Recently there has been substantial press attention paid to underfunding in state and local government pension plans, a longstanding problem made more urgent by recent troubles in the larger economy. There has of yet been comparatively less attention given to a similar (though smaller) set of mounting financial risks associated with private-sector worker pensions covered by the Pension Benefit Guaranty Corporation (pbgc). Yet here, too, public policy corrections are required to address underfunding and avoid another taxpayer-financed bailout we can ill afford.
pbgc’s latest annual report shows a net deficit of over $23 billion, of which roughly $21.6 billion is attributable to its single-employer insurance program. What’s worse, pbgc estimates its exposure to reasonably possible plan terminations at approximately $170 billion. While these figures may appear small relative to the potential losses in state and local government plans, they reflect underfunding in employer-provided plans that is proportionally comparable.
Elected officials face a fundamental value choice about whether employer-provided pensions should be funded and insured only by plan sponsors, or whether the dollars of others (i.e., taxpayers) should be tapped to fill the gap. Either policy requires substantial changes to existing law. If the former policy is not enforced so that pbgc is kept solvent, the approaching cost of the latter course should be disclosed. Each scenario requires unattractive actions but is nevertheless preferable to the current situation, in which taxpayers face a growing financial risk that is disguised rather than forestalled by the existence of the pbgc.
The financing shortfall
Employer-provided, defined-benefit pensions (both single-employer and multiemployer) are insured by the pbgc, a federally chartered corporation. When an insured pension plan terminates, the pbgc assumes the plan’s assets as well as the responsibility for paying promised benefits up to a cap set in law.
To simplify the pertinent issues, I will focus primarily on the single-employer pension insurance system, where the vast majority of the pbgc financing shortfall is concentrated. Its current $21.6 billion deficit is nearly the largest on record, falling just behind those posted in 2004 and 2005. While the size of the deficit fluctuates from year to year, it has remained persistently significant over most of the last decade (that is, more than $10 billion in each year since 2003, inclusive). Indeed, in each of the last eight years pbgc’s liabilities have been measured as being at least 15 percent larger than its assets, and usually much more.
These figures don’t capture the full extent of pbgc’s potential difficulties, as evidenced by its estimate of “reasonably possible” termination exposure — i.e., underfunding in plans with below-investment grade credit ratings — at roughly $170 billion. pbgc ’s multiemployer program is also presenting increased risks; in 2010, “reasonably possible” exposure in multiemployer plans suddenly rose from roughly $300 million to approximately $20 billion.
The financing hole in America’s pensions is now of such a size that we cannot expect it to close unless someone pays substantially greater costs out of pocket than is currently the case. In October 2009, pbgc provided congressional staff with estimates indicating that to comply with existing law, pension contributions by plan sponsors would need to rise from roughly $50 billion in 2008 to more than $250 billion annually by the mid-2010s. Others’ projections show considerably smaller figures but nevertheless agree that substantial future contribution increases will be required. Recently enacted funding relief could delay the full effects of these requirements, but that does not change the reality that markedly increased contributions will ultimately be necessary.
A critical public policy question, therefore, is to what extent the cost of filling the shortfall is to be met within the current pension insurance system, and to what extent risks and costs will be shifted to those now perceived to be outside of it.
Why the inadequate finance?
The pension insurance system was weakened when recent financial market declines both depressed pension-plan asset values and undermined the financial health of plan sponsors. From 2008 to 2009 the size of pbgc’s net deficit roughly doubled, in part due to a decline in interest rates (which increased the present value of plan liabilities), and also due to an increase in plan terminations. In 2009, pbgc reported that it had become responsible for paying benefits to more than 200,000 additional workers, the third-highest increase in its history and roughly nine times the number of new participants for which it had assumed responsibility in 2008. Over the same year, pbgc’s exposure to “reasonably possible” terminations drastically increased, from $47 billion to $168 billion.
Though this represented a sharp downturn in pbgc’s outlook, its finances were on an unsustainable course even when financial markets were at their peak. The reasons were rooted in longstanding flaws in federal pension law. Although the 2006 Pension Protection Act (ppa) addressed many of these problems, their effects continue to permeate the pension system. The ppa provided up-front pension funding relief to most plan sponsors while establishing tighter funding targets for the long term. But many of the ppa’s critical reforms are being phased in only gradually, and its near-term funding requirements were further relaxed in subsequent legislation. Regardless, any positive impact of the ppa is being swamped by the financing deterioration resulting from worsened market conditions.
The following is a capsule list of factors that have contributed over time to pension underfunding. More details about all of these factors can be found in Pension Wise (Hoover Institution Press, 2010).
1. Inaccurate measurements of plan assets. Plan underfunding can only be successfully addressed to the extent that contributions are based upon accurate, up-to-date measurements of plan assets, liabilities, and any gap between them. Prior to the ppa, sponsors could “smooth” asset valuations by blending more recent values with older valuations from the previous four years. Even under the ppa, the plan sponsor still has two years to fully recognize any discrepancy between a plan’s actual and its previously projected plan assets, though assets cannot be valued more than 10 percent differently than current market value. Such smoothing has often delayed recognition of underfunding, causing delays in contributions that would otherwise be required. Smoothing was often rationalized as being necessary to limit the volatility of and to prevent pro-cyclicality in contribution requirements. This rationale, however, conflated two separable concepts — the accuracy of pension funding measurements on the one hand, and the policy for determining contributions on the other.
2. Inaccurate measurements of plan liabilities. An accurate projection of plan liabilities requires both that future benefit payments be accurately estimated and that they are appropriately discounted into their present value. Accurate benefit payment projections in turn require accurate estimates of the number of individuals who will be receiving benefits, the amount and form of those benefits, and the time over which they will be paid. The ppa required plan sponsors to use updated mortality tables, addressing a problem that had long persisted before the legislation. The ppa also now accounts for the empirical phenomenon of the “rush to the exits” — i.e., the likelihood that individuals in a troubled plan will claim benefits at the earliest possible age and in a lump sum when available (both choices drain a plan of assets in the near term). The ppa’s transition period, however, has the effect of postponing recognition of most of these “at-risk” plans. As for liability discounting, what must be assessed is the degree of risk in pension benefits as well as the time period over which they will be paid. Current law discounts these liabilities according to a yield curve of corporate bond rates, which the sponsor has the option of smoothing over two years. A corporate-bond yield curve is potentially an appropriate means of discounting. It is, however, appropriate only to the extent that worker benefits are at risk whenever corporate financial health is also at risk; that is, when benefits are not backstopped by the taxpayer. Furthermore, it is accurate only to the extent that smoothing and other interest rate specifications do not introduce a distortion of up-to-date market conditions. Additionally, the ppa also allowed certain politically favored industries (e.g., airlines) to employ arbitrarily higher discount rates in their pension funding calculations.
3. Inadequate funding targets. A core principle of the ppa is that a pension plan’s appropriate funding target is 100 percent, with plan sponsors being given seven years to amortize contributions to address any shortfall. The ppa did not establish the 100 percent funding target immediately, however; even the first sponsor payments pursuant to full funding were not to be required until at least five years after the ppa’s enactment. Moreover, additional funding relief was enacted in 2010 that, while it did not waive the 100 percent funding target, reduced contribution requirements in relation to it. While it is understandable that Congress would choose to provide additional funding relief at a time when plan sponsors face unusual economic difficulties, the relief took immediate effect in a funding environment that was already quite weak.
4. Unfunded benefit increases. Prior to the ppa, employers had the latitude to increase benefit promises without fully funding these increases before a plan was assumed by the pbgc. This had the effect of worsening underfunding in some terminating plans. The ppa clamped down on such practices, imposing various limitations respectively upon lump sum payments, unfunded benefit increases, and in some cases even benefit accruals, though these safeguards have yet to fully take effect.
5. Loopholes and special preferences. Pre-ppa law contained enormous loopholes, many of which made significant contributions to pension underfunding. Among these was the “credit balance” rule. Specifically, any contribution made in excess of minimum statutory requirements was not only counted once in adding to a plan’s funding percentage but a second time in reducing the amount of future contributions otherwise required. Moreover, plan sponsors were permitted to assume that the credit balance had appreciated in value even if the actual investment had not. The ppa corrected the worst abuses of credit balances but permitted the basic concept to persist. It did not, however, reform various special statutory preferences for politically favored industries. It explicitly permitted airline plan sponsors to use longer amortization periods and to employ a higher discount rate to artificially shrink the size of plan liabilities. More recently, one domestic automaker was also permitted, while receiving government assistance, to “top up” affiliated union workers’ pension benefits, to circumvent the spirit of pbgc’s statutory benefit caps.
6. Inadequate premiums. Unlike other federal insurers (such as the fdic), the pbgc does not have the power to levy premium assessments sufficient to pay for the cost of the insurance that it provides. In 2005, the Congressional Budget Office calculated that pbgc premiums would need to be roughly 6.5 times higher to fund the size of anticipated claims. Equally important, not only is the level of premium income inadequate but premium assessments do not reflect the level of risk in a pension plan, arising from such factors as its investment portfolio, its funding percentage, and the health of its sponsor. Earlier this year, the Obama administration proposed that the pbgc be given increased authority to modify both the level and structure of premiums that sponsors are charged for pension insurance.
7. Limitations upon the national pension insurer (PBGC). The pbgc has only limited power to enforce contribution requirements. In one example from recent years, one airline sponsor simply stopped making statutorily-required contributions to its plan once the sponsor entered bankruptcy. As an unsecured creditor, pbgc was unable during bankruptcy proceedings to perfect a lien placed against the skipped contributions. pbgc also generally lacks the authority to regulate the investment policies of plan sponsors, or otherwise prevent them from taking actions that degrade its financial position. Moreover, pbgc has an ambiguous position in relation to the Department of Labor, in some respects appearing to sit within it, in others appearing to be independent. This occasionally calls into question pbgc’s latitude to take fully independent actions to defend pension plan funding. This is particularly problematic in circumstances where a presidential administration chooses to advance a conflicting policy priority, whether that competing priority involves providing direct taxpayer-backed assistance to the automotive industry or attempting to stimulate near-term job creation at the expense of pension funding.
8. Inadequate disclosure. Transparency and disclosure are effective spurs to stronger pension funding. Present funding disclosure requirements can only be described as inadequate. The 4010 form (pbgc’s primary source of detailed plan funding information) is only required of plans deemed less than 80 percent funded. This has the dual adverse impacts of both limiting information about some large plans that may pose a large aggregate risk to the pbgc and of stigmatizing the comparatively few plan sponsors who must file the information.
9. Barriers to funding up during good times. During the late 1990s, annual contributions to pension plans shrank (and in some cases were eliminated altogether) as plan assets rose during the stock market’s dot-com bubble. After the bubble burst, plan funding percentages declined to lower levels than they would have if sponsors had “overfunded” during good times. Pre-ppa law constrained the extent to which plan sponsors could do so, limiting the tax-deductibility of excess pension contributions to 100 percent of the plan’s “current liability.” The ppa increased employer flexibility to fund up during good times, allowing sponsors to fund to 150 percent of their target liability plus expenses based on future salary increases (a long-term reform of limited applicability in a recession environment). Current federal budget rules also remain a problem. Because employer pension contributions are tax deductible, the federal government can still “raise revenue” by reducing employer pension contributions, a tactic used as a budgetary offset for added federal spending in 2010. A better budget framework would recognize worsening pension underfunding as an increase in potential taxpayer exposure, rather than treat it as a cost-free source of financing for added federal spending.
10. Moral hazard and political economy factors. Current funding requirements and premium assessments are determined not by economic realities but rather by a political process. Elected officials remain under constant pressure to relax contribution requirements so that employer funds can be made available for new hiring or for more immediate forms of compensation (e.g., wages). Similar pressures are common throughout existing defined-benefit systems and have caused significant underfunding to arise in each of employer-provided pensions, state and local pensions, and in Social Security.
11. Periodic contribution relief. One specific manifestation of the moral hazard inherent in the current pension system is the recurrent congressional behavior whenever previously enacted funding requirements begin to bind “too tightly.” In the early 2000s, funding relief was first provided by allowing sponsors to discount using a widened corridor around Treasury bond rates, and then later by shifting to higher corporate bond rates. In 2006, the ppa provided additional near-term funding relief in exchange for a tightening of long-term funding standards. More recent funding relief has repeatedly been justified by difficult economic conditions. While the funding relief may have been defensible in each of these individual circumstances, the overall pattern has been unchanging; Congress has repeatedly shifted the risks of pension underfunding from plan sponsors to the pension insurance system as a whole.
Common moral hazards
There is periodic but largely separate public attention to the funding issues facing Social Security, state and local pension plans, and employer-provided defined-benefit pensions. All of these systems are underfunded; all suffer from inadequate accounting; and all face analogous moral hazards. Yet there has been comparatively little attention to the common factors driving underfunding in all three.
While defined-contribution systems also face their challenges, a fundamental misalignment of interests is not one of them. A worker’s ultimate defined-contribution benefit is a direct function both of the adequacy of plan contributions and of the rate of appreciation on investments. It is clearly in the worker’s interest to see that contributions are sufficient, that they are profitably invested, and that administrative expenses are minimized. If any of these elements is poorly handled, the worker’s ultimate retirement income will diminish. While the worker may lack sufficient information or education to assess these factors, the worker’s interest is at least straightforwardly aligned.
In a defined-benefit system, however, the interests of different actors diverge. Because the sponsor, rather than the worker, accepts the funding risk, it is in the sponsor’s interest to minimize his costs in providing a given benefit, while it is only in the worker’s interest to maximize his chance of receiving the full benefit, irrespective of who pays for it. The presence of pension insurance injects substantial moral hazard into this dynamic. When pension insurance is present, a plan sponsor can potentially reduce funding costs by investing in higher-risk securities (unless specifically prevented by regulation from doing so). If the upside returns are received, the sponsor’s pension liabilities are reduced; if the downside risk is realized, the pension insurance system is there to potentially absorb the loss. Pension insurance at the same time reduces the worker’s incentive to verify that pension contributions are both adequate and prudently invested.
The same problematic incentives exist in public-sector plans, including both state and local pension plans and the federal Social Security program. Elected officials responsible for controlling the finances of these plans have powerful incentives both to minimize costs faced by taxpayers while maximizing benefits paid to program participants. This leads naturally to two predictable results; first, to aggregate underfunding (an excess of promised benefits over contributed revenues); second, to rising pay-as-you-go obligations (that is, an escalation of obligations that must be financed by future contributors at the moment of benefit payment, as opposed to having been prefunded).
The adverse funding consequences of these hazards have resulted in roughly one-fifth to one-quarter of future benefits being unfunded in each defined-benefit sector, depending on the particular estimate. State and local plans have frequently employed aggressive liability discount rate assumptions that reduce near-term funding and increase the share of benefit payments that must be funded by future taxpayers. Similarly, the federal Social Security program continues to be operated essentially on a pay-as-you-go basis despite widely recognized population aging, a long-predicted decline in the ratio of workers to beneficiaries, and the recommendations of multiple bipartisan technical panels and advisory boards to shift toward partial prefunding.
Although the pbgc insures private-sector pensions, its operations are also greatly affected by political economy factors. Both premium assessments and funding requirements are established via a political process. The pbgc is not permitted to perform the equivalent of a private-sector insurer’s pricing of insurance coverage, nor does it have the legal authority to change premium assessments as financially necessary. And each time that legislators must vote on a bill to determine premium levels or funding requirements, they are subject to enormous political pressure to relax each of these relative to what is required to achieve a fully funded pension system. Unless the process itself is changed, this phenomenon should be expected to continue.
A striking parallel between the three major forms of defined-benefit pensions is the extent to which near-term funding relief has often been provided through the manipulation of asset and liability measurements in all three sectors. Asset and liability smoothing, artificially high discount rates, credit balances, and other techniques have delayed recognition of underfunding in employer-provided pensions. In state and local plans, the chief accounting tool by which liabilities have been understated has been the use of aggressive liability discount rate assumptions. In Social Security, Trust Fund accounting is the primary culprit, where Trust Fund assets are deemed to reduce future funding shortfalls even though these assets are simply a further debt obligation facing federal taxpayers. A classic example of the manipulation of this accounting occurred last December, when Congress cut payroll taxes by over $100 billion for 2011 but stipulated that additional debt would be issued to the Social Security Trust Funds just as though the additional taxes were being collected.
In each of these defined-benefit systems, the ongoing battle over transparent accounting is fought along predictable interest group lines. In the employer-provided pension world, plan sponsors often argue for asset/liability smoothing and higher discount rates, while the opposing case is made by those responsible for the health of the pension insurance system. With state and local plans, state officials and public employee unions both advocate high discount rates, while pushback comes from academics, taxpayer-watchdog groups, and federal officials concerned about pressure for a federal bailout. In Social Security, the beneficiary-advocacy group aarp has defended current Trust Fund accounting methods, while counter pressure has come from bipartisan technical expert panels and fiscal watchdog groups. Unfortunately, in each of these cases, those who have argued to disguise funding inadequacy have generally carried the day, contributing to the significant funding shortfalls in all three systems.
Avoiding a taxpayer bailout of pbgc
Regardless of the future structure of pbgc, certain reform principles should be observed. Specifically:
1. Limit asset smoothing to the extent practicable. Pension funding policy cannot be sensibly constructed if it is not built on a foundation of measurement accuracy. The only properly imposed limitation on the accuracy of asset measurements should reflect limits on practicability for plan sponsors. If requiring an asset measurement on a specified date proves to be particularly disadvantageous for a plan sponsor, this outcome is best avoided by allowing the sponsor to select from valuation dates within a narrow time window. To the extent that minimizing smoothing causes real asset volatility to be formally recognized, funding rules should be designed to limit volatility in annual required contributions — rather than relying on smoothed asset valuations. Asset and liability valuations should be conducted to produce the most accurate information about plan finances, not to steer toward a particular funding policy result.
2. In liability measurements, limit smoothing, use up-to-date mortality tables, and account accurately for worker behavior. Like asset measurements, liability measurements should be as accurate and up-to-date as is practicable. Specifically, liability measurements should reflect the most up-to-date mortality tables as well as the best available information about how and when workers are likely to claim benefits. This in turn requires that the pension insurer is able to ascertain the risk of a “rush to the exits” due to the sponsor itself becoming a credit risk. As with asset valuation, periodic fluctuations in liability discount rates are better handled by allowing a narrow time window during which a sponsor can choose a discount rate, rather than by smoothing with long outdated discount rates.
3. Eliminate special deals for politically-favored industries. Rules of play should be applied fairly across the board. There should be no special exemptions for politically favored industries as exist under current law. A private insurer would not be allowed to discriminate between covered entities based on political preference; the public pension insurance system should be bound by a similar ethic.
4. Enforce prohibitions on unfunded benefit increases by underfunded plans. Even where plan sponsors are suffering from economic forces beyond their control, little good can arise from permitting plan sponsors to promise benefits that they cannot pay. Recent funding relief legislation has permitted plan sponsors to escape restrictions on unfunded benefit increases by allowing funding valuations to be made based on pre-recession conditions. However faultless a sponsor with respect to the broader financial market decline, it should not then take the further irresponsible step of making further benefit promises that it cannot fund.
5. Increase disclosure of funding information. Sunshine and transparency are useful spurs to stronger pension funding. The 4010 submission requirements should be expanded to provide pbgc with the information necessary to gauge the likelihood of further financial hits on the pension insurance system. This would also help to destigmatize those plan sponsors who provide it. To prevent small employers from undue burdens and to focus information on the largest potential threats facing the pbgc, 4010 filing requirements should be based on aggregate plan underfunding rather than on a funding percentage. This filing threshold should be reestablished at a level low enough to enable pbgc to see at least the majority of the underfunding in plans for which it is reasonably likely to become responsible.
Three frameworks for reform
Elected officials should choose between three basic options for comprehensive reform of the employer-provided pension insurance system.
Option #. If pbgc is to remain financially viable without taxpayer funds, it must be provided with the resources required to close its funding shortfall. These must include enhancements of its financial, legal, and organizational tools. Among them:
1. Premium income. An insurance system cannot survive if it cannot adequately charge for the cost of insurance provided. The pbgc currently lacks the authority to charge premiums that are sufficient to fund anticipated claims. President Obama has recently proposed that pbgc be given the authority to raise premiums as necessary to prevent a government-financed rescue. Different methods of doing so are available that reflect different potential value choices. A higher flat-rate premium for all participants would reflect a judgment that the cost of filling the shortfall should be spread among the greatest number of plan sponsors (including those with well-funded plans). Alternatively, greater reliance on risk-based premiums would reflect a judgment that higher costs should be imposed on those sponsors who pose the greatest risk to the insurance system. A truly risk-based premium would provide greater incentives for full funding and should thus be a part of any solution. At the same time, it is unlikely that weak sponsors of underfunded plans will themselves be able to provide sufficient additional revenue to fill pbgc’s financing hole, necessitating an increase as well in the flat premium paid by all plan sponsors.
2. Adequate funding rules. The law should be revised to undo recent “complexity creep” and to once again provide all plan sponsors with a single amortization schedule (with 100 percent funding as the ultimate funding target). Under almost any scenario, future contributions by plan sponsors will ultimately need to be much larger than they have been in the recent past. Interim funding relief should therefore be limited only to that required to offset recent economic events, and not aim to avoid the funding increases eventually required even under more typical economic conditions. One way of implementing this would be to adopt a single “2 + 7” policy for everyone (two years of interest-only payments, followed by seven-year amortization). Under this framework, Congress would also repeal the special discount rates and amortization schedules for the airline industry, as well as other targeted exceptions to the funding rules.
3. Strengthened legal tools for PBGC. pbgc cannot adequately protect the pension insurance system with its current set of crude legal tools. Bankruptcy code changes could be enacted to give pension plan contributions the same status as wage payments, to move pbgc ahead in line over other unsecured creditors. pbgc could also be given an authority comparable to that of the fdic, to implement “cease and desist” orders to prevent plan sponsors from taking actions injurious to the health of the pension insurance system.
4. PBGC Independence. To fully empower pbgc to close its own funding shortfall Congress would clarify in law that pbgc is an independent regulatory agency and no longer a part of the Department of Labor. The current pbgc can never be wholly independent of White House direction because of the constitutional “unitary” nature of the federal executive branch. It could, nevertheless, wield a degree of de facto independence roughly comparable to that of the fic or Securities Exchange Commission. The goal would be to free pbgc to make determinations predicated exclusively on the health of the pension insurance system, without conscious subordination to the broader economic policy objectives of the presidential administration.
In sum, the basics of the approach in Option #1 are to provide pbgc with the necessary funding rules, premium income, legal tools, and organizational standing to address its own financing shortfall. If Congress is unwilling to permit this within the present pbgc system, it will need to choose between other alternatives.
Option #2: Eliminate PBGC and replace it with compulsory private insurance that would charge market rates to participants. Option #2 differs from Option #1 primarily in the structure of the pension insurance system; specifically, whether it is a compulsory system of private insurance or continues to be a government-chartered system. Certain core elements of reform would be implemented in Option #2 as they would be in Option #1:
1. Premium reforms. Just as with a strengthened pbgc, a private insurer would need to charge higher premiums than have been imposed to date and would likely increase its reliance on risk-based premiums.
2. Adequate funding. In a private insurance system, adequate funding would be facilitated not via government prescription but through the general requirement that the new system be self-financing.
3. Strengthened legal tools. Whether public or private, a pension insurer could be provided with stronger legal tools in bankruptcy proceedings, such as a higher claim priority given to pension plan contributions.
4. Independence from other government departments. This would be implicit in any private insurance model.
Choosing between Options #1 and #2 is therefore primarily a value judgment as to whether the above reforms are more likely to occur within a government-chartered or a privately administered system. If the conclusion is reached that persistent legislative tinkering will always prevent reforms under the current pbgc structure, then the private insurance model must be considered as an alternative.
The premium structure for pension insurance is of particular relevance when weighing whether to continue the pbgc or to transition to a private-sector model. A private insurer would be expected to assess premiums based in large part on its evaluation of risk, including sponsor default risk, underfunding risk, and asset-liability mismatch risk, among other factors. The public sector has generally shown an unwillingness to implement true risk-based pricing.
An important policy specification for any private-sector successor to pbgc involves whether it must not only sustain its future operations, but also carry the burden of an inherited pbgc deficit of roughly $23 billion. To require pension plan sponsors to make up this deficit is in effect to require that their future pension insurance costs are $23 billion higher than the value of the insurance itself. Some advocates of private pension insurance have therefore proposed that taxpayers shoulder the cost of financing the current pbgc deficit so that subsequent compulsory pension insurance can be given a “clean slate” free of legacy debt.
Though pbgc’s legacy debt is a thorny problem for any transition to private insurance, it is important to understand that the problem is not unique to the private insurance model. Whether pension insurance is publicly or privately offered, the same policy dilemma exists: namely, whether to require pension plan sponsors alone to make up the pbgc shortfall, or whether to pass the cost of that shortfall to general taxpayers.
The most practicable path for transition to a private-sector system might be via a two-stage process: During the first stage, reforms would be implemented to drastically curtail the size of the pbgc deficit. Only in the second stage would a transition be effected to a private-sector framework.
There are many reasons why a taxpayer bailout of employerprovided pension insurance is undesirable.
Both stages of any such two-stage process would likely need to be established in law from the outset to permit successful conversion to a private-sector model. If, alternatively, reforms succeeded in eliminating the pbgc deficit without being an explicit component of a private-sector transition plan, the fiscal improvement would itself eliminate much of the political momentum toward such conversion. Without establishing an explicit two-stage process, it appears very unlikely that such a transition would ultimately be facilitated.
Option #3: Unless and until the federal government requires that PBGC’s financing risks be resolved via Options #1 or #2, treat the shortfall for federal budgetary purposes as an obligation facing U.S. taxpayers. The policy objective underlying Options #1 and #2 is to enable the pension insurance system to meet its obligations without an infusion of taxpayer-provided revenue. If, however, policymakers are unwilling both to require and to empower the pension insurance system to be self-sustaining, there exists a substantial risk that taxpayers will ultimately be called upon to resolve the shortfall.
There are a number of reasons why a taxpayer bailout of employer-provided pension insurance is an undesirable policy outcome. First and foremost, it would be inequitable. Approximately 21 percent of American workers have been promised defined-benefit retirement income by their employers — meaning that the other 79 percent have not. To require these taxpayers to bail out pension promises made by others’ employers is to require the vast majority of taxpayers to subsidize a benefit that only a minority is eligible to receive.
Beyond this, the mere potential for a taxpayer-financed bailout injects additional moral hazard into a retirement income system already fraught with it. Pension benefits represent a particular form of compensation promised by an employer to an employee. To the extent that resources other than employer funds are used to fulfill that promise, not only specific sponsors but employers as a class would be compensating their employees with third-party (taxpayer) money.
For general taxpayers to bail out private-sector defined-benefit promises also renders it difficult to draw a clear line limiting potential taxpayer exposure. If defined-benefit promises are to be made good by the federal government, why should there not also be relief to workers who have only defined-contribution accounts severely weakened by the recent financial markets plunge? Once the federal government goes down the road of assuming responsibility for the retirement income promises made in private employment, there is no obvious limit at which the potential costs will be contained.
No matter which of the preceding reform models is adopted, a contribution funding schedule will be required.
Finally, the federal government is itself already in dire fiscal circumstances. Among other obligations, the federal government is responsible for financing the benefits of its own Social Security retirement program, currently projected to be insolvent over the long term. The deficit of pbgc is smaller than the deficits in state and local public plans or Social Security, and could theoretically be financed more easily; it nevertheless lacks the direct claim upon taxpayer resources that the Social Security program has. Unless and until the federal government has an effective plan for financing the obligations it has already taken on, it is not in a position to be shouldering the retirement benefit obligations of the private sector.
For these and other reasons, a taxpayer bailout of pbgc should be regarded as a “doomsday scenario.” If, however, policymakers decline to enable the pbgc pension insurance system to be self-financing, they are creating a risk that this doomsday scenario will come to pass. Policy Option #3 is to disclose the full amount of this risk to taxpayers for as long as it persists.
The procedures for bringing the costs of systems like pbgc onto the federal budget are not free of ambiguity; scorekeeping decisions of the Congressional Budget Office and the Office of Management and Budget sometimes diverge. To appropriately show the full extent of taxpayer exposure to the obligations of pbgc, the mission of budget scorekeepers would be to assess a policy outcome in which taxpayers bear the risks of financing benefit payment obligations. Any net negative cash flows projected would be scored as a general revenue obligation facing U.S. taxpayers.
Merely reporting this information within federal budgetary documents would likely be an inadequate public warning of the potential cost of a taxpayer bailout of pbgc. To secure the public attention required to deflect momentum from a bailout, periodic separate reporting of the size of the pbgc obligations facing taxpayers would likely need to be required, in a manner similar to the annual reports of the Social Security and Medicare Trustees. This could be required either of cbo, the Office of Management and Budget, the General Accounting Office, or of pbgc itself.
Any such reporting requirement should disclose not only the potential cost to taxpayers but also the potential losses in worker benefits. Unless pbgc’s statutory cap on benefits is waived, even a “taxpayer bailout” would not result in taxpayers making up the entirety of the pension funding shortfall. The remainder of the gap would be resolved by the loss of all worker benefits exceeding pbgc’s statutory limit. Reporting these projected worker benefit losses could add substantial cogency to these periodic reports.
Scoring the potential cost of a taxpayer bailout would have the benefit of deterring the use of pension funding relief as a budgetary offset for new federal spending. In the past, Congress has “raised revenue” by reducing requirements for (tax-deductible) employer pension contributions, and has used the revenue increases to finance its additional spending desires. This practice could be deterred if an increase in the projected pbgc deficit was simultaneously scored as a direct spending obligation facing taxpayers.
Questions that must be answered
The various tools provided to the pension insurer in Options #1 and #>2 are designed to be sufficient to close its financing shortfall. This, however, does not imply that it would be optimal policy to require that the shortfall be made up immediately and all at once. Indeed, to require this in the current economic environment could well be self-defeating. It would require rapid increases in annual contributions to pension plans, indeed potentially quadrupling annual contribution levels at a time when many employers are expected to struggle to recover from the recent economic downturn.
Accordingly, no matter which of the preceding reform models is adopted, a contribution funding schedule will be required, and almost certainly one in which employers’ near-term obligations are limited to levels substantially lower than ultimately necessary. This inevitably means that there will be some persistent risk of system insolvency, even after reforms are enacted, before they are fully phased in.
If past history is any guide, however, legislators’ relative weighting of near-term and long-term funding considerations will be far from optimal. Long before the recent financial markets downturn, legislators repeatedly valued near-term funding relief for employers above the long-term financing health of the pension insurance system. Evidence of this exists in the repeated delivery of additional funding relief, including the provision of higher liability discount rates in 2002 and 2004 as well as the near-term funding relief provided in the 2006 ppa itself. The further funding relief provided in 2010 was simply a continuation of a longstanding trend of relieving previous law funding requirements whenever they begin to have a significant effect on employer finances.
There are two important considerations to bear in mind when reforming pbgc to better make this judgment call: first, the difficult balancing act of weighing employers’ near-term viability against the pension system’s long-term funding needs; second, the demonstrated habit of the political system to subordinate long-term funding considerations to near-term exigencies. Reforming the structure of the pbgc system would only address the second of these considerations. It would not by itself provide answers to the first set of considerations, which must be carefully handled even from the pension insurer’s narrow perspective of self-interest.
These two sets of considerations are sometimes conflated in the public discussion about pension funding requirements. It is often argued, for example, that requiring employers to make significantly larger contributions to their pension plans would be counterproductive, because doing so might actually harm pbgc by triggering additional near-term plan terminations or, alternatively, because doing so would interfere with employers’ abilities to create jobs.
While these arguments can each sound similarly compelling to policy makers, it is important to remember that they are nevertheless distinct concerns. Whether an onerous funding contribution requirement would cause an otherwise viable pension plan to terminate would be a concern to any pension insurance system. But whether a funding contribution requirement interferes with other economic policy goals is a fundamentally different concern: It speaks instead to the question of whether broader economic policy goals should be achieved by weakening worker pension funding.
This leads our discussion to the following policy principles:
1. Funding requirements must become effective rapidly enough to bolster the health of the pension insurance system, but must still be implemented gradually enough to avoid counterproductive effects such as the avoidable termination of pension plans.
2. Funding requirements should not be manipulated to serve larger economic policy goals of the president and of Congress. The logical extension of such thinking would be that the more desperate the national employment situation, the more that pension promises to workers should go unfinanced. But the purpose of pension funding rules is not to serve larger economic policy objectives; the purpose of pension funding rules is solely to serve the policy goal of adequately funded pensions.
Many of the problems facing the employer-provided defined-benefit system arise from moral hazards common to all forms of defined-benefit pensions, including not only employer-provided pensions but also state and local systems as well as the federal Social Security program.
No matter how the pbgc system is structured in the future, asset/liability smoothing should be eliminated to the extent practicable, special preferences for politically favored industries eliminated, unfunded benefit increases prevented within underfunded plans, and funding disclosure significantly enhanced.
Within these common reform principles, policymakers should choose between three basic options for reform of the pbgc: 1) equipping pbgc with the tools required to close its shortfall without taxpayer assistance; 2) replacing the pbgc with compulsory private pension insurance, and; 3) unless and until one of the aforementioned approaches is adopted, fully disclosing the cost of a taxpayer bailout of pbgc within the federal budget as well as in a periodic, separate report.
Under all of these options, prudent decisions will be required to balance the competing objectives of long-term funding adequacy and near-term sponsor viability. But these decisions should only take into account the long-term health of the pension benefit insurance system and not remain a process of advancing broader economic policy objectives by exposing worker pension benefits to higher risks of underfunding.
America’s employer-provided defined-benefit pensions are significantly underfunded. And as with public-sector defined-benefit pensions, employer-provided pensions embody a significant risk to taxpayers unless this underfunding is addressed. Sound public policy would focus on ameliorating the underfunding in the employer-provided defined-benefit system and on containing the potential risk of a taxpayer bailout.
