Grassley Letter to GAO on Report on Cash Balance Plans

August 4, 2000

[1] The purpose of this letter is to request your assistance in reviewing three pending General Accounting Office's (GAO) studies each of which concerns or affects private-sector retirement plans or government insurance programs protecting them. On July 25, 2000, Senator Kit Bond and I wrote to you with other concerns about the GAO's pending report on the contracting and management issues at the Pension Benefit Guaranty Corporation (PBGC).

[2] There are two other reports to which I would like to draw your attention. The first was a [sic] requested on October 15, 1999, and asked the GAO to conduct a study on cash balance pension plans. From various briefings with your staff, it is our understanding that GAO intends to issue draft report concluding that a cash balance plan is not suitable for anyone at any age and that no workers ever benefit from a conversion regardless of the transition benefits offered.

[3] We understand that the GAO has based some of its conclusions about the effect of cash balance conversions on the calculations derived from flawed assumptions underlying a computer simulation model. That model assumes that workers never change their place of employment, but remain with one employer during their entire career. This grossly skews reality.

[4] The Bureau of Labor Statistics (BLS) in the U.S. Department of Labor found that in the United States median tenure in 1998 ranged from 7 years for workers 16 to 17 years old, to 10.1 years for those 55 to 64 years old. BLS data shows that tenure increases with age, but it does not indicate that long years of service are the norm for most workers. Consequently, if GAO bases its conclusions on a computer simulation model that assumes a life time employment with a single employer, by definition it is not possible to represent fairly the effect of a conversion from a traditional final average pay (FAP) plan to a cash balance plan or similar arrangement. The model would not reflect the actual effect on workers' benefits because it has no basis in reality. In addition, GAO's computer simulation model would never be viewed as inertly "Illustrative." Too many readers, both in Congress and in the public, would cite GAO's report as dispositive.

[5] The short service workers who are vested in defined benefit pension plans will not benefit from the "spike" that occurs at the end of a FAP formula. While some workers have a strong attachment to a single employer and will accrue many years of service under one plan, the vast majority of workers will have more than one, two or even three employers daring their carcer. One of the reasons that employers give for converting from a FAP formula to a cash balance plan is that a cash balance plan tends to change the accrual of pension wealth from one that primarily rewards long-service employees who have remained in the service of the current employer, to one that more evenly distributes pension wealth over the employee's entire career. While traditional FAP plans tend to be very generous for those few employees that work for one employer their entire career (the only possibility that appears to be assumed by the GAO model), it is less beneficial when employees work for more than one employer during their working lives (the situation for the vast majority of all U.S. workers). Thus it appears that GAO is using the small percentage of workers that are disproportionately favored by traditional final average pay plans as the benchmark for testing the efficacy of cash balance plans. Several studies, including a report released by CRS have demonstrated that the relative advantage of cash balance plans increase as labor mobility increases.

[6] The draft GAO report on cash balance plans has not yet been delivered to me. It is entirely possible that I will have no other concerns regarding its conclusions. Nevertheless, I am disturbed that GAO has decided to proceed with a report which can not possibly reflect fairly the benefit accrual patterns of more than a de minimus percentage of American workers. I am concerned that this inappropriate utilization of such a computer simulation model will lead the agency to make misleading conclusions and communicate them to Congress and the public.

[7] I have just received a draft report on "top heavy" rules that I requested on December 29, 1998. It is ironic that in this draft study, the GAO focuses on short tenure as one of the principal masons why the top-heavy rules are beneficial. As you know, the non- discrimination rules for retirement plans are extremely complicated. The top-heavy rules have been particularly controversial since they were enacted in the Tax Equity and Fiscal Responsibility Act (TEFRA) in 1982 and practitioners subsequently applied them to retirement plans. Many experts have reacted so negatively to these rules that they have repeatedly called for repeal of the top-heavy rules. For instance, eight years after TEFRA was passed, you testified on behalf of the American Institute of Certified Public Accountants (AICPA) before the Ways and Means Committee on the enforcement and administration of ERISA. Appended to that testimony was a series of recommendations including one to "Repeal the Top-Heavy Rules."

[8] That Testimony was echoed in 1998, by the Clinton/Gore Administration's ERISA Advisory Council's "Report of the Working Group on Small Business: How to Enhance and Encourage The Establishment of Pension Plans". That report also concluded that the top-heavy rules should be repealed. The ERISA Advisory Council's report argued:

. . . . the top heavy rules, originally enacted in 1982, do

[10] Norwithstanding the consensus of these experts, including the, AICPA, the GAO seems to be reacting to the "repeal the top heavy rules" movement. The authors seem to have concluded that the top- heavy rules are neither a disincentive to small business plan formation nor a burden to administer and fail to acknowledge that there are any weaknesses with the rules either in substance or in application.

[11] There are other potential problems with the draft. For example, the draft report details how top-heavy status varies by plan type. Figure 6 on page 19 states that 1,720 new defined benefit pension plans were started up in 1996. That total seems large given that defined benefit plans are generally declining. The number should probably be confirmed before this report is finalized. Is it possible that some of those plans counted as new defined benefit start-ups, may actually have been pre-existing amended plans, restated plans, or mergers of plans? Could these plans be one-participant plans? I raise this issue because the defined benefit plan data on the chart is somewhat counter-intuitive and could be called into question once the report is made public. It is important to know whether these are truly NEW defined benefit plans.

[12] Your analysis in Table 1 on page 11 claims to show an "actual top-heavy plan" (though there is no citation showing from where the data was obtained) and attempts to portray it as a typical top-heavy plan. This is a highly stylized plan however, and GAO appears to have gone out of its way to choose a cross-tested plan for purposes of a dramatic comparison. Even the ages of the non-key employees are ALL in their 30s. The oldest non-key employee is only 38, but the owner is a distant 60 years old. In most companies, the non-key employees will not all be crowded into one age group but will be distributed more evenly among various ages, including some close to the age of the owner. If all the non-key employees in this plan were in their 60s, the benefit distribution would be quite different with all of them having much higher account balances. In addition, the table shows two "key employees" who are not highly compensated (probably both family members), each of whom sustained reductions to their accounts on account of top heavy rules. The details of this example leave the GAO open to criticism that it has deliberately chosen a non-representative plan and is portraying it as if it were an example of most, if not all top-heavy plans.

[13] Mr. Comptroller General, the legislation that I introduced with Senator Bob Graham of Florida does, not repeal the top-heavy rules, but would only alter some aspects of the rules considered most unfair in their treatment of small business by pension experts. We recognized that there may be circumstances in which these rules are needed and so we stopped short of repeal. I am concerned that this report, as drafted, fails to highlight problems with the rules such as:

o the five year look-back for distributions from a plan, which

[15] I whole-heartedly support the independence of GAO in its review, analysis, and recommendations. However, Congress must be able to rely on the accuracy and thoroughness of GAO's work product.

[16] Mr. Comptroller General, as one who was formerly the highest-ranking executive in the United States Government charged with protecting and expanding pension plans, you have always been sensitive to the importance of balancing meaningful protections for participants with nurturing the growth of the private pension system. Given the upcoming retirement of the 'baby boomers,' expanding, encouraging and protecting the pension and retirement savings plans is vital to day-to-day lives of ordinary Americans. It is important that we accomplish both these objectives, or a majority of Americans could reach retirement age without either a pension or adequate retirement savings,

[17] Accordingly, I request that you review these three draft reports concerning the PBGC, cash balance plans and top heavy rules prior to their issuance so that you can assure me that they reflect appropriate assumptions, findings and conclusions, I would also appreciate a written response from you at your earliest convenience. Finally, I am available to discuss these matters with you. I believe retirement issues are of vital importance to Americans and must be treated with appropriate gravity by both the GAO and the Congress.

Sincerely,