Op-Ed ¦ May/June 1994
Is PBGC's Deficit Calculation Correct?
(If Not, Are Those 20+ Legislative Proposals Necessary?)

After 6 months of deliberations, the task force set up by Secretary of Labor Robert Reich to scrutinize the Pension Benefit Guaranty Corporation (PBGC) came up with a package of twenty legislative proposals, made public on September 30, 1993. In his announcement, Secretary Reich explained why an intensive effort was crucial: Even before the task force began its work, we knew we were faced with serious long-term problems. The financial security of thousands of Americans covered by single-employer pension plans is at risk, because their pensions arc underfunded. There has been dramatic growth in underfunding... expected to exceed $45 billion. . pos[ing] an unnecessary and unacceptable risk for workers and retirees. And PBGC, already facing a $2.7 billion deficit, is further jeopardized by this growing underfunding.

Does Secretary Reich have his facts straight? We sorely need to know; since the twenty proposed legislative changes he's offering are based on these numbers. The proposals threaten to impose a significant new overlay of complexity to pension law and regulation, and thereby add to the already considerable burdens plan sponsors carry. (Just the summary of the proposed changes is forty-six pages long.) If it turns out that the facts are not as stated, then less drastic remedies are clearly in order.

The $2.7 Billion Deficit
The 52.7 billion deficit is the most prominent figure cited by the PBGC to demonstrate the seriousness of its financial problems; it's calculated simply by deducting assets from liabilities. The specific amounts the PBGC has publicized over the years are shown in Table 1 (column 2). The numbers are derived by deducting total assets from total liabilities and adding in "probable net claims" (the PBGC's estimate of claims it will likely have to assume in the near future from plans expected to terminate because their sponsors are in dire financial straits).

Table 1
PBGC-Claimed Defecit Compared with True Deficit (in $ millions)
Cumulative Deficit
(as of 9/30)
PBGC-Claimed Defecit Probable Net Claims True Deficit
1985 $1,325 $464 $861
1986 $1,325 $2,145 $1,681
1987 $1,549 $312 $1,237
1988 $1,543 $108 $1,435
1989 $1,124 $242 $882
1990 $1,913 $1,111 $802
1991 $2,510 $776 $1,734
1990 $2,683 $999 $1684

Total liability
- assets
+ probable net claims

Total liability

These aren't actual claims - and including them substantially increases the deficit. But the PBGC doesn't attempt to justify this approach, even though other highly probable future transactions aren't included: premium and investment income and benefit and expense payout which, on balance, would reduce the deficit (more on this later). I have therefore in the last column of Table 1 shown the traditional "true deficit," equal to the PBGC-claimed deficit, less the probable net claims.

The result is a substantial reduction in the claimed deficit and limited variability since 1985. We can also see that the probable net claims more than doubled the true deficit in 1990 (from $802 million to $1913 billion) and represented a considerable 81% of the true deficit, on average, for the 3-year period 1990 -1992.

My office computed the PBGC's rates of success in predicting probable net claims that in fact did mature. The rates were 11%, 50%, and 35% in the years 1990 -1992, for an overall batting average of .320. Since this means that the PBGC was off the mark two-thirds of the time, the term "probable net claims" is inaccurate. And the impact of the PBGC's adding in the probable net claims is to put a more negative spin on the deficit than seems warranted.

There's another way to look at the S2.7 billion deficit: relate it to assets that are also cumulative. Figure 1 shows that the true deficit dropped as a percentage of the assets from 75% in 1985 to 27% in 1992, and its importance diminished accordingly by nearly two-thirds. At this point, it's only 21% of the total PBGC liability, and this may, in large part, be attributable to the substantial drop in interest rates.

The size, and particularly the trend, in the deficit, do not then support the assertion that the PBGC is asking the public to believe: namely, that the $2.7 billion deficit indicates a "threat to the ability of the PBGC to safeguard pension benefits." This doesn't mean that a threat couldn't materialize at some later date - few insurance schemes are totally invulnerable - but the PBGC has not proved its case.

Current Underfunding - $45 Billion
Secretary Reich held out another dire portent of long-term problems for the PBGC - its estimate of 45 billion in underfunding among all the 65,000 active defined benefit plans it insures. However, the $45 billion figure has only minimal value. It's only a stab in the dark, not calculated directly from the 65,000 plans (which differ substantially in benefit provisions and assumptions); the range of error in the PBGC's estimate is therefore quite wide.

The amount of underfunding is subject to interest rate fluctuations as well; long-term rates, for instance, are currently at their lowest, thereby leveraging the underfunding upward. It would be useful to see estimates that assumed higher interest rates as well.

There is also some implication that the $45 billion in underfunding lurks among us as a shadowy menace - which it isn't. For most plans, the underfunding simply represents the remaining past service obligation that an employer normally funds over a period of future years. This obligation is a necessary accompaniment if a plan is to provide benefits for service prior to the effective date of the plan or a benefit increase, and to attach a stigma to a sound actuarial method for budgeting plan costs is misleading.

Is it possible to determine which part of the $45 billion and future increases will emerge as claims against the PBGC? The agency believes that $12 billion is attributable to plans of sponsors in heavy industries in “difficult financial circumstances. This figure is most dubious: judgments are required on a host of factors, which can vary widely over each year for a variety of reasons. Consider the following and their related influences, the timing and magnitude of each being most difficult to predict:

•Probability that a plan will terminate: the economy, management's skills, technological change.
•Present value of the benefits the PBGC will have to assume: benefit increases will arise from plan liberalizations and salary increases, plant closings with enriched benefits, and early retirement subsidies; decreases will follow from layoffs and salary cutbacks; and interest rate changes play a role.

•Value of assets available to the PBGC: plan assets will vary with the actuarial assumptions and method used, contribution level, investment earnings, asset depletions due to lump-sum payouts and annuity purchases; corporate assets may or may not be adequate to make up an underfunding.

The PBGC also projects its future deficits over a 10-year period; the estimates depend on the factors just listed, in addition to: • Premium collections: legislated increases, the number of participants being added and subtracted in new, terminated, and existing plans.

•Investment earnings on the PBGC's assets.
•PBGC's expenses: expenses have been growing at a sizable rate but, it is hoped, will be stabilized in time.
•Legislation, regulation, and court decisions.

Let me repeat - the PBGC's batting average in predicting near-term terminations of shaky sponsors' plans has been a meager .320. And it's reasonable to assume that, as one goes out 2 to 10 years, the ability to make an accurate prediction is going to diminish rapidly. An example: the PBGC estimated in 1986 a deficit of $7.4 to $11.0 billion for 1992; the actual deficit amounted to $2.6 billion. Note that the estimated PBGC deficit for 2002 that appears in its 1992 Annual Report is in the range of $2 billion to $28 billion (or even higher - the agency doesn't regard $28 billion as the worst-case scenario).

I'm not convinced that Secretary Reich and his task force have made a persuasive case for adopting proposals aimed at either raising premiums or sponsor contributions

Two Other Indicators There are two other key indicators that the PBGC isn't in such bad shape. One is the cash flow generated by income items (premiums and investment income) and disbursement items (benefits and expenses). For example, in the period 1985 - 1988, income exceeded disbursements by a total of nearly $435 million ($109 million per year), while m the past 4 years, 1989 - 1992, the excess was $2.5 billion ($631 million per year).

That means that the average amount of annual income rose by nearly 500% - certainly indicative of strong financial performance. On an individual-item basis, premiums were up more than tenfold, from $82 million to $876 million, while benefits increased nearly fourfold, from $170 million to $634 million, and expenses (administration plus expenses) rose threefold, from $33 million to $99 million. Notable on the expense side was a 30% rate of increase in the PBGC's expenses in the period 1990 - 1992 (40% in both 1990 and 1992; 13% in 1991).

The second indicator is the ratio of assets to the sum of benefits and expenses. This provides a rough measure of the number of years that disbursements at the current annual rate could be made from existing assets. Figure 2 shows the trend of this ratio, from 1985 - 1992: it's risen by a hefty 51% (average rate of increase per year, 6%).

At the end of 1985, there was $5.69 for each dollar of payout; at the end of 1992, there was $8.59 - another indication of major gains in financial strength.

What is an adequate ratio? The goal for Social Security old-age benefits is about 1 year of benefit payments. Secretary Reich's task force on the PBGC proposes that defined benefit plans have $3 on hand for each dollar of payout. For the PBGC, the goal should probably be a bit higher - around $5 per dollar of payout, to build in some time to spot emerging trends, and then arrive at the best course of action.

Final Word
Based on my analysis, I'm not convinced that Secretary Reich and his task force have made a persuasive case for adopting proposals aimed at either raising premiums or sponsor contributions. There is some merit, though, in the proposals aimed at minimizing additional large claims, because they would make the system more fair to sponsors in general.

There is also a need to consider alternatives to the task force's proposals, and to make an effort to evaluate just how useful the PBGC has been in providing retirement security for all our citizens.

- David Langer is president, David Langer Company, Inc., New York City.

This article is based on his presentation at the Society of Actuaries' annual meeting, October 17-20, New York City; a shorter version appeared in Pensions and Investments, December 13, 1993.

Reprinted, courtesy of Contingencies, the magazine of the American Academy of Actuaries.
May/June 1994