DOCUMENT RESUME

ED 370 462 HE 027 167

AUTHOR Palmer, Bruce A. TITLE Planning For : Using Income Replacement Ratios in Setting Retirement Income Objectives. INSTITUTION Teachers Insurance and Annuity Association, New York, NY. College Retirement Equities Fund. PUB DATE Jul 93 NOTE 8p. PUB TYPE Collected Works Serials (022) JOURNAL CIT Research Dialogues; n37 Jul 1993

EDRS PRICE MF01/PC01 Plus Postage. DESCRIPTORS College Faculty; Family Financial Resources; Financial Needs; Fringe Benefits; Higher ; *Income; *Mathematical Formulas; Money Management; Ratios (Mathematics); Retirement Benefits; *Teacher Retirement IDENTIFIERS Consumer Expenditure Survey; Retired Persons; *Retirement Planning; Savings

ABSTRACT This paper presents a method for higher education faculty and staff to assess plan objectives by determining a retirement income replacement ratio to maintain the -based preretirement standard of living. The paper describes the RETIRE Project which researches income replacement using the federal government's annual "Consumer Expenditure Survey" to estimate individual savings and other expenditure variables incorporated into the Project's income replacement formulas. The income replacement ratio formula's components include preretirement gross pay, preretirement taxes, preretirement savings, postretirement taxes, work-related expenses, and net change in age-related expenditures. A central section looks in detail at how the ratio formula works with a single individual and a married couple under various conditions. Two detailed tables and two figures illustrate the analysis. Further sections discuss savings rates and their effect on income replacement ratios and the effects of taxing 85 percent of social security benefits (currently proposed by the Clinton Administration). A conclusion notes that research findings indicate that those retiring in 1993 need about 69 to 82 percent of final-year salary to sustain their preretirement standard of living. (Contains 13 references.) (JB)

*********************************************************************** Reproductions supplied by EDRS are the best that can be made from the original document. *********************************************************************** Research Dialogues

Issue Number 37 July 1993 A publication of External Affairs Policyholder and Institutional Research

Planning for Retirement: Using Income Replacement Ratios in Setting Retirement Income Objectives

BEST COPY AVAILABLE

U S DEPARTMENT OF EDUCATION oeice if Mew On. HPSCait,^ an, ."DrOvIY"e^1 "PERMISSION TO REPRODUCC THIS EDUCATIONAL RESOURCES INFORMATION BY CENTER tERIC) MATERIAL HAS BEEN GRAN 0 This document has been reproducedas received from the person or organization yiqinatinq it WMinor changes have been made to TIAACREF improve reproduction quality

Pools of view or opinions stated in this document do not necessarily represent TO THE EDUCATIONAL RESOURCES official OERI position or policy. INFORMATION CENTER (ERIC).- Research Dialogues

Issue Number 37 July 1993 A publication of External Affairs Policyholder and Institutional Research

Planning for Retirement: Using Income Replacement Ratios in Setting Retirement Income Objectives

In this issue ofResearch Dialogues, '86), which incorporated several impor- to achievePH incomereplacement through la present an article on the subject of re- tant changes in the federal allowing individuals and their families tirement income replacement ratios b.) laws affecting individuals. The most no- to continue their preretirement stan- Bruce A. Palmer. Ph.D.. professor of table changes, in terms of their effects on dard of living into their retirement risk management and insurance, Geor- income replacement ratios, were reduc- years. Its retirement income objectives ,,:ia State: nin-rsitl. Based on Profissor tions both in the number of income tax reflect the belief that most individuals Palmer:c extensive a fifk in this area, the brackets and in the overall marginal tax wish to avoid having to adjust to a lower article describes a method of daernannig rates. standard of living even when a lower standard may permit them to maintain a retirement income replacement ratio The enactment of TRA '86 and subse- a ((JO/rt./NG. life-style. designee' to maintatn the salary-based quent legislation created a need for new preretirement standard of lii ing. The and continuing research on income re- The RETIRE Project uses the fol- replacement ratio analysis can be used placement ratios and related retirement lowing definition of income replace- ky college and unhersity jachlty. income issues. To fill this gap, the ment in calculating replacement ratios': and administrators in assessing pendion Alexander and Alexander Consulting Preretirement gross salary plan objectites. and &v indiz.iduals in Group and the Center for Risk Manage- their financial planning jOr retirement. .Ilinus Prerecirement income taxes ment and Insurance Research, Georgia and FICA taxes State University, teamed together to ini- Introduction Preretirement savings tiate ongoing research, now known as the ,Ilinus 1. Work-related expenses duca-.ors, like most other American RETIRE Project.= The latest report of the workers, want to achieve a finan- Plus Aim Changes in age- E RETIRE Project was relmsed recently.' related expenditures cially secure retirement. The first step Earlier reports were released in 1988 and Plus p Postrenrement in attaining this goal is to establish 1991.1 income taxes appropriate retirement income objec- The cornerstone of the RETIRE Proj- Equals Replacement income tives. Today, income replacement ratios needed to maintain same are the most common approach to set- ect's research on income replacement is standard of living ting realistic retirement income targets. the U.S. government's annualConsumer Expenditure Surto.5These data are used Mathematically, the income replace- The relatively recent focus on in- to estimate individual savings and cer- ment ratio formula can be written as fol- come replacement (and income tain other expenditure variables that are lows: replacement ratios) began with the incorporated into the RETIRE Project's Replacement ratio = 1981 release of a report by the Presi- income replacement formulas. Federal PrRGP - PrRT - PrRS - WRE +/- NCASE + PoRT dent's Commission on Pension Policy and state income taxes, together with PrRGP IP(PP).1 The PCPP, formed during Social Security (FI(:A) taxes, are also in- President Jimmy Carter's administra- cluded in the formulas for determining where tioniddressed several public policy income replacement ratios. PrRGP Preretirement gross pay matters pertaining to retirement-relat- ed issues, including the determination A Definition of Income Replacement PrRT Preretirement taxes PrRS Preretirement sayings of appropriate income replacement Potentially, income replacement and ratios. income replacement ratios can be de- PoRT Postretirement taxes Heightened interest in the income re- fined in one of several ways. The mea- WRE Work-related expenses placement issue occurred in 1986 with sure that has received the most NCASE Net change in age- the passage of the Tax Reform Act (TRA widespread acceptance today is designed related expenditures

1001 Teachers Insurance and Annuo Associailon 3 Preretirement gross pay (PrRGP) retiring at age 65 in 1993. Income re- . As described more fully in a represents the individual's earnings in placement ratios are computed for ten later section, the savings variable (Line the year immediately preceding the separate levels of preretirement salary 6Table 1) plays a major role in both date of retirement. Prereurement taxes (i.e., final-year salary), ranging from the absolute size of income replacement (PrRT) include both federal and stare 515,0(K) to 590.000.' ratios and the shape of the replacement income taxes and Social Security The following discussion focuses ratio curve. (FICA) taxes. PoRT include federal and on Lines 12, 13, and 14 of Table 1. state income taxes but exclude FICA The decreasing pattern of Social Se- These lines display the gross income taxes, since it is assumed that the per- curity replacement ratios (Line 13-- replacement ratios, Social Security Table 1, and Figure I) is simply a re- son is fully retired. PrRS represent sav- replacement ratios, and net income ings amounts that are not replaced at flection of the underlying nature of the replacement ratios, respectiv:lv. Gross retirement.Work-related expenses Social Security program. As is widely income replacement ratios represent the (WRE) include items such as commut- recognized, Social Security's benefit for- aggregate percentage of preretirement ing costs, special clothing, union and mula is designed to favor lower-income salary needed at the moment of retire- professional dues, and other expenses individuals by providing them with mentCOmaintain the individual's cur- that arc higher for working individuals. higher income replacement ratios. This rent standard of living into the retire- Changes in age-related expenditures pattern of Social Security replacement ment period. Social Security replacement (NCASE) include changes in health ratios also explains the pattern of in- ratios are determined by dividing the care costs, housing costs, education creasing net income replacement ratios retiree's estimated annual Social Security costsind other expenditure categories as preretirement salaries increase (Line benefits by his or her preretirement where the consumption patterns may 1.iTable salary." Net income replacement ratios differ significantly between a working are derived by subtracting the Social Se- From a retirement planning perspec- status and a retired status.Collective- curity replacement ratio from the gross tive, lc can be argued that the greatest ly,c hanges in the selected age-related income replacement ratio. In essence, importance should be attached to the expenditures could either increase or the net income replacement ratio rep- net income replacement ratios. Howev- decrease after retirement. resents the shortfall in needai income er, employees of educational institutions The replacement ratio formula at- at retirement that must be met from an not covered by Social Security (and who tempts to measure the amount of in- employer-sponsored retirement plan or otherwise are not eligible for Social Se- come needed during retirement (i.e., from the individual's savings, invest- curity retirement benefits) should base the numerator) as a percentage of prere- ments program. their retirement income objectives on the gross income replacement ratios tirement gross pay (PrRGP). The sum It is interesting to examine the pat- of the variables in the numerator repre- tern of gross income replacement ratios (Line 12Table 1). Further, in this sents the amount of income needed in on Line 12Table I. The ratios de- case, the gross income replacement tar- retirement to maintain the individual's crease from the S15,000 salary level gets should be somewhat higher at same standard of living when taking through the 5.10,000 level and then in- lower salary levels than the ratios de- into consideration reductions in income crease over the remaining range of pre- picted in Tables 1 and 2.11 taxes and Social Security taxes, the elim- retirement salaries. This pattern is de- Alarrted Cuuple The only consumer ination of the need for further savings for picted in Figure I. unit configuration assumed in both the retirement, reductions in work-related Because income replacement ratios 1988 and the 1991 RETIRE Project expenditures. and changes in age-relat- were not derived for single individuals studies consisted of a married couple ed expenditure variables. in either the 1988 or the 1991 RETIRE with one earner. It was further as- sumed that the breadwinner retires at Current Income Project studies, comparisons with earli- Replacement Ratios er findings cannot be made in a com- age 65 and has a spouse who is three pletely consistent fashion. However, the years younger. In comparison with the The most recent RETIRE Project earlier single individual scenario, mar- Report bases its findings on 199; flattened u-shaped pattern of income re- placement ratios observed here for sin- ried couples (1) pay lower preretirement F1CA tax rates and federal income tax gle individuals is fairly consistent with and postretirement federal and state in- pro. isions and the 1990 Cuniumer Ex- similar results for married couples in the come taxes and (2) receive higher Social fs mhturc .1nritl data. Survey data fin. 1988 study, but somewhat inconsistent Security benefits. These differences are 1'191 and later had not been released with the income replacement results for reflected in the income replacement for public use when the underlying re- ratio calculations shown in Table 2. search was «inducted for the 199; married couples in the 1991 RETIRE Project Report. Prior to the release of RETIRE Project Report. The gross income replacement ratios the findings of the 1988 RETIRE Proj- (Line 12Thble 2) are higher for lower Indmilnal Table 1 contains the ect, it was commonly believed that in- salaries (up through 550,000) and slight- line-by-line calculations of income re- come replacement ratios decreased con- ly lower for higher salaries, in relation to placement ratios for a single individual tinuously over the range of increasing the ratios fiir single individuals. When

PAGE 2 :11(1,111C Table 1 ticker.. \ RM.". 05 Ratio Needed to Maintain Laving Standard Replacement Rano - single Figure 1 : Len1 11sta; Sri tints r.ixrsPreretirernent salan 51 4.04H) i 1-14i 420.000 21.5 j1.), it) 425.1.404) 1.91 4 ',Ink' $ 40,0181 2.2M 510,000 4,1160 540.000 N.-905,825 560,000 11.21- I i 11 S-0.000 1 .:151 1.586 580.000 16.5-44,-11 590,000 19. WI1.8-6 Replacement Ratio Stte 65 Retirement) s 9 IA.Ia., Prerrtittmcntretie,: 4,..148 atter-cal 4 nrs insonic stars' And kis.. int ..1nr Ls." $12.182 +.15 $ 15.86-i ` I $19.540 :.S.i (.99 $22.92,) .41:`,.; $29.019 1.50.: $;5,240 2.158 S-11.656 2.1,:ss $18,4.12 .211 S5.1.912 $61,513 I.271 S291 $ I 4.0 $1.6-1) 52.26s 4.7 1.558 1 ' L.irt, ($ amount4-vp-.0414ctirrs And ptcretircmcnt ..4,1:14; , .",..1,,,,, ..,.,: ot Lint '1 2 $ 4,8 i ; $9012 i-: S 1. -94 561- .2 5-9).. s ; 51.2n85, .928 2': 51.920 --: 51.68- s 4: 51.2-842.'61 h 1 ; s;.-5- 5-21 525 4 -, ,Att,ti.,t rrrrrtirrinrnt suniin.li.t :is 91.4r.,,,rvt:rrnirnt triltrai 444,..nie cast., ,5,,,,,,,- ,,,,,,,,I E,, tim.iiii ii I s ,i, 59 SI 4. 01-1 54. S V c $ 4-- '0 5:.1,, Stii-i2 5.:,.-5 :1, I .W1(1 52,-.8) 5.1.-.-III 56.1-s $ 4 4 58.11s I 1 4 5%14,4 4 4 5..9.91 i 511$5,1,928 Is 7.; ., P.Pis, II. 9rT8.- : l'res rhrtchrerwrr "Are 1,1 ON AI St" 5,, $so 591 51 s: 521- 505 S t .22, 52.02.5 565.1- ; 55./ 1 0. .. 2 (rms. 4444. rime repLit erns nt rano (i lanear ,...."1-^en,1114Line 1.1. $;; 44. "6.9. 4 ;4 I '2. 4.; .,,, sr. H: -1.99. "0.8'4 $2- ',I, 691r ; 3,4.869 69.-'4 SO in" -5.2'f SI$51.25- 4,591, -9.5'1 $1 4.6-4-1 81.-Vi ,4 4. ss .44.4 4 411.44, ti. 49.2 W SW:4 ei i 4 4 512 .044 49 .51 2.Sh I 2 2 '; S14.1.11) 26 w ; SI 4.5-2 226'4 19 4'4 SI 1.02o 1- 0'5 1, 2 : .5.2s 11 $' ~ $ 45 '4,9 ss $00 $.5 S,"44 S'S 4811 $85 I,.. :. ,,,,,4 trr,1,rrili i i. S ; 1" 115Is 5'4.1 Pr 511.-51 521 129 12 )14 $41.5 is 5810.6o) 1.; S-I9.)4S 4 $50 651 60 : Preretirement Salary in Thousands IIPort inn ',an .- In, MIL' Rs plat ellIellf RM., Table 2 A,,,,. h; A orker. Ag,. n2 5.pouse Rani, Needed to Maintain lasing Standard Replacement Rano - Couple Figure 2 1Prerrtirement a Nla ..11 St, irlt salln \ 'a \ trs $14.00/ I I 48 320.1114)\ LIrrIl'll I "Wick' ,1.5 i Illc 411 Vl ant' I arner.; $25.04)0 1,915 5111,1410 2.29s 4411.0011 5.1161) 460,000 5,82s $0.0041 1.4.11 5'0.000 4,586 580,000 .1,-31 590.00 1.);-6 \ ge hi Retirement:N.84A 4rking ,,,, .A44, (421 1 lai ink r,ti.iiinalr :at, ,,i'... 1.A/A 2,115 2.865 1.465 6,151 8,645 11,082 1 4,559 4,0148 1408.1 Replacement Ratio PrerrtirrmentIx. slate .1114I al,'knal cax Inst.file ill. fast, .I/Il 51..I15' 4 IN') $16," -0 145 $20,15 5 520 S21.I 45 4, 444 -05 S 41 501 S i 1,0-1 S58,528 1,196 $44,856 2,06- $51,- 55 2.597 358,622 $65.5 5.522 ' ',S 11.. . 44 . . I .. : ' s i'`1, 4 S. 4" i' -: i SI '425 52.1 49 5 4.15S r 4 $ 4.919 SI.SSS r sr11.11..rt s SS i.. ' 54 ni".. 51. 4-'1 Si SI 5)). $1.01% , 541,1 ' ' 114rt tax rrcrttirerm or 041 tlard .4 4 ,4,4- ',nit .1 r ., di, .., l 9 is, , S1S._ -, So 521." .' 5 _,.2'., 44..,, 50 5)1.si," 5 i ,-' S 411.w, S 4 -.Sr, . $5 1.25' so, ih I 1484 z Pit. tiArtecarmr ..4,14i,41, 'hIrra; cases .4' ...Ill $95, 505,, S14 Sli 31.-21 S28o S.lii- Sol9 $1.19)$1.-S8 $2.156.91- -0 54.44.658.,), 12 : t in Sw.41m in, hetorr"me replai4 aa 411...n4c.9 einem,,,rrr, nehir-1 ratiolit ilS.2 ", S18.55- 5:1 ss in 546.8"5 S i 1.626 554.s i i $64.41 S-4 4155 (.1 tile 11 i 4il me 14' ... ' . 41 ''' i "rr 2' i t.5, : -I im j -ON ; '4 8.1 I V) '6.51 -9.2'1 81.2 ' "r ir :., ...! .11111C I t .. I S. 8. Pio II / 518 it,: I , 515 6iis 26 515.-25 ; 518 -61 i 1. ,IS aniiii, i ,t,ir E. :(.11,171 rg, it ii .1',1 r:11 ; .11i it i'M 'I 5 25 i 20 5', ..r. 's I44 SO. Preretirement Salary In Thousands4 441 ;4 45 Olt iSS $611 Sh5 Si"11 $"5 S1111 $1,S 4.. 4., r- i 4.1 4r, . 51 ill .1. h $2, 1(.. 1 19 I S 4 ss 8; I 2 I III P..111,1 ,,11 Piit 3 PAGE 4 6 piotted, the gross income replacement ratios depict a flattened u-shaped pat- Table 3 tern (s(. e Figure .2) whose shape is con- Comparison of Savings Rates* sistent with the income replacement in 1988, 1991, and 1993 Studies ratio pattern observed earlier under the Ages 50-64 single individual scenario. Preretirement Salary 1988 Study 1991 Study 1993 Study

Social Security benefits (LineI SI5,000 A. -0.8q 2.-0 Table 2, and Figure _2)are 4-.5 percent S20,000 higher for married couples. reflecting $25,000 3.222i the extra Social Security amount 5;0,000 1.9'; payable to a 62-year-old spouse based 540.000 10.5c; on the earnings history of the bread- 550,000 winner. As a result, net income replace- 560,000 12.V; 5.. -; ment ratios (Line 14) are considerably 5-0,000 * * 10.(r; 6.1"; 6.8e; smaller (-Or married couples. particular- $80,001) 1 L-I'; ly at lower preretirement salaries. $90,000 As a percentage of alter-tax income Savings Rates and Their '* Salary level not examined Effect on Replacement Ratios benefits. The 199$ RETIRE Project, The savings variable exerts great in- come replacement ratios (Line 12 Table 2) are generally higher than the however, also examined the effects iluence on the size of gross income re- of the Clinton proposal on income placement ratios.I- In addition, the comparable ratios in the 1988 study but lower than the ratios derived in replacement ratios. The findings -c r ular" relationship between savings indicate that increasing the taxable and replacement ratios suggests chat the 1991 study. Because of signifi- portion of Social Security benefits care should be exercised in the interpre- cantly reduced savings rates, together with changes in taxes and work- and from 50 percent to 85 percent causes a tation of a particular set of income re- corresponding increase in gross in- placement ratios. "Circularity.' in this age-related expenditures, the 1993 come replacement ratios of 3.5 toII case means that as preretirement sav- gross replacement ratios at 550,000 percentage points for single individu- ings decrease, preretirement consump- and above are higher than the ratios als at preretirement salary levels of tion increases and derived gross income derived in both the 1988 and the replacement ratios become larger, 1991 studies. In factit the highest S60.000 and above. For married cou- ples. increases of 2.8 to 4.- percentage thereby creating a need for additional levels of preretirement salaries exam- points are observed for preretirement savings; similatly, as savings increase to ined, the 1993 gross income replace- ment ratios are substantially greater salaries of S-0,000 and above. At meet a higher replacement ratio, prere- lower preretirement salaries, no tirement consumption decreases, and than the ratios calculated in the earli- change or little change occurred in derived gross income replacement ra- er studies. nos become smaller, creating a lesser gross income replacement ratios be- Effects of Taxing 85 Percent cause the dollar amounts of available need fbr savings. of Social Security Benefits exemptions and deductions (for in- Table contains the savings races The Clinton administration has come tax purposes) substantially re- derived in the 1988. 1991, and 199$ proposed taxing up to 85 percent of duced, or eliminated, the portion of studies. It is evident that savings rates Social Security benefits for higher- Social Security benefits that was taxed. computed in the 1991 study are con- income retirees. Currently, no more siderably smaller at lower and middle than 50 percent of Social Security Conclusion: The Use of Income Replacement Ratios levels of preretirement salaries, in benefits is taxed. Under both current comparison with the savings rates cal- and proposed legislation, Social Secu- The research findings contained in culated in the 1988 study. The 1993 rity benefits are not subject to federal Tables I and 2 indicate that individu- RETIRE Project shows a considerable income tax if the aggregate of one- als:couples retiring in 1993 need ap- reduction in savings rates at the highest half of the Social Security benefits to- proximately 69 percent to 82 percent salary levels examined. The overall de- gether with "modified adjusted gross of final-Year salary (adjusted for cline in savings rates measured in the income" is below a threshold amount postretirement inflation) to sustain RETIRE Project is consistent with (525,000 for individuals. S32.000 for their preretirement standard of living other studies indicating significant de- married couples). I into their retirement years. After ac- clines in individual savings in the Unit- The calculations of gross income counting for estimated Social Security ed States. replacement ratios in Tables1 and 2 benefits, additional retirement income At salary levels through S f0.000, are based on the current federal in- needs range from approximately .26 the I90 ', RETIRE Project gross in- come tax treatment of Social Security percent to 66 percent of final-year \X'ashington. D.C. Government Printing Of- tOr single individuals and approx- ment ratios are partly afunction of fice, annual). imately 12 percent to 60 percent of pre- taxes, calculated ratios will be likely to The RETIRE Prolect Report also presents income retirement salary for married couples. change any time there is a significant replacement ratios calculated under a second Mr- based on the assumptions used in the change in FICA taxes or federal or state mula, reterred to as die Tax and savings Model This formula ignores changes in work-related ex- 1993 RETIRE Project Report. (and local) income tax laws. penses (WRE1 and age-related expenditures Third, postretirement inflation is not (NCASE) and simply considers savings (PrRS) Gross and net income replacement and differences between prerettrement and postre- ratios provide valuable guidelines when factored into the income replacement t irement taxes iPrRT and PoRD. used as management and financial plan- ratios provided here. Instead, these ra- Preretirement savings (PrRS) are measured by ex- ning tools in the larger context of retire- tios represent estimates of the percent- amining the savings behavior of a group of con- ages of final-year salaries that are needed sumer units whose -reference person- (i.e., head of ment income planning. They provide household) (a) is working and (b) is between the employers with useful benchmarks in during the first year of retirement. To ages of 50 and 64. Other preretirement expendi- setting their contribution rates and ben- maintain the same sto. dard of living ture behavior is measured in a similar fashion. during subsequent ye-ars of retirement, Postrenrement expenditure behavior is measured efit formulas. They are also of consider- through the examination of Consumer Expenditure able value to individuals interested in the retirement income amounts must be Surtey data of consumer units where the "reference establishing their own retirement in- ad justed to compensate for any increases person- (at is retired and (b) is between the ages of in the cost of living. 60 and '74. Unfortunately, the consumer expendi- come objectives. However, several criti- ture data do nor permit the calculation of income cal issues remain that may limit the Finally, the net income replacement replacement ram- under the more-preferred kin- gitudmal approach where expenditure and savings overall effectiveness of these ratios as 1 ratios displayed on Line 1-i of Tables patterns of specific individuals and families are measures of needed retirement income. and 2 assume that anticipated Social Se- monitored over an extended period of rime. in- First, the income replacement ratios curity benefits (Line 13) will beavail- cluding periods of retirement aS aS active par- able when the worker retires. In ticipation in the labor force. presented here should be used simply as 'In the tabular presentations of the income replace- guides for individuals in establishing essence, it is assumed that the Social Se- ment rano calculations. WRE and NCASE have their own retirement income objectives. curity program will remain financially been combined into a singular variable. This as solvent and that it will not be trans- necessary bvcause some of the individual expense The RETIRE Project's findings are categories in the Comumer Expenditure Surrey data based on average savings, expenditures, formed into a welfare program where do not permit the separation of the expenditures and income taxes of subsamples of con- only low-income retirees and their ben- in terms of whether they are work rela...1.1 or age sumers from whom data are collected in eficiaries are eligible for retirement related. benefits. If it is perceived that Social Se- 'Consumers with incomes. in excess of $100,000 the Consumer Expenditure Survel Conse- are "top coded" at $100.000 an the Lumumer Ex- quently. the calculated income replace- curity retirement benefits will not be penditure Sone . Consequently, the RETIRE Prot- ment ratios themselves represent aver- available in the future (for whatever ea makes no attempt to derive income replace- reason), or payable only at significantly ment ratios for individuals and families with ages. and should not he viewed as the incomes of S100,000 or above. exact amount of retirement ir-:ome reduced levels, then retirement plan- o Social Security benefits are estimated by assuming needed by any specific individuat Thus, ning decisions should focus on gross in- that the individual's grew at an annual rate while published income replacement ra- come replacement ratios, in lieuof net equal to the increase in national average wages. replacement ratios, in establishing re- The basis Mr this statement is that the RETIRE tios are guidelines, their availability Project assumes that all retirement income comes should not be viewed by individuals as tirement income objectives. either from a tax-deferred retirement plan or from eliminating the need to make replace- social Security benefits. It is Mrther assumed chat the income trim, the retirement plan is fully tax- ment ratio calculations based on their End notes able, and the Social Security benefits at lower own specific savings and expenditure President s Commission on Pension Policy. (//i- salary les els will not be taxable. behavior. Worksheets for calculating in- inev AtxTfmard a National Retirement !mount- Pot- I2The RETIRE Project defines "savings" to include dividualized income replacement ratios hliW.Lshington, D.C.. Government Printing Of- (1) net acquisition of stocks and bonds; (21 net in- fice. 1)81). vestment in farm or business; (3) net change are available from several public sources. =RETIRE is an acronym for the Rairee Income in savings and checking accounts; (1) net change in money owed the consumer unit, Second, because of the circular rela- Replacement Project. The RETIRE Project is part of the Retirement Financing Program ar the (5) net c 'ange in U.S. savings bond holdings: tionship between preretirement savings Center Mr Risk Management and Insurance Re- 6) contributions to retirement plans; and and income replacement ratios and the search, (;eorgia State University. Atlanta. Georgia. ) surrender proceeds. Admit- Bruce A. Palmer, 1993 GSU,AACG RETIRE tedly, this is not a perfect measure of savings: in- variability in these ratios observed over deed, a perfect measure may not even exist! This time (as described earlier), it is recom- Prsuer Report (Atlanta: The Center for Risk Man- agement and Insurance Research. Georgia State particular &Caution is selected because data on these seven items are included in the 001511111er mended that retirement income objec- University. 1993). ryPendituri Nunn. Although somewhat limited ' Bruce A. Palmer. fht I mpa.tf 'fax Rilorm tives not be established on the basis of in value as a measure ot absolute savings, this defi- income replacement ratios calculated Wage Replaamern Ratios (Atlanta: The Center Mr Risk Management and Insurance Research. nition of savings can still be of substantial use .1.5 with respect to any single year. Rather, (ieorgia State I'mversity, 1988): Bruce A relative measure ot savings to depict change.. in s,wings ratec that oecur over time. it is important to develop income re- Palmer, 1991 GSU AACG RETIRE ProInt Re- `For an extensive treatment ot the tederal income placement targets based on replacement port (Atlanta. The Center tor Risk Management and Insurance Research. Georgia State Unwersi- taxation of Social Security benefits, see, tor exam ratio research conducted over a multi- ple, Burton T. Beam.Jr.ind John J. Mc Fadden, iv. 19911 year period. In addition, it must be rec- Emplosie SRI ed. (Chicago. Dearborn Fi- I* S. Department ot Labor, Bureau ot Labor nancial Publishing, Inc.. 1992), c2 ognized that because income replace- Statistics. tootowr I:AA/la:turf Sun Intertren %us--

Teachers insurance and Annuity AssociationCollege Retirement ,puities Fund 730 Third Avenue, New York, NY 10017-3206