1994-1996 Advisory Council on Social Security

Report on Social Security Advisory Council Meetings on April 21 and April 22

Members Present:

Edward Gramlich (Chair), Fidel Vargas, Robert Ball, Gerald Shea, Marc Twinney, Ann Combs, Thomas Jones, Joan Bok, Sylvester Schieber, Edith Fierst, and Carolyn Weaver

April 21

The focus of the meeting was the discussion of an illustrative plan to restore long range solvency developed by Robert Ball. (Copy attached.) This plan would not just eliminate the long-range deficit, but would provide fiscal balance for each year during the 75-year period and result in a contingency reserve of 200 percent of the next year's outgo at the end of the period. The plan recognizes the need to save for the retirement of the baby boomers by going to partial advance funding with investments in the private sector as well as government bonds. It is intended to illustrate that "great big changes" are not needed to solve the long-range financing problem in Social Security.

In response to questions from several members, Steve Goss (OACT) said that under this plan benefit payments would exceed income at the end of the 75-year period and the balance of the trust funds would be drifting downward, but at about half the rate of the current system. Under current assumptions, the trust funds would again show a long-range imbalance after 10 or 15 years. The trust funds would reach a high point in the 2015-2030 period with about 5 years' worth of reserves and would end the 75-year period with about 2 1/2 years' worth of reserves.

It was clear during a general discussion of the package that the Council was divided. Some members liked it because it was a package of modest changes that, coupled with changes in investment policy, would build public confidence and preserve the current system. However, others thought that a more fundamental change was needed to secure Social Security for today's young people.

Many of the members were uneasy with this package because the system would again become out of balance and another financial crisis would occur in 10 or so years. There was general concern that the system would be "under fixed" and that this would lead to a further decrease in public confidence. There was discussion of the possibility of some kind of "failsafe" or automatic governor to take account of the inherent problems in predicting economic and demographic trends over a 75-year period. Two possible failsafes--a future increase in the FICA tax, if needed, and automatic indexing of retirement age to life expectancy changes--were discussed.

The members decided to leave for a future discussion the questions of what the target should be in terms of financing and whether fundamental or incremental changes were needed, and focus the rest of the meeting on the specific items in the illustrative plan. The discussion began with item 2--correction of the Consumer Price Index (CPI) used to measure inflation--because there was agreement that the savings shown for this change were both questionable and controversial.

Chairman Edward Gramlich stated that, although there seems to be general agreement that the CPI is flawed, there is no agreement on the degree to which it overstates inflation and no indication of when or how it will be adjusted. It seems risky to count on a saving by the Social Security trust funds based on an unknown change in the way the increase in prices is measured.

Dave Lindeman, Executive Director of the Advisory Council, indicated that the 0.25 percent of taxable payroll saving attributed to the change may be overstated because in the long-range the CPI and another measure of price increases called the GDP deflator are assumed to rise at the same rate. The deflator eliminates a lot of the bias that is found in the CPI and so the long-range estimates do not necessarily reflect the full overstatement of price increases found in the CPI.

Steve Goss indicated that there was a basic difference of opinion among knowledgeable people about how to estimate the saving due to a change in the CPI measure. In response to a question from Sylvester Schieber, he said that the 0.25 percent saving reflected the highest saving and that the other extreme would be a saving of as little as 0.03 percent.

There was general agreement that there was at least a reasonable doubt about the amount of saving that should be counted from this kind of change.

The Council then turned to a discussion of the first item--a change in investment policy for the trust funds. The specific proposal is to gradually invest the excess trust funds in a mixture of 1/2 government bonds, 1/4 public stocks, and 1/4 corporate bonds. This would result in increased earnings for the trust funds and in some real savings for the baby boomers retirement through the investment in stocks and corporate bonds. The Council was troubled about the specific mix of investments and the mechanics of managing the investments without "political" interference, but with enough flexibility to make reasonable financial decisions.

After a lengthy discussion, the Council decided that the current Trustees could pursue a passive investment strategy based on an index representative of the widest U.S. equity market as part of their fiduciary responsibilities. This strategy would be similar to the strategy of many large corporate retirement systems. In order to answer questions about specific indexes and possible risks, the Council decided to have some work done using a simulation model. Thomas Jones offered the assistance of his employees who have expertise in this area. The Council agreed that if the simulation showed no significant problems, they would be sympathetic to this kind of approach to investment of the trust funds.

Chairman Gramlich asked if there were any objections to the third item in the illustrative package--coverage of newly hired State and local workers--that had a program rationale rather than a political or financial rationale. Mr. Schieber said that universal coverage is desirable from a policy viewpoint, but can be politically unpopular in States that have large noncovered populations and could cause financial problems for these States.

Fidel Vargas said that,speaking as a representative of local government, he would oppose this proposal because it would have an unfavorable financial impact on local government. However, he would not let this consideration stop him from supporting a package of proposals including this provision that he was convinced would solve the Social Security financing problem for future generations. (He is not convinced that this package would do that.) Gerald Shea said that, although his constituency would normally oppose this proposal, they would be willing to support it in order to solve the financing problem. (He indicated that this package was moving in the right direction.) Chairman Gramlich summed up the discussion by indicating that all of the members present were sympathetic, or at least partially sympathetic, to this proposal under certain circumstances.

The Council turned to a discussion of the fourth item--taxation of that part of benefits that exceeds what the worker paid in payroll taxes. This proposal would maintain the current law income thresholds that must be met before benefits are taxable, but would compute the part of benefits that are taxable on an individual basis. The 0.48 percent of taxable savings estimate includes reallocating the income from the current taxation of benefits provision back to the OASDI Trust Fund from the HI Trust Fund.

Virtually all of the Council members agreed that Social Security benefits should be taxed "like other retirement payments." However, there was disagreement about the use of the thresholds and whether the proceeds from taxation should go to the trust funds rather than the general fund. Six members indicated that they supported elimination of the thresholds and seven members indicated that they supported earmarking the revenues for the trust funds rather than letting them go to the general fund.

In response to a question, Steve Goss indicated that about 1/4 of the 0.48 percent saving is new revenue and 3/4 is revenue that is currently going to the HI Trust Fund. The members discussed whether they should "take credit" for savings that were, in effect, just a reallocation from the HI Trust Fund. Although all members supported the proposition that revenue from tax on Social Security should not be credited to the HI Trust Fund, they could not agree that they should count any revenue currently going to the HI Trust Fund against the OASDI long-range deficit. Chairman Gramlich was particularly opposed to counting this revenue.

The Council returned to the question of what they were trying to do--just solve the financial shortfall by making incremental changes or make more fundamental changes in order to assure a sound system for all future generations. Mr. Vargas and Carolyn Weaver indicated that the Council first had to decide what they wanted Social Security to be in the future before they could consider specific ways of getting there. Mr. Jones, on the other hand, indicated that the Council should first see if there was a series of incremental changes that would solve the problem before considering more fundamental changes that would basically change the current system. Chairman Gramlich said that the illustrative proposal developed by Mr. Schieber, which the Council would consider on Saturday, would allow consideration of some of these issues and that further discussion should be postponed until then.

The Council then decided to begin discussing increasing retirement age. Mr. Lindeman and Mr. Schieber raised the question of how increases in retirement age can interact the benefit formula and real wage growth to result in "notches"--i.e., decreases in benefits from year to year for newly eligible retirees. The Council agreed that they wished to avoid such notches and briefly discussed the possibility of building in a failsafe mechanism to assure this outcome before adjourning for the day.

April 22

At the beginning of the meeting, Mr. Schieber presented a Social Security reform package that he had developed. This plan includes several elements:

  • Raise normal retirement age to 68 over the period 2000-2017 and increase the earliest eligibility age (now age 62) to age 65 over the same period.
  • Replace the current benefit structure with one that provides a flat-dollar benefit at a level that would balance the system at the current payroll tax rate. (The amount would be wage-indexed before initial eligibility and price-indexed thereafter.) Workers with 35 years of coverage credit would get 100 percent of this amount. Workers with 10 years of work would get 50 percent of the full flat benefit and each additional year of work over 10 years would accrue an additional 2 percent so that a full benefit would be earned with 35 years of earnings.
  • Spouse's benefits for non-insured spouses would equal one-third of the basic benefit paid to the worker. Retirees with a spouse would be required to purchase (through an appropriate reduction in their primary benefit) a survivor benefit equal to 75 percent of the primary benefit.
  • To compensate for the increase in the earliest eligibility age, a special program would allow a worker to purchase (either at age 62 or through pre-tax payroll deductions during his or her working years) an indexed annuity payable for the period between age 62 and 65. This program would also allow the person to purchase additional indexed annuities up to 150 percent of the poverty line, payable throughout his or her retirement years. The assets from these prefunded annuities would be separate from Social Security. They would not be part of any Federal fiscal operations and would be managed by an independent management board.
  • A transition would be accomplished over 2000-2020 for newly eligible retirees through a phased-in blending of the benefit calculated under the new formula and the benefit calculated under current law.

Mr. Ball indicated that he had serious concerns with providing a benefit structure based on a flat benefit if it were to be financed by the current regressive payroll tax. He was concerned that high earners would pay much more in taxes than low earners, yet get the same benefit. He said that this result would be inequitable and that public support for the program among high earners would evaporate. Mr. Schieber responded that the rates of return for high earners are already inequitable and that it is just a matter of degree. Ms. Weaver noted that cutting rates of return is inevitable under any reform plan if the plan is financed with the current level of payroll taxes.

Chairman Gramlich noted that Mr. Schieber's flat benefit could be modified so that higher earners could receive a greater benefit than low earners. Mr. Lindeman suggested that one approach would be to increase the 90 percent factor in the benefit formula and decrease the 15 percent factor. Alternatively, the rule could be that the worker gets the higher of a specified percentage of average earnings, or the flat benefit.

Ms. Weaver and Mr. Vargas generally expressed support for Mr. Schieber's flat-benefit plan. In addition, Ms. Weaver asked that more consideration be given by the Council to a double decker plan--i.e., a plan that provides a flat benefit plus a benefit directly proportional to the worker's earnings. She said that one of the advantages of a double decker scheme is that the adequacy and equity components of the program are clearly distinguished, so the program would be more understandable. Mr. Lindeman, noted that there was some congressional interest in double decker plans and that the plan advocated by the World Bank was a variation of a double decker--a flat benefit supplemented by a privately managed savings plan.

Edith Fierst, Marc Twinney, Gerald Shea, and Thomas Jones said they had problems with going to a flat-benefit plan like that suggested by Mr. Schieber. Mr. Jones said this would be a fundamental change in the program and could only be justified if the current system is "broken". However, he did not think that was the case, and that incremental change seemed more appropriate.

Ms. Fierst said that the element of the Mr. Schieber's package that would allow a worker to contribute to a special savings plan to supplement Social Security benefits was intriguing. Several members of the Council also expressed interest in this idea. Ms. Weaver said that under such a savings plan, the public could be assured that they are getting what they paid for. Chairman Gramlich agreed that this idea warrants further study and consideration.

Mr. Shea said that if the normal retirement age were increased beyond age 67, he was not sure he could support other additional benefit cuts because of the disproportionate impact. Rather, he thought a modest payroll tax increase would be better. Ann Combs said that, given the financing problems of the HI Trust Fund, which could require an HI tax increase, she did not think it was a good idea to also increase the OASDI tax. Ms. Weaver noted that the Social Security tax is already a bigger burden than the income tax for many people.

Ms. Combs suggested that consideration be given to a "middle ground" approach that could include elements from both Mr. Ball's plan and Mr. Schieber's plan. Although she said she was not claiming "ownership" of such a plan, it could include: (1) raising the normal retirement age to 68 and the earliest eligibility age to 65, (2) permitting the purchase of a "bridge annuity" between ages 62 and 65, (3) taxing benefits to a greater degree, (4) covering newly-hired State and local government employees, (5) some form of a flat-benefit plan, and (6) a means-tested cost-of-living adjustment.

Chairman Gramlich also suggested that, in addition to this "middle ground" approach, consideration be given to the plan offered by Mr. Ball, as modified to eliminate both the correction for the CPI and the transfer of money now earmarked for the HI Trust Fund. He suggested that, for purposes of comparing it to other plans, the payroll tax be further increased under such a modified plan to make up for the reduction in revenue due to the modifications.

Ms. Weaver raised a question about whether the Council was going to study the problems facing the Social Security disability program. Chairman Gramlich indicated that the group had previously agreed to not to do an in-depth review of the disability program. Ms. Weaver said she had many concerns about the current disability program, but acknowledged that the disability program was very complex and that it would likely take 6 months for the Council just to get a good understanding of the intricacies of the program. Chairman Gramlich noted that, by law, the Council was supposed to submit its report by the end of the year and that he did not think they would have time for a comprehensive review of the disability program.

With respect to agenda items for future meetings, Chairman Gramlich indicated that further discussion was needed on the idea of investing trust fund assets in equities. He then asked Council members for other items. Ms. Weaver offered three program changes that needed further discussion: a double decker benefit structure, providing an optional individual retirement account (IRA) as a substitute for part of Social Security, and the "companion proposal" to reduce disability benefits if the normal retirement age were increased.

At the end of the meeting, Chairman Gramlich introduced Chris Bender who has developed (under a contract with the Council) a user-friendly computer software package that can rapidly assess the effect on Social Security retirement benefits of many different OASI program changes. Mr. Bender explained that the package will provide data on the replacement rates (benefits as a percent of preretirement earnings) for career low, average, and maximum earners that would result under a myriad of possible program changes that the user could specify. Mr. Bender demonstrated the software and said that he would make copies of the software and the user manual and provide them to all interested Council members.


Robert Ball's Illustrative Plan to Restore Long-Range Solvency
1995 Trustees Report Deficit 2.17
Changes that do not affect benefits or contributions:
1. Build fund and change investment policy 0.70
2. Correct inflation measure (0.2) 0.25
3. Cover newly-hired State and local workers 0.21
1.16   -1.01
Shared Sacrifice
4.   Tax that part of benefits that exceeds what
     worker paid  
5. Extend computation from 35 to 38 years 0.28
6.   Increase maximum benefit and contribution base
     in two $5,000 steps 
7.   Increase contribution rate two-tenths of
     1 percent each
8.   Eliminate hiatus in present law schedule for
     increasing the normal retirement age   
Total 1.43
Interaction -0.10 0.32
9.   Reduce spouse's benefit and provide widow's
benefit equal to 75 % of couple's benefit 
Surplus 0.17
Robert M. Ball April 21, 1995