...increased in several steps during the 1980s before reaching $19 per year per participant in 1991. The flat-rate premium remained at its 1991 level in nominal terms (and thus obviously declined substantially in real terms) until it was raised to $30 in the budget bill enacted in early 2006. variable-rate premium was added in 1987, assessed initially at $6 per $1,000 of unfunded vested liability and capped at $34 per participant per year. The premium rate was increased to $9 per $1,000 in 1991, and the cap was removed in 1996. (84) Firms that contributed the FFL amount in the immediately preceding year are exempt from the variable-rate portion of the premium in the current year. (85)

Many analysts have criticized the current structure of premiums as failing to apply economically appropriate prices to the risk that plans actually present to the PBGC. (86) To be sure, premiums are not completely insensitive to risk under current law; the variable-rate portion of the premium does penalize some underfunding. Even so, the current structure of premiums bears little resemblance to the economically fair structure for two reasons: First, and most obviously, only one risk factor--the level of plan assets relative to plan liabilities--is taken into account in determining the premium, yet the risk confronted by the PBGC also depends on the financial health of plan sponsors and the extent to which plan assets have been invested so as to immunize liability risk. (87) The second major flaw in the current premium structure is that even the one risk factor recognized is priced only very incompletely. Precise figures are difficult to come by, but one set of calculations suggests that, in 2003, the variable-rate premium was assessed against only about a fifth of current-liability-basis underfunding; the rest was exempted under the FFL-related provision noted above. (88)

Several analysts have argued that, in addition to being too insensitive to risk, the current structure of premiums is simply too low, generating too little revenue given the current scale and structure of risks presented to the PBGC. Steven Boyce and Richard Ippolito simulate a detailed model of the DB sector and conclude that a private provider of current-law insurance would demand about twice as much premium income as the current premium structure generates. (89) VanDerhei concludes that the pricing prevailing as of his writing was even more inadequate: He estimates that levying an actuarially (but not economically) fair risk premium against actual exposure in 1984 would have boosted variable-rate premium income to the PBGC by a factor of about 4 1/2. (90) Finally, Christopher Lewis and George Pennacchi, as well as the CBO, estimate that both fixed-rate and variable-rate premiums should be boosted by a factor of about six in order to reflect the full economic cost of current-law coverage. (91)

Although premium revenue clearly seems too low given the current structure of risks presented to the PBGC, it bears emphasizing that premium revenue would not necessarily be substantially greater, and could be substantially less, in a reformed system. Boyce and Ippolito note that if sponsors were bound to a more stringent set of funding rules, and therefore presented less risk exposure to the PBGC, a fully priced set of economic premiums could, in fact, generate much less revenue than is provided in expectation under current law. (92) The same would be true to an even greater degree with both tighter funding rules and more stringent portfolio investment requirements.

DETERMINATION OF INSURANCE PREMIUMS: PROPOSED REFORMS. The premiums assessed by the PBGC should both allow the insurance program to cover its administrative costs and provide full compensation for all risk presented to it. Plans with sufficient assets, structured to immunize the risk of plan liabilities, present essentially no risk to the PBGC and so should pay no premium beyond the amount required to cover its administrative costs; these can be recovered by means of a successor to the current flat-rate premium. With 44 million insured participants, the PBGC could have covered its administrative expenses of $263 million in 2004 with a flat-rate annual premium of about $6 per participant. (93) As many observers have noted, a premium based on number of participants subsidizes plans with older workforces and more-generous benefit provisions at the expense of new plans and plans with younger workforces. To address that concern, a base premium could alternatively be tied to insured liability. For example, given the $1.55 trillion in PBGC-insured liability in 2002 (the latest year for which data are available), expenses that year of $222 million could have been recovered with a charge of about 14 cents per $1,000 of insured liability.

Before the design of a risk-based premium can be taken up, two questions of general principle must be addressed: First, should the government merely aim to assess actuarially fair premiums--and therefore only recover expected costs as they would be calculated using the Treasury rate--or should it be required to levy an additional charge to compensate for the fact that its losses occur disproportionately when financial resources are especially valuable? Private insurers acting in the place of the PBGC would demand compensation for such systematic risk, and basic principles of financial economics suggest that the PBGC should do the same. (94) The reason is not that the government should behave as if it were risk averse itself, but rather that a government seeking to tally the full cost of its programs should recognize that other participants in the economy are risk averse. An implication of that risk aversion on the part of others is that the government should take account not only of the average amount of resources it is appropriating from the private sector but also of the circumstances under which it is doing so. Programs that cause resources to be appropriated disproportionately when times are bad--and hence when resources are especially valuable--should be scored as more costly than programs that appropriate the same amount of resources in expectation but without correlation to the business cycle, and more costly by an even wider margin than programs that appropriate resources disproportionately when times are good.

A logical and appropriate consequence of a charge for market risk is that the PBGC would be expected to run a small surplus on average, once its insurance is fully priced. Pennacchi addresses the same issue in the context of the Federal Deposit Insurance Corporation: "An outcome of setting fair rates is that the FDIC will make profits, on average. That is, premiums less insurance losses must be, on average, positive in order to compensate taxpayers for having to fund large net insurance losses during economic downturns." (95) As Pennacchi notes, the presence of a surplus might incorrectly strike some observers "as evidence of excessive, rather than fair, insurance premiums." As noted by the CBO, (96) the "problem" of the surplus could be addressed by having the PBGC pay the general fund of the federal government a reinsurance premium, set at the estimated amount that the PBGC has collected from plan sponsors as compensation for risk. Alternatively, if the PBGC comes to be viewed as backed by the full faith and credit of the Treasury, a simpler method for dealing with the expected profit would be to abolish the trust fund and the revolving fund, eliminating the distraction of the balance in those two funds and putting the focus where it should be, namely, on the question of whether premium rates have been set appropriately on a prospective basis. (97)

A second matter of general principle is the question of whether--and to what extent--premiums should be adjusted to the individual characteristics of plan sponsors or reflect only aggregate conditions. Boyce and Ippolito argue that premium rates should reflect only the average probability of bankruptcy across the population of insured entities. (98) They argue that tuning premium rates to the financial health of individual sponsors would effectively tie premiums to precisely the risk being insured against. Just as an insurance company is not allowed to boost the premium under a term life insurance contract if and when the health of the policyholder has deteriorated, so the PBGC should not be allowed to boost the premium for pension insurance when the creditworthiness of the sponsor has declined. Taken to its logical conclusion, premiums tied to the health of the sponsor could substantially reduce or eliminate the value of the insurance. The Boyce-Ippolito argument is sensible with respect to the risk factor to which they apply it, namely, bankruptcy risk. However, as was noted earlier, two other factors--the degree of underfunding and the allocation of plan assets--are important as well in determining the overall risk that an individual plan presents to the insurer, and, unlike bankruptcy risk, these factors are unambiguously under the control of the plan sponsor. Failure to adjust premiums to those two risks at the level of the individual plan would unnecessarily continue to give scope for moral hazard to distort the behavior of plan sponsors.

Taking these considerations on board, what could be done to move the system toward rational risk-based pricing of default insurance? Thus far two detailed empirical models of the DB sector have been developed: one described by Boyce and Ippolito and housed at the PBGC, the other described by and housed at the CBO. (99) Although neither model may be ready to bear such pressure today, both are highly credible, and either might ultimately provide a suitable basis for setting risk-based premiums, especially once the reasons for some important differences between the two models are better understood. (100) Research, development, and refinement of these models should continue so that either could become a vehicle for setting premiums.

In moving toward market-based premiums, careful thought would have to be given to the design of the transition. Going "cold turkey" to market-based premiums could force some firms teetering on the brink of bankruptcy, with deeply underfunded plans, over the edge. One way to minimize unnecessary bankruptcies would be to phase in market-based premiums linearly over the same five-- or seven-year period allowed for amortization of underfunding: In the first year, assuming a five-year phase-in period, the premium paid would be equal to one-fifth of the market-based premium plus four-fifths of the current-law premium, and so forth. After five years, when sponsors should have eliminated their current underfunding and redeployed their assets to immunize the risk of their liabilities, the pricing of insurance would fully compensate taxpayers for the risk they are bearing, which would, at that point, be quite small. Another possibility would be to delay the implementation of risk-based premiums altogether until the end of the first five-year (or seven-year) amortization period.

The Third Locus of Reform: Transparency

Transparency is considered here with respect to four separate constituencies: workers, financial markets, the PBGC, and taxpayers. Transparency should be improved for each. (101)

REQUIREMENTS FOR DISCLOSURE TO WORKERS: CURRENT LAW. Under current law, "only participants in plans below a certain funding threshold receive annual notices regarding the funding status of their plans, and the information plans must currently provide does not reflect how the plan's assets are invested." (102) Sponsors are required to report in their annual Form 5500 filing the proportion of plan assets invested in securities issued by the sponsor. However, as the GAO has noted, this information is neither timely nor "readily accessible to participants." (103)

REQUIREMENTS FOR DISCLOSURE TO WORKERS: PROPOSED REFORMS. Care should be exercised neither to overburden workers with complex and difficult-to-understand financial material, nor to overburden sponsors with onerous compliance requirements. Two pieces of information that should be easy for participants to understand and straightforward for sponsors to prepare are, first, the present value of accrued benefits (useful to the worker in comparing the plan with a defined-contribution plan) and, second, the fraction of accrued benefits that would be funded if the plan were to be terminated immediately. After the initial phase-in period of five or seven years, most sponsors should be in a position to report a funding ratio very close to 100 percent. Exceptions would include sponsors that have granted improvements in plan parameters or increases in shutdown or flat-dollar benefits, and that are contributing the minimum required amounts to amortize the unfunded liabilities generated by those improvements or increases. If plan assets are being used to immunize liability risk, capital gains or losses should not be a source of material deviation from 100 percent funding, although other departures of actual experience from actuarial assumptions could generate deviations from 100 percent funding.

ACCOUNTING REQUIREMENTS: CURRENT FASB REGULATION. As noted above, current accounting requirements allow sponsors to assume a high rate of return on pension assets, commensurate with a heavy allocation of assets to risky assets, but then to smooth the associated volatility in actual returns over five years. In effect, the rules create a form of accounting arbitrage. Coronado and Sharpe show that investors seem to pay more heed to the smoothed figures reported in the body of the financial reports than to the fair market values reported in the footnotes. (104) Similarly, Franzoni and Matin show that firms with severely underfunded pension plans tend to underperform the overall stock market. (105) Since 2003 some information about asset allocation has been provided in the footnotes to the financial statements; however, this information is aggregated across all plans sponsored by the firm and provides only a very coarse breakdown of assets, without, for example, any clarity about the duration or quality of fixed-income securities. More recently, FASB has proposed to move within a year to requiring that fair-market-value measures of assets and liabilities be reported on the balance sheet, and to reconsider the appropriate treatment of pension expense in the income statement over the next two to three years. (106)

ACCOUNTING REQUIREMENTS: PROPOSED REFORMS. The general thrust of today's financial accounting requirements, which feature accrual-based concepts and allow finns to report a net pension expense (accrued expenses less returns on assets), is sensible and economically appropriate. Within...

NOTE: All illustrations and photos have been removed from this article.




Looking for additional articles?
Search our database of over 3 million articles.

Looking for more in-depth information on this industry?
Search our complete database of Industry & Market reports by text, subject, publication name or publication date.

About Goliath
Whether you're looking for sales prospects, competitive information, company analysis or best practices in managing your organization, Goliath can help you meet your business needs.

Our extensive business information databases empower business professionals with both the breadth and depth of credible, authoritative information they need to support their business goals. Whether it be strategic planning, sales prospecting, company research or defining management best practices - Goliath is your leading source for accurate information.