Military Pensions: How Scandalous?

410 March 19, 1985 MI.LITARY. PENSIONS HOW. SCANDALOUS INTRODUCTION In a moment of well-publicized exasperation, Budget Director David Stockman declared that military pensions are alscandalous.va The remark triggered a barrage of attacks on Stockman by some of the biggest guns in the U.S. military, active and retired.

Stockman, indeede, might not have been judicious in his choice of language, but it appears that he is on target if what he means is that U.S. military pensions are out of control and are costing the taxpayer too much.

Military retirement costs have climbed more than si xfold in 'the last 15 years. The Pentagon will spend $18.3 billion next year to provide military pensions that are at least twice as generous as the best private sector retirement plans. The typical nondisabled military retiree receives an annuity of $13, 2 26 which is fully indexed for inflation. Since the average retiree begins collecting these payments at age 43, he has plenty of opportunity to supplement them with second career private earnings and, eventually, Social Security benefits and even a second p ension. By allowing officers and enlisted men to retire at half of basic pay after only 20 years' service, the Military Retirement System MRS) encourages highly trained and skilled personnel to leave for private sector opportunities just when they may be needed most.

At one time, extremely generous military retrement benefits rightly were justified as an appropriate means of compensating men and women for serving in the armed forces at very low pay private sector wage, but they will be rewarded later, with hefty pensions payable at young ages. This rationale, of course dissolved a decade ago when the all-volunteer force concept raised military pay to levels generally comparable to the private sector. What did not dissolve was the system paying far above av e rage retirement benefits Men and.women in uniform, it was said, may not be earning a 2 As such, higher salary levels under the all-volunteer force as well as greater reliance on more experienced support personnel and skilled technicians, have eroded the m a npower assumptions on which the generous benefit levels of MRS were once based. The retirement system costs too much, narrowly concentrates its benefits on those who need them least, and aggravates manpower problems in the senior career force. Despite thi s , numerous proposals by study commissions1 to reform military retirement and trim its swelling costs largely have been ignored. If the Pentagon must cut costs, the place to start may not be in needed weapons systems or training or readiness. The P.entagon must be willing to curb retirement benefits THE MILITARY RETIREMENT SYSTEM I The military retirement system evolved from modest origins I before the Civil War. The original purpose of nondisability retirement, after 30 years' service, was to provide finan c ial security for members disqualified by advanced age from active duty From 1915 to 1948, retirement eligibility after 20 years service gradually was introduced for officers and enlisted members in each service branch It was adopted initially to induce mo r e enlisted men to remain in the service, but later was advanced as a tool to eliminate over-aged officers following World War II.3 MRS was designed for an environment of modest inflation and a relatively small career professional force. The apparent gener o sity of military pensions was at first justified as compen sating for relatively low wages In 1962, automatic cost-of-living adjustments (COLAS) to keep pace with inflation were introduced. From 1969 to 1976 each COLA included an extra one percent kicker above the consumer price index (CPI) increase. From 1977 to 1981, COLAS were granted twice a year, rather than annually.

With adoption of pay comparability principles for the all volunteer military in 1972, current military compensation became about equiva lent to private sector levels. But there was little corresponding adjustment downward in the generous level of retire Department of Defense, Fifth Quadrennial Review of Military Compensation Vol I, Uniformed Services Retirement System, January 1984; Presi d ent's Private Sector Survey on Cost Control, Vol. VI, Management Office Report on Federal Retirement Systems, 1983; Report of the President's Commission on Military Compensation, April 1978; and Defense Manpower Commission Defense Manpower: The Keystone o f National Security, Report to the President and the Congress, April 1976.

The 20-Year Military Retirement System Needs Reform (Washington, D.C General Accounting Office, 1978 p. 2.

Ibid., pp. 4, 5 Report to the Congress by the Comptroller General of the United States 3 ment benefit During periods of high inflation, substantial increases in basic pay levels and the number of military retirees combined with lucrative COLAs to boost MRS outlays rapidly over the last decade and a half MRS costs rose from $2. 85 billion in FY 1970 to $11.9 billion in EY 1980 and $17.3 billion in FY 19

85. They are projected to reach $43.5 billion by the turn of the century. In constant dollars, military retirement costs quadrupled over the last 20 years. Future real growth in o utlays will be a slower rate (1.1 percent yearly up to FY ZOOO), but will continue. Since there is no trust fund for the retirement obligations, the Pentagon has a long-term unfunded liability of $709 billion as of FY 1983.5 Beginning in EY 1985, MRS is t o be funded on an accrual basis, which means that annual Department of Defense budget authority must reflect the amount of money that would have to be set aside to build a trust fund large enough to finance the future liabilities of the system. This accoun ting change will reflect for the first time the future retirement system cost of today's active duty manpower decisions.

RETIREMENT BENEFITS The military retirement syst em dispenses benefits to 1.4 million retired veterans and survivors. Nearly 80 percent of the beneficiaries are former active duty personnel who completed their "full careerst' and retired without disability In 1982, Congress enacted a partial half-COLA l i mitation. For the next three years, military retirees under age 62 were to receive COLAS equal to half the projected rate of increase in the cons'iumer price index. How ever, lower inflation than expected in FY 1983 meant that these "younger retirees rece i ved nearly 85 percent of what they would have received that year in the absence of the limitation. Congress later decided to freeze COLAS for all retirees in FY 1984, and 'repeal the half-COLA provision for N 1985 retirees who have entered military servic e after September 7, 1980, will receive pensions based on their average pay in the three highest years rather than on final base pay.

The present value of future benefits to current military retirees, and reservists and active duty personnel expected to qualify for future pensions was $709 billion as of September 30, 19

83. Since MRS operates on a pay-as-you-go basis, with no funding set aside in a pension trust fund, the National Taxpayers Union considers the system's unfunded lia bility to be the full $7 09 billion. Pentagon actuaries, however, claim that the present value of future government payments for annual military pension benefits should offset this figure in the amount of $146 billion.

Thus, the Defense Deparmtent says the retirement system's unf unded lia bility, as of N 1983, was only $563 billion. Preliminary calculations for FY 1985, based on new economic assumptions different from the civilian economy, place the unfunded liability at $532 billion Another legislative change enacted in 1980 pro v ides that all future 4 These annuitants can qualify for immediate payments of 50 percent of base pay upon completing 20 years of active service MRS offers higher benefits for those who stay longer--2.5 percent of base pay for each additional year of servi ce up to 30 years and a maximum 75 percent annuity.6 retirees leave active duty upon completing 20 years of service.

Many, quite naturally, would rather retire than, in effect, llwork for half pay But over one-third of all Since there is no minimum age cri terion for retirement, the number of years that military pensions are paid to recipients is much greater than for private plans.

MRS produces relatively young, well-compensated retirees who typically receive retired pay for over 35 years. The average age of all nondisabled military retirees is 55, and over 70 percent of them currently are under age

60. During 1983, the average age of new, nondisabled military retirees was 43.1 (enlisted 42.1, officers 45.8 They started immediately receiving an average an nuity of $13,416 21,588 for officers 10,404 for enlisted). Future COLAS will increase its nominal value in line with inflation.

Since service members have been covered under Social Security since 1957, retirees are also eligible for Social Security benefi ts earned while on active duty, with no reduction in military retire ment pay C.OMPARIS0NS WITH OTHER SYSTEMS The military retirement system provides much more generous benefits than are available in most non-military pension plans.

Lifetime MRS benefit e arnings are at least twice as high as those under the best private sector plans, and significantly above Base pay is the only element of military compensation which retirement benefits are' computed. However, it represents only a portion of basic military compensation (BMC), which also includes a quarters allowance, a subsistence allowance, and the federal tax advantage accruing to such non-taxable allowances. Thus, 50 percent of base pay is equivalent to 38 percent of BMC, and 75 percent of base pay conve r ts into 57 percent of BMC It should also be noted that beginning in September 2000, the High-3 pay average will begin to replace final basic pay in computation of pen sions for new retirees just completing twenty years of service change will start to take effect in September 2010 for new retirees with thirty years of service The 5 retirement benefits for civil service retiree Most foreign countries' military retirement plans are also considerably less generous, especially for 20-year retirees.8 WHO BENEFIT S Only 12 percent of active duty recruits will serve long enough to enjoy the generous nondisability benefits of MRS. Of those who do collect MRS rewards the young and able-bodied best.

Enlisted retirees age 65 or older average over $1,200 a year less in b enefits than enlisted retirees under age 60. Average pensions for retired officers age 65 or older are,nearly $700 lower than annuities for retired officers under age 60 Permanently disabled military retirees average annuities that are 2,500 lower than th o se for nondisabled retirees, and about 11,500 below the average pension going to 130,000 healthy retired officers still under the age of 60 Officers benefit much more than enlisted personnel under the military retirement system. Their pensions are twice a s high, on average, because they directly reflect base pay levels. Officers also stand a better chance of reaching retirement eligibility than the typical enlisted man.

The average colonel retiring from the military in FY 1983 began drawing an annual pension of $30,7

20. The average sergeant first class retiring at the same time began receiving a 9,600 annuity. Both pensions are equally indexed for inflation, so the proportional difference between them remains,canstant In terms of the ko-called "normal cos t"--the percentage of an employee's salary that must be set aside yearly to fully fund future benefits before retirement--the price of military retirement amounted to nearly 51 per cent of base pay, or 35 percent of BMC "salary," as of FY 1983 into accoun t the effect of employer Social Security contributions as well, the normal cost of MRS was about 40 percent of payroll. By com parison, the normal cost of a "good" private pension plan which also includes Social Security coverage was about 14 percent.. Som e analysts argue that other private retirement features, such as stock options profit sharing, and higher-yielding investment opportunities, should raise the private pension normal cost to 20 percent of payroll. Depart ment of Defense, Fifth Quadrennial Re v iew of Military Compensation, Vol 1: Uniformed Services Retirement System (January 1984 p. VII-35 Congressional Budget Office, Modifying Military Retirement: Alternative Approaches (Washington, D.C U.S. Government Printing Office, 19841 pp. 25-29; Stateme n t of Dr. Kenneth J. Coffey, Associate Director (Military Personnel), before the Subcommittee on Military Personnel and Compensation House Committee on Armed Services, "How the U.S. Military Retirement System Compares with Other Systems July 14, 1983, Appe ndix V, Table 2.

Modifying Military Retirement, p. 29; "How the U.S. Military Retirement System Compares with Other Systems," Appendix V, Table 1 Taking 6 The difference between the typical officer versus enlisted retiree can be shown in several other ways A sergeant first class, retiring on January 1, 1984, after 20 years of service began receiving a yearly pension of 9,2

52. The average pension for a colonel retiring at the same time, after the same 20 years of service, was $21,7

56. Colonels tend to serve longer in the I military than sergeants; colonels retiring in Fy 1983 averaged 28 years of service as opposed to 22 years for sergeant first class retirees. The average pension for a colonel retiring on January 1 1984 after 30 y ears of service, was $37,4

64. If one calculates the current lump sum value of lifetime benefits available, a I sergeant first class retiring after 20 years of service will receive 263,0

15. A colonel retiring after 20 years will receive 629,1

25. A colo nel retiring after thirty years can expect a I pension income stream of $836,469 I Are these generous pension benefits needed to make up for inadequate salary levels? The typical enlisted man who retired in January 1984 at age 41 after 22 years of service was receiving 21,228 in final base pay. The typical officer at the same time after, after 28 years of service at age 50, was drawing final base pay of $49,9

80. This, of course, does not include other generous non-cash benefits that are part of total military compen sation.

More than two-thirds of all "military retirees" under age 60 I are earning additional income in civilian jobs. But the immediate availability of annuity payments upon retirement after 20 years of military service is sometimes justified compensation for retirees' loss of civilian earning power. To the extent that military experience is not fully transferable to civilian employ- ment, IIyoungtl military retirees suffer second-career penalties. I In general, military retirees earn less in post-retirement occupa tions than civilians of similar age and experience. But the difference is in the range of 10 to 20 percent and does not take into account voluntary choice of greater lei~ure.~ Total second career income including military retired pa y , however, exceeds the average income of counterpart civilian workers.1 Coopers & Lybrand, Militazy Retirees and Separatees' Post-Service Earnings unpublished, January 1984), pp. 38-46; Patricia Munch Danzon, Civilian Earnings of Military Retirees (R-2353 -MRAL, The Rand Corporation, March 1980), p.

37. Another analyst, comparing the earnings of military retirees with those of nonretired veterans of comparable age and educa tion, found that "retirees' second career earnings losses appear small on the avera ge (and for many subgroups of retirees, there does not seem to be anv second-career earnings loss at all Richard V. L. Cooper Military Retiiees Post-Service Earnings and Employment, R-2493-MRAL (The Rand Corporation, 1980 Cooper's -study suggested that mi l itary retirees who did sustain iosses often appeared to do so by choice the five- or six-year transition period following retirement, when mili tary retirees appear to be at a genuine disadvantage relative to their non-retired peers, any second-career los s can, on the average, be attri buted to voluntary decis'ions made by the retiree Except during lo Modifying MilitaG Retirement, p. 16; Report of the President's Commis sion on Military Compensation, April 1978, p. 28 7 Early military retirement also allow s an estimated 175,000 retirees to collect pensions while working in second careers, as civilians for the federal government. Many, in fact, work for the Pentagon. It pays well, whether it is called retirement or merely changing positions within the govern m ent WOWER Greater retirement benefits were once justified as a means to offset the higher pay in comparable civilian jobs. Increases in military pay under the all-volunteer force since 1972 have narrowed the salary gap almost completely, however, and elim i nated this rationale for generous MRS benefits.ll Defenders of the current retirement system justify it as an integral part of the compensation structure needed to meet military manpower objectives. MRS advocates claim that it helps maintain a young and v igorous force and also helps retain key mid-career personnel.

But the system distorts personnel management in two ways.

It causes retention of more servicemen than required up to the 20-year point at which benefits are vested, often Iflocking in unproductive personnel past the twelve years of service point.

Vesting at 20 years also creates too strong an incentive for some of the most highly trained and experienced personnel to leave shortly after qualifying for retirement benefits (the so-called Ithump at 20 MRS has also proved a clumsy, expensive, and ineffective way to attract and retain young service members.12 All-or-nothing retirement at 20 years of service historically is tied to the concept of retaining a young and vigorous force.

But most enlisted men and officers in the modern military spend the bulk of their career time in support-type (administration and communications) rather than combat-related (tactical and infantry jobs.

With the military continuing to shift from sole emphasis on physically demanding tasks and toward a greater emphasis on technology, more service members remain productive well beyond-20 years of service. For a growing number of military jobs, aptitude maturity, and experience may count more than the physical charac teristic s associated with Ityouth and vigor.1113 Keeping more I l1 Modifying Military Retirement, p. 24. l2 Report of the President's Commission on Military Compensation, p. 55 l3 A General Accounting Office (GAO) survey of 1975 retirees found that 81 percent of e n listees had never been assigned to combat-related duties during their entire career, and 66 percent of the service career of officers had been spent in non-combatant duties. At their time of retirement, 93 percent of enlisted personnel and 66 percent of o f ficers were working in support-type, non-combat jobs. The Twenty-Year Military Retirement System Needs Reform, pp. 9, 10. 8 personnel for'as long as 30 years of service would still mean most military members would leave active duty by the age of 50 before severe age-related decline in physical and mental capacity generally appears.

But the 20-year retirement policy combines with an aggressive lfup-or-outV1 promotion system to drive out unnecessarily officers and enlisted men still capable of performing many professional and technical non-combat jobs effectively. The cost of replacin g them is a largely hidden expense of the retirement system that is passed onto other personnel categories of the Pentagon budget I The military's competitive "up-or-outl' promotion system keeps most members from serving full careers. Pentagon force I stru c ture policies do not permit officers to complete full careers if they are passed over for promotion system say it is more concerned with the member's relative stand ing among his peers than individual ability to perform effective ly.14 lfup-or-outvr promo t ion system ''failure oriented inconceivable that a service member who has been screened many times during his- Service life by other promotion boards, selection boards and evaluations is suddenly of no further value simply because his service does not hav e enough promotions to go around.15 While spurring an exodus of members after 20 years of service the present retirement system increases the number of mid-career personnel--beyond twelve years of service. These members can be separated involuntarily up to about 18 years of active duty. But the services are reluctant to use this option because of the loss of retirement benefits it imposes. Thus, retirement at 20 years also "locks in" unproductive personnel who have passed the twelve year mark.

The retiremen t system provides few incentives to remain in the military early in a career.16 The pull of potentia1,pension benefits does not begin to take strong effect until after 10 to 12 years of service reason for joining the military. The retirement system is of l ittle help in meeting expanding needs for skilled journeymen with 4 to 12 years of service--who make up 60 percent of all career personnel Critics of the promotion The 1976 Defense Manpower Commission report called the finding it Recruits rarely list MRS as an important Concern over career retention in general reached a peak in the late 1970s, but has been-.eliminated by a series of substantial increases in military pay since then a less reliable guide for compensation policy, remain positive.

Career reenlistment rates l4 Ibid pp. 16, 17 l5 Defense Manpower Commission, Defense Mappower: The Keystone of National Security, op. cit.

Martin Binkin and Irene Kyriakopoulos, fi (Wash ington, D.C The Brookings Institution,.1981 pp. 69, 70; Modifying l6 Military Re tirement, p. 22. I 9 Enlisted career forces are projected to grow beyond current levels and gain more seniority over the next four years, even. as improvement in the civilian economy continues.

Although cutting retirement benefits would reduce the number and average experience level of career service members to some degree, the losses would be modest while the opportunities for saving are considerable,17 particularly during a period when all the military services have some latitude to reduce total compens a tion levels in return for increased funding of other elements of defense programs.

Differentials in active duty pay raises, targeted bonuses and involuntary separation severance payments are all much more efficient and flexible tools than across-the-boar d retirement benefits to meet changing force requirements and encourage members to serve in exceptionally demanding or hazardous duties.18 Career force losses due to retirement benefit reductions can be overcome, with net savings, by offering bonuses only to enlisted personnel in critical skills at the first- and second-term points of reenlistment. The worst way to solve personnel shortages is with the across-the-board, deferred incentives of the military retirement system servicemen and yields more immedi a te results Current compensation is not Ildiscountedll by I THE WRONG OPTIONS Nine major studies in the last 16 years have recommended comprehensive changes in MRS. They generally urge 1) revision of the retirement formula to reduce benefits for less than 30 years of service, 2) reduction or delay of cost of living adjustments and 3) early vesting of benefits.

Congress has shied away from fundamental revision of the system, but instead has made a number of piecemeal modifications aimed primarily at reducing short-term costs through slowing annual COLAS (cost of'living adjustments Modification of the COLA mechanism offers an across-the-board !'spread the pain device that pays off immediately in reduced current year outlays.

COLA reductions have a smaller ini tial impact on manpower levels than multiplier adjustments, because their full effects accumulate over time. But placing the greatest impact of benefit reductions later in a retired military retiree's life, when he is less financially able to deal with it , is deceptive and unfair. Using COLA modification is also an erratic tool for,budgetary cost cutting, given the unreliable history of consumer price index CPI) projections It yields much less during periods of low inflation Congressional Budget Office, Re t irement Changes and Military Force Manning A Sensitivity Analysis (1984 pp. 5, 13, 16 Ibid pp. 4, 13, 16; Modifying Military Retirement, p. 50; Binkin, pp 70, 77 l8 10 Variations on COLA adjustments for military retirement benefits include one-year freeze s (adopted in FY 1984 and under consideration again this year), CPI minus 3 percent, and setting accompanying wage increases as a ceiling on annual cost-of-living boosts (the COLA cannot exceed the annual increase in the salary base).

POLICY RECOMMENDATION S Reducinq Working- age" Pens ion Long-term reform of MFtS requires reduction of the ''working age" pension and an end to all-or-nothing vesting centered around the 20-year service mark. The best approach involves broadening the base of military retiremen t beneficiaries while phasing in deferral or actuarial reduction of pensions currently being paid to servicemen in their early forties. The standard for normal military retirement should then be based on 30, not 20, years of service In 1978, the President' s Commission on Military Compensation also called the Zwick Commission) recommended setting eligibility at age 55 for receipt of military pensions after 30 years of service, at age 60 for 20 to 29 years of service, and at age 62 for 10 to 19 years of servi c e bility rules for civil service retirement, and is a laudable long-term objective This would correspond with eligi- I Getting from here to there is a sticky problem. The best transition device is actuarial reduction or discounting over time any receipt o f pensions before the normal retirement age. The actuarial reductions should be phased in over ten years (at 10 percent a year of the full reduction required), for retirees younger than the eligible age who still wish to receive immediate annuities. Curren t military members whose pension benefits are already vested under current rules (20 years or more service would be fully grandfathered into the old retirement system.

Once the full transition was complete, the option of immediate but actuarially reduced, annuities would remain available.

Such elimination or actuarial reduction of immediate pensions would affect military retirees at an age when they are likely to have civilian job earnings nished by offering substantial cash bonuses to those involuntarily separated from the services beyond ten years and less than 30 years. Length of service criteria for retirement should also reflect the type of duty performed, with extra credit (perhaps 50 percent) for hazardous, combat-related roles Its impact should be f urther dimi Half -COLAS Other indirect or backdoor means to reduce ''working age pensions include providing only half-COLAS (50 percent of the full cost-of-living adjustment) to retirees under 62, reducing 11 the multiplier (currently 2.5 percent) in the b enefit formula for years of service less than 30, and raising the minimum number of years for retirement vesting from 20 to 25 Providing new retirement beenfits for military members with 10 to 19 years service would improve equity and personnel manage men t flexibility It would also provide a carrot to balance the stick of benefit reductions for retirees in the 20 to 29 years category.

To ensure efficient retention, incentives for current members in the 5 to 15 years of service range, the new benefit featur e should be phased in over a 10-year period Two other unattractive reform alternatives involve lump-sum costly withdrawal payments and members contributions.

COLA Cap If political considerations prevent a head-on assault against the working-age pension, COLA modification may be possible.

The National Committee on Public Employee Pension systems has proposed to index only a base portion of military pensions equivalent to the primary Social Security benefit payable to a person with maximum covered earnings. In current dollars, this would me an that roughly the first $10,000 of pension benefits would be adjusted annually for inflation. Only the basic "cost of-living, in effect, would be indexed.

Another backdoor too1,for reducing retirement benefits involves changing the wage base upon which they are calculated.

All members joining the military after September 1980 would have their pensions calculated on the basis of average base pay received over their last three years of service--the'so-called High-3.

Near-term budget savings could be achie ved by making the change retroactive to all other servicemen on active duty, whose pensions are currently based only on their final base pay The expedient of leaving the wage base for deferred annuities without indexation for inflation should be resisted, however, as inequitable. In fact, a more logical basis for benefit adjustments is to replace COLAS with indexation parallel to the annual increase in the average military pay base. This would help stabilize pensions as a constant percentage of payroll. Mi litary retirees would neither gain at the expense of active duty soldiers nor fall behind them.

Several other military retirement reforms could be desirable but would have little significant effect 0 Integrating Social Security Benefits with military retir ement has been proposed periodically to receive both, their MRS annuities would be reduced by a portion of the amount of their Social Security benefits attributable to military service. The Grace Commission recommended a reduction of 1.25 percent per year of military service multiplied by the primary Social Security benefit. It would be difficult, however As soon as members became eli'gible . 12 to trace and calculate these benefits. The burden would also fall more heavily on lower wage enlisted retirees t h an on officer annuitants 0 Curtail Double Dipping Excesses. Double dipping would be largely corrected by actuarial reduction or deferral of working age military pensions, because such measures would take most of the premium out of working-age military ret i rement. Adoption of another Zwick Commission proposal would solve any remaining problems posed by former military members who go to work'for the federal government. The Commission proposed that no military pension could be received by them while they were employed by the civil service. They could then exercise an option. If otherwise vested in the civil service retirement system, they could apply active military service toward the time credited to civil service retirement. Or former military members otherw ise vested in MRS could choose to apply years in the civil service toward their military pension eligibility. This approach recognizes the federal government as the common employer of both groups of employees.

Last'year's Fifth Quadrennial Review of Milita ry Compensation QRMC) insisted that any reduction in retirement benefits must be tied to a form of reallocation into immediate compensation to protect manpower incentives. However, its proposal for an early withdrawal option in which future members eligib le for retirement could receive lump sum payments in lieu of pre-30 year reduced annuities 3 percent annual reduction for each year under 30 is too costly and ill-advised.

The QRMC proposal would allow a member with 20 years of service to choose early with drawal of roughly twice his annual base pay, instead of an immediate, but reduced, annuity. The early withdrawal payments would create a surge in retirement outlays within ten years and leave a budgetary bulge for another twenty years, before the full eff e cts of annuity reductions could produce net savings.lS Making service members contribute to the military retirement system is even more counterproductive It offers the illusion of short-term savings which could be largely negated by its effects on militar y recruiting and retention.20 already contribute to their retirement through the Social Security tax on base pay further reduction in take-home pay for the military and have the short-term effect of a pay cut retirement benefits for a few 12 percent of all personnel) is far less than the reduction in current pay for everyone, a contri butory system's disincentive to recruiting and retention would quickly generate pressure for'an offsetting pay raise to maintain Military personnel Contributions to a retireme n t fund would mean a Since the present value of l9 Modifying Military Retirement, pp. 56, 57 2o L' Ibid pp. 67, 68 13 manpower levels. In addition, some of the short-term savings from contributory retirement would later be reduced, first by refunds to memb ers separating before retirement and then by tax credits to those drawing initial benefits representing previous taxed contributions (offset partially, in turn, by cashflow advantages to the Treasury).

KEEPING PROMISES AND THE SPECIAL CHARACTERISTICS OF MI LITARY LIFE Excessive military pensions are built on a shaky foundation of unfinanced promises. Private workers receive smaller benefits primarily because their employers have to fund them fully under the federal government's ERISA standards without any a c cess to taxpayers' pockets. But private retirees have an explicit contrac tual right to established annuities. Such deferred compensation is a form of property entitlement formulas, even though their pensions are called deferred compensation. Congress has the right to change their retirement benefits as a matter of policy. Military-members may be shocked to learn they have no vested property right in future annuities.21 Military retirees receive only transfer payments, granted Promises to current retirees and veterans nearing retirement eligibility, of course, should not be broken. Reasonable grand fathering provisions should be part of any major reform of MRS.

But further delay in correcting the system's excesses could create a financial breaking point and a panic atmosphere where such considerations are discarded.

The so-called "X factorI'--the special sacrifices and rigors imposed by military life--is often used as a final justification for generous early retirement benefits. But the President's Commissi on on Military Compensation concluded We find no compelling evidence that the calling to a military career is sufficiently unique to justify the prior to 1945 when the military functioned without the current system we reject the notion that the current sy s tem is a fundamental underpinning of the miJitary way of life.22 The Defense Manpower Commission was unequivocal in recommend current system Finally, in view of 'the many years ing that there should be no explicit payment made to all or most 21 Congressio nal Research Service, Restructuring the Civil Service Retire ment System: Analysis of Options to Control' Costs and Maintain Retire ment Income Security (Washington, D.C U.S. Government Printing Office January 1982 pp. 95-98.

Report of the President's Comm ission on Military Compensation, p. 179 i I I I 14 service members which is specifically designated as compensation for the X factor, noting that: in the private sector, there are many occupations or assignments 'entailing relatively onerous working condi tions which employees must be induced to voluntarily accept Volunteers are attracted by paying differentials only to those individuals who experience the more onerous working conditions and only for the period during which these conditions are e~perienced CONCLUSION The military retirement system if not, as David Stockman said, "scandalous, is nonetheless too costly, inefficient, and inequitable. Studied extensively for several decades, its basic flaws remain uncorrected. In a time of mushrooming federal b u dget deficits, the bills are coming due for lush early retirement policies, extravagant COLA formulas, and Ifpromise now, pay later financing I Such retirement benefits no longer can be justified as a necessary recruiting and retention incentive to offset inadequate is no longer a problem for certain groups of military servicemen, with resulting shortages in skilled, experienced personnel, is just that--a problem of low needs should be directed at the problem-setting pay or other cash incentives for specif ic requirements at competitive levels.

Promotion policies should also reflect manpower needs, rather than the implied dictates of the retirement system.

It is much wiser to make necessary payments once, for recog nized personnel needs, than to commit to a lifetime scheme of recurring benefits (estimated of half a trillion dollars for MRS that only indirectly influence military recruitment and retention military salaries. Pay comparability for the all-volunteer force I Any remaining difficulties with low p a y pay, not inadequate retirement benefits. Solutions to personnel 1 The military services, which have resisted MRS reform vigorous ly until now, must start making cost-conscious tradeoffs in deciding what is necessary for the nation's defense. Hardware re adiness, current compensation, and retirement payments all compete for increasingly tight Pentagon appropriations.

Too often, it seems as if retirement is first in line. It now must move back in that line.

Prepared for the Heritage Foundation by Thomas Miller A Washington-based attorney and writer 23 Defense Manpower, p. 341.