May 3, 2006 | WebMemo on Social Security
Almost twenty-five years ago, Congress failed to recognize that the financial services industry had changed, and that Savings and Loans (S & Ls) were no longer competitive. However, strong lobbying discouraged Congress from requiring the industry to come up with sufficient new capital. Instead, Congress decided that this was a temporary problem and allowed S & Ls to use an artificial accounting system that made them appear to be healthier than they actually were. When the industry inevitably did not revive, hundreds of S & Ls failed and cost the taxpayers several hundred billion dollars.
This same sort of wishful thinking has now reappeared in the Pension Security and Transparency Act (H.R. 2830), which is being considered by a House-Senate conference committee. Pension funding rules are already horribly complex and generate results that are often very far from reality. This bill was supposed to make pension funding rules more realistic, but it is increasingly likely the bill may make matters even worse.
One of the worst features of the legislation is that the airlines, and perhaps other failing industries as well, are likely to receive 20 years to fund their pension plans instead of the seven years that would apply to other pension plan sponsors. This unjustified move will only delay the inevitable failure of their pension plans, while providing a precedent that will make it harder to deal with any politically powerful industry. The direct parallel between this provision and the special treatment accorded to S & Ls in the 1980s seems to have escaped Congress's notice.
Other well connected employers would also receive special treatment. These include Smithfield Foods, where any new rules would be delayed until 2014; rural electric cooperatives, where the effective date of new funding requirements would be delayed until 2017; and interstate bus lines, which could make smaller contributions to their pension plans.
Other provisions that are supposed to provide more money for pension plans are likely to be phased in so slowly that any good they do will come far in the future. Important sections dealing with requiring full funding and eliminating smoothing and credit balances need to take effect rapidly if pensions are to be fully funded within any time close to seven years. Strong provisions in these areas are essential if accurate information on the financial health of pension plans is to be available to workers, retirees and investors. These provisions will be discussed in a separate paper.
It would be a mistake to judge the final pension bill by focusing only on specific provisions. Moreover, simply removing the special interest language will not ensure a good pension reform bill. Pension funding is so complex that seemingly insignificant provisions can nullify the additional contributions that good provisions would normally cause. While this paper will focus on the most important individual issues, the overall effect can only be known when the entire bill proposed by the House-Senate conference committee is examined.
Congress needs to recognize that pensions can only be paid with cash contribution, not promises, no matter how sincere. Thus the final measure of the bill should be how much more money goes into corporate pension plans. The pension reform bill should result in strict funding requirements and higher contributions to under funded pension plans. If it does not, it should be vetoed.
Today's pension funding is often a fantasy
Today's pension funding rules are both extremely complex and easy to manipulate. According to the Government Accountability Office, 62.5 percent of the top 100 companies with defined benefit pension plans, on average, made no cash contribution to their pension plans between 1995 and 2002, a period of strong economic growth.
To make matters worse, the rules also allow companies to hide the true health of their pension plans. The U.S. Airlines pilot's pension plan was over 94 percent funded in five of the six years before it was taken over by Pensions Benefits Guaranty Corporation (PBGC) in 2002. During those six years, the airline contributed a total of $157 million to the pension plan. However, once the plan was taken over by PBGC, it was found to have only enough assets to pay 35 percent of the benefits that it had promised, a gap of $2.2 billion. Similarly, Bethlehem Steel's pension plan was judged 84 percent funded shortly before that company's bankruptcy, even though it had only 45 percent of the assets needed to pay promised benefits. The PBGC was left to cover the plan's $4.3 billion shortfall.
Some of the difference between what was reported and what was found after PBGC took over the plan was due to expected future company contributions to the pension plan that did not happen. However, that would not account for the low company contributions found by the GAO.
An appropriate discount rate is important
The funding of a defined benefit pension plan is measured using a "discount rate." A plan is assumed to be fully funded if the assets that it currently has can be expected to grow at a certain interest rate until the resulting level of assets then equals the total amount of pension payments that the plan promises to make in the future. For example, if a fund will owe $1,000 in 30 years and assumes that its assets will earn an average of five percent every year after inflation, it must have $231 today in order to be fully funded. (Invested at a five percent interest rate, $231 will grow to $1,000 in 30 years.)
The discount (interest) rate used to measure a plan's funding is crucial. If a plan assumes that its assets will grow at seven percent a year instead of five percent, it needs only $131 today to be fully funded, rather than the $231 it would need if it used a five percent rate. On the other hand, if a plan uses a discount rate of only three percent, then it must have $412 on hand today to be fully funded.
The discount rate has no actual relationship to how much a pension plan's investments are earning. While the law requires that plans make prudent investments, these investments can change over time and are greatly affected by short-term swings in the stock, bond, and property markets. The discount rate is intended to measure whether or not the plan has sufficient assets to meet its obligations over a long period of time. For that reason, defined benefit plans used the rate for 30-year U.S. Treasury bonds until 2004. Since then, they have used a 4-year weighted average of corporate bonds instead of the rate of interest the plan is earning on its investments. Since the discount rate substitutes for the volatility of pension plan investments, the use of further accounting devices such as smoothing to hide changes in asset value accomplishes little more than to distort the financial health of the plan.
Congress should also require plans to value assets on either the actual market value or the average of their values over the past 12 months. This would provide a much more accurate picture of a plan's financial health than the current standard which allows assets to be valued using a weighted average of their value over the past five years.
The discount rate-and yield curve-contained in H.R. 2830
H.R. 2830 would make permanent the use of a blended rate of corporate bonds that began in 2004. Despite the claims of some industry representatives, it does not reestablish the discount rate based on the 30-year Treasury bond. While the law that allowed the use of the corporate bond rate expired at the end of 2005, both the House and Senate versions of H.R. 2830 contain language that would retroactively extend that law, and so the corporate bond rate would be used in determining the funding status of defined benefit pension plans if H.R. 2830 is signed into law. Since Treasury bonds have paid unusually low interest rates over the last several years, and since the Treasury stopped issuing 30-year bonds from 2001 until recently, the corporate bond rate more accurately reflects how adequately a pension plan is funded.
For 2006, the same law that applied in 2004 and 2005 would apply. Starting in 2007, both the House and the Senate bill would require the use of a corporate bond interest rate determined by the average age of an individual company's workforce. This is known as a "yield curve."
Some industries and companies have workforces that are older on average than others. Since these companies will have to begin paying their workers' pension benefits sooner, the health of their pension plans is a significant factor in their ability to remain in business. If a company's pension plan is under-funded and the company has to make significant payments to it, that company is at a higher risk of bankruptcy than the same company with a younger average workforce. Rather than using a uniform measure for all companies, it is much more prudent to use a discount rate that is customized to reflect a particular company's workers.
In H.R. 2830, both the House and Senate versions use a discount rate starting in 2007 that is based on the average interest rate paid on investment grade corporate bonds. The Senate bill requires the use of a rate based on the 12 month unweighted average rate, while the House bill uses a weighted average over a three-year period. Of the two versions, the Senate version is likely to result in a more accurate estimate of a pension plans funding.
In both cases, the yield curve is divided into three segments, and a plan is assigned a segment based on the age of the company's workforce. Since companies with older workers will begin to pay out significant levels of pension benefits sooner than companies with younger workers, both bills would require companies with older workers to use a segment based on a shorter-term corporate bond rate. Normally, short-term bonds of all types have a lower annual interest rate than longer-term bonds do. This lower discount rate means that those companies would have to have proportionately more assets available to pay pension benefits. Companies with younger workers could use a segment of the yield curve based on a longer corporate bond rate, which would allow them to have proportionately less cash and other assets available today since they will have a longer period before they must pay significant amounts of pension benefits.
The yield curve is an important reform that will improve pension funding. Using a yield curve would allow workers and investors to better understand a company's overall financial health. It also should allow earlier identification of problem companies so that changes can be required before they become critical.
The discount rate that would be in effect if H.R. 2830 is not passed
If H.R. 2830 does not become law, funding status would again be based on a measure was used from 1987 through 2003, the weighted four-year average of the returns of the 30-year U.S. Treasury bond. Although going back to the old Treasury bond-related rate would be less accurate, the lower discount rate would force companies to put more money in their pension plans than the rate that was used for the past two years. While a proper reform of pension plan funding would be preferable, going back to the old rate is better than a weak bill that gives special treatment to specific industries.
Special treatment for airlines-and perhaps others
One of the biggest mistakes that conferees could make would be to include in the final bill the ill-considered special treatment for airline pension plans that was contained in the Senate version. This provision would allow airline pension plans to fund their pension promises over a 20-year period instead of the seven years that would be required for pension plans sponsored by other industries. Supporters claim that the provision would actually save PBGC money since PBGC guarantees of the unfunded pension promises would be frozen as of the provision date. Thus, the 2-year period would make it less likely that PBGC would have to take over the plan. However, details of the provision show that these savings would not necessarily occur.
The special treatment gives the airlines two choices: they could either freeze the pension plan so that no new benefits are credited to employees, or they could allow employees to build pension benefits but pay for those new benefit promises on an expedited basis. In either case, the current unfunded pension promises could be funded over 20 years using the plan's interest rate rather than the rate (see above) that other pension plans would be required to use. Since the plan's interest rate would almost certainly be higher than the yield curve used to calculate other plans' unfunded liability, airline pension plans would not only have longer to pay for the benefits, they would have to contribute less money to be considered fully funded. The ability to use a different rate in calculating their payments may seem like a small item, but it could substantially change the amount that airlines would have to pay.
On top of that, even if airline pension plans do freeze their benefits, they will continue to pay out full promised benefits to current retirees and close to full promised benefits to employees who retire early within that 20-year period. Meanwhile, a significant reduction of their under funding would not occur for many years. Thus, if the airline filed for bankruptcy again-and many of them have filed for bankruptcy a number of times-the pension plan could be even more severely under funded than it is now.
A second special treatment is given to airline pilots. Under current law, a pension plan member will receive pension payments that are smaller than the full PBGC guaranteed pension benefit (currently $47,659 a year) if he or she is retired and under age 65 when either the plan is taken over by PBGC. (If his or her pension pays less than the PBGC guarantee before the plan failed, then it does not change after PBGC takes over.) That reduction also applies if a worker who is still working when PBGC takes over the plan retires before they reach age 65. The special treatment for airlines provision allows airline pilots to receive the full PBGC guaranteed pension benefit if they retire at age 60. Thus, a steel worker who retired after 30 years of work at age 62 would receive a maximum of $40,980, but an airline pilot who retired at age 60 would receive the full guaranteed amount of $47,659.
To make matters worse, the special treatment for airline pilots appears to be retroactive, so that PBGC would have to recalculate the benefits due to all retired pilots regardless of when their pension plan was taken over by PBGC. While this could be a drafting error and corrected before the final bill is ready, either PBGC will have a large administrative burden or future airline pilot retirees will be treated better than those whose airlines filed for bankruptcy before 2005.
Special treatment for industries with the high risks of pension-plan default is not the way to deal with a changed business environment. While airlines' pension plans are extremely expensive, those plans are only one of many challenges that the airlines face. To make matters worse, the airline industry is only the latest to face massive failures due in part to poorly funded pension plans. In the last few years, most of the steel industry went into bankruptcy and passed its pension obligations to PBGC. Airline failures are already being followed by major bankruptcies in the auto parts industry and eventually there could be the bankruptcies of the auto manufacturers themselves. All of these industries face or faced expensive pension plans, and it is difficult to justify why one industry should be treated better than another. Most importantly, why should an airlines employee receive better treatment than did a steel worker whose plan failed in 2003?
The only thing worse than the airlines provision would be if Congress granted special treatment to any company that runs into trouble funding its pension plan. Such a move would effectively transfer the cost of paying pensions from companies to PBGC, and guarantee a massive taxpayer bailout of that agency.
While there are major differences between the S & L industry in the 1980s and the remaining airline pension plans, the general situation is similar. Granting any powerfully connected industry special treatment will almost guarantee that eventually taxpayers will have to pay billions of dollars that should have been paid by the companies. Failure to learn the lessons of the past can have real consequences.
Eliminating the special treatment for airlines and other companies will not guarantee that H.R. 2830 is a good bill, but leaving those provisions in the final bill would guarantee that it is a bad one. No pension bill is worth signing into law if it includes special treatment for airlines and other industries. If Congress cannot take a responsible course of action, then President Bush should veto the bill.
David C. John is Senior Research Fellow for Retirement Security and Financial Institutions in the Thomas A. Roe Institute for Economic Policy Studies at The Heritage Foundation.