“’Short’ Falls: Who’s Most Likely to Come up Short in Retirement, and When?” and “Consumer Engagement Among HSA and HRA Enrollees: Findings from the 2013 EBRI/Greenwald & Associates Consumer Engagement in Health Care Survey”

“Short” Falls: Will Baby Boomers and Gen Xers have enough money to live on when they retire, and if not, when will they run short? New modeling by EBRI finds that those in the lowest-income brackets are most likely to run short, many in the first year of retirement. But some in all income brackets—including the highest—may also run short at some point during their retirement. HSA/HRA Consumer Engagement: Which type of health plan is more likely to get workers involved in their own health care: Health savings accounts or health reimbursement arrangements? The two account-based types of health insurance are similar, but a new report from EBRI finds that people with HSAs are more likely to engage in cost-conscious behavior related to use of health care services than are those in HRA.

“Short" Falls: Who’s Most Likely to Come up Short in Retirement, and When?

• This Notes article provides new results showing how many years into retirement Baby Boomer and Gen Xer households are simulated to run short of money, by preretirement income quartile.

• Under a variety of simulated post-retirement expense scenarios, the lowest preretirement income quartile is the cohort where the vast majority of the retirement readiness shortfall occurs, and the soonest. When nursing home and home health-care expenses are factored in, the number of households in the lowest-income quartile that is projected to run short of money within 20 years of retirement is considerably larger than those in the other three income quartiles combined.

• Extending the results to a maximum of 35 years in retirement (age 100, assuming retirement at age 65), 83 percent of the lowest-income quartile households would run short of money and almost half (47 percent) of those in the second-income quartile would face a similar situation. Only 28 percent of those in the third-income quartile and 13 percent of those in the highest income quartile are simulated to run short of money eventually.

Consumer Engagement Among HSA and HRA Enrollees: Findings from the 2013 EBRI/Greenwald & Associates Consumer Engagement in Health Care Survey

• Health reimbursement arrangements (HRAs) and health savings accounts (HSAs) are very similar, though there are some key differences that may produce different incentives related to using health care services, and thus, different consumer engagement experiences.

• Adults with an HSA were more likely than those with an HRA to exhibit a number of cost-conscious behaviors related to use of health care services. Those with an HSA were more likely than those with an HRA to report that they asked for a generic drug instead of a brand name; checked the price of a service before getting care; asked a doctor to recommend less costly prescriptions; developed a budget to manage health care expenses; and used an online cost-tracking tool provided by the health plan. Adults with an HSA were also more likely than those with an HRA to be engaged in their choice of health plan. Individuals with an HSA were more likely than individuals with an HRA to report that they had participated in a health-risk assessment, health-promotion program, or biometric screening program when it was available.

Figure 5 Figure 7 Figure 3 ® ® th Figure Figure Figure Figure 31 1 5 Endnotes assumptions show, th Appendix A: Brief Description Demogr Appendix B: Brief Chronology as than In that Cost- By the 80 s igned Percen ? ? 1 in . . “All or 2 the C T T T 0 to th 5 aphi he ne he June 201 eons stimon init year in (1 t o9 percent c Nothi e CDHP fw ial stakehol c De Differen mo ious Use of y. Joint DOL retirem terministic ng? A de 3 g e lo l EBRI Issue Brief w )r of t oup a n ce we der s use Exp ent, 73 s h ,st-income / they m s SEC Public H ose in of th ad n Expenses, to a d Health Care Se e percent d the e model Persp nal us (VanDer qua tyz se al e ho of t s eari cond-income e of EBRI’s of the EBRI rtile is o wer ctive on R No have h w ng on T he hei, eli Stoch e the lowest ig nibi June 2 terest a most d a e litrvices rg a tir astic y -income quart f e et Dates F n Retireme o ed ment Re 0 vulnera account, suc r parti 1 Retire 3a) u in Health ile determ cipati woul quarti sadi ed RSPM to p unds ble, ne Co d face on in ment Sec lin e nt Security Projection Model while h ss.” . house sts How ing as a a de EBRI Notes, a simi what lon W hea fined co h ould T rg olds ovid perc evit llar th ue a direct would hav y and lon savings account (HSA rity Projection Model situation. Only aentage of ntrib rget-D no. 11 (Empl utiate F on pl comparis ge -te f ru u an ur tu n m care n short o 7 d o re retir impact s Like yee Be perc on of ) or ent o are t ly Imp s retirement enefit f the like ment mo health h fa t e ney, c ht lose y Demographics, by Type of Health Plan, 2013 Consumer Engagement Among HSA and HRA Enrollees: “Short” Falls: Who’s Most Likely to Come up Short in Years Years in Retirement in Retirement Before Before Baby Boomers Baby Boomers and Gen and Xers Gen Run Xers Short of Years in Retirement Before Baby Boomers and Gen Xers Cost-Conscious Years Av in ailabili Retirement tDecision y and Before Use Making, of Cost Baby by Information, Boomers Type of Health and 2013 Gen Plan, Xers 2013 10 1 ® ® ben incom Rese Future 401(k) efits und e arch Institute, Novemb adequ er specific t ac Accumul y in Septem y apes tions ber 20 er of ? 2012): 11 DC and DB r (T-160), 18 June 2 10 (Va–2 nD 3. erh etireme ei, Septe 00 nt plan 9. www mber s. .e 20 bri.or 10), an g/pdf/pub d walicat s later us ions/te ed stimon to compute R y/t160.pd eftireme nt reimbursem Demogra cohorts woul Finally, b in th but on iggest risk e thir ly 3 Figur pd-in 1 hics en pe factors across the entire d run ecome quartile may e t arran rcent o 6 provi short of x gement fdpla those es t in s h money in e and (HRA o results of in th me of 2 e secon perc ), the retir with incom ent a di simulation a d efment a -income of those f rollover e sp erenc ectrum. e n i in t d w provi quarti nwh co h her en, EBRI n e s s l highest-i e e ion th umer en wo only uld 80 perc at t spe fac ga nco hey ci ge em f a ically c could ment ent of the averag e quartile similar situat betw use to hose ar een n pay for m e simulate ot to rely i individu on. Only e det on e e als w d to dic 15 rmini a percent al run replaceme i expenses or th an stic short of ex HSA a of t penses are nht-rate ose in money n t d an he The ? ge Tnerati he Retirement Secu on born betw rity een Proj 1948–19 ection Mo 64. de l (RSPM) grew out of a multi-year project to analyze the future economic The One of the 2013 CEH basic objectives CS found evidof the Re ence that a tirem dule ts nt S with ecurity an HS Projection Model A were more likely than (RSPM thos ) is to simulate the percenta e with an HRA to exhibit a ge of the Money, by Various Percentages of Deterministic HSA Retirement HRA Expenses Run Short of Money, by Preretirement Income Quartile Run Short of Money, by Preretirement Income Quartile Retirement, and When? Findings from the 2013 EBRI/Greenwald & Associates Run Short of Money, by Preretirement Income Quartile 90% Savings Shortfalls (RSS) for Baby Boomers and Generation Xers in October 2010 (VanDerhei, October 2010a). ability to HRA. Ind target paid, within the thi but non rd-inco as a m a deca take ividude. e e m als their ac of t asure of success in RSPM, e quartile with he stocha an count HS and 6 with t A wer stic health percent h eem sho more car of fo li those i uld t e kely than r several costs are hey c n th thos reas h pai e h ange ie w ons: d ghest-income jobs. by the ith aIndivi n hous HRA duals quart ehold. to be with ile In mal are this onl e: O simulated to r ycase, a com a f ne-h lexi alfb ( le 5 spendi 2p pe u an short o rison wit rcent) ng accoun of h f mo Figure ney t 5 well-being of the retired p Gender opulation at the state level. The Employee Benefit Research Institute (EBRI) and the popul number ation of cost-conscious at risk of not hav be (100% ing haviors relat retirem of nursing home ent ed to income use o adequa or home-heal f health te to car cover e servi th-care average expenses and un ces. Specific costs in each case): ally, those with insured an HSA health wer c e are more Scenario: Scenario: 80% 90% of of deterministic deterministic retirement retirement expenses, expenses, June 2014 • Vol. 35, No. 6 Scenario: 100% of deterministic retirement expenses, ? . “Increa . “ TT he June 201 he Ex sing D pected Impact e 3fault D EBRI Notes e of Automatic E ferral R article (Van ates in Au scalati Derh tomatic Enr on of 40 ei, June 2 o1(k) Contri llme 013 nb t 401(k) ) used RSPM to sho butions on R Plans: The Impact on Retirem etirem w that 2 ent Income.” 5–27 pe EBRI Notes, rce ent Savin nt of Bab gs no. 9 y 2 57% th th In pr Consumer Engagement in Health Care Survey esenting these results, EBRI does not favor or oppose any specific modification to the current retirement system. The generation born between 196 Fe 5– ma 197 le 4. 48% 61% By Jack VanDerhei, Ph.D., Checke E d wh methe ployee r plan Be woulne d Low cov fit R er est car esearch e Preretirement Institute Income Quartile (FSA indivi shows the by the ) dwer 2 uals Milba 0 e no impact of y wit e nar. By th k Memo h t inclu an HSA de ignor rial F e d i wer 35 n t ing the un e year i he C d male, , wDHP o stochastic costs whe rki n 100% compared r ng egrou tire wiof nursing th the ment, 7 p. with office of the gover 6 3 perc home or 9 percent n th ent of ere is the home of individuals with a nor of Orego lo 20 w health care costs per est-income cent r n, set out in the la eductio qu anartile HRA n in t househo (Figure he te 1990s to se det e 5 lrm ds wo ). inistic costs. uld hav e if this e run 100% of nursing home or home health care costs costs ( likely tha includ n tho ing se with an HRA long-term-care costs) to report th 100% at ag of nursing at they e 65 or old asked home or erfor a throu home generic ghout health care costs re drtirement ug instead of in specific a bra income nd name groupings. (52 percent H RSPM SA 56% ? Boomers a In October 201 (Emplo Success in Plans With yee Be nd Gen Xers 0 testimon nefit Rese Automatic Escalation. warch y bef ho wore t Inst ould hav itute, Septe he Se e ha nate d ade Health, Educ m ” EBRI Notes, ber 20 quate re 07): 2– tia rement inc no. 9 (Emplo tion, Lab 8 or an ome 78%*** ye under retur d Pensio e Benefit Res ns Com n assumptio earch Institute, Septemb mittee on “T ns ba he W sed on obbly er th ? Because 80% very few households annuitize all (or even most) of their individual accounts in retirement, a Where the world turns for the facts Male on U.S. employee benefits. 52 39 R By the ather 2 , EB 0 R year in I’s missi retirem on remain ent (a s to p gainr, assuming retir ovide objective an ema elnt at ysis th age at c 65), more an inform than decisi 4on in ma 5 (8 ki1 pe ng brcen y oth t) e of rs. As th the loe west- 90 70% % 90% short of money, and 41 perc situatio 90 n cou % ld be evaent of thos luated for the st e in the ate. second-in The resultic ng a ome quartile nalysis (Va would fac nDerhei an e a d Co similar situatio peland, Septemb n. Only er 200 231) percent also provides i vs 3 . 49 percent HRA); nformation check on e the distri d the price butio of a s n of the likely ervice befo re nuge mb tting er o care f year (4 s b 1 pe eforce re th nt oHSA vs. 3 se at risk r 4 perce un sho nrt HRA) t of mo ; aske ney ad s a By Paul Fronstin, Ph.D., Employee Benefit Research Institute historic Stool: Retirem 2012): 12– al aver 22. a ent (In)securit ges were simul y in America,” th ated to end up e mode running sh l waort o s used to an f moneyal in retir yze the relativ ement if toda e imp y’o s historica rtance of em lly lo plw o yier- nterest replacement-rate focus wou Age ld overlook the potential risk of outliving their income (longevity risk). See Appendix A for a brief technical description of RSPM. Additional detail is provided in VanDerhei and Copeland various desi Because Diff Although income erenc . “Measurin quarti the 2 es in pe gn base sample l rcent or e and pro the a house g Reti ge less o hcomposition of g rem olds ram (n ent Income Ade would ru ational sam fmodificat those in n sho in the ion a p div le quac second-, rt of mo ) i ilduals ternatives ar nclu y: Ca d with ed onl ney, c thir lcul an a d- and hi te y ing o HSA and HRA 18 mpared deRe bat 0 ialistic In n e ghest divi d (both with 38 perc du -ials in a come Re n were reforms come fou CDH quartil ent plac nand status quo d. o P ef S m th es are sim and p ent Rates.” ecif ose in 39 ically, 7 i the n 52%* ulat divi )indi , it second-income EBRI Issue Brief, e duals i d to vsiduals instructive t run with shor with an a t of no. n o focused primarily on simulated retirement wealth with a comparison to ad hoc thresholds for retirement expenditures. of those in theA third-in sked for gene come ric drug quartil instead e an of brand d n 1ame 1 percent drug of those in the highest-income quartile are simulated to run short well doctor to r as the p eco ercentag mmende les ofs p -cro eretirem stly pres ent co criptions mpensation (40 perc th eney t HSA vs would ne . 38ed in t percee nrms of t HRA) a ; dev dditio elop nal savin ed a bud gs in get orde to ma r to nage have provi rates ded retir were ass ement ben umed to be efits and Soc 21–34 a permanent con ial Secd uirit tion. y (Va nDerhei, October 20 27 10b). 16 Introduction 70% 49% 80 80 80 %% % 67% Retirement and health benefits are at the heart of workers’, employers’, and our nation’s . “I297 (Emp s Working to Age loyee Benefit Rese 70 Really the Answ arch Institute, Septemb er for Retirement Income A er 2006). dequacy?” EBRI Notes, no. 8 (Employee Benefit keep in m HDHP, an oversample o HSA wer money in quartil (2010).e that e the mor ind w f who’s most e io rst year o like uld face ly tha a si f individuals fn ret lik tely to com milar situatio hiose wit rement, h with 22 a e u n percent a CDHP or HDH n p HRA to . short Only of t 19 in r be i per e h ntir ose in tch e ent of e ment, P was 21– the those 34 added. whe low -year n e in , st-income -and the old age The oversa thir why. d-i quar grou nmple in co p, w tile me quartile woul hile cluded 1, in d hav divi an duals e0 a d 8 62 de percent indiv with ficit an idua bas of HRA ls ed o with n ? While th 60%e annuity purchase price relied upon in a replacement-rate target does depend on an implicit of money eventually. 35–44 22 34 a 50, health 7 care expenses 0, or 90 percent (32 percent probability of HSA retir vs. 22 perc ement income ent HR ade A); q an uacy. d used an online, cost-trackin 1 g tool provided by the ? The April 2001 EBRI Issue Brief (VanDerhei and Copeland, April 2001) highlighted the changes in private pension Will Americans have enough to live on when they retire? As the pace of Baby Boomers crossing the traditional “Short” Falls: Who’s Most Likely to Come up Short in economic security. Founded in 1978, EBRI is the most authoritative and objective source of ? ? T TRese he November he August 20 arch Institute, August 20 13 2010 EBRI Issue Brief EBRI Issue Brief 12): 10– (VanD e 21. xperh a nded u ei, August 2 pon earl 01 ie 3r ) used RSPM work by EBRI to provide th to analyze the e first results of a ne Obama administration’s w a CDHP. were more lik determi those innistic a the In addition hi ely than ghest-i nd stochastic to bein n those with comeg stratified, th costs. By the quartil an This survey e ar HSA t e sie base sample 10th mulated o be in was mad year the 35 to r in r ue e n short o –44 p was also tirement, ossib age le with supp fgrou mo 2 wei 9 g ney by p perce hted by . The n ort from: r the twentieth year. t of e wer ge lo nde w e e no di r, age, st-income fferenc educati qu es in t artile on, re he househo gion, percentlds age assumption with respect to 45–54 (at least some) future market returns, 31 it does not typicall 29 42% y account for the 70% 4 . “Defined Ben 70 70 %% efit Plan Freezes: Who's Affected, How Much, and Replacing Lost Accruals.” EBRI Issue Brief, no. 291 health pla 60n % (27 percent Talked to doctor HSA about vs. 21 percen prescription options an t HRA) (Figure 1). d costs plan participation for defined benefit (DB) and defined contribution (DC) plans and used the model to quantify how threshold age A brief information on these critical, complex issues. chronolog of retir y ofe RSPM ment accelerat is contained es, so has th in Appendix e fr B. equenc Lowes y t-Inc with ome w Quh aich this rtile question is posed. Unfortunately, the HSA HRA 39% fiscal simulati year (F on mo Y) 2014 b del that estimate udget prd the impact of oposal to inclu chang de a cap ing 401( on tax-d k) plan d eferred retir esign e varia ment savi bles angs nd athat ssumptio would limit ns on the retirement 55–64 21 21 income, a that w 90 Retirement, and When? would ru Percen ere n sho n in d r t o eiace/et f rt of mo ther th Deterministic hnicity e ney 45– t but only 5o 4 refl ag Expenses, e co ect th 3 hort o perc e aent of ctual r the No p. 2 55 proportio those Stoch –64 a in astic g n the e co s in t second-in Health hort. T he popu Co here come quartile, lation a sts were g also es 21 no di only –64 wit ff1 erenc perce h priva es by mar nt of t te hehalth ose in ital sta insur tth us. ance e VanDerhei potential and Co inv pelan estment risk ass d (2010) descri ociated wit be how househol h “risky” asset allocations (investment risk). ds are tracked through retirement a ge and how their 50% . “Retirement Readines (Employee Benefit Rese s Ratings a arch Institute, March 200 nd Retirement Sa 6). vings Shortfalls for Gen Xers: The Impact of Eligibility for much the importance of individual-account plans waSe s e cox npec d ted to increase because of these changes. 4 answers provided are as diverse, and sometimes disparate, as the projection models that produce those results. 5 These ? values Blue Cross an continue 60% d Bl to i ue nc Shi rease u Me ald ritaAssociation l n Stil tatall us households either run short ? of money HealthEquity or there are no surviving retirees. By amou incomnts accu e adequmulate acy. Un d in sp til rece entl cified retir y however, there ement accounts to was extrethat necess mely limiteda ev ry t ide o provi nce o de the n the imaximum a mpact of auto nnm uit atic y permitted 60% coverage. Figure third-income 4 provi Th 60 quartile, d e %es th CDHP e resu and le oversa lts of a ss than 1 mple simulati was perc wei on w g ent of hteh dere those by gen only in d 90 er, age, the percent highest-i in of com the ne come quartile , an avera d rac ge e/et deterministic hnicity. Mor are simulat ex e ed ipenses ar nto run formation short of e pai cad n be retireme Health VanDerhei Plan nt inc and Choice ome/w Copeland ealth Deci is simulate (Ap sions ril 20 01)d used for the fo the ea llowi rly version ng cLo oLo mponents: wes wes of t- Inc t-the Inc om o m model e e Q Q u a ura tirlt e ilto e highlight the changes in private-pension- Participation in a 401(k) Pla 7 n.” EBRI Notes, no. 6 (Emplo Third yee Benefit Research Institute, June 2 41%** 012): 9–21. 50% th EBRI focus ? Previous EBR es solel Chec I research ked pricy e on emplo of ser Not h vas ice m befo demonstrate arried re ge yttin ee g benefits research — no lobb care d that one of the biggest f 20 inancial o 20 ying bstacles or advocacy in terms of . maintaining EBRI Employee Benefit Research Institute Notes (ISSN 1085 ?4452) is published monthly by the Employee Benefit Research Indivi Consumer Engagement Among HSA and HRA Enrollees: the 35 duals . “Projections of Futu year wit in h retir an HSA ement wer (age re Retirement Income e more likely 100, assu tming r han those with an Secur etirem ity e: Impact of Lon nt at a HRA to ge 65), be Hispan g T 83 eperc rm Care Insuranc ent o ic or fAsian. Specifi the lowest-i e.” 200 nc c 5 America ally, 1 ome quartile 3 perce n Soci nt of ety for a tax-q contribution ual es ificalati ed defi on (Va ned be nD ne erhe fit plan u i and Luca nder current la s, 2010). w. ? With the assistance of the Kansas Insurance Departme Sec Sec ond n ond t, EBRI was able to create the E 34% BRI Retirement Readiness found in but money. plannone participation Fro of t nstin he stochastic h (20 for 1 define 3). d 44%*** e benefit alth care cost (DB) and s are defi paid by ned co t nthribution e househol (DCd. ) plans In this case, and useda comparison the model to with quanFi tify gure how 3 much th Highest-Income Quartile The 2013 CEH ? Chevron CS found that adults with an HSA were more likely tha ? n tNational hose wit Ru h aral n HRA to Electricbe Cooperative Associ engaged in their ation 50% 40% Institute, 1100 13 St. NW, Suit M e arried 878, Washington, DC 20005-4051, 80 at $300 per year 80 or is included as part of a membership One explanati EBRI on for stand the dis s alone in em parity among t ployee he resu benlts fro efits rese m the ar various ch as an models li independ es in en the t, nonprofit, and no different age cohorts npartisan 50% retireme 50TM % nt income adequacy for households who might otherwise have sufficient financial resources at . “Retirement Income A on Aging/National Counc dequac il ony for Bo Aging Joint C omers an onferenc d Gen Xers: Eviden e, March 2005. ce from the 2012 EBRI Retirement Security indivi households duals wit wo huld r an HSA un short of mo were Hispan ney a ic an nd almost hal d 8 percent fwer (47 percent Th eTh Asian, com ird ird ) of t pared hose in with the 6 se pecond-income rcent Hispanic quartil and 4 e percent would face the importance ? Social S Rating eof (RRR curity individual ) based o . -acco n a fu unt ll stochastic d plans was expecte ecumul d to ation mo increase del t be hat took i cause ofn these to accou chang nt the hous es. ehold’s longevity risk, shows the Findings from the 2013 EBRI/Greenwald & Associates impact of ignoring the stochastic costs when there is a 10 percent reduction in the deterministic costs. subscription. Periodicals postage rate paid in Washingt on, DC, and additional mailing offices. POSTMASTER: Send address choice of health plan, when they had a choice. Specifically, those with an HSA were more likely than those with an ? ? T In F 40 he December %ebruary 2011, the mod 2013 EBRI Notes Race el /Eth wan s use i article (Van city37% d to anal Derh yze the im ei, Decem pactber 20 of the 2008– 13) used 200 RSPM to exp 9 crisis in the fina and the a ncianal l an ys d real estate is in the organization. ®It analyzes and reports research data without spin or underlying agenda. All findings, examined. For example, some models limit their analysis to households already retir40%* ed, while others focus on th retirement age is the risk of long-term care costs for HiHi ghe ghe st a prolonge s in t-In cocme ome q uar Quti arti le d perio le d. As with the annuitization experience ? Deseret Projecti Mutua on Modl e l. ” EBRI Notes, no. 5 (Employee Benefit Researc ? Optum h Institute, May 2012): 2–14. Whil Asian among i By the 20th year in a similar situ e pane Al I skat ed n 40 n ternet surv ion. Only doc divi % toduals r retirem to rec 2 om wit eys are 8 men e percent h nt, 32 d an less HRA. co nonran stl percent of t y pr es hcose in do ripti of t m, studies on dr the h uge lowest thirhav d--income income e demonstrated that quartil quartie an le house d 13 such h perc olds surveys, when c ent of would hav those e ru in the n short o areful highes ly design f mo t- ney. ed, changes to: EBRI Notes, 1100 13 St. NW, Suite 878, Washington, DC 20005-4051. Copyright 2014 by Employee Benefit post-retirement investment risk, and exposure to potentially catastrophic nursing home and home health-care risks. White 74 80 38% . Testimon 40%y. U.S. Congress. Senate Special Committee on Aging. Retirement Planning: Do We Have a Crisis in HRA to report June 2 markets on reti wheth 0 that they 13 EBRI Issue Brief. er on fi rement inc ha nan d talked to cial ome data, Rather tha adeq friends, family, option uacy (Va ns tr , ying to reflect th or tre nDer an nds, hei, F d colleagues a are reve ebru e real- ary 2 aling 0w 1 bo 1 out the r). ld an va d riatio reli plans able n in DB ac (44 — the re perc cruals, the bas ent H aso SA vs. 32 n EBRI informatio elin p e ercent n is Consumer Engagement in Health Care Survey, 6 40% p. 19 30% households Research still worki Institute. All nrights g, but reserved, old enou Vol. gh35, that reasona no. 6. bly accurate projections regarding their future wages and VanD ? erhei cited above, Defi ned contri and Copeland inbution the real (200 (DC) balances. 3) worl . d few retirees have long-term 100% care insurance policies in place that would cover the obtain results comparabl Although 4 percent or In contrast, o income quartinly 4 le are percent simulate less o e wi of those fd to r th those in the ran Bla un short dom-digit-dia c in t k he second-incom second-, of money l telepho thir eve d-, e quart n ntually. e and hi surveys. ile ghest wo 4 Ta u-lid ylor, (2003), ncome face a simil quartil 5 for exam ar situation, es are sim ple, ulat a pr ne d o ovides th d to nly 1 run perc shor e res ent of t of ults The first state-level RSPM results were presented to the Kansas Long-Term Care Services Task Force on July 11, the gold standard for private analysts and decision makers, government policymakers, the media, and . “Modifyi Amer 30% ica ng the F ? Resue lts From the E deral Tax Treatment of 401( BRI-ERF Retirement Secur k) Plan Contri ity Projectio butions: Project n Model ed Imp (T-141), 27 act on Particip Jan. 200 ant Account 4. HRA); use ana dl ot ysis in the her websites t previous an o learn alysis used the m about health pla edin an accru choices (4 al ra 1 p te in the sample (1.5 percen ercent HSA vs. 32 percen t of final compensation per t HRA); consulted with 30% prospects for accumulating retirement wealth are obtainable. Other models attempt to analyze the prospects for all Individuals with an HSA were more likely than those with an HRA to 90% be highly educated. One quarter (26 percent) of ? potential An April 2ly cat 011 article i astrophic ntroduc fina Hispanic ed a ne ncial imw pact me of thod of an this exal posur yzing e the results from RSPM (VanDerh . The 13 refore, any attempt 6 37% to simply add t ei, April 2h 011). Rath e cost of er from a money in those innum the the be th f 30 r of surv iird-i rst year o % ncome eys th f qu retartile a at wer irement, e n co d less 34 ndu percent cted at than 1 of t the perce h sa ose in nme time t of t the hose in low usin e t st-income g the he h same ighest-inc quar que tile ome quartile stionnaires would hav bo a ere simulat th via t a deficit ele bas e pho d to run en d o e n 2002 (VanDerhei and Copeland, July 2002), and the results of the Massachusetts study were presented on Dec. 1, the 30% Talpubli ked to c do. ctor about treatment options and costs 7 Balances.” EBRI Notes, no. 3 (Employee Benefit Research Institute, March 2012): 2–18. year of participation) as the stylized value for the baseline counterfactual simulations. The new research computed their employer’s HR staff (39 percent HSA vs. 26 percent HRA); consulted with 32% an insurance broker to understand See VanDerhei (June 2012) for details. ? Individual retirement accoun Asiant (IRA) balances. 80% 8 th 4 indivi The VanD working house Impact d erhe uals i, Jack, and Nevi wit of h h an olds, in Nur HSA h sing cluad n Adams. “A Little Hel Home dina gra g those and duate whose Ho de me gree, relative Health p: T compar he Impac yo -Care uth ed makes accur w t of On-line Expen ith one s-e fift s Calcu a h (19 te, lon lators a percent g-ten rm d) F of predictions so inanc indiv ial Ad idua visors on Setti ls mewhat with an HRA. ng than simp 20% ly computing an overall perceA T ntage of the simulat A G L A N C ed life paths in Ea particular cohort that would not have long-term care insurance into a replacement-rate methodology will vastly underestimate the potential severity The Employee Benefit Research Institute (EBRI) was founded in 1978. Its mission is to and o determi short of money by that po nlin nistic a 200 e. He 2 (Van 20 foun n% d stochastic Derh d tei an hat th int d Co e use costs, as illustrated . By the pel of and, Decemb demo 35th year in graph er 2 in ic retir Fi wei 002). gure e gment, a h t 4. ing By th alone thir e 1 was d 0 of the suffic year i loiwest-i n ent re to tirement, just over hal ncome quartil bring almost all e house of tfh h e r (52 perc olds would esults ent) 20% the actual final-average DB accrual that would be required to provide an equal amount of retirement income at age plan choices (31 percent HSA vs. 11 percent HRA); asked health plans to send information in theHS mail A (28 percent Other 1 6 20% probl e . “ matic. A T Adeq ax Reform Opti uate R nother etir ex eplan ment-Sav ons:ation for Promoting R ings contribute to the Targets: Evide e wide ran tirement Sec , to encourag ge n of ce u rit from the 2 results lies ye, an .” EBRI Issue Brief, d to enhanc 013 R in the e tireme e the developmen basic o no. 364 ( nt Confide bjective of Emplo nt of sound empl ce S the yeu e Be par rvey n t,” icular mo efit Researc EBRI Notes, oyee ben del h e used no. 3 fit Figure 2 simu sufficient retire 20la %tes the same scenario ment income to pa as Fi y for the si gure 1 in mulate term d e s of xp de ens te es, the ne rministicw e m xpe ethod com nses, bupute t this tim d the perce e the mo ntadel assumes ge of of this exposure. 8Introduction have ru from the onl of the lo n short west-i ine ncome of mo survey c ney, quartil lose to the repli wh e house ile only 4 holds perc es would ru from the paral ent of those n sho in rt of lel telephone s themo second-in ney, but only u crvey. He also foun ome quartile, 32%* 7 perc ** ent of 1 perc those d t ent o hat i infn the t some cases, hose in second-in the thi come rd- EBRI explores the breadth of employee benefits and related issues. VanDerhei 65 as (February w De o ve ul lop d be ed bud 20 produc get 14to ). med b ana Educ ge y h at t eal he an ion th carenu exitized v penses alue of the projected sum of the 401(k) and IRA rollover balances. Who we are ? Defined benefit (DB) annuities and/or lump-sum distributions. HSA vs. 10 percent HRA ? RSPM was expan)de ; ad to a na nd talked to t programs and so tional mode heir doctor or l—the first nation und public policy other he a through objective alth l, micro-simulati professional a on, retireme resear bout ch and the plan o nt-i educati ncome- on. EBR ptions adequ (19 per I is the only acy mod cent e l, Finally, individuals with an HSA were more likely than those with an HRA to be in the highest-income group. One- 10% 22% HRA “Short” F (Emplo Institute, Nove alls: Who’s Most Likely to Come up Sh yee Benefit Rese mber 2011). arch Institute, March 2013). ort in Retirement, and When? by Jack that any for the ana hous cost lysis. eh of nursi olds tha ng ho t would me me or et that requir home health-c ement are more t expenses are han a sp not ecifie borne d perce by t ntage of times in he household. As the simul expecte ation. d, t he 10% EBRI studies the world of health and retirement benefits — issues such as 401(k)s, IRAs, retirement income prope In 2001, a quartilnsity w e quarti woul hae d face ndfu lie g, an htin l o a simi d l g f em (mea ess th plo lar ni an 1 yng t situation, an ers started o percent he propen o d ffferin sity for a those only 2 g health in percent certa the r high e in of t imbursem type of eh st-incom ose in person to ent th e quartile arran e third-income gements (H be o are nlin simulated to r e) r quart RAs)— educe ile and l d th a th un short o e remai en-n ess than ew type o nin f1 m g per gaps, oney fcent 10% High school graduate 23 33 10% private, nonprofit, nonpartisan, Washington, DC-based organization committed exclusively to 9 HSA vs. 5 built i perc 10 n part from administr % ent HRA) (Figure 2). ativ While e 401(k) d indivi ata. T duals he initia with an l resu HSA lts wer were e more prese likely nted at the EBRI Decemb than those with an er 20 HRA to 03 po hlic ave y quarter (27 percent) of those with an HSA had household income of $150,000 or more, compared with 16 percent In a simi VanDerhei lar ve , Ph.D., in, a EBRI life-cycl e smoothing model was rejected for the development of RSPM, given the extraordinarily This ? a T diff he ere Januar nt sety of 2014 scenarios EBRI Not than es article (Van those modeled Derhe in iA , Janu ppendix ary 201 B of4) use VanDerhei d RSPM to model th (February th 201e likel 4). In ih thood th at case, at 401(k) a percentage of incom households e adequacy, in any incom consum e er-drive quartilen that r benefits, un short o Socia f mo l Security, ney within tax treatment of both retireme a particular time period is small nt and health er in College graduate 51 47 eventual but i of those n oth in t ly. er c he ases it did not highest-incom . Perha e quartile public ps th ar poli e e simulate most striking cy research d toand dif run fere edu short of ncat ce ion on in demographics money by economi the c securi betwee 10ty year. and em n te leplo phone an yee benefi d ot is nlinsu e es. VanD health erhe . pla Te i, Jack, and Crai n. stimon The mos 0%y. U.S. Cong t preva g Cop lent HRA ress. Senate F eland. “T plan he Impact of Defe des inanc ign e Committ then hrring R ad eea. Tax Reform deduct etiremibl ent Age o e of at l Options: Prom n e Ret ast $1,000 irement Incom oting R for em etire eployee-on Adeq ment Security uacy.” ly cov EBRI (T- erage ? ? Net housin As explored in g eq the June uity. 2011 EBRI Issue Brief, the RSPM allowed retir 27%* ement-income adequacy to be assessed at forum (VanDerhei and Copeland, 2003). vis This ited h Notes ealth arti Us plans cle ed onlbegins in’ we e cosbs t-trwit ac ite kin s h g to le a tool brie prar ovid f i n ed n atro bb yo hea u dt uction lththei planr plans or of the various metho attended a meetin ds of qua g wher ntifyi e nhealth pla g retireme nnt choices were income adequacy, among individuals with an HRA. low l consistent evels o 10 f “optimal” savin or 20 per 0 1cent 2 3 re g4duc s that 5tion 6 approa 7 was 8 9tach provi 10 ken 11 ac 12ro d 13ss es 14 bo fo 15th r low 16 categories 17-incom 18 19 20 e i of 21 n divi expen 22 23 duals at 24 ditures. 25 26 re 27 In tirement. 28 th 29is 30 No 31 te Wh s 32 art ile th 33ic 34leere 35 , it is littl was e partici benefits, pants curcost rentl y manage ages 25ment –29 w , o worker a uld have sufficient 4 nd employe 01r( k) accumul attitudes, a policy reform p tions that, when c ropo ombi sals, ned and p with Socia ension l assets Figure 2 than i0% n Figure 1, where those cost Graduat EBRI’s member e degree s are cont ship includes a emplated. cross-section of Moreov 26 21% er, the differences ill pension funds; businesses; trade associations 19 ustrate the significance; 0% 0% surveys was t and a ? tax Issue Brief, 170), 15 S This -pref 0% Notes herred accou e un ept. 201 der-re no. 358 (Empl article pr n 1. provides n t esentation that coul oyee Benefit Res e d of w r beminorities i e tsults sho apped eby ar w n ch Institute, June 2 wor ing ho online kers and the w ma sampl ny ye es. ir ars i 0families to 11). nto retirement pay out-o Bafby Boomer -pocket healt and Gen X h care er retirement ages later than 65 (VanDerhei and Copeland, June 2011). ® 0 1 2 3 4 5 6 7 8 9 1011 1213 1415 1617 18 1920 2122 2324 2526 2728 29 3031 3233 3435 explained, th th e differences were not statistically significant. along with decided to a onl and description y tak fundi 0e the 1 ng. 2 of EBRI’s R reductions 3There is wi 4 5 6 for 7 etirem de 8thspread e9 e compo 1nt S 0111 ere curity 21 nen cog 3t1 41 the nPition rojection Model 51 house 61 that if employee be 71h81 old9 2is02 likely 12 (RSPM 22 to32 have 42 )n . T e 5 fits his is fol 2a62 fair data exist, EBRI kno 72 amount 82lowe 930 d3 13 by a of 23 con 33 series trol 43 5over w of s it. (i.e., Trie 0d to 1 fin2 d th3 e co 4st o Hous 5 f he6 alt labor un ehold h7 care s 8 Ie nc 9 ions; h rvom ic 10 esY e be 11 ear efore g alth 12 s in 13 car Retir etti14 ng e prov em c15 a ent re16 (A iders and insur s17 sum 18 ing 19 Retir 20em 21 ent ers; 22 at government org Age 23 24 65) Fou 25 nd i 26 nfo27 rmati 28 a onizations; n29 30 31 32 and service firms. 33 34 35 doubt that so Securit me house y benefits, could repl holds may be ace 60, 70 able to subsist at or 80 perce lo nt of their w-income l preretirem evels inent retir ine come o ment as a n an i result nflatioof the n-adjuste support d basis. from By the that ignorin ? 2T 0he basic mod year in g these im retirem e portan l wae s su nt, 56 t costs makes bseq percent uently mo of t in any difie he d lowest a for testimon ccurat -income e simulatio y fo quarti r the Sen n o lef house retir ate Speci ement holds a in l Comm would hav come adequac ittee on Agi e run short o y. ng to qua f mo ntif ney. y A household is considered to run short of money in this model if aggregate resources in retirement are not sufficient expenses. T households he Medicare are simulate Prescription d to Dru rung short of , Improvement, TM 0% money, by 10 and Moderni % preretireme 20% zation Act o n 30 t inc % ome quartile. 40f% 2003 incl 50% uded a 60 provision % 70% to allow 2 Health Status and Healthy Behaviors the Summary deterministic expenses). Less than $30,000 4 2 results from t . “T . “The EBRI Retireme he Importanc he RSPM e of Defin focun sing on t Readi ed Be the ness R nefi percentage of Ba t Plans for Reti ating: Years Retire in Retirem rement Income ment Income P by-Boomer ent (Assuming retir and em Adequ repar ent Gen-Xer at ation age ac65) y.” and E ho BRI Notes, useholds simu Future Prospe no. 8 (Emplo cts.” lated to ru EBRI Issue yee B n sh enefit ort ? In a July 2011 EBRI Notes article (VanDerhei, July 2011), RSPM was used to provide preliminary evidence of the means-tested programs (including early eli®gibility for Medicaid), from a public policy perspective EBRI chose to Y Year ears s in in Retir Retirem ement ent ((A As ss sum umiing ng Retir Retirem ement ent at at Age Age 65) 65) However, o the ben nly Source: 10 efici EBR percent a I R l i em tirement pact of a mandator of Security those i Projection n th Model e s y co e vcond-income ersion ntrib 2030. ution of 5 p quart ercile, ent of compe 3 percent i nsn ation (Va the third-i nDer ncome hei, 20 qu 04). artile, a nd fewer to meet average retirement expenditures, defined as a combination of deterministic expenses from the Consumer References indivi ? d T uals he wit Februar h certai y 201 n high 4 EBRI Issue Brief -deductible hea (Van lth plans to Derhei, F co ebruar ntribut y 2014) focus e to a health savin ed on ho gw s th account e proba (HSA). Overall, bility of not runn 26. ing sh 1 ort Availability and Use of Cost $30, Inform 000–$49, ation 999 9 14 An individual’sSource: healt EBR h status I/Greenwald may also in & Associates Consum flue er nce Engagement how en in Health gaged h Care Surv e ey or , 2013. she is in health insurance and health-care-use EBRI delive Source: EBR rs a stead I/Greenwald & Associates y stream of invaluable research and anal EBRI’s work advances knowledge and unders Consumer Engagement in Health Care Survey, 2013. tanding of emplo ysis. yee benefits and their Rese Brief, no. 344 ( arch Institute, August 20 Employee Bene 11): 7–1 fit Researc 6. h Institute, July 2010). th of money The results Although 4 percent or impact of the “20/20 ca inpr re esented in tirement, less o Figures 1 throu as w f those in eps” on ll as an projecte estimate of the gh 6 show that second, th d retir how ement accu ird-, so th and hi on e years aft mue lghest a r ret tions pr of r irement this -ein tirement come oposequartil d befor b is expect y thes are sim e Nati e Baby Boom ed to onal Commiss ulat take plac er and ed to run ioe. In that Gen n on F shor Xer iscal t of establish a threshold that would allow hou ®®seholds to afford average expenditures throughout their retirement, while 10 than 1 perce Source: Source: Differenc nt of t Differenc EBR EBR e betw I R h I R e ose in ebetw e tirement een tirement een HSA HSA t Security Security and he h and HRA Projection HRA Projection igh is statis est-inc is statis tic M M a odel ll odel tic y s ai®ome quartile ll gnific y v ersion v sersion ignific ant at 2030. a2030. n*** t at p<0.01; *** p<0.01; wo ** p<0.05; uld fac ** p<0.05; * p<0.10. * e p<0.10. a similar situation at that point. By the 35 year in Expenditure Survey (as a function of income and marital status) and some health insurance and out-of-pocket, Fronstin, Paul. Source: "Findings From EBRI Retirement Security the 2013 EBRI/Green Projection Model version 2030.wald & Associates Consumer Engagement in Health Care Survey." of money in retirement var $50, ies 000–$99, with resp 999 ect to longevity, investment return, and p 33 36 otential long-term health care costs million individuals See ? VaUnder nDerhei a vari (Fe wit bet ruar h priva y of simul y 201 te insura 4) ated for importance to an post-ret dditio ce, renprese alireme th det e nation’s econo nating 1 il n t ex on the pe 5 percent nse sc impact myenarios, th of among policy of t longevity he market, e lo m risk west p akers, w an ere eit d the news reretir long her -term em in media, and ent in an care HRA or come costs. the public. It an H quartil SA- e is Consumer-driv ? The model en hw ea alth s enha plan nce s (CDHPs d to allo) ar w ae desi n analgned to ysis of the promot impact of annu e heightened sensitivit itizing defined contri y to cost in bution an people’s d indivi dec dual isions decisions. The 2013EBRI CEHCS publications found that i inclu ndi de viduals in-dep with th cove an HS rage A wer of key issues a e more likely tha nd tren n those ds; summ with an HRA to aries of rese report arch that there are di Resp ffe onsi rent perspecti bility and Rev form. es on the flexibility of individuals in retirement to adjust lifestyle and/or spending, results households money in the ru fn short o irst year o f mo f ret ney vary tr irement inem Figure endously 2, 3 by: 8 perc ent of those in the lowest-income quartile would have a at the same time accounting for the potential impact of uninsured long-term care costs. retirement, 57 percent of the low $100, est-inco 000–$149, me quartile 999 households woul 24 d have run sh 26 ort of money, compared with just . “ . “Cappi The Impact of PPA on Reti ng Tax-Preferred Retir ree ment ment Income for 4 Contributions: Prelimi 01(k) Participants.” nary Evidenc EBRI Issue e of the Im Brief pact of the Nation , no. 318 (Emplo al Comm yee Ben ission efit health-relat EBRI Issue Brief, ed expenses, plu no.393 (Em s stochastic ex does this b ployee Be penses y co nefit Rese nducting from n arch and p ursing Institute, Decemb ublishing policy re home and home ersear 2013). healt ch, analysis, h care (at le and ast until special reports on the point in retirement (e.g., nursing home costs). eligible plan (Fronstin, 20 the cohort where th 13). e vast majority of the retirement readiness shortfall occurs, and the soonest. When What we do retirement account (IRA) bal findings and policy andevelo ces at reti pments; rement age (V timely factsheet anDerhei and C s on hot topics; opeland, 20 regula 04). r updates on legislative and about their health care. Yet the ability of people to make informed deci th sions is highly dependent on the extent to they were in excellent or very good health and that they exercise at least four days per week (Figure 6). Individuals are presented deficit based ounder n deterministic and three separate thresho stochastic costs alone. By the lds of deterministic expense (80, 10 year in retirem 90 and 100 perc ent, 64 percent ent of of average lowest-income $150,000 or more 27 16 employee benefits issues; holding educational briefings for EBRI members, congressional and 11 percent o Rese on Ffiscal R arch Institute, June 2 those in esp the onsi second-income bility and R 008). eform Recom quartile, m 3 end peations.” rcent ofE those BRI Notes, in the thir no. 7 (Emplo d-income quartile, yee Benefit and 1 Researc percent h Institute, of such ex ? pe The August 20 nses are covere 11 EBRI Notes d by Medic article (Van aid). This version o Derhei, Aug f the mo ust 20del 11) use is co dnstructed to RSPM to anal simulate r yze the impact of defi etirement inned b come enefit regulatory developments; comprehensive reference resources on benefit programs and workforce nursing home and home health-care expenses are factored in, the number of households in the lowest- which they have access to useful information. with an HSA ? Preretir we ement re also -income quarti less likely than le. those with an HRA to be obese, though they were more likely to smoke Haviland, Amelia M., Neeraj Sood, Roland McDevitt, and M. Susan Marquis. "The Effects of Consumer-Directed Health Dec federal agen lined to ans cy w s etraff, and the news media; 3 and sponsoring public opinion survey 6 s on employee expenses), as well as with and without nursing home and home health-care expenses. quartil ? e ho Additi use ona holds l refin woul ements d run short o were introduce f money, but on d to evalualy 11 te the im percent pact of purchasi of those in ng lo the s ng-term c econd-income are insu quart rance on retirem ile would face ent those in the highest- July 2011): 2– in6. come quartile. HRAs and HSA plans in ac s are very issue hievs ing retir ; and simimajor survey e lar, thou ment income a gh ths ere ar d oe f qua pu e some blic cy for Bab attitudes. key di y Bo ffer omers an ences that may d Gen Xers. provide different incentives related adequacy, as noted above. Alternative versions of the model allow similar analysis for replacement rates, standard-of- Results income quartile that is projected to run short of money within 20 years of retirement is considerably larger Source: EBRI/Greenwald & Associates Consumer Engagement in Health Care . “ERISA At 30: The Decline of Privat benefit issues. e-Sector EBRI’s Ed Defined Ben ucation and Re efit Promse ises an arch F d Ann und (EBRI-ERF) performs uity Payments: What Will It Mean the charitable, ?” cigarettes. References Plans o n Episodes of Health Care." Forum for Health Economics and Policy 14, no. 2 (September 2011): 1-27. a similar situ incom at eion. Only adequacy 3 (V pe anD rcent o erhei, 2005). f those in th e third-income quartile and 1 percent of those in the highest-income EBRI meetings present and explore issues with thought leaders from all sectors. to using health care services, and t Surveyhus, , 2013.different consumer engagement experiences. An HSA is owned by the living calculations, and other ad-hoc thresholds. The 2013 CEHCS asked if an individu educa al tri tiona ed to l, andfin scientif d the c ic func ost of h tions of the Instit ealth care services ute. EBRI before gett -ERF is a tax-ex ing cempt organizatio are. Individuals n ? The than those in percentage of the oth avera er thr ge e deterministic e income quarti costs assumed paid les combi Figure ned. 8 by the household. One of the primary outputs of RSPM is the production of Figure RRRs 4 for various subgroups of the population. The RRR is . “Retirement Income A EBRI Issue Brief, no. 269 (E dequacmplo y: Alter yee Be native nef Tit Resear hresho th Figure Figure lds an ch Institute, Ma 6 d the Importanc 4 y 200 e of Future Eli 4). gibility in Defined Quan ?tif In Septemb ying Retir ee r, it ment was us Income ed to supp Adeq ort testimon uacy y bef Figure ore the Senate F 2 inance Committee (VanDerhei, September Figure 2 Copeland, Craig, and Jack VanDerhei. “The Declining Role of Private Defined Benefit Pension Plans: Who Is Affected, and 80 quartil Percen e are si t o mulated to r f De EBRI terministic regula un short o rlyExpenses, provides f money by co 100 ngre the Percen ssion 10al year. t testimony of Stocha , sti anc d Health briefs po Co licy stm s ake rs, member organizations, Taylor, Humphrey. "Does Internet Rese supported b arch ‘W y ork’? Comp contributions a and g ring Onl rants. ine Survey Results With Telephone Surveys." individual with the high-deductible health plan and is completely portable. There is no annual use-it-or-lose-it rule with an HSA ? The model were mor wa Years s used to eva e likel in y than t Retirement lua hte the impact o ose with Before an HRA f defined Baby to report Boomers ben that efit freezes on p th and ey tried to Gen a Xers rticip find cost ants b Run yinform s Short imulati ation. Forty of ng the minimum -four Years in Retirement Before Baby Boomers and Gen Xers Informed Years Indiv Decision iin dual Retirement Participates Making Before for in Health W Baby ellness Plan Boomers Program Choice, and by Offered Gen Type Xers of by Health defined as Contri the perc bution R enta etireme ge of s Years nt Plan imulate s.” in d l Retirement EBRI Notes, ife-paths that no. 4 (Emplo Before do notBoomers run short yee Ben of money e and fit Res Gen earch in Xers re Institute, tirement. A April 20 previous EBRI 11): 10–19. Several di 201 ffer 1) in a ent ap nalproach yzing the p es have otenti be al imp en used to act of various t quantify ypes of ta and model r x-reform o etirement ptions on retirement income. T income adequacy. Some heith is wa er s (1) ® How.” In Rob and the medi ert L. Clark and a on employe Olivia Mitch r benefits. ell, eds., Reorienting Retirement Risk Management. New York: Oxford Figure 5 provides the results of a simulation where only 80 percent of the average deterministic expenses are paid by Appendix . “Can America Afford Tomorrow's Retirees: Results From the EBRI-ERF Retirement Security Projection Model. ” The basel Internatio ine version of nal Journ the a l of Mark model used et Researc for this an h 42alysis , no. 1 (August 200 assumes all 3). workers re tire at age 65, that they immediately ? Whether or not nursin Money g hom , by e an Percentage d home health-c of Deterministic are expenses are Retirement included iExpenses n the simulation. associated ? Extending the with an HSA, as results to a any money maximum o left in th f e 35 accou years F nit at t gur in re eh tire 6 e end ment (ag of the year e 100, au assuming tomatically roll retirement s over an at agd is e 65 ), 83 employer-contribution rate tha Run Short t would b of Money e needed to fin , by Preretirement ancially indemnif Income y the empl Quartile oyees for the reduction in their percent of individuals with Plan, aRun n Employ HSA r am Short on eport er g , Indiv of a em dMoney that on iduals g they Those , by tried to With a Choice Preretirement Offered find the a Wof cost of he ellness Income Health Program, alth Plan, Quartile th care 2013 services before getting care, Run Short of Money, by Preretirement Income Quartile publication used the 2014 version of RSPM to show that the RRRs for Baby Boomer and Gen Xer households ranged expanded in the November 2011 EBRI Issue Brief (VanDerhei, November 2011). Even ignorin model only th g e accum the cost o EBRI ulatio issu f nes u n side rsing hom press of the rele e e or as q e home h uation s on ne an ealth wsworthy d th -c en are expenses, rely on developm some type by the 20 ents, of and i pr y eretirem s e ar in among retirem ent i the mo nce ome re nt, more t st widely qu place hm anent oted two- University Press for the Pension Research Council, 2010: 122–136. the ho . “T usehol EBRI Issue Brief, he Impact of Modif d, but 100 perc no. 263 (E yent of ing the E EBRI Issue B the mplo x stoc clus yeion of Em e Be hastic hea riefs nefit Researc are period plo lth c yeeare Contri icals providing exp h costs are Institute, Novemb butions for Retirement met. In t ert evaluati er 200 his ca 3). Savings Pl ons of emplo se, a comparison wit ans yee benefit issues and From Th a xabl Figure 1 e begin drawing benefits from Social (No nursing home Security and defin or home health ed benefit pla care costs ns (if anyin each case): ), and, to the extent that the sum of their Scenario: Healt 90% h Stof atu deterministic s and Health retirement y Behavioexpenses rs, percent of the lowest-incom Scenario: e quartile 80% by hou Ty of pe s deterministic eholds of Health 8would run Plan, retirement short of 2013 expenses money and almost half (47 percent) of available for expecte future d retire use. Bot ment incom h individuals a e under var nd ious ra employers te-of-return ass are allowed to umptio contribute to ns (VanDerhei, March an HSA. 200 Distributions 6). from an compare About the d wit 20 h13 37EBRI/Gr percent a em en Scenario: ong i wald ndivi & As du 100% als sociates with of an deterministic Co HR nsumer A (Figure retirement Eng 3). S agemen imilarly, i expenses t in ndivi Health duals Car withe an Sur HSA wer vey e more sources on employee benefits by all media. between 56.7 and 58.5 percent, depending on age cohort. When the results were analyzed by preretirement wage rate measure as a threshold for success, or (2) make use of a so-called “life-cycle” model that attempts to thirds (69 percent) of the lowest-income trends, as well quartile as cr househ itical analy olds wo ses of uld ru emplon short o yee benefit po f mo licies an ney. On d proposals. the other h EBand, RI Notes onlyis a (100 However, Our perc Income: Resu ent even when of deterministic and stoc 100 lts Frpercent om the 201 of av 1 Retirem hastic costs) er Low age est det ent C preretirement ermini shows the im onfid stic enc costs are e Surv income quartile pa ect of a y.” paid by EBRI Notes, ssu the ming ho a no. 3 (Emplo usehold an 20 percentd n red yee Ben ursing uctio en i fit Re home n th search e and Endnotes ? A March 201 90% 2 EBRI Notes article (Va but no nursing home nDerhei, March 201 or home 2) used n health care ew survecosts y results to update the analysis of the expenses and uninsured medical ex but no penses nursing home exc by T eed th ype o e pr f Healt or home ojected, a h Plan health care fter , 2013 -tax annual costs income from those sources, VanDerhei, Jack. “What Causes EBRI Retir but no nursing home ement Readineor home ss Ratings to V health care ary: Res costs ults from the 2014 Retirement Security HSA can those in be made the at any second-income time. An individu quartile al ne wou ed not ld face a be covere similad r situation. by a high On -deduct ly 28ibl percent e health of those plan to in t with he dra third- w money The Employee Benefit Research Institute (EBRI) and Greenwald & Associates created the EBRI/Greenwald & likely tha . “Kansas Future Retire n tho se with an HRA EBRI directs ment Income A to report th members aat they nd ssessment Pro other constit found the i ject.” A project of uen nformatio cies to n th the informatio the EBRI Educatio ey were looking for. Nea n they need an n and Re rly d search F 80 unde 53% percent rtakes und of and new ® monthly periodical providing current information on a variety of employee benefit topics. quartile, the RRR varied from 16.8 percent for the lowest income quartile to 86.4 percent for the highest income 17 smooth/sprea perc ? ent o Later that 80 f 80 d those in % some ty % year, Visite the an upd d pe heal second-income o th f p consum lated ans' web ve sitrsion of the mo eption s to lear quart -n base about ile d thut eir wou del ility plans w ld a over the f s dev ace elo a simila decisi ped to e r situation, on-maker’s nhance the EBRI inte as lifetime. would 5 perce ractive Bal nt of those i lpark E$timate n the 1 60% ® 83%*** home heInstitute, March 201 alth-care expenses 1): 2–1 are inc 0. luded (Figure 1), only the households in the lowest-income quartile eventually end deterministic c potentia osts. 35 l imp % act of various types of tax-reform options on retirement income. 45% immediately begin to withdraw money from their individual accounts (defined contribution and cash balance plans, as More informati Projecti on ab on Mod out the d e 81%*** l, ” EBRI Issue Brief, ata can be found in the no. 396app , Februar endix a y nd i 201n HS 4. FA ronstin, 2013. HRA income resea quartilre ch an on a d 13n perc ongoi ent of ng ba thos sis. e in the highest income quartile are simulated to run short of money from the HSA. This CHECK OUT EBRI’S WEB SITE! means that indivi EBRIef duals is a weekly who do not roundup of use all t EBRI resear he money in ch their and insights, as HSA during thei well as updates on survey r working years can s, indivi Associates Consumer En duals the Milb with an ank Me HSA morial F foun gagement d th und. Jul e in in H format ye 16, 20 alth Care ion they 02. Surv wee re y seeking, (CEHCS) to compared with examine issue 67s percent surroundi th among indivi ng consumer-dire duals wit cted h publications quartile. by providing Monte Carlo simulations of the replacement rates needed for specific probabilities of retirement-income third-income quartile and 1 percent of those in the highe HS st-income A HRA quartile. By the 35 year in retirement, 70 percent 80% up with a majority of the households r Health Stat unni us ng short of money during retirement. studies, litigation, legislation and regulation affecting employee benefit plans, while EBRI’s well as IRAs). If th EBRI ere maintains an is sufficient mo d ney t analy oz pay ®es the mo expensst escomp without rehen tapping sive databa into the ta se of x-quali 401(k) fie-d i type ndi vidual program account s in the s, 2 eventually. use it to pay out-of-pocket expenses when they are retired. health . “A care, Post-Crisis Assessment of Reti including the cost of insuranc ree ment Income A , the cost of care dequac , satisfacti y for Baon wit by Boomers a h healthn care d Gen Xers.” , satisfact EBRI Issue Brief, ion with a health no. an HRA. EBRI’s Retirement S ? T he May 201e2 curity EBRI Notes Projection Mo article (Va delnDer tak he es a i, Ma differ y 2012) pr ent—an ovid d ed 2 argua 012 bly u updates for the nique—perspprevi ective, dra ously pw un bl fro ishe m d RRRs the 70 70 % % 49% T he specific q .”The Rouestio le of Soci n wa al S s as ecurit follo yw , D s: Do efines yo ed Benefit ur em s, an ployd er Private Retire offer any of the ment Accounts following we ll in the F ness progr ace of the Retirem ams? ent adequacy under alternative-ri sk-management treatments (VanDerhei, September 2006). Although 4 percent or of the lowest-income quartil less oe house f those in the holds would ru second-,n sho third-, rt and hi of money a ghestn -id ncome 20 percent quartil of those es are sim in t ulat he second-incom ed to run short of e . “Massachus Atten etts F ded a u m ture eeting Retirem whe Ex Blog rc ee yllent our supplements our regular pub ent Income Assess choices were explained ment Project.” lications, offering commentary 21% A project of the 13% EBRI Educatio on questions received from 72% n and Research world. Its computer simulation analyses on Social Security reform and retirement income adequacy 30% those balances are assumed to be in 3vested in a non-tax-advantaged account where the investmen 39% t income is taxed 50% This Notes 354 (Emp article lo provides yee Benefit Rese new results sho arch Institute, Febru wing how ma any years i ry 2011). nto retirement Baby Boomer and Gen Xer care plan, as rea well sons for choos as the RSS. ing a plan, and sources of health information. The 2013 CEHCS is comparable with objective of its original applications. Rather than focusing on an individual’s projected ability to achieve a designated Crisis,” 70% EBRI Notes, no. 1 (Employee Benefit Research Institute, January 2014): 8–21. Very news reporters, policy good makers, and others. EBRI’s 50 43 Fundamentals of Employee Benefit Figure EBRI’s website is easy to use and packe 7 summarizes the results for the households in the lo d with useful information! Look for west-income quartile from Figures 1, 3 and 5. The results money in quartile woul F the first year o und an d face d the a simi Milba f ret lar nk Memo situation. Only irement rial F (Figur und e 5), 6 . Dece perc 34 percent ent o mber 1, 200 f those in of2. those i the thi n the lo rd-income west-iqua ncome quartile rtile and 1 per would hav cent of those e a defic in it are unique. 66% ? Health-risk assessment, where you answer a questionnaire and then a medical professional examines your health In contras ? RSPM t, an HRA was sig isn an ifica en m tlployer- y enhanced for the M funded healtha pla y 200 n t8 EBRI po hat reimburs licy for es employe um by allo es w f in og r qua autolifi matic enro ed medical llme e nx t of 401(k) penses. as ordinary Participation income. In in Wellness dividual acco Programs unts a re tracked until the point at which they are depleted. At that point, any net 63%*** findings from households ar 60 60 te simulate % he 200 % 5–20 d to 07 run EBRI/Commonwealt short of Programs money, by offers a straigh h Fun pred ret tConsumerism forward, basic explan irement income quartile. in Healt ation of h Car Figures e emplo Surveys a yee ben 1, 3 a ned t nfit progr d h5 e incl 200 ams in the u 8de -20 nursing 12 percentage of his or her preretirement Good income as a proxy for retirem 25ent income adequacy, RSPM 35 grew out of a multi- 44%* Consumer Engagement Among HSA and HRA Enrollees: Findings from the 2013 based on the h show that igh . T est-in edete eve stimon come quartile n rministic an if y. U.S. Cong this cohort d s are simulat wer t ress. S ochastic costs in t e a enate H ble to ed to run sustain a l ealth, Educ h short of e first ye ife atio st money yle assumin n, Labor ar of retirem event and P g eu a 20 percent r nt (assumin ensi ally. ons Comm g e re ittee. ductio tirement Tn i he W n at av obbly Stool: a erage determi ge 65). By the nistic ? The June 2012 EBRI Notes article (VanDerhei, June 2012) introduced severity categories in the RSS projections for these special features: Talked to friends, family, colleagues about the plans . “Ho histor w Much W y to iden oul tify d it T any co ake? Achievi nditions you ma ng Retir y ha ement ve or that Income Equ you might be at risk ivalency betw of develo een Final-Av ping. erag e-Pay Defined Benefit HRAs are ty partici pic pants ally set with up as the pote notional ntial for autom arrangements. L atic escalati eftover on of co funds at ntributi the end of eac ons to be include h year d (Va can nDer be hei a carried over nd Copelanfo d, r housing equity is assumed to 25% be adde privat d to e and re pu tirement blic sectors saving . Th s ie n tEBRI Datab he form of a l ook on Emplo ump-sum dyee Benef istributionits (nis o a st t a reatist verse ical Employers an . “Oregon Future Retire d insurers offer a n ment Income A umber of different types of ssessment Project.” A project of wellness benefits—p the EBRI Educatio rograms d 32% en an signed d Re to promote health search Fund and Fair or Poor 3 10 home and 60ho % me health-care expenses in the simulation, whereas Figures 2, 4 and 6 provide results if the impact of CEHCS. year project to analyze the future economic well-being of the retired population at the state level, focused on EBRI/Greenwald & Associate costs, the per EBRI make Retirecentage of 40 me %s information freel nt (In)security in A households t merica hat y av would ru (T s Consumer -166), 7 Oct. 2010b. ailable to al n shorLo t of mo wes l. Engagement in Health Care Survey, t- Inconey me Qby the uartile 35th year of retirement would only By 10th year Gen Xers. in retirement, 47 percent of lowest-income quartile households would run short of money but only 11 Plan Accruals and Voluntar referen y Enrollm ce work o ent 4n 0 emplo 1(k) Plans in the yee benefit pr Private Sector ograms and work force-related issues. .” EBRI Notes, no. 12 (Employee Benefit 2008). Lowest-Income Quartile future use (at the employer’s discretio Smokes Cigaret n), allowin tes g employees to accumulate funds over time. In principle, at least, annuity mort and to prthe Milb even gage t disease. ank Me (RAMmorial F Th )). If e 20 all t u 13 nd. Septemb h CEHCS e retirement savin examin er 20ed avai 01. g s are e lability xhauste and pa d and if rticipation the Soc in tih al Se ree ty curity a pes ofn welln d defiess pro ned ben greams: fit 50 50 % % these potentia • ? EBRI’s entire lib Programs for i lly catastroph mprovin rary of research publicat ic ex g your penses hea is i lth, like for w gnored. eight loss, ions starts at the m walking or other e ain W xercise, nutriti eb page. Click on on, stress mana EBRI gement, identifying the point at which individuals would run short of money and perhaps become a financial obligation of the 90 Percent EBRI of Deassume terministic s a publi Expenses, c service re 100 spo Percen nLo siwes bility t t-Inc of o to m Stocha emake Quartile sti its finding c Health s co Co mpletely sts accessible at www.ebri.org Second Paul Fronstin, Ph.D., EBRI 41%*** perce decreas nt of e from thos 83 e in t peh rcent e second-i to 64 p ncome ercent. qu F artile igurewoul 8 summa d face a simi rizes the r lar sit esul utation. On s for the ho ly 5useholds percent o in t f th he ose in t lowest-incom he third- e Second Research Institute, Decemb Yeser 2013): 11–23. 18 9 Used other websites to learn about your choices 48% this provid 50 es a % n incentive for individuals to make health care purchases responsibly. However, an employer is not payments ar The 20 . “Retirement Sa 13 surv e not suff ey was 20% co icient vinnd gs Shortfalls for T u tco ted pay withi expense n thoda e Unite s, the i y’s W dn o State divi rker du s.” s betw al is EBRI Notes, designated as een Augus no. 10 (Emplo t 8 an having ru d Augus ye n short o e Ben t 20, e 201 fit Re f mo 3, throu search Institute ney at t gh a hat 13poi - , nt. a healt ? hT -risk he August 20 assessment, a 12 EBRI Notes health-promot article (Van ion pro De grhei, Aug ram thatu inc st 20 luded a 12) prov num ideber o d additi f dif ona ferent l evid types o ence on f bene whethfits, an er deferrin d a g 32% smoking cessation, and so on. ? Issue Briefs Additi — ona so that all deci l modand ificatio EBRI Notes ns sw io ens re add that ed for a Pe for our in-depth and relate to emp nsion Rese loyee Third arc ben hnonpartisan periodicals. Cou efits, ncil pr wheth eser entatio made n that in in Cong volve ress d a “ or bo winne ard rs/losers” rooms or state. EBRI a Figure 3 provi nd d th es th e Mil e resu bank lts of a Memorial simulati Funon w d, worki heren only g wit Second 90 h th percent e office of the gover of the averagne or deterministic of Oregon, set expenses ar out in the lat e pai ed by . “A Behavioral Model for Predicting Employee Contributions to 401(k) Plans.” North American Actuarial Journal Third quartile from Figures 2, 4 and 6. No The Contact EBRI results show th Publications, (2 at if this co 02) 659-0670; hort 82 were abfax publication le to su 91 stain a orders to 20 percent r (202) 775-6312. eduction in income quartile and 22 percent of those in the highest-income quartile are simulated to run short of money by the 10th First-Y required to m ear October 201 Shortfalls 40 40 ake the % % 0a un ): 2 use -9. d balance available to a worker when he or she leaves. minute biometric Intern retireme screenin et survey. 30nt to age 70 % g. Thew nationa ould provid l or ba e retireme se sample nt income a was dra dw 4 equac n from I y for the v psos’s onlin 9 ast majorit e panel y of Bab of Inter y Boom net users ers and Gen who have families’ homes, are based on the highest quality, most dependable information. EBRI’s Web site posts . “The Impact of a Retirement Savings Account Cap,” HiEBRI Issue Brief, ghest-Income Quartile no. 389, (Employee Benefit Research analysis of defined benefit freezes and the enha Th nce ird d employer contributions provided to defined contribution plans 1990s to see if this situation could be Subscriptions to evaluated for EBRI the Issue Bri state.e The fs are included resulting ana as part of lysis focuse EBRI membership, or as part of d primarily on simulated a the househol (Novemb d, but er 20 100 01). percent of Exerci the ses stochastic health c Hare costs are ighest-Income Qua rmet. tile In this case, a comparison with Figure 1 ? Health reimbursement arrangements (HRAs) and health savings accounts (HSAs) are very similar, though year. average ? deter Biometric scre 40% ministic costs, and if enings, which are meas the financial 3 urem implic ents or blo ations od from nurs work to determin ing home and home e your health s 39%** he tatus inclu alth-care dcosts ing blo are od • Orders/ Visit EBRI’s blog, or subscribe to the EBRIef e-letter. Consulted with your employer's HR staff about you choices agreed to Figure 1 sho Xer part sall .w s the r icipate in reseae rsults a ch research surveys. finding ssuming that s, publications, 100 Two t percent an hous d n a of e nd ws the adu alert simulate ltss ag . EBRI also exte es 2 d det 1 -64 erministic who h nds a d h expenses are 5 its e ealth du insur cation ance t met; in and pu h other rough blic an service 15% Institute, August 2013). Never $199 annual subscription to EBRI Not 1e 2s and EBRI Issue 20 Briefs. Change of Address: EBRI, at the time the defined benefit plans were frozen (Co Highestpel -Incoan med and Va Quartile nDerhei,26% 2010). (100 retirement wealth perc . “Retirement Income A ent of deterministic and stoc with a comparison to dequacy for T hastic costs) ad oda hoc y’s W thresho shows the im orkers: lds for How retirem Certa pact of a in, Ho ent e ssu w x Muc penditur ming h W a ill es. 1 It Cost, and H 0 p ercent redo uw ct iDo on i es Elig n the ibility there are some key differences that may produce different incentives related to using health care services, ignored, t pressur he percenta e, cholge o esterol, f how useholds eight, heig wh ho are t, etc. projected to run short of money in retirement would decrease Ultimately, an HSA creates a stronger financial incentive than an HRA for workers to be more engaged in their health 30% The survey fo . “The C role 30 und that han %to improving Ameri ging F indivi ace of Private Retir duals with can an s’ HSA wer efinan ment Pla cial e mor n kn s.” owle EBRI Issue Brief, e lik dge ely tha thro nugh individuals with no. 232 (E its award-mplo win an HRA to ning yee Be publi nef report that th c it Researc service h cam Institute, ey h paign ad 1100 13th St. NW, Suite 878, Washington, DC, 20005-4051, (202) 659-0670; fax number, words, 100 percent of the average ex 1 day per penses (based on week, on averagepost-retireme 16 nt income) for 16 components likely to be employer or purchased directly from a carrier were drawn randomly from the Ipsos sample for this base sample. This • Subscriptions EBRI’s reliable health and retirement surveys are just a click away through the topic boxes at for Participation in a Defi ® ned Contribution Plan Help?” EBRI Notes, no. 9 (Employee Benefit Research Institute, By the 20th year in deterministic c ? The Septembe 30% osts. retirem r 2012 ent, 59 EBRI Notes percent article (Va of the lowest nDerhei, Se -income ptem quarti ber 20 le12) house anah ly olds zed the imp would hav act of increasi e run short o ng the f money and thus, d 20% ifferent consumer engagement experiences. . “Reality Checks: A Comparative Analysis of Future Benefits from Private-Sector, Voluntary-Enrollment 401(k) Plans 3 substantially: ? Also in 2 from 70 009, a ne percen w subr t to 3 outi(202) 775-6312 3 n e pe wrcen as ad t by t ded to allo ; e-mail: he 35th w sim year of retir subs ula criptions@ebr tions of various st ement. i.org yles of target-d Membership Information: ate funds for a Inquiries care becauseC thhoosetoSave e account is owne and d the by th companio e worker a n site nd co ww mpletely w.choo porta setosav ble e.o upo rgn job c 31%*** hange. Haviland, et al. (2011) participated in April 2 a h 001). ealth -risk assessment, health-promoti 2–3 days per week, on average on program, o 34 r biometric s 35creening program when it was s encount See ample ww was ered o w.glob s Co tratifie n ns a a r ult loed pe in w d gular i ith on by an pa gen ins ba nel uran sis (e.g., d s.co c er, a e brm/h okg ere, o to foo re m une der gi ds on, income, , housin tand your cg, tran hoices and sprac ortation). e. The In addition to response rate was these relat 37.2 perc ively ent (32 percent predictable for With the assistance o 10% f the Kansas Insurance Department, the Employee Benefit Research Institute (EBRI) was able to the top of the page. September 2010): 13–20. 11% but only 25 defau percent o lt-contribu f those tion rate for a in the se utomatic enr cond-income ollment 401(k) quartile wo puld fac lans with a e a simil utomar situat atic escal io an, as wo tion of contributi uld 12 ons perc . ent o f regarding EBRI membership and/or contributions to EBRI-ERF should be directed to EBRI vs. Stylized, Final-Average-Pay Defined Benefit and Cash Balance Plans.” EBRI Issue Brief, no. 387 (Employee comparis 20% on with participant-d 4–5 idays rected in per w veek estments (VanD , on average erhei, June 2 25 009). 17 and citations within 20% it found that HSA plans have a greater effect than HRA plans on the use of health care services TM available. Four-fifths (81 percent) of individuals with an HSA participated in a health-risk assessment, compared with ? Adults with an HSA were more likely than those with an HRA to exhibit a number of cost-conscious behaviors expenses, the stochastic costs arising from nursing home and 10home 0% health-care expenses are assumed to be covered the 4 base sample or national sample, and 44 percent for the oversample). As a nonprobability sample, traditional Although 5 percent or create the E 20%BRI Retireme less o nt Rea f those in the diness Rating second-, (RRR thir ) based on d-, and hi a ghest full stochastic dec -income quartil ues are sim mulation mo ulat del ed to that tak run shor es into t of VanDerhei, Jack, and Lori Lucas. “T President Dallas he Impact of Auto-enrol Salisbury at th lment an e above d Auto address, ( matic Contri 202) 659-0670; bution Esca e-mail: latio salisbur n on Retir y@ebri.org ement Each of the six analyses with results presented in Figures 1 through 6 show the same stark conclusion: The lowest those in In theor the y, a ra third-i ndom sam ncomep qu le of 2,000 artile More anyi d 5 than elds pe 5 a statistical pre days rcent o , on faverage those cin th ision e hi of plus o gh 13est-income r minus 2.2 per quarti 12 centag le. By t e po he 35t ints ( h year i with 95 p n ercent Benefit Research Institute, June 2013a). http://www.ebri.org/publications/ib/index.cfm?fa=ibDisp&content_id=5229 EBRI is supported by organizations from all industries and sectors that appreciate the value of and t • h . “T e cost Need a number? Check out the he Impact of Automatic Enrol per episode. lment in 401( EBRI Databook on Employee Benefits. k) Plans on Future Retirement Accumulati ons: A Simulation Study 66 perc ? ent o The November f individuals 2012 witEBRI Notes h an HRA (Figure article (Van 4). Nearly Derhei, Novem two-thirds ber 20 (63 perc 12) recl ent a) ssified the R of 28%*** individuals RR w s to provid ith HSan HSA A e additional 90% related to 10% use of health care services. Those with an HSA were more likely than those with an HRA to report survey-margi in years ? In April 20 when n-the mo of-error 10, the mod Asked del hea estimates do simulates th lth ple ans l w toa ss end compl not a yo eir u info ex ep rtel m istence. ply. How ati yon re- in param the mai ever, l eterize had t d h we surv ith 401(k)-pl ey used a an desi probability sam gn parameters for spo ple, the margi nsors that ha n of d account the household’s longevity risk, post-retirement investment risk, and exposure to long-term nursing-home and money in Income Ad the first year o 5% equacy f ret .” EBRI Issue Brief, irement (Figure 1), no. 34 40 percent 9 (Employe ofe B those i enefit Res n the lo earc west-i h Institute, November 2 ncome quartile would hav 010); and DCIIA e a deficit BMI retireme confid preretir enc ement nt, 6 e) of w 4 perc income hat the resu ent o quarti f th lts e low lew is th oul ed be st-inco e coho if th m rt wh e entire p e quartile ere th o ho p eu vast la useholds tion ag majority of es 21 woul 10%–6 d hav 4 the shortfa with e priv run sh ate h ll o ort of mo ccurs, and the soonest. ealth insura ney nce and coverag 35 percen eWhen wa t of s 10% unbiased, reliable information on emplo 10% yee benefits. Visit www.ebri.org/about/join/ for more. Based on Plan Design Modifications of Large Plan Sponsors.” EBRI Issue Brief, no. 341 (Employee Benefit Editorial Bo informati ard:on on Dallas L those subst . Salisbury, publisher antially a ; Stephen Blakely bove the thresh , editor old; clos . Any views e to the threshol expressed in this p d; and su ublication and th bstantiall ose o y be f th loe author w the threshol s should d. participated in a health-promotion program, compared thwith 48 80 perce % nt of individuals with an HRA. Over four-fifths . “W that they 10 hat % a Sustain asked for ed L a owgeneric -yield Rate E drug i nn viro stead nment Me of a bran ans for Retire d name; check ment Income A ed the price deq of uac a s y: Res ervice u HRA lts F before rom the 2013 getting adopted automatic-enrollment provisions (VanDerhei, April 2010). error for the national sample woul Underw d have eight been ±2.2 percent. 4 3 home h based on eRese alth dete -c arch Re rministic an are risks. The port (Novemb d sfirst state-l tochastic costs. By th er 201 e0). vel RSPM result e 10 y s wer ear in e retir present eme ed to nt, 6 the Kans 3 percenas Long-Term t of lowest-inc ome quartile Care Services • Instantly get e-mail noti?cations of the latest EBRI data, surveys, publications, and meetings surve nursing yedhome with and home hea complete accuralth-care cy. There are expenalso oth ses are factor er possed ible s in ources of (Figures 1, 3 error in a and 5), ll surv the num eys that may ber of h be more serious ouseholds in the This artic those in not be the ascribed to the officers, le ex second-income amines how c quart o trust nsumer e ees, ile wou menm ga blers, or dgement face a other sponsors of the E simila variesr si for tuation. individua m On ploye ls enroll ce again, t e Benefit Research ed in he Hupper-i SA-eli Institute, the EBRI Educ gn ib come le plaquartil ns and HRAs. es are l ation and eThe ss ® Research Institute, April 2010). http://www.ebri.org/publications/ib/index.cfm?fa=ibDisp&content_id=4495 Note that in Figure 1, while 5 percent or less of those in the second-, third-, and highest-income quartiles would run (83 percent) of individuals with an HSA participated in biometric screening, compared with 72 percent of individuals EBRI Retirement Security Norm Proj al ection Model. ” EBRI Notes, no. 3 (Emplo 36 19%*** yee34 Benefit Research Institute, June care; asked a doctor to recommend less costly prescriptions; developed a budget to manage health care Task Forc Research Fund, e on Talke July d or to their your 11, d staffs. oc2002, and the resul tor or Nothin other heal g her th pe rin is to be co ofessional ts of a about nstr the separat ued as an attem plans e study pt to aid or for th hi e state o nder the adoption f Massachusetts were pr of any pending legislation, esented o regulation, n households ? The March 201 would r 0% un short of mo 3 EBRI Notes ney, article (Va but only nDer 15hei a perc nd Ad ent of ams those , March 2013) in the second-in used a mo come quartile dified versio would fac n of RSPM to e a similar and seminars by clicking on the “Notify Me” or “RSS” buttons at the top of our home page. than theoretica 0% l calculations of sampling error. These include refusals to be interviewed and other forms of nonresponse, the affect lowest-i ?ed: A complete ncome Only 19 0% quart 0% percent ly ile update that of di those i versi s projected to onn of the nation the third-income run short al model of m quartile w oa ne s y w pro and 8 duce ithin percent 2 d for the Ma 0 years o of t fy r h 2 ose in eti 01 rement 0 EBR the hi I polic is cons ghey forum a st-income iderably l nd use arger quartile d in HSA sample includes only individuals who have an HSA. Data from 5% the 2013 EBRI/Greenwald & Associates Consumer 0% 0 1 2 3 4 Overw 5 6 7 eight 8 9 10 11 12 13 14 15 16 17 18 33 19 20 21 22 23 2431 25 26 27 28 29 30 31 32 33 34 35 The short of mon sample w eas divided into y in the first ye three ar of retir grou eps: me those nt, more with tha a nconsumer-driv 2 in 5 (43 percent en he ) of alth tho plan se in t (CDHP), he lowe thos st-income e with a hi qua gh rtile - with an or interpretative HRA. 2013b): 2– ru01234 le, or as 12. legal, acco 56789 unting, actuarial, o 1011 r1 other such prof 213141516 essional advice. 17181920212223242526272829303132333435 expenses; an 00 d used an 11 22 33 44 onli 55 66n7e 7 cost-trackin 88 99110 0111 1 112 21 g tool 13 3 114 41 provi 15 5 116 61d 17 71 ed 18 8 by the h 119 9220 0 221 12e 22 2alt 223 32 h pl 24 4 2an. A 25 5226 6227 7 dults 228 8229 9with 330 0331 1an HSA 332 2333 3 334 43 were also 35 5 . “Falling Stocks: What Will Happen to Retirees' Incomes? The Worker 1 Perspective,” Presentation for The Economic Dec. situation. Only 1, 2 assess the pro 002. There 6 perc after, ent o babilit RS f t y that res PM was hose in the pon expan de thi nt hous d rd-i ed to a ncome ehol natio ds qua wnrtile a oal u model, ld nn od 2 t run short of mone wit perhc the ent of init those ial r y i e i n sults present n retirem the hient if the ghest ed at -iny come d the id, i EBRI n quartil fact, -ERF e Health-risk assessment Health-promotion program Biometric screening effects of question wording and question order, and screening. While attempts are made to minimize these factors, it is than those in the Jul tyh 20 e oth 10 e EBRI Issue Brief r three income quarti (VanDer les combi hei and C ned. ope Indeed, land, 201 as the results across 0). multiple scenarios and are simulat Engagement ed in to run Healt short of h Care Survey money (CEHC eventS u) ar ally. e used for the analysis. Factors a th ffecting use of health care services, Obese 19 24 would, based on deterministic and stochastic costs. More There’s lo 0%over, by t ts more! 10% he 1 0 20% year in 30 reti % rement 40% (assu50 ming r % etirem 60% ent at deductible health plan (HDHP), a 6 nd those with tr Year ths in aditio Retirem na ent l he (Assal um th ing cRetir ove em ra ent ge at . In Age d 65) ividuals were assigned to the CDHP or Years in Retirement (Assuming retirement at age 65) more likely than those with an HRA to be engaged in their choice of health plan. Individuals with an HSA were Crisis of 2008: What Will Happen to Retir Year Year s ees in s Retir in ’ In Retir em comes? 2 em ent ent (As (A sum ssum ing 0ing 0 Retir 9r APPAM Fall Confer etir em em ent ent at at Age age 65) 65) ence (November 2009). Policy Forum i accumu nlate the Decemb amount the er 2003. y said would be required in the 2013 Retirement Confidence Survey. are simulated to run short of money by the 10 year. impossible to quantify the errors that may result from them. Don't know 7 9 EBRI Notes is registered in the U.S. Patent and Trademark Office. ISSN: 1085 ?4452 1085 ?4452/90 $ .50+.50 choice of health plan, use of cost informati on, and participation in wellness programs are examined. Source: Source: EBR EBR I/G I/G reenw reenw ald ald & & Associates Associates C C onsum onsum er er Engagement Engagement in in H H ealth ealth C C aa re re Surv Surv ey ey , , 2013. 2013. th HDHP g age 65), roup nearl ify 3 i they n had 4 (7 a 2 percent deductib ) of le o th f at e l lowest-i east $1, ncome 000 fo quart r individual co ile households verage or woul$2,000 d run short o for family c f money, w overage. hile To fewer be more likely than individuals with an HRA to report that they had participated in a health-risk assessment, Source: EBRI/Greenw ®® ald & 1100 13 Associates StC reet onsum NW · Suite 878 er Engagement in Health Care ®® Source: Source: Differenc Differenc EBR EBR e e betw betw I R I R eeen e tirement een tirement HSA HSA and Security and Security HRA HRA is Projection Projection is s statis tatis Visit EBRI on-line today: tic tic M a M a o llll odel del y y s signific ignific vv ersion ersion a an nt t at at 2030. 2030. *** *** p<0.01; p<0.01; ** ** p<0.05; p<0.05; * * p<0.10. p<0.10.www.ebri.org Source: Source: EBR EBR I R I R ee tirement tirement Security Security Projection Projection M M odel odel vv ersion ersion 2030. 2030. Survey, 2013. Washington, DC 20005 health-promotion program, or biometric screening program when it was available. ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org ebri.org No No No No No No No No No No No No No No No No Nottttttttttttttttte e e e e e e e e e e e e e e e es s s s s s s s s s s s s s s s s • June 2014 • Vo • June 2014 • Vo • June 2014 • Vo • June 2014 • Vo • June 2014 • Vo • June 2014 • Vo • June 2014 • Vo • June 2014 • Vo • June 2014 • Vo • June 2014 • Vo • June 2014 • Vo • June 2014 • Vo • June 2014 • Vo • June 2014 • Vo • June 2014 • Vo • June 2014 • Vo • June 2014 • Vol. 35, No. 6 l. 35, No. 6 l. 35, No. 6 l. 35, No. 6 l. 35, No. 6 l. 35, No. 6 l. 35, No. 6 l. 35, No. 6 l. 35, No. 6 l. 35, No. 6 l. 35, No. 6 l. 35, No. 6 l. 35, No. 6 l. 35, No. 6 l. 35, No. 6 l. 35, No. 6 l. 35, No. 6 15 14 13 20 10 16 19 18 12 17 23 24 7 4 2 9 3 (202) 659-0670 www.ebri.org © 2014, Employee Benefit Research Institute ?Education and Research Fund. All rights reserved. www.choosetosave.org A monthl y newsletter from the EBRI Education and Research Fund © 2014 Employee Benefit Research Institute ebri.org Notes • June 2014 • Vol. 35, No. 6 ebri.org Notes • June 2014 • Vol. 35, No. 6 ebri.org Notes • June 2014 • Vol. 35, No. 6 ebri.org Notes • June 2014 • Vol. 35, No. 6 ebri.org Notes • June 2014 • Vol. 35, No. 6 ebri.org Notes • June 2014 • Vol. 35, No. 6 ebri.org Notes • June 2014 • Vol. 35, No. 6 25 11 22 21 5 8 6 Cumulative Probability Cum Cum ulative ulative Pr Pr obabilit obabilit yy Cumulative Probability Cumulative Probability Cum Cum ulative ulative PrPr obabilit obabilit y y Cumulative Probability