INTRODUCTION: PBGC INSURANCE OPERATIONS AND SOURCES OF FUNDING

The Pension Benefit Guarantee Corporation (PBGC), created under the 1974 Employee Retirement Income Security Act, is a quasi-governmental insurer offering coverage to participants in defined benefit pension plans. While PBGC is considered a federal agency under the US Department of Labor, its primary funding is based on income generated from premiums charged to employer sponsors of defined benefit plans. This agency operates as an insurance operation that depends upon having premium income adequate to meet future loss obligations to retirees in failed defined benefit pension plans. However, if PBGC deficits were to reach the point of insolvency, the federal government would have an implicit obligation to meet PBGC's obligations in much the same manner as it did with the FSLIC in the 1980s. Given the economic downturn over the last decade, PBGC obligations have increased to the point where increases in flat and variable rate premiums have been insufficient to meet claims costs from failing defined benefit retirement plans. Since current PGBC rates have not been adequate to meet loss experience for the past several years, the main concern is what may be an appropriate premium rate to charge in order to credibly cover future claim experience. This article seeks to examine this question by using a retrospective reserve model to test what combination of flat and variable rate premiums might have been sufficient to cover PBGC claims costs over the last 15 years from 1994 to 2009.

This study will examine the adequacy of premiums in relation to single-employer insurance offered by PBGC, the largest insurance program and one that has generated the greatest losses to the PBGC fund. As a federal corporation, PBGC is governed by a Board of Directors consisting of the Secretaries of Labor, Commerce and Treasury. It provides insurance to 44 million workers and retirees in 29 100 private defined benefit plans. The largest number of workers and retirees, 33.79 million, are members of single-employer plans.1 Consequently, the focus of this study will concentrate on the coverage having the largest loss potential, single-employer plans involving individual US companies offering defined benefit plans to workers. During the 2010 fiscal year, PBGC received US$2.231billion in premium income plus $7.594 billion from investments against plan termination losses and actuarial adjustments of $9.421 billion. Net losses on insurance to the single-employer plans amounted to $517 million and the capital position ran a deficit of $21.594 billion.1 As noted in PBGC's 2009 Annual Report, ‘The agency's deficit remains a cause for concern and is a reflection of the long term challenges confronting PBGC’.2 Under these circumstances, the data demonstrate that premiums were inadequate for meeting PBGC claim costs leading up to 2010. At issue is what premium should have been charged given this type of experience and how might that information be used to revise current rates to meet future PBGC insured loss obligations.

Table 1 provides a historical perspective on the long-run financial position of PBGC since 1985. Annual increases in the PBGC capital deficit from 1995 onward grew at a faster rate (32.56 per cent) than during the entire 25-year period from 1985 (11.81 per cent). In the period from 1985 to 1995, PBGC experienced deficits of between −$315 million and −$2.897 billion. Following a brief period from 1996 to 2001, when PBGC ran capital surpluses between $869 million and $9.704 billion, the agency faced a growing trend in deficits thereafter with the latest 2010 shortfall reported at $21.594 billion. Particularly telling is the fact that the latest deficit comes after similar shortfalls in 2004 and 2005 when flat and variable premiums were increased in an effort to shore up the PBGC fund. While the PBGC fund deficit decreased at a rate of 7.34 per cent in the period from 2005 to 2010 (that is, $23.305 billion to −$21.594 billion), the lack of a significant reduction in fund losses shows that the higher current premiums may not be adequate to sustain the fund. In 2010 the fund generated a 12.1 per cent investment return, up from the −6.5 per cent return in 2008, but less than the 13.2 per cent return in 2009.1, 2 Even with the two years of above average investment returns in 2009 and 2010, premium income along with investment income was not sufficient to significantly reduce the $21 billion deficit to the PBGC fund.

Table 1 Net financial position of PBGC's single-employer program from 1985 to 2010
Full size table

An actuarial and financial approach to PBGC funding requires that the present value of future premiums meet the present value of future obligations to workers and retirees in defined benefit plans under coverage. Three critical variables in maintaining the viability of an insurer is the ability to (i) adjust premiums to fully reflect the risks assumed with coverage; (ii) alter underwriting and benefit structures to avoid poorer than average risks getting coverage at rates below the true cost of claims (that is, adverse selection risk) and (iii) generate investment returns that will grow funds that can be used to meet future claim liabilities.4 Table 2 provides historical information on PBGC premium rates and revenues since 1985. Initially, PBGC charged a flat premium rate per insured worker of $2.60 to $8.50. By 1988, PGBC started charging a flat rate, plus a variable rate based on the level of under funding in vested plan benefits. Rates in 1988 were $16 per worker with an excess variable premium of $6 per $1000 of unfunded benefit up to a maximum of $34 per worker. These rates gradually increased to $19 per insured worker and $9 per $1000 of unfunded benefit without an upper bound on unfunded amount. Table 2 examines the historic premium rates for PBGC coverage from 1985 to 2009.

Table 2 Pension Benefit Guarantee Corporation historic premium rates and revenues for 1985 through 2009
Full size table

From 2005 to 2009, flat premium rates went up from $30 to $35 per worker, while the variable rate formula remained the same. As a consequence the contribution of flat premiums to overall premium revenues to PBGC grew from 45.8 per cent in 2005 to 61.8 per cent in 2009, and variable premiums as a percentage of total premium revenue declined from 54.2 per cent to 38.2 per cent in that same period. If variable premiums are used to differentiate those plans with higher risk by charging correspondingly higher rates for coverage, the 2009 variable premiums do not appear to fulfill this purpose in light of the lowered contributions to PBGC revenues. With the latest changes to PGBC premiums, rates are at their highest levels, yet premium revenues and investment returns have been insufficient to reduce the agency's net financial deficit. Consequently, if PBGC, as an insurance operation, were to have adequate premiums, capable of credibly covering future claim costs, rates would have to increase to reflect losses above what was expected. Despite increasing the flat premium from $2.80 to $35, and variable rates from $6 per $1000 to $9 per $1000 of unfunded pension liabilities, PBGC continued to increase losses in its net financial position resulting in a negative $21 billion balance by the beginning of 2010. One way private insurers can analyze how much they may need to increase rates to cover losses is to use retrospective premium analysis to determine how a higher rate might have fared in generating loss reserves. Such an investigation permits the insurer to determine a revision in current rates that may more fully reflect what is required to meet future claim obligations.

RETROSPECTIVE PREMIUM ANALYSIS OF THE PBGC FUND

The retrospective premium method of analyzing the PBGC fund is based on viewing insurance coverage as consisting of a pool of policyholders who pay premiums in order to be indemnified against future losses from defined benefit pension plans. These policyholders pay both flat and variable premiums invested in a fund to pay future losses. The PBGC policyholders pay a flat premium per participant and a variable premium if the plan has an unfunded pension liability. The retrospective model assumes that the fund from premiums will be invested at 5 per cent as long as accumulations remain positive, if not, the investment return is zero.5 Actual claim experience over the study period is counted against accumulations from paid-in premiums and investment return. Premiums paid for the year are used to offset claim experience. If claims in a given year are higher than premium income, funds are taken from the PBGC fund. If the PBGC fund balance is insufficient to meet current claims, paid losses will be assigned to the PBGC fund to produce a negative amount. If the PBGC fund is positive at the beginning of the year, a 5 per cent investment return will be added to the balance.6 If the fund is in a deficit position at the beginning of the year, there will be no investment return at the end of the year. Claim costs, premium income and investment return will be recorded at the end of the year. Flat premiums may be increased on a per participant basis and applied to those policyholders that were paying premiums over the study period. Since it is not possible to tell how variable premiums were assigned to individual single-employer plans, the present study considers aggregate changes to variable rate revenues that may be developed by altering variable rates.7 For example, if variable rates were to be doubled over the period, the model projects a doubling of historical revenues each year from variable rate premiums.

Table 3 provides a retrospective premium analysis using the highest flat premium rate charged to date by PBGC, $35 per participant, starting in 1993 and continuing until 2010. Under this model the PBGC net financial position becomes positive in 1996, remains so until 2004, and then turns negative ending with a deficit of $8.7 billion.

Table 3 Examination of the PBGC loss reserve based on actual net claims, a $35 flat retrospective premium, 5% return on reserve assets and maintenance of variable premiums from 1993 to 2010
Full size table

Table 4 considers the impact to the PBGC net financial position from a $35 flat premium and a doubling of revenues from variable rate premiums. Under this scenario PBGC's net financial position becomes positive in 1995, and remains so with the exception of 2005 and 2009. Even though net claims in 2009 is substantial amounting to $7.9 billion in losses, under this scenario PGBC's net financial position ends with a deficit of $791 million, an amount which could be made up from premium and investment income in succeeding years. Consequently, this premium structure might be adequate on a long-term basis when viewed in terms of PBGC's ability to cover losses incurred historically.

Table 4 Examination of the PBGC loss reserve based on actual net claims, $35 flat retrospective premium, doubling of variable premium revenues and a 5% return on reserve assets
Full size table

Table 5 examines the effect that a $50 flat premium would have on PBGC net financial position, leaving variable premiums in tact from 1993 to 2010. Under this scenario the PBGC fund extinguishes its deficit position by 1994 and remains positive to 2010 leaving a funding balance of $2.1 billion. By increasing the highest historical flat premium by 42 per cent, PBGC with a $50 premium is able to meet annual claim costs and still have some left over for unexpected future claims. This result reinforces the perspective that current PBGC rates are too low to meet current and future losses to the fund based on historical claim experience.

Table 5
Full size table

THE NEED TO MAINTAIN A POSITIVE FUND BALANCE TO FACILITATE RESOLUTIONS

From a microeconomic, insurance perspective, PGBC offers a unique form of coverage where rates must be sufficient to cover current claims and still have funding for unexpected losses in the future. Although the policy insures pension benefits for workers participating in defined benefit plans, the premiums are paid by firms sponsoring the pensions. The insured party is not the policyholder or premium payer for the coverage. The insurance is owned by the corporation, on behalf of the insured workers covered by PBGC. Consequently, the managers of the firm decide whether to continue their defined benefit plan and pay premiums, or terminate the pension. Insured workers have little control over these decisions. If a plan terminates due to bankruptcy PBGC's insurance coverage provides guaranteed benefits to retirees, and the claim liability is based on the value of the plan assets, the level of benefits defined within the terminated plan, the limits of PBGC coverage, and the extent corporate assets subrogated to pay guaranteed benefits. in most of these cases plan assets are considerably less than the actuarial value of the plan's future pension obligations. Under such circumstances PBGC recalculates benefits to workers and recognizes a future insurance claim liability based on the limits of coverage, the value of the transferred assets in the terminated plan and the present value of future benefits.9 When PBGC receives pension assets from a terminated plan, there can be a priority to a retirees claim to enhanced benefits. Current employees covered under PBGC insurance, retirees receiving fixed benefits before PBGC plan assumption, current employees with vested benefits less or more than the PBGC maximum benefit limits, as well as participants with unvested benefits, all have varying priorities of claims to ERISA law.10 Difficulties in handling assets acquired from terminated plans are significant including, but not limited to: (i) the cost of managing physical assets until a sale can be made; (ii) determining a fair value of the assets quickly; (iii) maintaining the value of the assets while an appraisal is being made of whether to dispose or retain the property; and (iv) dealing with litigation costs associated with enforcing PBGC's right to the property. Without adequate funding to resolve pension bankruptcies, the eventual costs of selling assets may increase substantially reducing the value of PBGC owned assets. These problems may raise PBGC settlement costs thereby adding to the fund deficit due to an ability to quickly, efficiently and effectively address asset sales and claim payments.

One illustration that may serve to highlight loss settlement challenges from an inability to dispose of pension assets on a timely basis is the case of the United Airlines bankruptcy and pension termination. PGBC became an unsecured creditor in United Airlines when the company shifted $10.2 billion in unfunded pension liabilities to the agency in December of 2002. PBGC reached an agreement, during the United Airlines bankruptcy proceedings, to receive a $5.6 billion claim on the new United Airlines. In February 2006, PBGC sold $2.5 billion of this claim to hedge fund investors and banks for $450 million or $.18 on the dollar. Under PBGC's maximum benefit cap, some of the 120 000 United workers saw large cuts in their retirement income due to the significant drop in the value of plan assets from 2002 to 2006.11 Further highlighting this problem are past bankruptcies where PBGC received such diverse assets as: ‘diamonds, a hog slaughtering facility, oil wells, a restaurant, interest in a nuclear fuel reconditioning partnership, and water rights in the Mojave Valley’. While the agency has hired a special assets manager to dispose of or manage PBGC bankruptcy assets, the main issue remains as to how funds can be deployed to this activity when PBGC is running a deficit of $21 billion.11 The recent declared bankruptcy filing of American Airlines in fall 2011 once again brings into focus the costs that may attend airline pension plan failures and the difficulties PBGC faces in liquidating assets to reduce termination costs.12

PBGC premiums charged on single-employer defined benefit plans have been inadequate to meet claim experience resulting in a fund deficit of $21 billion. Despite increasing premium rates for the past several years, PBGC's net financial position has been falling since 2008. A retrospective analysis involving an adjustment of premiums to reflect historical loss experience shows that the current PBGC rate structure is inadequate and that significant increases would be needed to cover losses from the period 1993 to 2010. By setting the premium rate at a level that is at least sufficient to cover past loss experience, PBGC could seek to experientially rate its coverage. Under such an arrangement the agency could alter rates downward after the fund achieves a certain level of surplus and then increase rates at times when current claims costs lower the fund from its target level. Further research into an experiential rating system for PBGC would have the added benefit of allowing rates to be set on a pro-active rather than a re-active basis.