A detailed look confirms that older Americans (50 or above) spend less in retirement, and that home-related expenses remain the top spending category. But health costs are the second-biggest expense for older Americans, and data show that demographic sub-groups such as singles, blacks, and high school dropouts are outspending their resources in retirement, according to a new report by the nonpartisan Employee Benefit Research Institute (EBRI).

PRE- AND POSTRETIREMENT EXPENSES: Before retirement, people pay FICA taxes, incur work-related expenses, and set aside money for retirement. But after retirement, most people have different financial obligations, and, as a result, retirees may still be able to maintain their level of preretirement well-being with very different income levels. Studying income, expenditures, and wealth-holding patterns together provides a more complete idea of how people are doing in terms of being able to afford retirement than arbitrary estimates such as income replacement ratios.

UNIQUE DATA: This Issue Brief examines the expenditure patterns of the older section of the population. It uses data from the Consumption and Activities Mail Survey (CAMS), a supplement to the Health and Retirement Study (HRS), conducted by the Institute for Social Research at the University of Michigan, contains detailed expenditure data on 32 categories, and follows the same group of individuals over eight years In addition, the income and wealth data available in the HRS are used to establish the financial standing of older households.

DECLINING EXPENSES: Household expenses steadily decline with age. With the age 65 expenditure as a benchmark, household expenditure falls by 19 percent by age 75, 34 percent by age 85, and 52 percent by age 95.

HOME EXPENSES: Home and home-related expenses remain the single largest spending category for older Americans. On average, those over age 50 spend around 40 ? 45 percent of their budget on home and home-related items.

RISING HEALTH CARE EXPENSES: Health-related expenses are the second-largest component in the budget of older Americans. It is the only component which steadily increases with age. Health care expenses capture around 10 percent of the budget for those between 50–64, but increase to about 20 percent for those age 85 and over.

DEMOGRAPHIC GROUPS: Singles, blacks, and high school dropouts do not have a sound financial standing in retirement. Their expenditures exceed their income and they hold very little financial wealth. The bottom income quartile, which includes mostly these demographic groups, has the weakest financial standing in retirement.

LONG-TERM CARE AND PRIVATE HEALTH INSURANCE: Long-term care and some form of private health insurance coverage have a significant effect on increased spending by older households.

“Expenditures” Defined 5 section for To ana Fig Figure with wealth Figure References ure at least o is i l6A sho 6B, 10 yze thi s n Me det ho cluw w dian s d n ails). , a regression s ed, s how e mem the Househol these “Expen me the ber dian e m age 65 or fdi fe e d Incom income, tures” ar dian cts decr framework is house over, t e me ease e t , Ho h hdi o e usehold Spen both i an e ld i h total of used e median n th xpe nwher ma e ndit ex botto gnit pen h e i ur ding, ee, an n m half al ud d come an itur th e e an Househo d me es x of p d in in end the d oth dia all c ter itu n l i d wea n m ategories ree was Nonho come r demogra s of statistic lth- $3, distribution i using Wealth, holdi de 15 p fi n 0 hi ned earli g al sign ,c characteristi but patterns the nifica ev Household Total e mea r. ery o nce. Fi f tnally, a age n was he But cs can bottom-inc grou LTC i $4, house be p is 89 ncont sur doing 4, h and ome old ance rolled is at in Sudipto Banerjee is a research associate at the Employee Benefit Research Institute. This Issue Brief was written with Expenditure Patterns of Older Americans, 2001 ?2009 ? th Home-related expenses include mortgage, property taxes, homeowner’s or renter’s insurance, rent, Figure 3 still remai retireme considere quartil the 90 for, to se Wealt e se pe nt in e i d n h par r rce s a f(Housi L e t tately for n ntile o T ired e import C insur rms of if ng a eith f ex worki ant ance bala npen d No er t determina n s has any si he res n cing their es it was nho g an using), in d retir pond nt of inco gnif $1 ee nt 0 d househol ispendi , 2 c m or th 262. ant a 0 e an 10$, e spouse reports dditio ng. d for th expen To ds n during a understan e ls ef es. To fect p Hal In eac the on t being retired. surv f d t of h o h h th e e magnitu ey years o el f the e I derly’s n fo come D ur s The d ex fe o 2 urv pe 0 istribution o median f01– nditur the ey years betw 20 ee. T ffect 07. an Not h nua of fe D n surprisingly, il LTC insura ff e een 20 income xt sectio erent A 01 of ret g n e n an ce discusses d 200 data (f ired rom on 7, assistance from the Institute’s research and editorial staffs. Any views expressed in this report are those of the author, Butrica, Barbara A., Joshua H. Goldwyn, and Richard W. Johnson. “Understanding Expenditure Patterns in Retirement.” By Sudipto Banerjee, Employee Benefit Research Institute Figure 5 Appendix C utilities, home repairs, home furnishings, housecleaning supplies, housekeeping and laundry services, Percentage of Population Who Anticipated a Drop in Spending in Figure 6A column households almost the i the reg Groups ncome revery age grou 1 o ession results. f is ........................................................................................................................ ga Appe $30, p an ndi 48 d wea x 0, w A p had )l, th hic nholdin o h is arou ex te that, pen gs show ditur nd whi es that 5le 7 that a on percepeo e-yea ent o xcee ple d fr in in wo thcrea the ei rking r in bottom se in come. I househo educ in n ation inc come quartile lds’ m all other survey years except edia rease n an sare struggli nual spending by income ..................... 12 ng, o 3 percent both f 2001, $53, in 548. the 50 their age Th points, e pre- ?64 and sho CRR Workin uld not be g Paper ascribed t No. 20 o the 05 ? offi 03. C cers, hestnu trustees, or ot t Hill, MA: Cen her spo ter for nsors of EBRI, Retirement R EBRI esearc -ERF, or h at Boston C their staffs. ollege, 2 Neith 0e 0r 5EBRI . Median Household Spending and Median Retirement vs. Percentage of Population Who gardenin Media g a n Hous nd yard su ehold Inc pplies,o and ga me, Hous rdenin ehold g an Appendix A d yar Spending, Hous d services. ehold Nonhousing Wealth, The two components of expenditure which do show a declining pattern across age groups are transportation expenses Median Household Income, Household Spending, Household Nonhousing median age LTC ins and grou postretirement years. ura an p sho nua nce coverag l wex s this penditu dissaving e incr re I nof retir eas all four e pattern. s s ep d surv en hou din Tey years, s h g eholds is is could be co by 10 perce both $31 grou ,ntage 365, ncerps re wh nin poi g ich n port beca ts. The is income around use those effect 8 that 0 of private healt close to r per isc less than ent of e th tireme e the h working insurance n ir ex t might pehouseho ndit bcov e ure. ex ep rage For lds’ ected is nor EBRI-ERF lobbies or takes positions on specific policy proposals. EBRI invites comment on this research. Figure 7 Househol Actua d Inlc lyo Experienced me (in 20 Figure 9 10a $)Drop, for W 2001 ork ?in 2009 g and Retired Figure 7, Median Household Income, Household Spending, Household Nonhousing Wealth, Household Total February 2012 • No. 368 Butrica, Barbara A., an Household T d Gordon otal W B.T. M ealth ermin. “A (Housing and Nonhousing), in nnuitizFigure 13 Figure 12 ed Wealth and Consumptio 2010$, for S n at Older A ingle M ges.” a CR les R Working Paper 4 Regression Results Introduction Although and enterta ever inm yone a ent e Wealth, Household Total Wealth (Housing and Nonhousing), in 2010$, for the x gpense e 65 o sr. Again, t over is cover his ised expecte by Medi d. car With e, it retir is w ement, ell kn daily own th transpor at Meditation n care covera eedsge (such as comm is not uting ? Food expenses include food and drink, including alcoholic beverages that are bought in grocery and other Median Household Income, Household Spending, to be saving retired similar to th median hous expe at en more to hd olds of a o iture , the of n“catc e-year $ ga 39, p h945. u betw in pcrease i ” wit een h th innco eir retir edu me an cation. ement d exp saving en Figure 2 diture s. A sleso, ems their l to be iqu ris id an ing ov d total er time, wealth poss holdin ibly becaus gs are n e of ot large Households Age 50 or Older, 2001 ?2007 Wealth (Housing a Median Household Spending in Each Category and Mean Percentage nd Nonhousing), in 2010$, for Different Marital Groups Age 65+ ............................... 14 Median Household Income, Household Spending, Household No. 2006 ?26. Chestnut Hill, MA: Center for Retirement Research at Boston College, 2006. Median Household Spending in Each Category and Mean compreh to work) fa ensiv ll to e: T a great here are exte Bottom c Int, an n acps on t ome half of the Income Distribuiton of Different Age Groups d with he nu in Ember o x creasing age a penditufre days spen nId ncdecl o t i mn a einin Gap nur g h sing hom ealth Nonhous peo e th ple ing W a become more t can ealtbe h cover T r e oe d ta stricted to th l W by M ealt eh dicaree , and Regressions Without Wealth Regressions With Wealth According to stores. Dinin economic th g out is not eory, “well incl Household Nonhousing Wealth, Household -be uing” is measur ded . ed better by consumption rather than by income. Furthermore, Where the world turns for the facts Median Spending, in 2010$, in Each Category and Mean Percentage on U.S. employee benefits. increas enough to ing h support them in c e $6a 0lth care ,000 costs. Als ase of o, both large e groups xpenditu hold re sh very ocks. little wealth that could support them through large 70% 66.5% 65.9% Share of Each Category in Total Spending for Individuals With and 6 Nonhousing Wealth, Household Total Wealth (Housing and 2001 1 Percentage Share of Each Category in Total Spending for Mean Regression Median Regression Mean Regression Median Regression However, C there indoors, opyright are co which this 80 perc Inf -payments even cuts e ormation: nent tertain co This re nsumption befor ment e eport h xpense itting replacement is copyright the ca s. Food and cl p. e shou For M d by the Em othing ld not edicar be e ex ploye Pa pe equa rt B nses (as a e Be services, ted nef with it R share an e gen se 80 arch Institut of total e perc rally ent 80 e perc x inco pen e (EBRI m ent of diture e replacem ). It ) the r may emain exent: penses be more Income Total Wealth (Housing and Nonhousing), in 2010$, Expenditure Income Gap Nonhousing Wealth Total Wealth Figure pre- an8, M d pos edi tretirement an Househol Share of Each Category in Total Spending, 2001 spd endin Income, Hous g requiremehold Spen ents are ver din y dif g, Househo ferent: Be ld fore r Nonhousi - etireme 2009, by Age Groups ng Wealth, nt, people Total H pay FICA ousehold taxes, incur expenditure shocks. Also not surprisingly, groups that are doing less well in retirement—singles, blacks, and high school Butrica, Barba ? Health ra A., Gordon expenses B.T. include out-of Mermin, and Richar -pocket (uni d W. nsured) Johnson. health insu “Do Healt ranc h Pro e costs, incl b To lems R tal Spee nuding M d duce ing Consumption edicare suppl at Older emental Separated* $9,114 Without Private Health Insurance, Age 65+ $8,431 $683 $2,788 $7,380 Conclusion Nonhousing), in 2010$, for Different Educational Groups Among $53,548 Individuals With and Without LTC Insurance, Age 65+ 2001Log of Total Expenditures Pre- are cover F or less flat across d igure an6B shows the same d pos ed by Medic tretirement ifferent ag are, m spe ne ndin uanin mbers e ggrg th r oup equ for th at 2 s. ire0 m e top hal perc entsent arefare t ver of tyh h dif e i e pa n fer come tient ent. P ’s re dist reretirem ribution. sponsibility, ent, Here a pen ople it is cl d ou pay t-of ear that FIC -pocA ket ex taxes in allpen years a , incur ditur work- n es can d at all used without permission but citation of t for Different Marital Groups Age 65+ he source is re Figure 11A quired. work-rel (Housing ated expenses, and and Nonhousing) Wealt set money asi h, in d 2010$, for Different e for retirement. But no Racia ne l Grou of these savings or ta ps Among Households A x requirements exist ge 65+ ... 14 Retirement and health benefits are at the heart of w 50 -64 65 -74 orkers’, emplo 75 -84 yers’, and our nation’s 85+ dropouts—are Ages?” CRR ge Working nerally ove Paper rwhelm No. 2 ingly 009 r-9. C eprese hestnut Hi nted in Figure 11B the lo ll, MAw-i : Center ncom for e gro Retireme up. nt Research at Boston College, 2009. Divorc insurance; ed out-of-pocket 31,437 costs on pr 20,767 escription and 10, nonpr 670 escription dr52, ug410 s; out-of-pocket cost of hosp 99,131 ital care, This study att 60e %mpts to document the income and expenditure patterns of Americans who a Total Income re retired or close to Log Income 50 ?64 $28,845 $27,230 .219*** No Pvt. Health Insurance Households Age 65+ $1,615.253*** $27,781 Pvt. Health Insurance .163*** $95,449 .200*** No LTC Insurance LTC Insurance Median Spending in Each Category and Mean Percentage related e rise very Expenditure Patterns of Older Americans, 2001 ages, respon x fast. pen dents’ earn sPeopl es, and e try to save ings are suffici mitigate for retirem such ee nt nt to cover health . But none spen thei of din r t ex gh risks by ese savings penses. As obtainin expected, or tax r g private equirements exists postretirement. the health margin insuran bet ?2009 wee ces, ge n income a nerally nd So known economic security. Founded in 1978, EBRI is Income % of total Expenditure Income Gap % of total the most authoritative and objective source of Nonhousing Wealth % of total Total Wealth % of total postretiremen Widow t. As a resu ed lt, 27, even 082with very differ 22,695ent post-retirem 4,387ent income lev 112, els re 950 tirees may still221, be able to 021 afford doctor service $50,000 s, lab tests, eye, dental, and nursing home care; and out-of-pocket costs for medical supplies. % of total expenses % of total expenses Median Spending in Each Category and Mean Percentage Share of Each Category in 65 ?74 20,912 $ Amount 23,060 (0.015) -2,147 (0.014) $ Amount 32,014(0.015) 102,553 (0.015) Figure retireme 9, M nt. T edi han Househol e results shod wIncome, Hous that while high-incom ehold Spen e ho din ug seh , Househo olds are m ld Nonhousi anaging tn heir g Wealth, a income an nd Tot d exal pens es well in $ Amount % of total expenses $ Amount % of total expenses Recommended Copeland, Craig. “Ho Citation: w Are New Sudipto B Retireaes D nerjo ee, “Ex ing Fin pa encial nditure P ly in Retirem atterns o efn Ol t?” der Am EBRI Issue Brief ericans, 20 , no. 01 - 3 20 009,” 2 (Employee EBRI Issue Brief Benefit , $ Amount expenses $ Amount expensesIncome $ Amount Nonhousing expenses Total $ Amount expenses Category Share of Each Category in Total Spending for Households even i as expenditure Medi f rgap eNever M tire po shrinks for eslicies ’ postretir arried . Als the oe , s m o 19, e more adva me retir nt 982 income ees nce is l ar d e e covere 31, a ss than g 954 e grou d 80 ps. by r percent Thei e2001 tiree hea -11, r li of t 972 quih dleir th insurance from an prer d total etirem we 28, alth ent 222 income holdin their own or gs (whic are h h it a spouse’s igh 69,is in 443 enou aggre gh previous to gate), information on these critical, complex issues. the consumption expenses needed to maintain their level of preretirement well-being. By Sudipto Banerjee, Employee Benefit Research Institute Log Wealth 75 ?84 15,922 18,688 ? -2,767 2001 ? 28,405.086*** 105,335 .075*** Total Spending for Households Across Different Income Quartiles and Ages 75+ 2001 Comparing retireme Household ? nt, lo Transp w-in Consumption Wealt come ortati h hou o (Housi n s exp eholds ar ng a enses nPatterns d No e st inclu nho rugglin de using), in car g. T of payments (prin hDifferent e 2 hig 01h 0$, -incom for cipal Dif e h Income Gap fere o and int useholds ma nt Educ Wealth er Groups est), veh ational intain Grou icle high l Wealth in pssurance, ve Amon evels of g wea hic lth, le but 50% Educational Level Income Expenditure Research Institute, 2007). 2003 2001 no. 368, February 2012. Couple $46,805 $33,958 $12,847 $206,654 $373,740 Across Different Income Quartiles and Ages 65 ?74 they may still employer. absorb l arge e maintain xpenditu 80 re sh pe ocks. However rcent of their prer , this do etirem es not ent mean consumptio that no n, provi one in t ded they he top ha have saved lf of the ifo ncome r retirem distribution ent; in fact, 85+ Home 12,010 $8,122 15,368 -3,358 39% $8,584 18,655(0.007) 34% 49,547(0.006) Home $8,136 37% $9,902 33% Home Smaintena eparated* nce, and gas $13,486 $22,Bottom Quartile 023 . 38% $21,616 $9,449 Second Quartile 2001 36% $407 $7,529 Third Quartile $190 35% $6,155 Top Quartile $3,220 39% Fig wheth ures Households er th 11A a esend wea A 11B gle 65+ th show Single levels ............................................................................................................ how m will be 18,780 edia sufn s ficie pen nt to su 21,800 ding (in pport 20 t 10 -3,020 hem thro dollars) i ugh n different categories varies a very advanc 30,265 ed ages 65,468 or in cas cross income ............. 15 e of catastrophic quartiles Figure 1 The study o EBRI focus Age f consumption es solel Food or y expen on emplo $39,94 ditur 5 2,954 e -.004** y patterns provi ee benefits research — no lobb d15 es 2003 a -.005*** better sense of retir 3,201 -.005*** ees ying ’ well or advocacy 12 -being than ar -.007*** bitrary faces th this study’s r e risk esults show tha of running shotrt of Bottom Quartile the m wealt edian h in household in the case o Second Quartile f u this nexgrou pecte p o d la f rre gtirees e ex Third Quartile pen is doi ditur ng e shocks. One well in retirTop Quartile e thin ment. B g tou note t ther from e are Food 2,954 14 3,265 9 EBRI Employee Benefit Research Institute Issue Brief (ISSN 0887 ?137X) is published monthly by the Employee Benefit Research Institute, Hurd, Michael, and Susann Rohwedder. “Economic Preparation For Retirement,” NBER Working Paper #17203 (July Food HS Dropou 4,432 t 13 $18,7803,457 $21,171 13 -$2,391 2,881 $18,54613 $71,720 2,122 12 Divorced Widowed 24,28819,204 22, 21,592 873 -2,388 1,415 48,273 44,667 140,317 134,589 $40,000 Appendix B Report availability: This report is available on the Internet at www.ebri.org for households with expenditure shocks is beyo members ages nd t $ he sco 65–74 and 75 a % of total pe 2 of this stud$ n y. d o But ver, re % of total for lo spectively. Both w-income househol $ fig % of total ures show ds that are some $ already stru expected trend % of total ggling, such s. For 50 ?64 $27,248 Med $29,496 i(0.001) an Total Ho -$2,248 usehol (0.001) d Annual $29,627 Expe (0.001) nditure, $114,213 (0.001) Figure 13 exa 40m %ines wheth Health er any such for 2,325 m of private he13 alth insurance covera 4,428 ge is associated 18 with expenditure. As with 11 ? 00 13Clothing th S EBRI stand t. NW, Suite expenses 8s alone in em 78, Washi inc ngtlude on, D clothin pl C,oyee ben 20005g a -40n 51, d efits rese at appar $300e pl er ar (inc ych a ear or luding jew sis an inde include edlry) as pend pand a art eof nt, nonprofit, and no a m lso personal car embership subscr eipt produ ion. npa P rtisan cts and eriodi- estimates such as “replace Health ment ratios.” $ % of total 3,139 And if income, ex 16 % of total penditures, an 6,007 d wealth- % of total holding 18 patterns can% of total be studied Figure Figure some demo 10, 6 (parti Me graphic dian cularly Househol grou 6B) is t ps td Incom hat are he discrete ju fac e, Ho ing usehold Spen mp in a diffic non ult r he ou tir ding, sing (or finan ement Househo . Also, lc it d ial) should be Nonho wealth bet using Wealth, emphasize ween th Household Total d e tahgat e groups 50 the older po ?64 pu an lation d Health HS Grad 2,954 11 30,081 3,314 26,377 14 3,7043,570 92,40418 206,961 3,415 22 2011). W idowed 33,408 27,144 6,264 143,001 260,121 2003 Amount Longitudinal Changes in Median Household expenses Amount expenses Amount expenses Amount expenses cals Race postage ra 65 te ?pai 74d in Was 22,821 hington, DC, an 26,474 d additional mailing o -3,653 ffices. POSTMASTER32,780 : Send address changes 116,395 to: EBRI Issue Brief, 1100 Transport 2,148 11 2,637 11 exampl PRE- A events w e, m N ill o D P edi n orga lO y an s make matter STR nization. It analyzes a peE ndi TIRE ng Amount g MENT o s worse. es upexpenses wit EXP Ther h i E nd rep n N e are cS ome, wit ES: $ Amount in o several rt Bef 2010$ s re h th osea re ke retirement, e hi , r expenses y demo A ch data cr ghest oss A - graphic inco without spin o $ Amount people pay FI gm ee quartile s 50+ groups t expenses r spendi h unde aCA taxes, incur work-rel t are rlying a ng almost three also not $ Amount gend doing w a. All fin expenses times ell in ated d the ings, Transport 2,416 11 2,947 11 LTC insuraservices. nce coverage, this figure shows that people with some form of private health care coverage spend more in together, Transport this can provi College and Above de4,247 a more compl 14ete 46,210 idea o 2,927 f how 35,902 people 12 are 10,308 doing i 2,091 n terms 253,464 of 10 being 419,649 able to 936 afford reti 7 rement. 65 faces h ?74. As ex igh Never M unc plaine ertainty r arried* d above, egar 22, d gene ing h 540rally, ealthpen expenses. s 22, ions are not 368 For ex in ampl clud 172 e, ed in the wea long nursing-home lth m 40, easures. But, 760 stays or prolo upon retirem nged use 81,038 of ent, lon if g-term Wealt 13th St. N h (Housi W, Suite 8 ng a 78Couple ,n W d No ashinnho gtonusing), in $47,918 , DC, 20005-4 2$39,944 0 01 51 0$, . Cofor Households i pyright 20 $7,974 12 by Employ n th ee B e B $190,718 enefit ottom Income Quartil Research Inst$356,971 itute. All e ri.................. 15 ghts reserved. No. 368. 2001 White (dropped) 75 ?84 Clothing 17,871 20,477 369Income, Household Spending, -2,606 4 517 32,566 3 123,855 Clothing 431 3 640 4 whether on financial data, options, or trends, are reve 2001 aling and reliable — the reason EBRI information is expen amount of retireme ses, an nt, an thod set aside m s d th e in t ey may he botto beo at risk o m-income ney for retirement. But f ru qua nning short o rtile. Each inc af f wealth te ome qu r retiremen at som artile s e t, most peopl p poi en$ ds the n 31, t i3n 65 re hi tir ghest share e e have differen ment. Severa of tl of hteir financial the income study’s on home Goda, Gopi Sh $60,ah, Ezra 000 Golberstein, and David C. Grabowski. “Income and the Utilization of Long-Term Care Services: every cate Clothing gory and overall739 as well. Aga4in, in 2009, 616 the me2005 dian total spending 4 369 for thos 3e with some private h 185 3 ealth 30% 2003 $30,480 Single 16,969 20,348 -3,378 21,764 80,995 someone rolls care servic ? Home Entertainment es c over his or ould deplet her e t expenses $5,399 he acc 401(kumulate ) or inclu DC 41% d w account de tr ealth ips a $6,377 ra balan nd vacat pidly. Thus, ce into 37% ions, an even tickets to mo IRA $8,000 if or takes a retirees hol vies, sporting, o 34% lump-sum d enough $9,778 distri wea r per bution lth to forming a 34% an fina d i nce t rts events; nvests it heir Black 85+ Entertainment 11,071 1,280 17,064 0.046 -5,993 10 .057* 2,240 22,130.106*** 11 48,419.098*** Entertainment Household Nonhousing Wealth, Household 1,625 10 3,102 12 Homethe gold stan$6,655 dard for priva 39% te analysts a $8,794nd decisi 38% on make $9,946 rs, governme 32% nt policy $13,356 makers, the media, and 35% Figure Table 11A, of M $3 Contents e 0,dian 000 Spendin   g in Each Category and Mean Percentage Share of Each Category in Total Spending Entertainment Separated* HS Dropou 3,484 $10,240 t 12 $21,459 $16,205 2,364 $22,150 -$5, 12965 -$6911,530 $18,686 $12,120 10 $83,121 638 $13,029 7 This obligation and important fin Evidence from home-rela Issue Brief s, and, as a resul dings: ted examin ex the Social Sec penses, Widowed es the consum t, retirees m althoug u 20,920 rity Benefit h th ptio e s a n patt y still be able h 25,498 are o Notch,” ern o f total f t NBER h e e older sect -4,579 to xpe Workin maintain their level of p nses spent g ion Pa of t pe on hom 53,699 rh #1 e U.S. 607 e a po 6 n(Jun d rpulation. eretirement related items e 2 152,237 01 T 0). h e well-being with very majority of th decreases wit e h insurance coverage and those without were $38,091 and $30,117, respectively. And people with private health Food 1,920 16 2,216 14 2,913 11 3,570 10 Other 923 (0.031) 92005 (0.029) 1,847 (0.032) 11 (0.027) in othe regular e r forms xpen hobbi s of es, this analy es an wealt d l h, it eisure e will sis cannot be qui inc pme lude qua nt (photo dn in tify thwhat e w graphy, rea ealt woh uld measur happen ding, c e. T a in mpin h is may ex the f g, etc. ace o ); plai f dinin su nch lar the g o measura gue t i healt n restaurants, c b hle jump in expenditure fi afes, an na shocks, ncial d Other 1,231 10 2,658 13 Food 2,807 16 3,201 13 3,841 14 4,432 9 the public. Total Wealth (Housing and Nonhousing), Other 1,600 8 1,477 9 1,354 11 985 11 Divorced HS Grad 26,634 31,710 23,738 30,469 2,896 1,241 103,235 41,958 217,376 113,673 for Households Across Different Income Quartiles and A 2005 ge 65 ?74 ......................................................... 16 different income levels. Studying income, exp higher income. The percentage share of total exe pe nditur nditur es, e spe and wealth-holding patterns together provides a nt on other basic goods (like food and health) also more goes Introductio households Hispanic Health n studie .................................................................................................................. 50 ?d 64 here ha $28,080 ve eith 2,261 er re$29,895 ac 21 -0.074 hed retir2,909 emen -$1,816 t age or a -0.057 20 re on the 4,864 $18,401 cusp of -0.061 19 retireme ............................ 4 nt. T 4,160 $74,558 he data -0.039 come 15 from the insurance coverage spend Total Spending more on healt 25,183 h and entertainment items as well. 30,459 Again, one obvious explanation for higher 20% Total Spending 26,910 39,119 Health 2,364 The Em 15 ployee Benefi 2,967 t Research Institut 15 e 3,831 (EBRI) was founded in 1978. Its 14 4,846 m 13 ission is to wealth for and they ar the 65 diners; and e not inclu ?74 a de ta ge ke-out food. d i gr noup. the calc ulation s provided here. Selden, Total Spending T., K. Levit, J. Coh 37,712 en, S. Zuvekas, J. Moell 30,805 er, D. McKusick, and R. Arn 0 26,121 ett. “Reconciling M 20,023 edical Expenditure Widowed College and Above 25,474 46,344 25,328 for Different Age Cohorts 40,482 146 5,862 226,832 61,767 414,871 163,976 ? Household coCouple nsumption $49,059 steadily decl $40,984 ines with age, $8,075 and successive birt $181,308h cohorts h $369,790 ave higher consumption. Transpor $50,0 65 00 t ?74 22,411 853 24,251 (0.046) 6 1,758 -1,840 (0.045) 10 2,167 27,318(0.047) 10 2,924 120,078 (0.049) 11 2003 d complete idea of how people are doing Health a own with nd Re inco tirement me, but m Stu od ne y (HRS y spen ) t o and t n en h in terms of b e Consum tertainmen pti t a o en n an ing d other able to d Activiti items ( afford retireme es Mail donations Survey , g (CAMS nt than ifts, etc.) rises steep ), arbitr which is a ary estimat supplem ly with es such ent of Data .......................................................................................................................... contribute to, to encourag2003 e, and to enhance the developmen ............................... 4 t of sound employee benefit health spen Transpor dingt is that the 1,610 private health 11 insurance prem 2,730 iums a 13 re inclu3,158 ded in health sp 12 ending 4,038 , and that th 11ose who buy Figure 11B, Median Spending in Each Category and Mean Perc 2003entage Share of Each Category in Total Spending Never Married 32,316 28,979 2005 3,336 60,627 133,745 Estimates from the MEPS a Single nd 21,783 NHA, 1996.” 24,378 Health Car -2,595 e Financing Rev 34,688 iew, 23 (1), (Fal 115,652 l 2001): 161 ?178. EBRI explores the breadth of employee benefits and related issues ? Clothing Other expenses inclu 123 de contributio (Only individuals observed in all four waves) 3 ns to 203religious, e 3 ducational, c 406 haritable, 3 or politic 591al organizatio 3 ns, and Years of Education 75 ?84 17,940 19,573 .030*** -1,633 .028*** 21,863.025*** 96,787.025*** Who we are Home $8,768 40% $11,189 39% Home $9,818 39% $12,441 38% Clothing 308 programs and so 4 und public policy 492 3 through objective 739 resear 4 ch and educati 985 on. EBRI is the only 3 income. O as inc ? ome re Declin n th $2placement r 0 e other ing h ,000 eal ha th limits nd, atimoney spe os.dif ferent activit nt on transportat ies and coinsumpti on and c oln of othi di ng it ffer ems rema ent goods, ins roughly which strongly af flat as a fect perc s the entage o declifne the HRS (described below). CAMS contains detailed spending information on 26 nondurable and six durable categories, this kind of Home coverage have $15,287 the income to 40% afford it. $11,173 But it mi 2007 ght 39% also be the $9,444 case tha40% t people wit $7,798 h private covera 43% ge utilize Expenditure Pattern Across HS Dropou Ages t ............................................................................................... $20,548 $21,851 -$1,303 $15,897 $79,640 ................. 5 Income, Expenditure, Widowed an 22,043 d Wealth 22,860Comparisons -817 M 37,737 edian Total Expenditu 135,648 re for Househo 10% lds Across Different Income Quartiles and Age 75+ ............................................................ 17 Entertainment EBRI studies the worl 295 d of health and retirement ben 5 953 8 efit1,891 s — issues such a 10 s 40 3,538 1(k)s, IRAs, retire 13 ment 85+ 13,483 15,167 (0.003) -1,684 (0.002) 14,386(0.003) 68,422(0.003) Food 3,081 14 3,413 11 cash and gifts Food to family and f private, nonprof riends 3,081 outside it, nonpar the tisan 13 hous , Washington, DC-b ehold (incl 3,555 uding ased organi alimony an zat 9d ch ion com ild-s mup itted exclusivel port payments y to ). Comparing Entertainment Income, 764 Expendit Age Group 8 ure, and 1,807 2001Wealth 2003 10 Across 2,954 2005Different 2007 14 Marital, 5,478 Racial, 15 and VanDerhei, Food in total Separat Jack. “Measuri ex ed*penditure. 4,622 n$26, g R 050 e tirement I 12 n $27, come A 563 3,697 dequacy: Calculat -$1, 13513 ing R 2,844 ealistic I $6,n 937 co 12 me Replac 2,157 ement Rat $38,429e10 s.” EBRI total and it fo expe llows ndit ure the s across dif ame HS Grad grou fep of rent income people ov quartil 32,651 er eight es. years. Figu 28,054 re Using t 11A com his in 4,597 pares the formation cou top 91,511 and pled with bottom inco 237,532 the in me quartiles come, wealth, across all their coverage more by going for regular check-ups, buying 2007 preventive medicine, etc., which might cause higher out-of- “Expenditures Other incom ” Defined e ade ........................................................................................................ quacy, co 616 nsum7er-driven b 923 enefits, Socia 9 l Security, tax treatment of both retireme 1,797 12 3,078 ...................... 5 15 nt and health Labor Force Status 2007 Health public 2,629 policy research and 13 education on 4,215 economic security a 15 nd employee benefit issues. The following section discusses the i Health ncom 3,299 e, expenditure, and w 14 ealth-holdin 5,073 g patterns of ho 15useholds with at least one Other 443 5 1,231 8 2,277 11 3,693 14 50 ?64 in 2001 Health $40,000 3,105 10 3,288 12 3,679 15 4,337 21 Education UNIQUE Figure 12, Divorc Me DAdian TA: a edl Groups Thi Hous s ehol Issue Brief 25, d S 107 pendin exami g in n Each Cate 27, es the 445 expe gory nditure patt and Mea -2,338 n P erns of th ercentage S e olde 23, h096 are of Eac r section of t h Cate hgory in e populati 77,588 Total on. It College and Above 49,912 42,757 7,155 208,604 426,298 Issue Brief, no. 297 (Employee Benefit Research Institute, 2006). years, and shows that t Couple he bottom-income $54,970 quartil $44,378 e of hou $10,592 seholds spends almost 73 $216,149 percent $418,733 of their total expenditure on health, and la Works full time (dropped) Total Spending bor-market 50 ?64 in $29,854 formation avail 14,627 $31,094 able in the HR 21,093 -$1,240 S, this study a28,088 ttempts to su $18,465 mmarize t 38,099 h $84,975 e consumption behavior pocket healthbenefits, cost spendi Transport ng. But manage looking ment at th 2,538 e , worker a se simple corr nd employe elations, it 12 r attitude cann3,034 ot be s, policy reform p determined w 11r hopo ethe sals, and p r private healt ensh ion asset s ? Home and home-related eEBRI’s member xpenses remain ship includes a the largest spendi cross-section of ng catepension funds; businesses; trade associations gory for older households. ; Transport 2,654 11 3,799 12 Expenditur Total Spending e Patterns of Retire 19,335 d and Working Households 26,532 ........................................................................ 33,026 48,978 ..... 6 member age 0% 65 or over. This simple comparison can give a sense of how retirees are doing in retirement and if they Transport 5,636 Income16 3,574 $55,407 $49,050 13 $49,013 2,390 $51,773 11 1,005 5 Widowed 33,178 33,645 -467 119,421 257,929 Expenditure Pattern Across Ages 2007 u Figure 7 ses data sho frw om s t income, ex he CSingle onsum pendit ption and A ure, 21,749 and wealth ctivities 24,065 -hol Mail din Survey (CAMS), a g-2,316 patterns of different marit 22,927 supplement to t al grou 72,837 ps he He for t alth and hose age Retir 65 or over. ement Spending for Individuals With and Without LTC Insurance, Age 65+ ..................................................... 21 Works part time 65 ?74 22,080 25,973 0.01 -3,893 2003 0.039 21,740 -0.009 119,778 -0.002 and fundi Clothing ng. There is wi Anti labor un cipat de ed a 830 spread ions; h Drop ealth recog care prov nition that if employee be 5 iders and insure 984 rs; government org Exper n iee nc fits data exist, EBRI kno ed a Drop a 4nizations; and service firms. ws it. basic items of the Americ lik $1a 0 e home, n ,00 el 0derly. foo It h d, and he as three prim alth, w ahile t ry objectives: he top qu artile spends only 58 percent of their total expenditure on Clothing 877 5 1,223 4 insurance coverage has any independent effect on expenditur 2003e. Clothing 1,304 5 1,007 5 830 4 622 4 Income, Expenditur ? Health car Never Married ee, and expenWealt d 29, itures 425 Expenditure h ste Com adily i parisons 24, ncrease 084 ................................................................................... 36,721 with age.42,435 5, 341 42,107 25, 42,345 409 174,112 ........ 8 are in danger of outlivi HS Dropou ng their assets. Figures 6 t $20,562 ?10 use $24,080 the same me -$3,518 asures of $13,355 income, ex$92,542 penditures, nonhousing Widowed 22,649 26,050 -3,401 44,243 184,515 Home $5,300 41% $8,344 40% $9,138 41% $13,615 40% 75 ?84 18,837 22,360 (0.36) -3,523 (0.031) 31,367(0.027) 136,238 (0.032) It is apparent Study (HRS), that cou conducted by Entertainment ples are do the Institu ing better than those single or 1,725 te for Social Rese 9arch at the widowUniver ed. 2,441 They sity of haveMichigan enou 10 gh inco , con m tains detailed e to support their Figure 1 plots the me Entertainment dian (the midpoint— 1,896 half above and ha 9 lf below) total house 3,496 hold annu 10 al expenditure (in 2010 Home $7,050 47% $9,282 37% $12,680 37% $17,140 34% similar items. For Figure 11B (those age 75 and over), the similar percentages are 77 percent and 62 percent. Entertainment Source: Employee Benefi 3,715 t Research Insti 11 tute Estimates from 2,749 the Health and R 11 etireme 1,725 nt Study (2000 ?2008) and the C 8 748 onsumption and 6 Figure 13, Median Household Spending in Each Category and Mean Percentage Share of Each Category in Total HS Grad 33,460 30,641 2,819 107,773 247,699 Source: Employee Benefit Research Institute estimates from the Health and Retirement Study Financial Wealth 90,620 91,693 108,805 118,894 wealth, an Food d to Source : tal household Employee Benefit 1,849 wealth. Research These measures a Institute estimates from 14 2,630 the Consum re descri ption and Activ 12bed ities bel M3,019 o ail w. Surv ey (2001 ?2009). 11 3,820 11 Treatment of Unemployed Defi 85+ ned Bene14,082 fit (DB), Defin 18,629 -.347*** ed Contributio -4,547 n (DC), -.214* and Individual 15,969 Ret -.303** irement Accou 3 47,108 nt (IRA) -.275** $30,000 Other 995 9 1,659 9 ? On averageOther , retired households1,244 spend about 80 percen 9 t of workin 2,903 g households, a 12 nd their earnings are about expenses, an expenditure d Food d they a ata on 32 cat lso hol 2,465 de hi gories, and f gh leve EBRI’s work advances knowledge and unders 14 ls ofo lillow quids 3,697 a the nd ot same her forms of group of i 13 wealth 4,266 ndividuals ov to su tanding of emplo pport th 12er eight year emselv 4,622 yee benefits and their es throu s In addition, the g 11 h periods of dollars) across ages. EBRI delive ? To examin Activities Mair le h S s a stead It cl urvey ow consum (2001 early shows that ?2007). y stream of invaluable research and anal ption patt hous erns ev ehold e olve with xpendit ag ure stea e, income, dily declin and other es wit ysis demogr h age.aphic The de charact cline is almost eristics. Other 1,593 7 1,538 9 1,422 10 853 10 College and Above (2000 ?2008) and the Consumption and Activities Mail Survey (2001 53,061 45,071 7,990 ?2007).234,701 446,073 Source: Employee Benefit Research Institute estimates from the Health and Retirement Study (2000 ?2008) and the Spending for Health Individuals Wit 2,420 Total Wealth h and Witho 19 ut Private Healt 3,022 193,138 208,647 h18 Insurance, A 253,667 4,586ge 65+ 283,463 17 ...................................... 4,613 13 22 (0.128) (0.123) (0.132) (0.139) Total Spending 28,205 35,508 Determinants Assets in Income and Total Spending of Household Wealth Measur 30,780 Expenditure: es in the RAND HRS Regression .................................................................... 40,071 Results 10 Health 2,020 importance to 12 th3,112 e nation’s econo 13my among policy 3,816 makers, 13 the news 4,693 media, and the public. It 11 57 percent of those of working households. Total Spending * Less than 50 observati EBRI publications 44,153 ons. include in-de 36,007 pth coverage of key issues a 27,833 nd trends; summ 23,783 aries of research Long-term large income and expendit wealth d ure shocks. Care ata Inavailable in th surance and e HRS Prare u ivate sed to Health establi Insurance sh the financia l standing of older households. linear, indicating that e Source: Employee Benefit Research Institute estimates from the Health and Retirement Study xpenditure falls at a constant rate with age. With age 65 spending as a benchmark, household ? Consumption and Activities Mail Survey (2001 Income represents total household incom ?2007). e, i.e., for couples households, it is the sum of respondent ? To study the income, expenditures, and wealth-holding patterns of the elderly to get a sense of how they are Partly Retired Transport 1,060 9 -.053* 1,787 -.045* 10 2,525 -.081*** 11 3,093 -.054** 9 $0 65 ?74 in 2001 2005 does this by conducting and p 2005ublishing policy research, analysis, and special reports on Transport 1,891 10 3,172 14 4,396 14 5,570 13 The last sect What we do ion noted t (2000 hat ?2008) and the Consumption and Activities Mail Survey (2001 LTC insurance coverage and pr 2005 ivate health insuranc ?2007). e coverage are highly correlated with Endnotes Comparing Income, Expenditure, and findings and policy develo Wealth Across Diff pments; timel erent M y factsheet arital, s Racial, a on hot topics; nd Educational Gro regular upda ups tes on legi .......... 11 slative and Appendix A, Regression Results ................................................................................................................ 23 As the average age of the American popul Workin ation inc g reases, long-term care (LTC) insu Retirrance is ed a growing area of concern expenditure falls by 19 percent by age 75, 34 percent by age 85, and 52 percent by age 95. Future retirement income ? Demogra Clothing and s phic groups suc pouse i 569 ncome. It i h as singles, (0.027) n 5cludes w blac711 ks, and ages anhigh d la (0.024) 5 bor school dro earnin 782 gs; capit pouts are outspen (0.027) al 3 earnings; 1,422 ding th defineeir r d be (0.023) enef sources in 3 it pensi ons, managing t Home heir finances a Income emplo nd if they $10,248 yee benef are iat ts issues; holding educational br $36,756 risk of outl 43%$35,231 iving their assets. $35,839 $11,558 iefings for EBRI memb $34,679 39% ers, congressional and Home $10,588 42% $12,280 37% Clothing 697 5 796 4 1,185 4 1,564 4 Home $18,267 44% $11,869 41% $10,213 40% $8,698 44% 1 $20,000regulatory developments; comprehensive reference resources on benefit programs and workforce For e DECLI highex r spendi amp NIN le, i G EXPE ng. n 2007, But sinc N the SES e m : people e Hou dian s incom eh with old e su e x ch insura of cou penses ples n ste ce hou a spen dily decline with age. seholds e d morex ce in e every cate ded theWith ir me gordi y the age 65 e , it is h an expe ard to nditur x tell penditure as e by if they $10,5 are 92, a simply and Comparing Co Entertainment nsumption Patterns 341 of Differ 6 ent Incom 1,043 e Groups 7 .................................................................... 1,908 8 4,236 11 13 Retired -0.014 -0.004 -.040* -0.021 for retir FICA (tehe Federal Insura es. According to go nce Co vernment ntributio esti ns Act) mates, 12 is a federally million older Am imposed payr ericans wi oll tax used to ll need LTC fund by Social Security retirement, 2020. However, in most adequacy stu retirement. dies shoul Food Not surprisin d explicitly take gly, federal agen the lo acc 3,081 o west-i unt of cy stnc aff, and th this ome gr de 14 ce news linin oup (bottom-income g e media; xpend and sponsoring public opinion survey iture 3,200patter quartil n ae) wh s done 12ic in h is Hur gen d an erally s on emplo d Rohwedder yee Entertainment annuiti Foodes, an 853 d income7 from 3,081 other ret 2,204 ireme 14nt savings suc 10 3,437 3,555 h as 401(11 k)-type 9plans 5,999 and indivi 14dual Expenditure 30,805 33,624 32,549 33,867 Appendi Food x B,S L oo urngitu ce : Emd ploy inal ee B Ch 4,313 enef anges it Res ear in M ch Ins 12 e tdi itutan Househol e es tim at3,555 es from H d ealt Income, Hous h and12 Retirem ent Sehold Spen t3,081 udy and Cons udin m pt14 g ion an , Househo d Activities 2,370 l M d ailNonhousi Survey. 12 ng ? To determine if long-term care (LTC) insurance and private health insurance affect the elderly’s consumption issues; and major surveys of public attitudes. Other 474 7 948 8 1,706 9 3,555 14 benchmark, h wealth their mier edia individuals n no ousehold expenditure fall nhousing who an cod n uld ebe t total expect w(0.022) e s by 19 percent by age 75, 3 ed to alth was $ spen216 d more. To ,149 an (0.019) fi d nd out if $418,733, 4 percent by age 85, these resp fact (0.022) ectively. In ors have a contra and 5 ny in st, singles de 2 (0.020) percent by age pendent (w eh ffect on ich benefit issues. EBRI’s Education and Research Fund (EBRI-ERF) performs the charitable, Health Health 2,152 3,171 12 14 5,152 4,323 16 16 Long-term survivors, disability, an cases LTC is Other Car no et covere Insura d hospital insurance. nce d by a 510 n Medicar d Private Health Insura e, an 5 d this car 1,482 e is nce ex......................................................................... pens9 ive and c1,995 an be indefin8 itely long3,993 or even perman 12 ... 13 ent. Since (2011). overwhelming retirement acc ly represente o Financial Wealth unts (I d by the RAs); aSocial S bove 120,996 groups ecurity , ap 113,703 Disabi pearslit to be y Ins 120,115 s urance; trugg129,829 lin Social g the mos Secur t. ity retirement benefits; Health 2,844 9 3,259 Appendix 12 D 3,746 16 3,427 20 Wealth, Household Total EBRI meetings Wealth (Housing and present and explore i Nonhousing), ssues with forthou Differ ght leade ent Age Co rs from all se horts .......................... ctors. 24 Figure 6B behavior. Total Spending 15,579 23,060 28,784 47,393 Not in labor force educationa -.000 l, and scientific func -0.0004 tions of the Institute. EBRI -0.031-ERF is a tax-exempt organizatio -0.043 n Transport 2,779 12 3,211 12 95. inclu the eld de se erly’ para s spted, endidivorce ng Transport , som de , reg and ression never mar s 2,963 were ried persons run wher Figure 10 ) e controls 12 don’t have wer en eough inc 3,727 luded income for t age, o cover th 11 income, eir re ace, e xpenses, and t ducation, heir Total Spending 19,286 30,065 38,132 52,262 Transport 6,593 16 3,828 Figure 8 Figure 14 4 2,489 10 1,511 8 Determinants of this is a large and potent Household Expend ially catastrophic Total Wealth iture: Regr health ession ex 244,233 pen Results diture r 255,106 ..................................................................... isk, peo287,121 ple withou 299,282 t LTC insurance may be forc . 18 ed to cut unemployment compensation; and government transfers and other sources of income such as EBRI regula Median Household Income, Household Spending, Household Nonhousing rly provides congressional testimony, and briefs policymakers, member organizations, ? Apart from the traditional determina (0.031) nts of spendin 2005 g, suc (0.027) h as income and a (0.030) ge, other major factors (0.027) that affect Median Household Inc supported b ome, Hous y contributions ehold and g Spending, Hous rants. ehold Nonhousing Wealth, 2 $10,000 Clothing 682 4 754 3 Clothing 711 3 957 3 Median Household Income, Household Spending, 2005 wealth labor-force status, self-r Replacement Clothing holdings are r ratio is the ratio of postre ePercentage latively epor 1,138 ted low. health Widows, of Population tire stat 4 ment earnings us, and in on the oth 830 Who A dicat er to preretireme ha on rs of LTC nd, tici 4 hol pate d mo ins d nt earnin an urance covera derate 640 Increase leve gs. For a discu ls o 3 in ge a f w Spending en alth, d privat ssion but 569 on e healt in don’t calcu have lh ation i 3nsura en of ough nce Figure Appendi 2 tak x C,e M s a closer edian Hous look ehold In at the expen Median Household Income, Household Spending, come, H dio ture usehold pattern Spen acro ding, ss ages, bo Househoth by ld Nonho breaking total using Wealth, househ Household Total old expenditur e into 75 ?84 in 2001 their spending and save more. and the media on employer benefits. Conclusion .................................................................................................................... Health Status Home alimony, lum $6,271 p sums from 44%insurance, $9,732pensio44% ns, or inherit $10,873 ance, or anyth 38% ing $13,799 .......................... 19 else. 36% Wealth, Household Total Wealth (Housing and Nonhousing), in 2010$, for the Entertainment 1,410 8 2,098 9 the spending Household T Entertainment patterns o otal W f r et ealth ired househo 1,602 (Housing and Nonhousing), in lds are lon 8 g-term care insura 3,235 2010 $ for S nce and some form o 10 ingle Females f private health Home $7,655 46% $11,459 40% $13,024 36% $21,407 40% HOME EXPENSES: Home and home-related expenses remain the single largest spending category for older Entertainment 3,555 Household Nonhousing Wealth, Household 10 2,370 10 1,434 8 756 6 Retirement vs. Percentage of Population current coverage. realistic replacement ratios please see VanD incom In ce to cover ouples househ their ex olds, penses. controls Fu erhei (2006). were added rther analysis fo(results sho r spouses as wwell. n in A Th ppen e de dix pende C and D nt varia ) sho blw e is a s that amo log of to ng singles tal Data differe Wealt nt cate h (Housi gories ng a and als nd No oHousehold Nonhousing Wealth, Household Total Wealth nho by susing) in eparatin 2 g 0 it into 10$, for diff Sin ere gnt ag le Males e groups ............................................................... 25 . The reported numbers are the medians (in Food Treatment 1,914 of Income Defin 17ed Benefit 2,844 $29,170 (DB) $25,102 13 , Defined $27,490 3,555 Con $26,540 13tribution 4,266 (DC), 11 Excellent (dropped EBRI issu ) es press releases on newsworthy developments, and is among the most widely quoted Other Other 735 948 7 7 1,541 3,496 13 9 References Food ............................................................................................................................... Top Half 2,465 EBRI Issue Briefs of the Income Distribution of Different Age Groups 15 3,555 is a monthly 13 periodical4,266 with in-depth 12 evaluation of em 5,504 plo ............... 20 yee benefi 10 t issues insurance coverage. In In com parti e cular, hav Expendit ing l ureong-term car Incomee Gap insurancN e h onhous as a signifi ing Wealt cant h effect To on tal W highe ealthr spending Other 1,304 6 1,259 8 1,084 8 711 7 Americans. On average, those over age 50 spend around 40 45 percent of thei ? r budget on home and home- Total Wealth (Housing and Nonhousing), in 2010$, Who Actually Experienced an Increase, 2001 ?2009 ? Expenditure is total expenditure on all categories (home, food, health, transportation, clothing, Health 2,252 19 3,618 16 4,676 17 4,870 15 household e and wi Figure dows 12 sho Very good , men xpendit ws the sou ar ure. consu e doin rces o (Housing and Nonhousing), in 2010$, for Different Racial Groups The g m n bett control for ption em Expenditure er ployee be patter than w inc n -.050** s of t oome is also n men. efits by all media. hose a 26,945 gin e 6 clude 5 or 27,073 -.035* d over as a lo who are wit g of 25,154 total h hous -.047** or 26,806 witho ehold incom ut LTC ie. Age nsurance. -0.03 and year There are s of Two sourc 2010 dollars) and perc es of data arenta e use ge d for share this stu of eac dy. Firs h catet is t gory i he n t Health a otal spen nd Re dingtirement . The dif Stu ferent dy (HRS categori ), whic es ar h e d a study escribe of d a above. and Total Spending Total Spending Individual and trends 27,397 28,891 Retirement , as well as crit Ac ical count analyses(IRA) of 33,674 41,164 emploAssets yee benef it in polici Income es and pro posals. EBRI Health 1,659 13 3,250 13 4,229 12 5,096 10 Our Total Spending 46,876 34,187 26,953 24,300 3 Income Expenditure Income Gap Nonhousing Wealth Total Wealth Endnotes Appendix ...................................................................................................................... D, Median $0 Household Income, Household Spend2001 ing, Household Nonhousing Wealth, .......................... 20 Household Total by retired households. related item Figure 1 is ba Transport s.sed on a cross-sectional data, so the consumptio 830 for Households in the 9 2,140 Bottom n path does not show the same group of 10 Income Quartile 2,824 9 3,637 people at different 11 18entertai %CH nment E ECK , and other O O (0.020) expenses com UT E EBRI’ bined). (0.019) S S WE (0.019) B B SITE TE! (0.020) EBRI directs membe Financial Wealth rs and Among Households Age 65+ other constit 108,155 uen 107,671 cies to the informatio 102,875 85,720 n they need, and undertakes new educatio several things n enteto note r linearly here: into the mode Notes l w is a monthly hile all oth p ee r va riodriables ical pr 2007 2007 oviding curr are include ent in d as c formation on a variety ategorical variable of emplo s. Two se yee ben parate sets efit nationally Transport representative sa 2,083 mple of U.S. 13 households 3,733 with individuals o 14 v5,593 er age 50. It16 is the most co 5,540 mprehe 12 nsive survey and Wealth Measures in the RAND HRS 50 55 60 65 70 2007 75 80 85 90 95 2001 Separated 8,935 15,207 -6,272 461 461 Wealth (Housing and Nonhousing) in 2010$ for Single Females ............................................................ 25 Good Clothing 388 -.078*** 3 593 -.074*** 3 770 -.053** 4 982 -.049** 3 Figure ages. Appendix B 8 takes a closer sho resea wPanel A : s the co look rch on a at di nsumptio Working n ongoi ffer topics. EBRI ent Households n path o ng ba racial sis. groups fF different undamen age 65 co tals ho of Employee Ben or rts o over a ver a perio nd sho d o ew fit Pr fs seven years. that ogr whit ams es offers a straightf are In this doing shobett rt perio orward, basic er tha d, the n other Some imme diate patt Home Home erns emerge Total Wealth from $11,070 $11,660 Figure 2. Firs 243,660 t, as in 44% 43%230,842 previous studies 255,781 $13,912 $16,092(Butri 263,050 ca, Gol 40% 40% dwyn and Johnson, 2005; Clothing 479 4 711 3 995 3 1,600 3 publications of re gressions Home were run, on $18,903 e without co Income 45% ntExpenditure rolling for $13,622 we Income Gap alth an 42% d the ot Nonnhousing Wealth her $10,666 with log w 15.8% 42% ealth as Total Wealth a con $11,376 trol. Economic 46% theory of older Americans in 50 ?64 the na $95,718 tion and cover $49,338 s topics like h $46,379 ealth, assets, income, an $221,628d labor-force stat $372,664 us in detail. It is a Divorced 18,780 22,686 -3, Ag 906 e 19,734 55,839 ? Net nonhousing wealth includes any real estate other than primary residence; net value of vehicles Entertainment 332 (0.022) 4 1,043 (0.020) 6 1,786 (0.021) 9 3,655 (0.021) 11 RISING HEALTH CARE EXPENS explan ES:ation Health of -rel emplo ate yee b d expenses enefit progr are the second ams in the privat -largest c e and public omponent in th sectors. Thee EBRI budget 16% EBRI maintains an Food Food d analy 3,081 3,294zes the most comp 13 13 re Income hensi3,555 3,555 ve databa Nonhousing se of 401(k) 10 12 Total -type programs in the Entertainment 673 85+ in 2001 6 1,979 8 3,911 12 6,512 13 racial consumptio ? groups Peopl n paths fo . e Their with me LTC insur r differe dian int co ncome e ance horts loo spend much xceed k relatively flat, s their m more than edian bu ex t anchored at a higher level tho pen se wi diturtho e, and ut it,t i hn ey hold large every category and overal for each successive cohort. T amount of non l. For housin exam g h e and ple, Butrica Food and Mermin, 20064,622 ), it was foun 11 d that home 3,564 and ho 2001 me-re 13 lated expenses rem 3,309 13 ain the lar2,465 gest spendin11 g category 65 ?74 64,080 41,240 22,840 304,251 483,744 suggests that Widow consumption ed i 18, s primarily 055 a function o 21,346 f income. But -3,290 for retired people, 42,982 accumulated wealth is likely 133,096 to be a biennial longitudinal survey with qu estionnaire waves in even-numbered years beginning in 1992. The initial sample DB: Other 356 4 747 7 2,050 10 3,555 13 EBRI’ Fair s s website is s easy to u -.085*** s se and pac cked with -.092*** u useful in -0.049** fo ormation! Look -.052** for owned; IRAs, stocks and mu Databook on E tual f mployee Benefits unds, checking, is a savings and money statistical reference work on em market accou ployee ben nts, CDs, efit programs Health 2,275 11 4,304 14 Other worl Health d. Its co 332 mputer simul 42,971 ation analy 1,422 se12 s on So 8 cial Secu Gap1,896 5,008 rity reform Wealth9 and retireme 13 Wealth 4,148 nt income a 12 dequacy of older Americans. It is the only component which Income Expenditure steadily increases with age. Health care expenses capture Source: Em ployee Benefit Research Ins titute estimates from the Cons umption and Activities Mail Survey (2001–2009). Health 75 ?84 White 2,853 45,639 Income $32,730 932,885 $28,837 3,117 $17,982 12,754 $3,892 $16,574 11 $17,087 3,445 $120,864 231,676 $19,994 14 $244,159 403,580 3,377 15 net cross-sect wealth to in 2 ional 00 fund their r diagram captur 9, people w etir ith LT e es this di ment. I C insurance n fference in cohorts as a rapidly decl 2007 covera , nonhou ge sihad ng w me ealdia th of n total white inin househ s w g consumption path. As a r as sl old sp ightly mor ending e th of $ an 47, 20 392, esult, it might time w s th hile t e hose for every ag Never M e group. arried In any 14, given 280 year, the p 20,e 646 rcentage of total -6,367 expenditures sp11, ent 121 on home-related 20, items rema 646 in Figures Total Spending 16,613 24,091 34,345 44,326 more important determinant of expen and work for ditu (0.025) re tha ce-r n in elacome. Me ted issues. (0.024) dian regression was (0.024) used along with mean r (0.023) egression consisted of individuals born between 1931 and 1941 and their spouses, regardless of their birth year. Newer cohorts PreretirementTransport —Not in Transport cluded in eith 3,010 3,245 er income or wealt 12 12h. 4,266 3,674 12 12 Total Spending 14% 19,039 31,292 39,478 62,775 governme are unique. nt savings bonds, Treasury bills, bonds and bond funds; and any other source of wealth these s Transportp pecial feat 6,612 u ures: 16 4,323 14 2,785 10 1,533 7 around 10 percent of th 85+ Black e budget for those between 5 29,958 17,14224,011 23,814 5,947 0–64, -6,672 but increase to 2,926 125,790 about 20 percent f 28,649 267,636 or those age 85 and 2001 Expenditure$12,541 18,808 $19,59017,985-$7,049 15,413 $8,229 22,578 $56,284 2003 nonhousin overestimate the slope of the cons without g wealth o LTC fin blacks surance sp . The umptio incom ent only n path fo e of blacks has $32,04 r any 8. givbeen en co 2007 less t hort. han their expenditures in all years except 2007, and almost the sa Poorme for every age group. Se -.165*** cond, health expenses i -.130*** ncrease steadily-.093** with age. In 2009, p -.095** eople between because spending can Clothing Clothing be very large for t 687 711 he wealthy or those who 4 4 experienc 1,007 815e health or oth 3 3 er expenditure shocks. Unlike have been added in the following years. The study is sponsore 2007d by the National Institute of Aging (NIA) and the Social Figure 1, Median Total Annual Expenditure, in 2010$, Across Ages 50+ ......................................................... 6 Clothing 1,185 4 830 4 711 4 593 5 minus Hispanic all debts (14,683 such as co17,997 nsumer loan-3,314 s) 2003 . 10,898 55,839 Separated* 14,651 Financial Wealth 20,062 46,420 -5, 33,520 411 30,300 485 21,306 20,020 over. Home $7,729 49% (0.043) $9,421 (0.037) 42% $13,093 (0.041) 41% $17,909 (0.037) 39% 2003 13,800 21,387 -7,587 8,861 83,561 they hold low levels of wealth. Also, if the median black household experiences any large expenditures arising out of ages 50 ?64 spent 9 percent Entertainment Entertainment of their total budget on healt Contact EBRI 1,422 1,732 Publications, (2 8 h 9 items, 02) 659-0670; while 2,548 3,293 those ag fax publication e 85 or over sp 10 10 orders to ent twice (202) 775-6312. as much • Postretiremen Home EBRI I’s entir t—Monthly pa e li$8,732 brary of rese yments from t 47%a arch publica h $12,566 e pension t tions starts ar39% e included in a at th $15,524 e m income. ain W W 40% eb page. $23,935 C Click on 41% EBR RI ? In individual categories, the differences are particularly large in health, entertainment, and other spending. In mean reg Entertainment ressions, med 12% ian 3,579 regressions ar 9 e robust to 2,714 outliers, so the es 10 timates are al 1,687 so more robust. 8 924 7 Security Administration (SS 50 ?64 $94,645 A) and admin $57,413 istered by the Institute $37,232 2003 for Social R $206,965 esearch (ISR) at th $367,821 e University of Michigan. 4 EBRI make s information freely available to all Divorced 15,228 18,205 -2,977 12,657 48,974 Total Wealth 89,104 89,440 84,485 71,628 Long-term Care Insurance Food 2,157 13 2,844 13 3,697 12 4,777 10 For Figure 5, the data from 2001 to 2007 are pooled as one single cross-section. Therefore, working households are Figure 2, Median Spending, in 2010 Subscriptions to $, in Each Cate EBRI gory Issue Bri and Me ean fs are included Percentage S as part of hare of Each EBRI membership, or as part of Category in Total a Other Other 723 948 7 7 1,718 2,963 13 9 Food 2005 2,844 15 19,430 3,555 23,163 13 -3,733 4,622 9,49312 53,388 4,835 10 catastrophic h ? ealth shocks, i Net total we t can r alth un shor includes t of net no funds in nhousin a very short t g wealth iplus me. Hispan value oics f prima are also ry reside not doing v nce minus ery mortga well in ge Expenditure (18 percent) o Other f their Pattern budget 1,185 on hea s of Retired lth 6 items. Health-r and 1,185 Working elated ex 7penses a Households 1,197 re the second 9-largest shar 912e of total 9 expenditure Issue e Briefs 65 ?74 White and 66,553 E EBRI Notes $35,25445,149 for our in-d $32,601 21,405 e ept $2,654 h and non npartisan pe $127,655 284,773 r riodicals. $264,150 457,498 Orders/ 2009, people without EBRI assumes a publi LTC inc surance service re spent spo only 60 nsibility to perce make its finding nt of what thoss co e with LTC mpletely acce insurance spent ssible at on h www.ebri.org ealth W Health idowed 19, 2,048 651 Source: Employee Benefit Research Institute Estimates from the Health and 14 25,387 2,828 -5, 14 736 4,266 49,214 14 5,925 138,737 14 DEMOGRAPHIC No (dropped GRO ) UPS: Singles, blacks, and high school dropouts do not have a sound financial standing in Total Spending Total Spending $199 annual sub 29,047 scription to EBRI Notes and 38,553 EBRI Issue Briefs. Change of Address: EBRI, Health 1,567 11 31,545 3,474 12 3,911 45,605 11 4,385 9 generally separate from retired households. Thus, the 80 percent consumption replacement means that the retired group Spending, 2001 -2009, by Age Groups .................................................................................................. 7 DC: Total Spending 75 ?84 48,515 47,006 38,281 35,779 8,725 32,420 205,535 28,348 398,860 Ap retirement, alt pendix A shows the reg hough 10and ot % better t Black her ression ho han me loans. It blacks: In results. 22,563 First, the regr does not i 2007, 26,478 Hispani nclu essions with -3,915 d cs’ e incom nonho e. usin c ontrol for g w 4,180 ealth weal was th show that almost three 57,083 ti both income an mes larger and d total for those over But the crucial age 75. part of t 2007 he data used in Retirement Study (2000 this 17,940 study com ?2008) and the Consumption and Activities Mail e 25,442 from the Consu -7,502mption a 9,396 nd Activities Mail 78,729 Survey (CAMS), which — so that all decisions that relate to employee benefits, whether made in Congress or board rooms or Figure 3 items. Th uses Never M Transport a sample arried e similar n of 1, 24, u 094 1,120 mber 201 individ for entertainm ual 7s (a su 26,075bset o 2,406 ent items was f the ori -1, 11 874 gina less than l HR 3,327 S sampl 50 percent. 253, e) who were o 433 11 4,286 bserv317, ed both 210 10 before and Yes .100*** .067** .083*** .048* 1100 13th St. NW, Suite 878, Washington, DC, 20005-4051, (202) 659-0670; fax number, 2009 retirement Transport . Their expendit 2,110 ures exceed t 13 8.9% heir income and th 4,130 2009 15 ey hold very little financia 5,637 16 l we6,208 alth. The bottom i 13 ncome consumes 80 percent of the working group, which 85+ 34,912 Survey (2001 27,456 ?2007). are two different group 7,456 2009 s at any 115,873 point in time. 246,477 Subscriptions Hispanic 24,688 31,051 -6,363 15,415 95,604 Preretirement—Not included in either income or wealth. • Visit families’ ho EBRI’s blo meg g s, are . based on the highest qualit 2005 y, most dependable information. EBRI’s Web site posts wealth net worth are im Clothing wasportant more th deter an tw mice that inants of sp 521 of bla endi 6 cks. ng at ol 593 der ages. Secon 4 d, bot 747h the me3 an and medi 992an regressions 4 show was started in 2001 as a sup Panel B : plRetired ement to the HRS. From (202) 775-6312 Households (0.027) ; e-mail: the subs partici (0.028) criptions@ebr pants in th i.o e rg 200 (0.027) 0 Me HRS, 5, mbersh000 ip Information: households (0.027) wer Inquiries e after retirement. In this sample, 66.5 percent of the respondents anticipated their spending to drop in retirement, and Figure 3, Percentage Home of Population Who A $12,155 nticipated a Dr43% op in Spending $15,976 in Retirement vs40% . Percentage of Clothing Home 524 $11,502 4 711 45% 4 1,067 $14,238 4 40%1,509 3 quartile, ? Home One o which includes mostly thes bvious explan $18,828 ation for high 47% e demographic gr er health $14,471 spendi oups, has the weakest ng for 44% those wit $11,755 h LTC insuranc fina 42% ncial standing e is that the $9,533 in retirement. LTC in 43% surance 2005 2005 8% Generally, defined benefit pensions and withdrawals from 401(k)s are counted as income and are measured by This is not sur Private Health Insurance S Entertainment eparat all resea prising, ed* eve rch finding n i 11, f everyo 364 233s, publi ne in t c 6ations, an h 9, ese a 936 g 1,274 e grou d news alert ps is cover 1,7 298s. EBRI also exte ed 2,331 by Medicare, 0 9 nds its e since Me 3,596 ddic ucare ation and pu coverage is no 23911 blic se t rvice regarding EBRI membership and/or con Income tributioNonhousing ns to EBRI-ERF should be dir Total ected to EBRI that, apart from income, age and education are also important determinants of consumption or spending. The selected at random an Food d mailed the CAM3,555 S questionnaire. I13 n couples households, t 3,964 he questionn 10 aire was sent randomly to Entertainment Food 667 3,413 7 2,356 14 11 3,810 3,697 11 11 6,233 13 5 65.9 percent actually experienced a drop in retirement spending. On the other hand, Figure 4 shows that, while only Populatio Food n Who Actually E 4,622xperienc 12 ed a Drop, 2 3,896 001 ?200912 ....................................................................... 9 3,555 13 2,844 12 50 ?64 $106,175 $60,604 $45,570 $195,954 $384,511 premiums are White included in $36,444 health spendi $31,562 ng. Other $4,882 wise, if it is assumed t $117,224hat relatively s $260,107 ick people buy LTC A regression analysis is a statistical tool used to investigate the relationship between two or more variables. More • Postretiremen EBR Divorc No (dropped Other role to improving Ameri I Ied ’s relia t—Regul ) ble h h 19, ar ealth an 356 606 withPresident Dallas draw c d r an als are e e 5s’ finan tirement su 24,445 in cluded in cial kn Salisbur 948 r rowle veys are ju y at th inc -4, dge ome, but e 9 839 above thro s s address, ( ugh its a t a c the 2,015 lick a account 202) 659-0670; w 29, ard w921 a a 10 balan -y through winning p ce is not e-mail: 3,555 ubli t the topic bo salisbur inc c116, se lurvice ded in 165 12 y@ebri.org cam w x xes at ealth. paign separate varia Figure 9 shows the i bles not ncome, included expenditure, in the weal an th measure. d wealth ho Thus, ldings for a re di tirem Gap fferent e ent plan ducat Wealth bal ion anacl gro e accum u wealth ps a ulated ge 65 or befor over. As e retirement comprehensive. Specifically, it caps the nuIncome mbers of days Expenditure covered during a nursing-home stay and excludes long-term Health Health 2,510 3,182 12 13 5,398 4,450 16 16 Other 296 4 1,185 7 2,015 7 4,148 11 directions of these effects are along expected lines. While income has a strong positive effect on spending, age has a one ofHealth the two spouses. Si 2,844 nce 2001, CAMS 9 has been 3,504 conducte 12 d every tw 3,692 o years, with 15 2009 the 3,006 latest round of 18 available 8.9 percent of the respondents anticipate 65 ?74 68,371 ® 44,753 d their retirem 23,618 ent spending to rise, 221,241 15.8 percent of th 460,994 e households actually LONG-TERM CARE AND Black PRIVATE 21,889 HEALTH 25,549INSURANCE: -3,660 Long-te 8,957 rm care and so 54,923 me form of private health Total Spending 17,846 24,652 35,251 53,068 specifically, it is used to determine the effect insurance, W Yes idowed 6% then this co 21,349 uld explain .040** the of one variab 22,052 higher hele o alth s n ano -703 .032* pen ther ( ding.such as ho But that 35, does not w pri .032* 501ce affects dem explain wa h 124, nd) y 0.028 people keeping o 036 with L ther TC ChoosetoSave and the companion site www.choosetosave.org However, if the account balance is rolled over into an IRA or distributed as a lump sum from the DC account and Figure 4, Total Spending the t Perce o op of the pa ntage Transport Transport of Population Wh 20,549 g ge. o An 2,748 2,907 ticipa34,589 ted an In 11 11 crease in Spen 42,068 3,605 3,383 ding in Retiremen 11 12 t vs. Pe 60,529rcentage of is not consi expected, pe der op ed as le wh a o atte part o ndfe househol d college are d wealth. B doing be ut u tte pr than on retirem those ent, if a who diperson d not: In rolls over all years surv his or her eyed, t pension, heir inc 401 ome (k), care costs, particularly t 2001 hose incurred for custodial servi $13,019 $16,979 ces. As a r-$3,960 esult, the healt $10,792 h expenses show $55,595 n here may still be Transport 75 ?84 5,759 45,550 14 40,410 3,887 5,141 13 2,712 245,569 10 453,185 1,511 8 strong negative effect on spending. But the results show that among the demographic factors, health status is a very data. For those between ages 55 Hispanic ? 26,978 64, the aggr 21,312 egate expendit 5,666ure in each of 21,484 32 categories (si 139,426 x durable and 26 non- exp insurance cov eEditorial Bo rienced hiard: e gher rage hav Dallas L retire . Salisbur e me a nt spe significant eff y, publisher nding. ; Stephen Blakely ect on increased spending by older households. , editor. Any views expressed in this publication and th ose of the authors should Never M Source: Employee Benefit Research Institute estimates from the Health and Retirement Study (2000 arried 22,025 (0.016) 26,479 -4, (0.017) 454 137, (0.016) 880 ?2008) and the Consumption and 292, (0.017) 497 insurance spend mor Clothing e in every category. 652 Another a 4 lternative expla 960 nation could be 3 that only people with high variables (such as income, po Source: Employee Benefit Research Institute estimates from the Health and Retirement Study (2000 Clothing pulation etc.) fi 640xed. The process in 4 volves computer mo 741 deling that use ?2008) and the Consumption and 3 s data on the underlying invested in Clothing 2 stocks, bonds, m 2 1,043 utual funds, m 3 oney market 830 accounts, 3 CDs, Treasury 622 bills 3etc., then533 it is included in 5 wealth. or de excee Populatio fin ds th ed con eir n expenditure t ribution 85+ Who Actually E (DC a 42,780 )n retireme d t xperienc hey ho nt acc ld h e 28,291 d an igh ount balanc In levels o crease, 2001 f 14,489 both non e into an ?2009 ............................................................... in housin dividual ret g an 133,900 d tota ireme l nnt et wea account lth. (I The 271,864 RA) or takes a data also show that lu 9 mp- conservative. Selden et al. (2001) show that 13 percent of out-of-pocket health expenditures arise from nursing-home not be ascribed to the officers, trustees, members, or other sponsors of the E 2007 mployee Benefit Research Institute, the EBRI Education and R Activities Mail Survey (2001 / Pseudo R 2003 ?2007). 12,498 16,398 -3,900 6,578 41,172 important determinant of spending. Expenditure drops signif 2007 icantly with every single drop in self-reported health status. durable items) of CAMS are very close to t 0.312 he same categories in 0.191 the Consumer Ex 0.338 penditure Survey (CEX), 0.205 which is the Activities Mail Survey (2001 4% ?2007). Entertainment 1,535 8 3,275 10 Entertainment 1,304 8 2,275 9 EBRI is supported by organizations from all industries and sectors that appreciate the value of variables and • Entertainment income o Need u dses a statistical algorithm to e a number? btain LTC i 3,022 Check out t nsurance, an 9 h h d th e stimate how much one variabl EBRI Da ey spend more mo 2,417 t tabook on E 9 ney Eon m mployee Be every e affects other 1,506 item (Go n nefits. 8 s. da, Golberstein, 889 and Grabo 6 wski, 2007 Research Fund, or their staffs. Nothing herein is to be construed as an attempt to aid or hinder the adoption of any pending legislation, regulation, White $37,391 $34,187 $3,203 $136,943 $316,558 sum distributi high school gron from aduates a are pen generally sion, 401(k doi)n o g we r other ll in DC ret acc irement ount : T plan heir and income cov invests it in ers st their ocks, bonds, expenses, and mutual they fun hol ds, dmoney stays. The HR N S sample is selected from th 8180 e noninstitutionalize 8180 d population, but res 8180 pondents are kept i 8180 n the sample if Separated* 11,827 10,108 1,719 864 864 Figure However, co 5 show nts the m rolling for edia wne a household inc lth lessens thome an e magnd me itude of dian hous the effect ehold of spendi healthng in on s 2010 dol pending to lars for some workin extent. B g an ud r trice a, tired benchmark survey on Other 2005 household consumption in 1,043 13,248 the Uni 17,708 ted St 7 ates. Ho -4,460 wever, 2,643 CAMS 9,702 reports hi 11 57,887 gher consumption Figure 5, Median Hous Other ehold Spending a711 nd Median Household I 7 ncome (in 2 1,896 010$) for Workin 9 g and Retired Other 1,185 5 1,185 7 1,197 8 711 9 IRA: unbiased, reliable information on emplo or interpretative 50ru ?64 le, or as legal, $113,123 accounting, actuarial, o $64,945 r other such prof $48,178 yee b essional advice. enefits. Vis ww $200,288 it w. www.e ebri.org b ri.org/about/join/ $432,863 for more. 2010). Therefore, it is not clear whether LTC insurance has any independent effect on expenditure. Black 29,311 28,756 555 6,787 55,787 market accou significant am Source: Employee Benefit Research Institute Estimates from the Health and Retirement Study (2000 Divorc n ount of ed ts, certifi bot cate h n 17, so of deposit nhousing an 830 (Cd total Ds), 25,872 Treasury net wealth bills, etc., to s -8,043 uppor it is t their inclu retir ded in 33, e ? m th 293 2008) and the Consumption and ee w nt. T ealth he low meas est-educat ure. 67,151 The ion sum group is mary they are institutionalized after the initial interview, so the proportion of the HRS sample ever admitted to a nursing 2%Total Spending Total Spending 30,117 32,048 38,091 47,392 households 6 age 50 or over. This figure uses four rounds of CAMS data (2001–2007). The income data come from the Johnson, an d Mermin (2009) have shown that medical conditions reduce nonhealth spending, particularly for low- expenditures Total Spending for older ho 2007 46,213 useholds (Hurd a14,531 nd Rohw 37,647 edder 19,516 , 2011). -4,985 31,728 9,063 58,849 25,765 Households A 65 ?g 74 e 50 or Older 70,776, 2001 ?47,838 2007 ............................................................................................ 22,938 239,919 475,332 10 Medicare Part B is a part of Medicare which helps cover medi • Instan ntly get e-mail noti?cations of the la atest EBRI cally necessary services, such as d data, surveys s, publicatio doctors' services, outpatient n ns, and meet tings Hispanic 27,600 21,584 6,016 19,322 121,422 Preretir Activities Mail Survey (2001 ement—Account bal ?2007). ance is included in wealth. Widowed 21,238 24,388 -3,150 40,237 165,003 Source: Employee Benefit Research Institute estimates from the Health and Retirement Study (2000 Source: Employee Benefit Research Institute estimates from the Health and Retirement Study (2000 ? ? 2008) and 2008) table mainly h (belo igh w sc ) shows how hool dropouts, these accou who do not nts are treat have incom ed in e su the construct fficient to support all ion of the the inco ir m ex e an penses, and d wealth measur wealth hol es in din RAND gs that home or Source: Employee Benefit Research Institute estimates from the Consumption and Activities Mail Survey (2001 long-term care facility is rather small. In the CAMS sample used for this study, only -2009). 3.65 percent reported 75 ?84 Source: Employee Benefit Research Institute estimates from the Health and Retirement Study 53,227 43,066 10,161 326,774 554,358 HRS, which income senior asks respondent s who are on ts to report he cusp of reti their incom rement. e duri ng the last calendar year. (The 2002 HRS data have the EBRStandard errors are reported in parentheses. For mean regressions the standard errors were clustered around each household. For I Issue Brief is register Source: Employee Benefit Research Institute estimates from the Health and Retirement Study ed in the U.S. Patent and Trademark Office. ISSN: 0887 ?137X/90 0887 ?137X/90 $ .50+.50 the median care, home health services, and other medical services. Part B also covers some preventive services. and s seminars by the Consumption and Activities Mail Survey (2001 and the Consumption and Activities Mail Survey (2001 c clicking on t the “Notify M M ?2007). e” or “RS ?2007). S S” buttons a at the top of o our home pa age. Never Married 21,943 22,983 -1,040 16,437 94,989 Figure 6A, Median 0% Househol (2000 ?2008) and the Consumption and Activities Mail Survey (2001 d Income, Household Spending, Household ?Nonho 2007). using Wealth, Household Total HRS dat are also a. insufficient to 85+ guara 39,620 ntee a comfor 34,377 table retirem5,243 ent. For example, in 175,600 2007, their median 365,644 expenditure exceeded spendi The househo ng at l regressions the standard errors were bootstrapped with 400 repetitions. e ld in ast one come a (2000 night i nd ?2008) and the Consumption and Activities Mail Survey (2001 wealth n a nursin measu g ho res are taken me or a long from -term care the RAND facili version o ty in th ?2007). e tfw HR o y S data e ars pr becaus ecedine it g thprovi eir 20 des 08 HRS income i nformation for 2001, and so on.) The 2009 round of CAMS data could not be used for this figure or any of the Postretiremen Source: Employ t—Regul ee Benefiar t Rewit search I hdrawals are i nstitute Estiman tes from cluded the H in inc ealth and R ome, an etirem d th ent S e tcurrent udy (2000 accou ?2008) and the C nt balance onsum is i ption and ncluded in Anticipated an Increase Experienced an Increase *p< 0.10, **p<0.05, ***p<0.01. Source: Employee Benefit Research Institute estimates from the Health and Retirement Study (2000 ?2008) and the Wealth (Housing and Nonhousing), in 2010$, for the Bottom Half of the Income Distribution of Different Also, the corr their m edian ie ncome lation betw by $3 ee ,518 n hea anlth a d thn ey d had o wealth nly has $13, be355 en docume in nonhnte ousing or d very we liquid ll in assets. Their the economic net total literature. we But alth in the 2007 intervie consistent measures of w. AcGiv tivitie es M n that ail S nu urvey inco rsin (2001 g-hom m ?e an 2007). d e stays wealth can be lo across all ng a waves. nd very expensive, undersampling the institutionalized other figures where income and ex penditure are compared, because the 2010 round of the HRS provided by RAND was wealth. Consumption and Activities Mail Survey (2001 ?2007). There e’s lots more e! Source : Employee Benefit Research Institute estimates from the Consumption and Activities Mail Survey (2001 ?2009). th Age Groups ....................................................................................................................................... 12 * Less than 50 observations. was $ more import 92,542, a nwhich t findi incl ng fro udes housin m the regr g essions is that L wealth. With falli TC insuranc ng housing e pric and es and rea private hea l estat lth insura e valu nes, it is ce continue likely that to be they population may underestimate the actual health-care 1100 13 expenses. Street NW · Suite 878 Also, the distributio n of health expenditure is generally not available when this study was done. “Income” includes wages and labor earnings, public or private pensions, Washington, DC 20005 important have experien dete ced a rminants o decline f sip n net endin w g ev eVisi alth en a t tafter co EBRI on- nd will ntrollin need a li line toda g dditio for incom nal yfun : w w e a d ww.ebri.or s for a nd oth comfor er facto g g table retirem rs, `though o ent. nce a control for very skewed, so the median may not show the entire picture of health expenses. For example, in 2009, for households annuities, government transfers, and lump-sum distributions from pensions or insurance and inheritance (see next © 2012, Employee Benefit Research Institute (202) 6 ?Educ 59-06 ation and Re 70 search Fund. All rights reserved. www.ebri.org www.choosetosave.org ebr ebr ebri.org ebr ebr ebr ebr ebr ebr ebr ebr A research rep ebri.org i.o i.o i.o i.o i.o i.o i.o i.o i.o i.oebri.org Issue Brief • February 2012 • No. 368 ebri.org Issue Brief • February 2012 • No. 368 ebri.org Issue Brief • February 2012 • No. 368 ebri.org Issue Brief • February 2012 • No. 368 ebri.org Issue Brief • February 2012 • No. 368 ebri.org Issue Brief • February 2012 • No. 368 ebri.org Issue Brief • February 2012 • No. 368 ebri.org Issue Brief • February 2012 • No. 368 ebri.org Issue Brief • February 2012 • No. 368 ebri.org Issue Brief • February 2012 • No. 368 ebri.org Issue Brief • February 2012 • No. 368 ebri.org Issue Brief • February 2012 • No. 368 rg rg rg rg rg rg rg rg rg rg Is Is Is Is Is Is Is Is Is Is Issue Bri Issue Bri su su su su su su su su su sue B e B e B e B e B e B e B e B e B e Bort from the EBRI ri ri ri ri ri ri ri ri ri rie e e e e e e e e ee e ffffffffff • • • • • • • • • • ff • Februa • Februa Fe Fe Fe Fe Fe Fe Fe Fe Fe Febr br br br br br br br br brua ua ua ua ua ua ua ua ua uar r r r r r r r r ry y y y y y y y y y 2 2 2 2 2 2 2 2 2 2 r ry y0 0 0 0 0 0 0 0 0 0 2 2 12 12 12 12 12 12 12 12 12 12 Education and R 012 • No. 368 012 • No. 368 No No No No No No No No No No 3 3 3 3 3 3 3 3 3 368 68 68 68 68 68 68 68 68 68 esearch Fund © 2012 Employee Benefit Research Institute 15 25 21 22 17 14 12 16 23 24 9 7 18 13 10 20 19 11 2 3 5 6 4 8

Volume 368

Pages 28

EBRI Issue Brief

Sudipto Banerjee

Health