Skip Navigation

Mr. Chairman and Members of the Subcommittee:

I appreciate the opportunity to testify this morning on methods of accounting for military retirement costs. As you know, the current method of accounting for military retirement in the federal budget is quite simple. The defense budget includes the outlays actually paid to those who are already retired. This is the only place where military retirement costs appear in the federal budget. However, it is likely that in the next few months the Executive Branch will recommend major changes in the way the federal budget accounts for military retirement costs. These accounting changes will not affect anybody's retirement benefits. But they should improve the way defense manages its manpower, increase the visibility of savings from retirement reform, and bring more consistency to accounting in the federal budget. In matters of budget presentation, we believe it is appropriate for CBO to state a clear view. For the reasons cited, we believe these accounting changes should be implemented.

What I would like to do this morning is to first describe these accounting changes to you. Then I'll discuss what advantages the changes offer, as well as some problems associated with them. I will also show you how they will affect the costs in the federal budget. If I may, Mr. Chairman, I would like to submit as a prepared statement a CBO report done last spring which describes these changes in detail. Then in my testimony this morning, I will cover the key points of that report.

There are three accounting changes that are likely to be proposed for military retirement. The first change would transfer the Defense Retired Pay appropriation out of the defense budget. The appropriation would probably be transferred to an entirely new budget function created to house the costs of the federal retirement systems. As I have said, this Defense Retired Pay appropriation is the one that is now in the defense budget to pay for the costs of those already retired. So the effect of this change will be that the defense budget will no longer include the costs of those already retired.

The second accounting change would add a charge to the defense budget. This charge would be the amount of dollars that must be set aside, or accrued, each year in order to pay the retirement costs of military personnel now on active duty. Stated another way, this accrual charge is just the amount you would have to put in a bank, at interest, so that you could pay the future retirement costs of today's military personnel. The addition of this charge to the defense budget will mean that the budget reflects full funding of future military retirement liabilities. Of course, in order to insure full funding, this accrual charge must take into account expected future growth in prices and wages and also future interest rates.

I want to emphasize that, under these accounting changes, the entire accrual contribution would be made by the Department of Defense. Whether military personnel should contribute to their retirement is a separate issue that is not addressed by these changes.

The third of the accounting changes would create a trust fund, similar to the civil service retirement trust fund. This fund would hold contributions by the Department of Defense and would pay future retirement benefits.

In summary, then, the effect of these three accounting changes would transfer out of the defense budget the cost of past retirees, while adding to the defense budget a charge for the retirement costs of today's military employees. And the changes would create a military trust fund. Of course, some of these changes could be made independently of the others, and 1 will note their separate effects as I go along.

I want to emphasize again that these accounting changes will not affect anybody's retirement paycheck. This raises the question of why we should make the changes. The answer is that the accounting changes offer a number of important advantages, and I would like to discuss them with you.

First, these accounting changes will improve the visibility of retirement costs to defense managers. Right now a shift in numbers of military personnel on active duty, either up or down, has almost no effect on retirement costs for at least 20 years. Because there is little change for many years, retirement costs are often ignored during the manpower planning process in the Department of Defense. However, if these accounting changes are implemented, then any shift in numbers of military personnel will immediately affect retirement costs in the defense budget. Again, this occurs because the accounting changes cause the defense budget to set aside money for retirement costs of today's employees, and hence the amount that is set aside changes whenever there is a shift in the number of today's employees. This immediate effect on the budget should insure that retirement costs are visible and are fully considered by defense managers in their planning.

These accounting changes will also affect the visibility of savings from any reform of military retirement benefits. During the next year or so this subcommittee may well be considering important reforms in military retirement benefits. To insure fairness, any reductions in benefits will undoubtedly be phased in gradually over a number of years, while improvements in benefits may be phased in more quickly. This means that retirement reform will increase outlays for several years. Thus, under current accounting procedures, passage of retirement reform actually increases the defense budget in the first few years. This may well make enactment of reform less inviting for both the Department of Defense and the Congress. However, under the proposed new accounting system, passage of retirement reform will lead to an immediate and substantial cut in the defense budget.

Let me give you an example. The Retirement Modernization Act, which has been proposed several times in the last few years, would increase retirement outlays by $50 to $150 million per year during the first few years after it is implemented. Thus undercurrent accounting procedures, passage of the Act would make the defense budget go up for several years. But passage of the Retirement Modernization Act would immediately reduce the defense budget by more than half a billion dollars per year if the proposed accounting procedures are in place. The reduction occurs because, under the new accounting procedures, the defense budget would immediately recognize that the Retirement Modernization Act reduces future retirement costs.

Another advantage of these accounting changes is that they will make accounting in the federal budget more consistent. Right now the federal budget accounts for civil service retirement one way, and military retirement quite another. The proposed accounting changes will greatly reduce the differences.

Still another advantage is that these accounting changes will make the defense budget a better indicator of the costs of today's defense. Right now the defense budget contains costs of past retirees. Those costs were determined by decisions made many years ago. On the other hand, the defense budget contains no costs at all for retirement liabilities being built up by today's military personnel. The proposed accounting changes will make the defense budget reflect today's activities, rather than past decisions.

There are, of course, some potential problems with these accounting changes, though many of the problems can be solved by careful implementation. One problem is that the estimate of how much must be set aside, or accrued, to pay future retirement costs depends on many assumptions about uncertain future events--for example, the rate of future inflation and the fraction of military personnel who stay to retirement. Under the new accounting system, changes in these assumptions can significantly affect the size of the accrual charge, and hence the size of the defense budget. But we should insure that this accrual charge is not used to manipulate the size of the defense budget by requiring that the charge be determined by an independent board of actuaries. Such a board is already set up by law for the civil service retirement system.

Another problem is confusion during transition to the new accounting procedures. During the changeover period it is likely that the defense budget will go down, even if there is no change in the level of defense activity. However, it will be relatively easy to restate past defense expenditures assuming that these accounting changes had always been in effect. This should insure that the pattern of change in the real costs of defense is not obscured by accounting modifications.

There may also be some problems if all three of the retirement accounting changes are not implemented together. For example, if the only change was to transfer the defense retired pay appropriation out of the defense budget, without adding a charge for future retirement costs, then none of the advantages I have mentioned earlier would be realized. Retirement costs would be even less visible to defense managers than they are now. Also, the defense budget would not be an accurate indicator of current defense activity. And accounting would remain inconsistent between civil service and military retirement.

In sum, while there are problems that are important to consider as Congress debates these accounting changes, it appears that most of the problems can be solved by careful implementation.

Let me turn now to the effect of the accounting changes on costs in the federal budget. I must emphasize that the numbers I am going to show are tentative. They may change as actuaries make more refined estimates, and they are quite sensitive to assumptions--not yet fully resolved--about future wage growth and interest rates. Nonetheless, I think the numbers--which are shown in the one-page handout before you--are reasonable illustrations of how costs will be affected if the accounting changes are implemented in the beginning of fiscal year 1979.

Let me first address changes in outlays. The main thing to note is that, in the federal budget as a whole, there is no change in outlays. These are accounting changes, and therefore they do not affect the total cost of checks that the federal government actually writes for retirees. However, there are some changes in. outlays of the agencies. The right hand column on the handout shows that, if all three accounting changes are implemented together, then defense outlays will go down by about $2.7 billion dollars in fiscal year 1979. This decrease is the net effect of two actions: transferring out the Defense Retired Pay appropriation, which cuts defense outlays by $9.7 billion, and adding an accrual charge, which increases defense outlays by $7 billion. While defense outlays go down, there are offsetting changes in other budget functions which insure that total outlays in the federal budget are unchanged.

The accounting changes have a different effect on budget authority. In the federal budget as a whole, the accounting changes result in an increase of $7 billion in budget authority in fiscal year 1979. This increase occurs because the federal government begins recognizing its liability to pay future retirement costs. Under current accounting procedures these future liabilities are ignored.

Mr. Chairman, I would like to summarize what 1 have had to say this morning. In the next few months the administration has said that it will propose major changes in the way the federal budget accounts for military retirement costs. These changes include transferring the cost of past retirees out of the defense budget and substituting a contribution that fully funds retirement costs of today's employees. The administration may also propose establishing a trust fund to receive contributions and pay retirement benefits.

As I have said, these changes will not affect the retirement benefits of any individual. But they do offer important advantages. They will make retirement costs more visible to defense managers and hence should cause these costs to be fully considered in the planning process. They will make savings from retirement reform much more obvious. And they will improve the consistency of accounting in the federal budget. Because of these advantages, the principles underlying these accounting changes have been recommended by the General Accounting Office, and most were recommended by the Defense Manpower Commission. There are, of course, some problems with these accounting changes. However, most of the problems can probably be overcome through careful implementation.

In sum, we believe that the concepts underlying these accounting changes would improve the federal budget and should be implemented. Of course, it is important to review the details of the proposals when they are made to insure that problems are avoided, and we hope to assist this Subcommittee in that review.

That completes my testimony. I would be glad to answer any questions.

This document is available in its entirety in PDF.