee Benefit Plans in a Nutshell 441 (3d ed. 2003) (citing Citrus Valley Estates, Inc. v. C.I.R. , 49 F.3d 1410 (9th Cir. 1995) ; Rhoades, McKee & Boer v. United States , 43 F.3d 1071 (6th Cir. 1995) ). See also Langbein, et al., supra , at 166 (stating that "[t]he courts have interpreted the ‘best estimate’ requirement (previously contained in IRC § 412(c)(3) ) in a fashion that gives actuaries tremendous leeway in selecting assumptions").
In Citrus Valley Estates , for example, the United States Court of Appeals for the Ninth Circuit considered an appeal from the Tax Court regarding the then-current "best estimate" provision of section 412(c)(3) of the Internal Revenue Code. Following the Second and Fifth Circuits, it rejected the Commissioner of Internal Revenue's argument that "by endorsing the use of conservative actuarial assumptions, the Tax Court effectively read the ‘best estimate’ provision out of section 412(c)(3)." 49 F.3d at 1414. In other words, the Commissioner's position was that "an assumption cannot be an actuary's ‘best estimate’ if it reflects a more conservative view of an anticipated plan experience than the actuary believes is likely." Id.
At the time, Section 412(c)(3) of the Internal Revenue Code required, as a prerequisite for deductibility, that the calculation of the minimum funding contribution
be determined on the basis of actuarial assumptions and methods which, in the aggregate, are reasonable (taking into account the experience of the plan and reasonable expectations) and which, in combination, offer the actuary's best estimate of anticipated experience under the plan.
26 U.S.C. § 412(c)(3). The 2006 PPA removed that Section; however, the PPA added Section 431, which incorporated similar language in subsection (c)(3), and, as stated above, it also added ERISA Section 1084. Notably, the "best estimate of anticipated experience under the plan" provision is still in place. However, the Internal Revenue Code, like ERISA, now requires that for minimum funding, the actuarial assumptions and methods must "each" be reasonable rather than reasonable "in the aggregate."
The Tax Court's review was too intrusive, said the Court of Appeals in Citrus Valley . Actually, the actuary retained great leeway:
This statutory scheme serves the dual but sometimes conflicting goals of guaranteeing adequate plan funding while preventing taxpayer abuse. Within the range of reasonableness, Congress assigned the task of balancing these goals to actuaries. We will not narrow the statutory gap between the Scylla of underfunding and the Charybdis of tax penalties.'... So long as the actuary's funding decisions fall within the range of reasonableness, the substantive provisions of section 412(c)(3) are satisfied.
This means that the ‘best estimate’ provision of section 412(c)(3), properly construed, is essentially procedural in nature... The ‘best estimate’ language is ‘principally designed to insure that the chosen assumptions actually represent the actuary's own judgment rather than the dictates of plan administrators or sponsors.’
Id. (internal citations omitted). It further noted that "[t]he mere fact that the challenged assumptions fell on the conservative end of the acceptable range does not render them invalid as a matter of law. Conservative assumptions result in a higher level of initial plan funding, which helps ensure that IDB plans will be able to deliver the promised retirement benefit when due, clearly one of ERISA's most important goals." Id. at 1414-15.
Similarly, in Vinson & Elkins v. C.I.R. , 7 F.3d 1235, 1238 (5th Cir. 1993), the Fifth Circuit adopted the procedural view of the IRC "best estimate" provision, and rejected the Commissioner's assertion that it imposed an additional substantive hurdle:
The statute refers to the actuary's best estimate, not that of a court or of outside experts. Further, by entrusting actuaries with the task of determining plan contributions, and by granting the latitude inherent in the statutory reasonableness test, Congress intended to give actuaries some leeway and freedom from second-guessing. See H.R. Rep. No. 807, 93d Cong., 2d Sess. 27 (1974) U.S. Code Cong. & Admin. News 4639, 4670, reprinted in 2 Subcomm. on Labor of the Senate Comm. on Labor & Public Welfare, 94th Cong., 2d Sess., Legislative History of the Employee Retirement Income Security Act of 1974 , at 3115, 3147 (Comm. Print 1976) (rejecting imposition of uniform actuarial methods and assumptions): "[A]ny attempt to specify actuarial assumptions and funding methods for pension plans would in effect place these plans in a straitjacket ... and would be likely to result in [unreasonable] cost estimates." Adding a second, more rigorous level of substantive review via the best estimate test would frustrate that goal. Moreover, a second substantive test would render the reasonableness test superfluous.
Id. See Wachtell, Lipton, Rosen & Katz v. C.I.R. , 26 F.3d 291, 296 (2d Cir. 1994) (citing Vinson & Elkins and stating that the "best estimate" requirement is "basically procedural in nature and is principally designed to insure that the chosen assumptions actually represent the actuary's own judgment rather than the dictates of plan administrators or sponsors").Generally, it should not be difficult to establish that the actuary's stated conclusions represent the actual exercise of his or her best judgment. Of course, the case law leaves open the possibility of proof that an actuary succumbed to improper influence. See Huber v. Casablanca Indus., Inc. , 916 F.2d 85, 93 (3d Cir. 1990), abrogated on other grounds by Milwaukee Brewery Workers' Pension Plan v. Joseph Schlitz Brewing Co. , 513 U.S. 414, 115 S.Ct. 981, 130 L.Ed.2d 932 (1995) (noting that "an actuary selected and paid by the trustees" may be biased in favor of the pension fund, and finding that a withdrawal liability calculation was not the actuary's best estimate when there was evidence that the Board of Trustees pressured the actuary to revise his initial withdrawal liability calculation and instead calculate withdrawal liability based on assumptions that were in the best interests of the Fund). Here, however, there is not so much as an allegation, let alone proof, that this Plan's actuary was improperly influenced by the Plan trustees or anyone else.
Minimum funding cases, I recognize, must be applied with care in the separate context of withdrawal liability. On the limited issue of procedural deference to actuaries' professionalism, however, I find those cases persuasive. They are, moreover, consistent with Concrete Pipe's general discussion of the role of the actuary as a buffer between the Plan's natural low-rate bias and the selection of an appropriate rate.
The "best estimate" provision, then, does not embody a substantive standard so much as it commits the issue to the judgment of the actuary. The Arbitrator thus proceeded correctly in his deferential analysis of the question of whether the actuarial assumptions chosen were reasonable in the aggregate. And I think that the deference owed to the actuary extends to an actuary's determination, in a particular case, that a lower rate may be required at the withdrawal stage than is required at the funding stage.
I reiterate that in arbitration stage, Manhattan Ford had the burden of overcoming the presumption of correctness by proving by a preponderance of the evidence that the actuarial assumptions were, "in the aggregate, unreasonable (taking into account the experience of the plan and reasonable expectations)." 29 U.S.C. § 1401(a)(3)(B)(i). Legislative history reveals that the intent was to avoid a case-by-case assessment in which the plan as decision-maker would have to prove the reasonableness of its own judgments:
The series of presumptions prescribed by the Multiemployer Act were intended by Congress to ‘ensure the enforceability of employer liability. In the absence of these presumptions, employers could effectively nullify their obligation by refusing to pay and forcing the plan sponsor to prove every element involved in making an actuarial determination.’
Eberhard Foods, Inc. , 831 F.2d at 1260 (quoting H.R. Rep. No. 869, pt. I, 96th Cong., 2d Sess. 1, 86, reprinted in 1980 U.S. Code Cong. & Admin. News 2918, 2954). In discussing ERISA's presumption of correctness for a plan's determination of UVB, the Concrete Pipe Court emphasized the "technical nature of an actuary's assumptions and methods," and the fact that "[a]lthough plan sponsors employ them, actuaries are trained professionals subject to regulatory standards." 508 U.S. at 632, 113 S.Ct. 2264. Thereafter, the Court discussed the aggregate "sense of reasonableness that must be disproven by an employer attacking the actuary's methods and assumptions." Id. at 634, 113 S.Ct. 2264.
Since the methodology is a subject of technical judgment within a recognized professional discipline, it would make
sense to judge the reasonableness of a method by reference to what the actuarial profession considers to be within the scope of professional acceptability in making an unfunded liability calculation. Accordingly, an employer's burden to overcome the presumption in question ... is simply a burden to show that the combination of methods and assumptions employed in the calculation would not have been acceptable to a reasonable actuary. In practical terms it is a burden to show something about standard actuarial practice, not about the accuracy of a predictive calculation, even though consonance with professional standards in making the calculation might justify confidence that its results are sound.
Id. at 635, 113 S.Ct. 2264. To phrase it differently, "[t]he employer merely has a burden to show that an apparently unbiased professional, whose obligations tend to moderate any claimed inclination to come down hard on withdrawing employers, has based a calculation on a combination of methods and assumptions that falls outside the range of reasonable actuarial practice." Id.
Over five and a half years before Concrete Pipe , the Sixth Circuit expressed a similar view in EberhardFoods , 831 F.2d at 1263. It stated: "the employer is not entitled to what we as judges would say is the best or most reasonable method of calculating withdrawal liability. Rather, the employer is only entitled to complain if he proves that the actuarial assumptions applied by the trustees in the aggregate are unreasonable." Id. (emphasis in original). Furthermore, in Combs v. Classic Coal Corp. , 931 F.2d 96, 99-100 (D.C. Cir. 1991), also decided before Concrete Pipe , the District of Columbia Circuit adopted a similar view of the actuary's paramount role:
Actuarial valuations are based upon and reflect the experience of the plan, the professional judgment of the actuary, and the theories and expectations to which the actuary ascribes. Great differences of opinion exist as to actuarial methods. Congress, therefore, created the statutory presumption in favor of withdrawal determinations expressly to forestall endless disputes ‘over technical actuarial matters with respect to which there are often several equally ‘correct approaches.’ S. 1076, The Multiemployer Pension Plan Amendments Act of 1980: Summary and Analysis of Consideration , 98th Cong., 2d Sess. 21 (1980). In the absence of this presumption, ‘employers could effectively nullify their obligation by refusing to pay and forcing the plan sponsor to prove every element involved in making an actuarial determination.’ H.R. Rep. No. 869, pt. I, 96th Cong., 2d Sess. 1, 86, reprinted in 1980 U.S. Code Cong. & Admin. News 2918, 2954.
That standard, although deferential and procedural, is not toothless. At the hearing, the parties' witnesses agreed that the concept of a "best estimate" implies that an actuary is at least attempting to choose a figure where the likelihood that the investment return will be exceeded is equal to the likelihood that it will not-a 50/50 proposition. (Arb. Hr. Tr. 59:18-61:12, 66:5-:10, 118:4-:12.) However, they disagreed as to whether use of the Segal Blend struck that balance, a factual issue as to which the Arbitrator heard testimony. In the following section, I analyze the factual record compiled in the arbitration.
2. The record before the Arbitrator
The Pension Fund actuary, Ms. Gleave, testified that under the Actuarial Standards of Practice ("ASOP"), "every actuary needs to look at the purpose of the measurement before you know what assumptions you can use." (Arb. Hrg. Tr. 85:17-:20.) "For purposes of funding," she said, an investment-return assumption "represents the anticipated experience of the multiemployer plan's actual invested assets." (Id. at 87:24-88:5.) Ms. Gleave testified that "for purposes of the ERISA funding requirements," 7.5% represented her "best estimate of the anticipated experience" of the Pension Fund's assets. (Id. at 84:25-85:3, 89:22 to 90:2.) At her deposition, she explained that for minimum funding, she would confine herself to a long-term interest-rate assumption because for purposes of minimum funding, the plan's assets are invested and expected to earn returns over the long-term. (Gleave Dep. 52:2-:10.)
Ms. Gleave was referring to ASOP No. 27, entitled "Selection of Economic Assumptions for Measuring Pension Obligations." Under Sections 3.3(a) and 3.6.3(a), ASOP recommends that actuaries consider the purpose of the measurement when selecting economic assumptions, including the investment-return assumption. (ASOP No. 27 at 4, 7.)
See also Arb. Hrg. Tr. 87:11-:17 (acknowledging that 7.5% represented her "best estimate of the potential return outcomes," considered "for purposes of the funding of the plan"); id. at 112:6-:12 (stating that for purposes of funding the plan, the 7.5% assumption was her "best estimate of the long-term returns on this fund's assets."); id at 117:18-:23 (recognizing that 7.5% is her best estimate of how the plan's assets will perform over the long term).
As to her withdrawal liability calculation, Ms. Gleave explained that she did not find it appropriate to use the 7.5% investment-return assumption as the discount rate for unfunded liabilities. (Arb. Hr. Tr. at 92:2-:9.) Instead, she used the Segal Blend, so called because it is a blend of the 7.5% and the PBGC rates. (Id. at 92:7-: 19.) Ms. Gleave testified that the Segal Blend "reflects anticipated experience for the portion of the liability that uses the funding interest rate" of 7.5%. (Id. at 99:3-:6.) In other words, the 7.5% portion of the Segal Blend reflects "anticipated experience under the plan," (id. at 99:21-:24), while the PBGC portion of the Segal Blend "does not reflect the anticipated periods of experience under the plan." (Id. at 99:25-100:4.) Accordingly, the discount rate that results from applying the Segal Blend reflects the "best estimate of anticipated experience under the plan," but only "in part." (Gleave Dep. 57:13-:17).
Ms. Gleave was here referring to her deposition testimony. (Gleave Dep. 7:10-:12.)
See (Gleave Dep. 58:2-:8.)
Mr. Levy similarly testified that the Segal Blend takes into account the actuary's best estimate of future experience of the plan "[i]n effect, for the unfunded portion, which is the portion the ... withdrawing employer is going to pay off over up to 20 years, we use the funding assumption." (Arb. Hr. Tr. 142:5-:8).
The difference, Gleave explained, is explained by the differing purposes of the two calculations and the differing nature of the risk as to each. She stated that she viewed withdrawal liability, not as a simple actuarial forecast, but as a one-time "settlement" between the Pension Fund and the withdrawing employer:
[F]or purposes of ongoing funding, I use the 7-and-a-half percent. For purposes of withdrawal liability, which I view to be a settlement with an employer at the point of withdrawal, I will use the Segal Blend assumptions, and that is my best estimate assumptions for withdrawal liability.
(Arb. Hr. Tr. 119:19-120:2.) She then elaborated on the concept of a "settlement with the employer":
[In the context of withdrawal liability,] you have one shot to assess the employer, settlements, at the point of withdrawal. It's done using this blend, which is reflective of current market conditions, because you have no ability to go back to that [withdrawn] employer and get additional money should the plan incur losses. Likewise ... on the upside as well. There's no ability, after the one assessment is done, that's the settlement at that point in time for that employer
as to what their obligations to the plan would be.
(Id. at 120:5 to :16; 122:24-:25.)
The Pension Fund's expert, Mr. Levy, elaborated on the concept of an employer's withdrawal as the equivalent of a final settlement, taking into account both funded and unfunded liabilities. Such settlements, he stated, are made in the United States at market value. (Id. at 135:13-23, 158:15-:18). Thus Levy fortified the rationale behind use of the Segal Blend rate:
We said, to the extent that [the plan is] funded, we could treat this as a settlement and what it would cost you to settle with regard to the assets that were already there, because you can only settle what you can - what you have assets to settle with. You can't settle the liabilities that are going to be paid off over 15 or 20 years into the future.
So under the circumstances, admittedly, we came to the conclusion that we ought to use a market settlement, which is the PBGC rates and market value of assets, to the extent funded, and [for the unfunded portion] we could use the [investment-return] assumption as [a] long-term proxy over the next 20 years for what you could do when the money came in that wasn't coming in immediately, that was being paid by the employer over quarterly payments for up to 15 or 20 years.
(Id. at 136:16-137:9; see also id. at 139:19-:20; Levy Report ¶ 10 (explaining that "[t]he foundation of the Blend is that the portion that is already funded should be valued at market (because that portion could actually be settled), whereas the unfunded portion that will be funded over a period of years into the future should be valued on funding assumptions").) In short, the unfunded portion posed unique risks, requiring a conservative response from an actuary.
Levy explained that the withdrawal situation was unique in that it represented, not an ongoing funding relationship, but a one-time transfer of risk from the withdrawing employer to the continuing employers and participants. For that transfer, he implied, a premium must be paid; the transfer of risk of the investments' performance "has a price." (Arb. Hrg. Tr. 141:19-:20, 143:11-:19. See id. at 160:11-:16.) The plan trustees may make various investment choices in the future; they may even hypothetically choose to pay a premium to offload the risk to a third party through an annuity or otherwise. Those possibilities, he explained, should not affect the actuary's choice of the Segal Blend, because as of the withdrawal date, the withdrawing employer has shed its own risk by transferring it to the Fund:
See Moore, Understanding Employee Benefits Law at 31 (stating that in a defined benefit plan, "the employer is said to bear the investment risk").
Mr. Levy's Report provided an explanation of settlements:
Legally, there are only two ways that a multiemployer plan can settle a pension benefit obligation. One, payment to the participant of the lump sum present value of the obligation, is not available except with respect to small benefits, and therefore is inapplicable. The other is the purchase of an annuity from an insurance company. Thus, the market value of a plan's vested benefit obligations can only be determined by reference to the rates charged by insurance companies to pension plans that wish to settle some or all of their obligations. Unfortunately, insurance companies do not publish or otherwise make available their rates for assuming benefit obligations from pension plans. They do, however, furnish information on their actuarial basis for actual bids in the open, competitive marketplace to the Pension Benefit Guaranty Corporation (PBGC). PBGC then reviews this information and publishes interest rates consistent with the information provided by the insurance companies. (Levy Report ¶ 6.)
Whether or not [the Plan trustees] choose to get rid of that risk by buying annuities doesn't affect the fact that the risk has been transferred. And the markets say, when you transfer a risk, there is a price to be paid for doing so. Somebody else is taking the risk for you, then the price is not the same as if you're taking it yourself.
Mr. Levy further elaborated on the functioning of the Segal Blend:
[It] is looking at what the market in the form of annuity purchase rates would charge you to transfer that obligation. It isn't looking at a portfolio as such. It's looking at what it would take to have somebody else take over the responsibility. That is what ... the market would charge for transferring that liability from the plan to an insurance company.
(Arb. Hr. Tr. 159:17-:24.)
(Id. at 143:11-:24. See also id. at 144:7-: 17.)
Mr. Levy therefore described the use of the Segal Blend as reflecting a "risk-adjusted expected experience." (Id. at 160:4-:5. Emphasis added.) The anticipated experience has been adjusted "to reflect the fact that the risk has been transferred," he explained:
It's really a risk adjustment that says, this is a final settlement with the withdrawing employer, and to the extent that it is possible because the assets are there to make a final settlement of the withdrawing employer's obligation, that should be done at market value.
Whether or not the trustees choose to, in fact, buy annuities or set those obligations, they are still taking the risks related to those obligations and the withdrawing employer is not.
(Id. at 153:3-: 12. See Levy Report ¶ 8 (explaining the logic of the risk transfer theory).)
Manhattan Ford's expert, Mr. French, on the other hand, opined that use of the Segal Blend was unreasonable. Because the discount rate is "such an important assumption," he said, use of an incorrect rate was sufficient to render the assumptions "unreasonable in the aggregate." (Arb. Hrg. Tr. 50:13-:19, 55:24-57:7.) He reached that conclusion based on the language in Section 3.6 of ASOP No. 27 and his view that the "best estimate of anticipated experience under the plan" "has to be the [investment] return on the assets, because there is no other experience that you could be looking at." (Id. at 29:2-:6, 39:5-: 18.) Under Section 3.6 of ASOP No. 27, he testified, the 7.5% investment-return assumption should have been used to calculate the Pension Fund's withdrawal liability:
Because there is a pool of assets that is clearly dedicated and must be dedicated to it and because the statute doesn't point you to something other than this. It in fact says you must use the anticipated experience under the plan, so the statute itself is directing you to the investment return.
(Id. at 39:12-:18.) Mr. French objected that the Segal Blend is not tied to the Plan's actual investment intentions for the future:
It doesn't even purport to represent the anticipated experience under the plan unless ... the plan has said or represented that it is, in fact, going to purchase annuities or it is, in fact, going to take those withdrawal liability payments and put it into a separate investment vehicle that is only invested in corporate bonds, and I have never seen that situation.
(Id. at 49:10-:18.) Here, said Mr. French, the Pension Fund is "already overfunded on the actuary's best estimate assumptions" and Manhattan Ford "is making it even more overfunded" by having to make withdrawal liability payments. (Id. at 52:8-: 10.) He admitted that he was aware that some actuaries use the Segal Blend or another form of blended rate to calculate withdrawal liability, but said that he was unaware of the justification for such use. (Id. at 72:6-:11.)
Mr. French rejected the Pension Fund experts' concept of withdrawal liability as a settlement, or the Segal Blend as reflecting a "risk-adjusted" approach to the Fund's experience. According to Mr. French, when an employer withdraws and therefore permanently ceases its obligation to contribute, it is not settling any benefit plan liability because it never had an obligation to pay the plan's benefit liabilities; that obligation belongs to the pension plan's trustees. (Arb. Hr. Tr. 172:17-:19, 173:19-174:1.) An employer "can't settle something he never had an obligation for," he testified. (Id. at 172:19-:21.) Mr. French did, however, recognize that if the plan does not perform as expected under his investment-return assumption, there may be consequences for those remaining in the Plan. These include increased contributions, reduced future benefit accruals, and so forth. (Id. at 175:10-:19.) According to Mr. French, once the employer has withdrawn, the plan trustees have the option of using the employer's withdrawal liability payments to buy annuities; "They could go and put a separate fund and guarantee that this isn't a problem. But they've chosen not to do that. They want the benefit of that extra return." (Id. at 182:25-183:5.)
Mr. Levy similarly acknowledged that if the Pension Fund cannot meet its obligation of paying benefits, one possibility is that the employers may engage in a mass withdrawal (id. at 168:24-169:22)-the proverbial rush for the exits.
3. Analysis
My analysis of the record starts from the premise that protection of Plan participants, so long as it reflects professional actuarial judgments and not the self-interested bias of the Plan itself, see Section IV.B.1, supra , is a permissible, indeed a paramount, goal. See IUE AFL-CIO Pension Fund v. Barker & Williamson, Inc. , 788 F.2d 118, 127 (3d Cir. 1986) ("Courts have indicated that because ERISA (and the MPPAA) are remedial statutes, they should be liberally construed in favor of protecting the participants in employee benefit plans."); Pittsburgh Mack Sales & Serv., Inc. v. Int'l Union of Operating Eng'rs., Local Union No. 66 , 580 F.3d 185, 194 (3d Cir. 2009) (describing the purpose behind ERISA and the MPPAA as "ensuring that pension funds will be adequately funded, even when employers withdraw from them, and that the employees who are relying on those funds will be protected"). Based on the testimony, the Arbitrator upheld the Pension Fund actuary's "best estimate" that the context of withdrawal liability permitted application of the Segal Blend discount rate, not the funding rate. The evidence fully supported the Arbitrator's decision that the actuary permissibly took a conservative risk-adjusted stance. The Arbitrator permissibly accepted the actuary's position that withdrawal liability represents a one-time settlement of obligations, not an ongoing calculation of contribution rates. Of course there was expert testimony to the contrary, and another arbitrator could have found it persuasive. But "[w]here there are two permissible views of the evidence, the factfinder's choice between them cannot be clearly erroneous." Anderson , 470 U.S. at 574, 105 S.Ct. 1504.
ERISA, to be sure, does not use the words "settlement" or "risk-adjusted expected experience." The statutory standard of "reasonableness," however, is open-textured and deferential. It is capacious enough to permit an actuary, in his or her professional judgment, to adopt such an approach. And the record was adequate to demonstrate that the Plan's actuary validly adopted such an approach here based on reasonable professional judgments. As the Arbitrator pointed out in his Final Award, a valuation is "a snapshot," but "[o]ne cannot rely on the fact that the Plan is fully funded now to say that the Plan will be fully funded for all time." (Final Op. at 14 (citing Arb. Hr. Tr. 120:17-122:23).) As emphasized by the Pension Fund, a withdrawn employer's obligations are fixed as of the withdrawal date; the ongoing investment risk is transferred to the pension plan and the remaining employers. (Def. Brf. at 13, 15.) It was therefore not unreasonable, as found by the Arbitrator, for the actuary to use a conservative model that recognizes the different nature of future risks with respect to withdrawing employers and continuing employers. (See Levy Dep. 66:5-:20.)
The Arbitrator did not clearly err in finding that Manhattan Ford failed to meet its burden to rebut the reasonableness of the actuary's approach. I do not overlook Manhattan Ford's argument that the use of the Segal Blend results in a withdrawal liability figure that is expected to overfund its share of the benefits. It may be true, moreover, that the PBGC-rate component will not turn out to correspond to actual fund investments. (For that matter, the Fund could someday decide to change its investment mix in a way that undermines the 7.5% rate as well.) The "best estimate of anticipated experience under the plan," however, may reflect factors other than the Plan's commitment to some future investment portfolio. The testimony of the Pension Fund's actuary, Ms. Gleave, and that of its expert, Mr. Levy, amply support the Arbitrator's finding that the actuary's use of the Segal Blend represents their professional judgment and satisfies professional standards. I am not persuaded by Manhattan Ford's characterization of the Arbitrator's reliance on the risk-transfer theory as "factually misguided." (Pl. Br. at 13.) The situation, says Manhattan Ford, is symmetrical; while the remaining employers bear the risk of underperformance, the withdrawn employer bears the opportunity risk of overperformance, and those risks are "equally likely to occur." (Id. at 13-14.) One short answer to that contention is that ERISA, a remedial, pro-employee statute, may properly be regarded as not considering those risks to be equivalent; an actuary, and arbitrator, or a court might permissibly err on the side of safety for the plan participants.
At the hearing, Mr. French, Manhattan Ford's expert, testified that as of the snapshot at the time, the Pension Fund is "110 percent funded, and it's also got a very healthy contribution rate." (Arb. Hrg. Tr. 176:13-:15.) He opined that because the Pension Plan "is so well-funded and has such a good contribution rate, so much money coming into benefits" (id. at 176:18-:20), the likelihood that the 7.5% investment-return assumption would be achieved or exceeded is "probably better than" 50%. (Id. at 177:19.) However, as noted by the Arbitrator, he "did not opine on how much better that likelihood was." (Final Op. at 14.) Furthermore, upon further questioning, Mr. French also acknowledged that the plan's funded state "absolutely" could change. (Arb. Hr. Tr. 177:20-:21).
One such published standard, as the Arbitrator found, is Section 3.6 of ASOP No. 27. Section 3.6, entitled "Selecting an Investment Return Assumption and a Discount Rate," states in relevant part:
The investment return assumption reflects anticipated returns on the plan's current and funded assets.
The discount rate is used to determine the present value of expected future plan payments. Generally, the appropriate discount rate is the same as the investment return assumption. But for some purposes, such as SFAS No. 87 or unfunded plan valuations, the discount rate may be selected independently of the plan's investment return assumption, if any. In such cases, the discount rate reflects anticipated returns on a hypothetical asset portfolio, rather than on the plan's expected investments.
(ASOP No. 27 at 5). The "best-estimate range" is defined in Section 2.1 of ASOP 27 as "the narrowest range within which the actuary reasonably anticipates that the actual results, compounded over the measurement period, are more likely than not to fall." (Id. at 2.)
Manhattan Ford argued that because the two specified exceptions did not apply, the Pension Fund actuary was required to use the same discount rate and investment-return assumption. There was no indication in the record that section 3.6 precluded the use of the Segal Blend, however, and a subsequent version of that ASOP explicitly stated that its list of examples was not exhaustive. (Final Op. at 15.)
Manhattan Ford notes that in 1979, the American Academy of Actuaries "urged Congress to amend the pending withdrawal liability legislation to direct mat ‘PBGC assumptions’ be used to value vested benefits, because the normal investment-return assumption is "inappropriate" in this context." (Pl. Brf. at 15 (citing Hr'gs on Multiemployer Pension Plan Amendments Act of 1979 (H.R. 3904) Before Task Force on Welfare & Pension Plans of Subcomm. on Labor-Mgmt. Relations of H. Comm. on Educ. & Labor , 96th Cong. 1498, 1502 (1979).) Manhattan Ford asserts that because Congress did not make that change, Congress "signal[ed]" its rejection of the risk-transfer theory. (Id. ) Congress's failure to act is weak evidence, but at any rate it does not signal an intent to require use of the funding rate only.
At any rate, the Arbitrator could properly rely on the testimony regarding the withdrawn employer's transfer of its risk in a one-time settlement transaction. Viewed in light of the remedial purposes of ERISA, the symmetry is a false one; post-withdrawal, the Pension Fund remains responsible to pay benefits to the withdrawing employer's employees, but the withdrawing employer does not. The Fund must guard against the risk of capital loss, or at least of falling short of a 7.5% return, without the backstop of possible additional contributions from the withdrawing employer. The Fund could offload some or all of that risk-for example by purchasing an annuity or investing in safe corporate or treasury bonds. (This may be referred to as a "risk-free" option, though the risk is not really zero.) That, however, would have a cost, and that hypothetical cost is a proxy for the dollar value of the risk that the Fund is taking on. The risk-free PBGC rate and the "settlement" theory espoused by the actuary here reflect that cost and that dollar value.
On that score, by the way, I note that the Pension Fund has used the Segal Blend for some 25 years, and consistency has a virtue of its own. See (Final Op. 6, 12-13; DSMF ¶ 7; Joint Stip. of Facts ¶ 11) Presumably if there were prior withdrawals (the record does not say), Manhattan Ford would benefited, at least indirectly, from those employers' pay-in of withdrawal liability in an amount dictated by the Segal Blend.
--------
The risk-transfer and settlement models of withdrawal liability recognize a more complicated reality than the one embodied in minimum funding levels. The Arbitrator did not have to accept those alternative models, and a different arbitrator could have decided the case differently. This Arbitrator, however, did not clearly err in accepting the risk-transfer theory and settlement-based concept behind the Segal Blend. The testimony of Ms. Gleave and Mr. Levy, their depositions, and Levy's Report, if accepted by the Arbitrator (as they were), provided a sufficient basis for the Arbitrator to find that there was a good faith application of reasonable actuarial assumptions and practices.
There was no error in the Arbitrator's legal conclusions or clear error in his factual conclusions. IV. Conclusion
For the reasons outlined above, Manhattan Ford's motion for summary judgment (ECF no. 22) is denied, and the Pension Fund's cross-motion for summary judgment (ECF no. 23-2) is granted. The Arbitrator's Interim and Final Awards are therefore affirmed in accordance with this Opinion.
An appropriate order follows.