Chapter 4
The Regulatory Framework—Benefit Insurance, Minimum Funding Rules and Accounting Standards Affecting DB Plan Finance
One way to think of a defined benefit plan is as an annuity company run by the employer-sponsor exclusively for its own employees. The employee is the beneficiary of the employer’s DB “life annuity promise,” and all the risks that an insurance company/annuity carrier undertakes when it sells a life annuity are, in a DB plan, borne by the plan sponsor.
The difference between an annuity purchased from an insurance company and the annuity provided by the employer’s DB plan—and the reason that, mid-20th century, DB plans became the primary means of providing corporate retirement benefits—is that, while insurance companies are subject to regulatory overhead and, critically, restrictions on what they can invest in, DB plan sponsors are not.
The ability of corporations, in DB plans, to pursue a modern portfolio theory (MPT) investment strategy and (theoretically at least) earn the higher rates of return MPT promises, made employer-provided retirement benefits cheaper. And that reduction in the cost of retirement savings (a theme we will come back to again and again) meant that many more participants could be covered under many more retirement programs at many more employers.
This was, in many respects, a great gain for sponsors and participants. There was, however, a flaw in this “DB value proposition.” While, in the first instance, the employer-plan sponsor bore all the risks associated with the pension promise, and the participant was in a (theoretically) ideal position as the beneficiary of a promise of a retirement income for life, if the employer could not deliver on that promise, the participant was out of luck.
Several major pension defaults—most famously, that of the Studebaker car company in 1963—led Congress in 1974 to enact ERISA. ERISA wrote into law a number of policies—for example, with respect to minimum standards for retirement plan design, fiduciary conduct and participant disclosure—that we will be discussing in detail in subsequent sections. But at the time, and for purposes of our discussion of DB plans, the most important policy innovation introduced by ERISA (and, in many respects, its centerpiece) was a new set of rules for DB plan benefit security and plan finance.
The Pension Benefit Guaranty Corporation
ERISA created a federal corporation—in some respects similar to the Federal Deposit Insurance Corporation—to insure pension benefits, the Pension Benefit Guaranty Corporation (PBGC). And it created a set of minimum funding standards.
It’s important to understand that, under ERISA and as a fundamental part of the “ERISA vision,” PBGC benefit insurance—which DB plan participants typically are unaware of except in the (generally infrequent) situation in which their employer goes bankrupt—is primary. ERISA’s minimum funding rules are, in effect, a solvency regime designed to limit the circumstances in which the PBGC will be stuck holding the bag.
It’s not even clear how much the original drafters understood this critical relationship. ERISA’s original minimum funding standards were quite lenient—allowing sponsors to fund “unfunded past service liability” (unfunded pension liabilities already on their books) over 30 years or, for plans in existence when ERISA was enacted, 40 years.
At the same time, PBGC insurance benefits were (relatively) generous, with a dollar limit of $750 per month. All DB sponsors were required to fund the PBGC by paying annual premiums. Initially, all plans paid the same (relatively low) premium of $1 per participant per year.
These new regulatory regimes introduced by ERISA—PBGC benefit insurance, funded by sponsor-paid premiums, the risk with respect to which was mitigated by minimum funding requirements—were intended to work together to secure the employer’s pension promise. And let’s note—because it’s an interesting moving part of all this—that if either sponsor premiums had been set high enough to adequately underwrite PBGC’s insurance risk, or the minimum funding requirements had been set high enough to guarantee that DB plans would have adequate assets to meet the promises they represented, then this system would have worked.
Indeed, it did in fact work for the most part. Nevertheless, it’s fair to say that as originally designed ERISA’s federal retirement income insurance scheme was inadequately underwritten. We are not going to rehearse the history of how the PBGC insurance system was abused by sponsors “laying off” liabilities on an inadequately financed PBGC. Suffice it to say that by the early-2000s there was a bipartisan consensus that something needed to be done.
The Pension Protection Act of 2006
Currently, DB plan benefit security and minimum funding are governed by the regulatory scheme introduced by amendments to ERISA in the Pension Protection Act of 2006 (PPA).
In what follows we’re going to briefly summarize that regulatory scheme—critically, the PBGC premium and minimum funding rules applicable to plan sponsors and designed to secure the DB benefit promise. But as we do so, let’s not lose track of the conceptual “ball” we’re following.
DB plans emerged in the second half of the 20th century as the primary private employer retirement arrangement as a “cheaper” way to finance the delivery of retirement annuities to employees. Under these plans, employers bore all the financial risks, but the employee bore the one great risk that the employer might not deliver on its promise. Federal regulation was introduced to secure that promise, through a scheme of federal benefit insurance, financed by sponsor-paid premiums, the risk with respect to which was mitigated by federal minimum funding requirements.
As we just said, the original (1974) version of this federal program to secure private pensions was inadequately underwritten. And, even after 30 years of tweaks to it, there was a bipartisan consensus that it was still dangerously inadequate.
Plan A: Tighten the Minimum Funding Rules
There were (as we have noted) two different ways to address this inadequacy. Policymakers could either increase PBGC premiums, so that PBGC’s insurance was adequately funded. Or they could tighten ERISA’s funding requirements, so that fewer plans failed.
Policymakers (initially at least) chose the latter approach.
Radically oversimplifying, rules for minimum funding have two major moving parts:
- The valuation assumptions used to determine the present value plan liabilities. The critical assumption in this regard is the interest rate used to discount plan liabilities, but assumptions about mortality—how long participants will live and thus how long benefits will be paid to them—also matter.
- The amortization period: how fast a plan funding shortfall must be paid off.
With respect to moving part (1) (valuation assumptions), prior to the PPA, a plan’s actuaries were only required to use assumptions that were “reasonable,” allowing plan actuaries some latitude in how they valued liabilities. The PPA changed this, requiring that the actuary use a valuation interest rate assumption derived from a high-quality corporate bond yield curve and mortality assumptions specified by the IRS.
We reviewed the implications for liability valuation of using a market interest rate yield curve in the preceding chapter. The principle “tweak” to this approach introduced by the PPA was to allow sponsors to average market interest rates over 24 months. Generally, this rule was intended to smooth out short-term interest rate spikes, but in the context of declining interest rates, it also has had the effect of marginally increasing the valuation interest rate relative to market rates.
With respect to moving part (2) (amortization period), the PPA required that (after a phase-in period) funding shortfalls be paid off over a relatively tight schedule of seven years.
Mugged by the Financial Markets
The PPA was adopted in 2006, to be effective for 2008, and requiring, for many employers, under its new, tighter funding rules, substantial increases in contributions to their defined benefit plans for 2008.
In September 2008, what has come to be called the Global Financial Crisis struck the US and international markets. Equities collapsed—as we noted, the S&P 500 lost 37% of its value. Over the next several years, as the Federal Reserve embarked on its Quantitative Easing program, medium- and long-term interest rates also decreased dramatically.
These two markets—for equities and, most significantly, interest rates—were, under the PPA, the critical determinants of a sponsor’s net funding position and its minimum funding obligation. With interest rates crashing, cash-strapped employers were faced with the prospect of huge increases in the cash funding requirements for their pension plans. Some employers believed they were faced with a choice between laying off workers or defaulting on their (legal) obligation to fund their plans.
… And by Congressional Budget Exigencies
The cash squeeze that the triple whammy of the PPA, the Global Financial Crisis and Fed Quantitative Easing put on DB plan sponsors also had consequences for the Congressional budget—which was also being squeezed mercilessly by the economic slowdown. The increased pension contributions mandated by the PPA meant increased tax deductions for contributing sponsors, draining tax revenues.
This is another theme we are going to come back to again and again: many Congressional decisions with respect to retirement savings policy are made based on budget concerns, without regard for the consequences to retirement savings.
Plan B: Relax Minimum Funding Standards and Increase PBGC Premiums
In some ways, what happened next was inevitable. Nevertheless, few foresaw it.
It turns out that, for purposes of the Congressional budget, PBGC premiums have one transcendent virtue—notwithstanding that they are only available to pay the unfunded pension benefits of insolvent pension plans, they count as tax revenues for the general budget. As a bonus, they aren’t a “tax” and therefore don’t offend Republicans.
And, fortuitously, loosening ERISA minimum funding requirements actually increases tax revenues—as sponsors contribute less and therefore deduct less.
The idea of combining these two policies captured the imagination of policymakers, and in a series of bills—funding highway legislation and closing year-end budget gaps—over the period 2012–2016, Congress proceeded to dramatically increase PBGC premiums, especially on underfunded plans, while dramatically reducing funding requirements, by allowing plans to use 25-year (rather than 24-month) interest rate smoothing.
The Current Minimum Funding Regime
Currently, and through the year 2020, the valuation interest rate used to determine the present value of plan liabilities is 90% of the average of rates (on high-quality corporate bonds) over the prior 25 years. Because—as we said and will discuss in detail below—interest rates have been declining since 1982, use of a 25-year “backwards-looking” average of rates produces significantly higher rates than current market rates.
Let’s go back to the example we used in the last chapter. At the current market interest rate (4.5%), the present value of an obligation to pay $1,000 25 years from now is $333. 90% of the 25-year trailing average of those rates is 7.92%. Using that 7.92% rate, the value of that $1,000 liability (payable in 25 years) is only $150, a more than 50% reduction in value.
The reason 25-year averaging of rates produces such an artificially high valuation interest rate is—as we just noted—the long-term decline in interest rates since 1982. Going back 25 years means averaging in double-digit interest rates.
We will return to the problems presented by the long-term decline in interest rates in detail in Chapter 8.
Under current rules, the use of artificially high interest rates to value liabilities is phased down (to 70% of the 25-year average), over the period 2021–2024. Thus, it is likely that, by 2024, plans will be back to using market interest rates (subject to 2-year smoothing) to value liabilities for purposes of ERISA minimum funding requirements.
The Current PBGC Premium Regime
Under current rules, all corporate DB sponsors pay an annual per capita (per participant) premium to the PBGC of (for 2018) $74.
Sponsors of underfunded plans generally (and oversimplifying) pay a premium equal to (for 2018) 3.8% of unfunded benefits. This “variable-rate” premium obligation is, however, subject to a cap equal to $523 times the total number of participants in the plan.
Remember, when ERISA was enacted, the only premium sponsors paid was $1 per participant.
Not the Worst Result
Whatever sort of muddle the policy changes we have just traced present, the solution Congress hit on does represent a legitimate policy alternative. As we said above, the DB system could be made secure by either mandating solvency via increased contributions by employers to their plans or by increasing the premiums those employers paid to the PBGC.
The latter strategy—aside from the (somewhat fortuitous) benefits it produced for the federal budget—had the effect of encouraging those employers who could do so to fund benefits rather than pay the dramatically increased PBGC premiums on unfunded benefits. Also, as we’ll see, and less benignly, it has led many employers still sponsoring DB plans to look for ways to get out of the system.
Managing PBGC Premiums, Not Funding, Now Drives Policy for Many Employers
What just happened? What started out as a funding solution to the benefit security crisis confronting policymakers in the early 2000s morphed into a PBGC premium solution. As a result—as we will discuss in detail below in our section on strategies for managing DB liabilities—sponsor focus has shifted from managing the cash demands of DB finance to implementing strategies to reduce the PBGC premium tax.
Federal Regulation of DB Finance, in Sum
We have just sketched the principal regulatory concern preoccupying policymakers and regulators with respect to defined benefit plans: the security of the employer’s pension promise. That concern is generated by the basic fact that, while the employer generally bears all risk under a DB plan, the employee-participant bears the (at times catastrophic) risk that the employer will not deliver on its promise.
The solution developed under ERISA and “perfected” by the PPA was to create a benefit insurance system (the PBGC), funded by premiums paid by plan sponsor, and to mandate plan funding at a minimum rate.
These rules increased (perhaps not unjustifiably) the burden DB plans imposed on sponsors.
The Accounting Regime, Very Briefly
Before we move on to consideration of federal minimum standards for participation, vesting and accrual and Tax Code nondiscrimination rules, we must consider another, critical element of DB plan finance: accounting.
Let’s begin with a first principle: that a DB plan represents a legal claim against the sponsor’s assets and income. If the plan does not have enough assets to pay liabilities, the sponsor is—absent bankruptcy and liquidation—obligated to make up the shortfall. As such, the plan has an effect both on the sponsor’s cash position, its ability to pay liabilities when due, and its profitability, the net of its income and expenses.
Cash—money coming in and going out—matters to a company because, if it doesn’t have enough cash, a company seizes up. A DB plan sponsor with cash “issues”—concerned about the ability to meet payroll or pay the next loan installment—may be concerned with the cash demands of ERISA’s minimum funding standards that we have just discussed.
But a company’s cash flow does not reveal much about how “good”—how profitable—the company’s business is. To determine that, the analyst needs a notion of the company’s income and expenses.
Generally, in normal times at least, there are more companies that are concerned with the latter issue—the company’s profitability—than there are concerned about the former—the company’s ability to meet immediate cash demands.
Expense, in this context, is an accounting concept, in which actual or anticipated obligations of the company are matched to the right period. Generally, and oversimplifying, a company’s income statement shows its profits for a period as the net of income and expenses for that period. Thus, companies that care about profitability care about the effect of DB plan sponsorship—and the DB plan expense—on their income statements.
The expense chargeable for any given period with respect to an employer’s DB plan is strictly a function of accounting rules—for a public company, the rules adopted by the Financial Accounting Standards Board.
We’re not going to review FASB DB accounting rules in detail. But, because these rules probably have been a bigger factor in the disappearance of DB plans than, e.g., the ERISA regulatory regime or the recent dramatic increase in PBGC premiums, we are going to (briefly) summarize them and discuss how they have affected corporate retirement policy.
How DB Liabilities Are Valued for Financial Statement Purposes
As we discussed briefly in Chapter 3, FASB uses a discount rate derived from a high-quality corporate bond yield curve to value plan liabilities. Thus, a plan sponsor’s balance sheet shows a mark-to-market value of its DB plan’s liabilities (priced at current interest rates) minus the market value of the plan assets. Expense (aka the “cost” of the DB plan for the relevant period) is determined based on this mark-to-market number, but subject to several rules that “buffer” the effect of short-run changes in valuation (due, e.g., to a significant decrease in valuation interest rates). Those “buffers” include:
–Smoothing of, and credit for an expected rate of return on, asset values.
–Limited recognition of gains and losses (resulting, for example, from short-run changes in interest rates).
–Delayed recognition of gains and losses and “prior service cost.”
As we’ve indicated, the issue of cost and its effect on the income statement is probably more significant than the issue of net under/over-funding, because earnings (as in “earnings per share”) represent the basis for most corporate valuations. Earnings also frequently determine executive bonuses. This dual significance probably explains why it has been harder for FASB to move to a “true” mark-to-market regime with respect to earnings than it has been with respect to net under-/over-funding.
Whether for good or ill (and there are those who see it both ways), as a practical matter the market-based elements in FASB’s corporate pension accounting regime were enough to fundamentally change corporate retirement policy. FASB rules—however imperfect—have, in effect, translated the secular decline in interest rates (about which we will have more to say in Chapter 8) into losses on corporate financial statements. With a consequent intense dis-affection for DB plans in the CFO’s office.
* * *
All of the foregoing represents a (in some respects very general) summary of:
–The current state of DB plan finance.
–How, in general, these plans work and the risks they present for the sponsor, the participant and other stakeholders.
–A regulatory framework that is designed to secure the pension promise for the participant under a system of PBGC benefit insurance guarantees, financed by sponsor premiums, the risk with respect to which is mitigated by a minimum funding regime.
–An accounting framework that transmits—pursuant to a relatively conservative valuation regime—the costs and liabilities of the pension plan onto the sponsor’s balance sheet and income statement.