“Characteristics of the Population With Consumer-Driven and High-Deductible Health Plans, 2005–2014,” and “Examining the New Income Measures in the Current Population Survey”

CDHPs: People of have so-called “consumer-driven” and high-deductible health plans tend to be older, more educated, and with higher incomes than those with traditional health plans, according to new research from EBRI. <A href="http://www.ebri.org/pdf/PR1126.CEHCS.27May15.pdf">Press release.</A> Income Measures: A new analysis by EBRI shows that the new questions being used to measure income in the Census Bureau’s Current Population Survey found a significant amount of income from IRAs and 401(k)-type plans was being missed across all levels of income for Americans age 65 or older. <A href="http://www.ebri.org/pdf/PR1127.CPS.28May15.pdf">Press release.</A>

Characteristics of the Population With Consumer-Driven and High-Deductible Health Plans, 2005–2014

The populations of adults within consumer-driven health plans (CDHPs), high-deductible health plans (HDHPs) and traditional health plans were each split about 50–50 between men and women in 2014.
CDHP enrollees were less likely than those with traditional coverage to be between the ages of 21 and 34 in 2014, and more likely to be ages 45?54.
CDHP enrollees were more likely than traditional-plan enrollees to be in households with $150,000 or more in income in every year except 2006, 2009 and 2010. They were also more likely to be in households with $100,000–$149,999 in income in most years. They were roughly twice as likely as individuals with traditional coverage to have college or postgraduate educations in nearly all years of the survey.

Examining the New Income Measure in the Current Population Survey

The U.S. Census Bureau’s Current Population Survey (CPS) is a primary source of income data for those whose ages are associated with being retired. In response to research showing that the survey has misclassified and under-reported certain types of income, the 2014 CPS included a redesigned set of questions aimed at better capturing income from individual retirement accounts (IRAs) and 401(k)-type plans, among other goals.
This article finds the new measures of income in the CPS identify significantly more income (and a much larger percentage of income) coming from IRAs and 401(k)-type plans. Compared with the estimated amount under the traditional-income questions for 2013, the redesigned questions have resulted in an estimated total annual income 9.1 percent larger for those ages 65 or older, an aggregate amount of almost an additional $133 billion. Retirement income is 27.9 percent larger, an aggregate difference of almost $71 billion.
However, Social Security remains the overwhelmingly predominant source of income for those ages 65 or older. The redesigned CPS still finds that over 60 percent of individuals in the two lowest-income quartiles receive more than 90 percent of their total income from Social Security.

Figure 3 Figure 7 Figure 5 Figure 4 F Fig igur ure e 2 1 CDHP enrollees were rou from IRAs a Endnotes sample for this b nd 40 a1(k se sample. This sam )-typ ge hlplans un y twice as lik der p the le elwas stratifi y as in traditio dividual nal m ed Figure 1 by gender, a e s wi asure, compa th traditio ge, region, nal covera red with 2. inco ge 1 tpe m o have e, an rcent u d rac colle nder th e. geThe response or post e new m gradu easur ate rate e Characteristics of the Popul Examining the New Income Measure in the Current ation With Consumer-Driven and Percentage Distribution of Retirement of Aggregate Income Plus IRAFi , 401(k) Income, by gure 2 Social Security Income Quartile, Income From Percentage of Retirement Income From IRA, 401(k) Income, Selected Health Status Indicators, by Type of Health Plan, 2005–2014 1 CompariHo son of useh th oe ld Tr In aco dime, b tionaly a T nd N ype o ew f Healt Totalh A Plan nnua , 2005–2014 l Income Measures Selected Demographics, by Type of Health Plan, 2005–2014 ( was 36.3 educatio Figure 4) ns in perc . Th e nearly a ent pe(26 rcenll ye percent for tage ar in s of th the se the bas e surve cond in ey sample or (Figu comer e qu 3). a nation r I tiln e 2 w014, a al sam s also hi 22p pe le, and 48 gher rcent o (6.f2 CDH percent for percent vs 0. P enroll the over ees 5 perc had gra ent). sample). d In t uate As a h d e t eg hrird non- ees For example, s ee Bruce Meyer, Wallace K. C. Mok, and James X. Sullivan. “The Under-Reporting of Transfers in Household Traditional IRA, 401(k) and New Income, by Measures of Income Income Quartile, Traditional in the Current Population and New Measures Surve of y, High-Deducti Population Survey by Compari b Income Quartile, Traditional and le Health Plans, 2005–2014 son of IRA, 401(k) Aggre New gat e Income Measures in Income, by Income the 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 in th2005 e Curre2006 nt Popul 2007 ation S2008 urvey, I2009 ndividua 2010 ls Ages 2011 65 or O 2012 lder, 2013 2014 and proba and 49 fourth b ility sa percen incmple, tr ome quartiles, t had coll aditio egen degr a the l survey mar perc ees, enta comp g ges of r are in od with 18 f error ecipient esti perc mates di s of this ent an time d d no 27 percent, t apply. How of income res wer ep ver, ha e ectively, more d th than of tr e s 1 u aditional rvey use 3 percent -plan da a ge Surveys: Its Nature and Consequences.” Harris School Working Paper #09.03, 2009 and John L. Czajka and Gabrielle 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 May 2015 • Vol. 36, No. 5 Income Current in the Current Population Indiv Population iduals SurveA ySurv g , Indiv es 65 or Older, eyidu , Indiv als A iduals ges 65 2013 Ag or es Older, 65 or Older, 2013 2013 Quartile, Traditional and New Measures of Income in the Current Self-Rated Health Status Less Than $30,000 Male by In come Quartile, 2013 By Paul By Craig Co Fronstin, Ph. peland, Ph.D., D., Employee B Employeee Be nene fit fit R Research esearch Inst Ini stitute tute, an d Anne Elmlin ger, Green wald & As sociates points h Denmead. “Inc enrolle probabes. HD ility sa igher mple, t HP enrol under ome Data for Policy Analysis: A Comparative th he m le n ees a ewere r w g measure. in o also mor f error foer the likel nationa y than tr l aditional sample Assessment of woul -pland enrolle have Eight Surveys.” bee es to n ± h2 ave colle .2 percent. Mathematica ge or gradu Reference No.: ate degrees. 10% a E10 xc0% ellent/very good 30% Population Survey, Individuals Ages 65 or Older, 2013 Traditional 15% 12% 15% 14% 11% 14% 11% 8% 10% 10% a a 49% 49% Traditional Tradit ional 50% 48% 50% Incom e Q50% uarti le 50% 50% 49% 48% b 42% 54% 49% 56% 59% 59% 58% 60% 56% 55% 6302-601, 2008 available at http://www.mathematica-mpr.com/~/media/publications/PDFs/incomedata.pdf HDHP 12 21 14 (in billio 10 ns of do 12 llars) 4* 9* 9 7* 10 b b HDHP 41 53 Traditiona 54l Income Mea 51 sure 57 New Income57 Measure 55 54 57 54 HDHP c 51 48 48 50 46 44 48 46 45 44 Furthermore, To e xamine th 90 9% %IRA an e issues me d 401(k nti)-type oned above, plan inc tT h ome also acco re sample aditional IncoF me wa irs Mea ts unted divided into sure fo Ne r a S w Inc ec la ond othree rg mee M r sh eas grou ur aere of ps: T to hithose wit rdtal re 87.3% tire hme aF c nt our o i nsumer-drive tn hcome for ea n ch Traditional Income Measure New Income Measure 8.6% $40 CDHP c 11 11 9 7 5 4* 5* 5* 5* 4* 2 c 25.4% Health-Status Diffe Introduction CDHP rences 57* 55* 58* 62* 63 63* 63* 65* 61* 63* For a complete explanation of the changes to the survey, see Jessica L. Semega and Edward Welniak, Jr. “The Effects of CDHP 59 51 56* 52* 52 47 45 46* 45 49 All Quartile Quartile Quartile Quartile income Introduction health pla quarti n (ClDHP), e. The those percent wia th a ge ohigh-de f retirem ductibl ent in ecome health fro plan m IRAs an (HDHP), d 401(k and t) h-ty ose pe pla withn traditional s in the first health cov quartile sho erage. wed $30,000–$49,999 25% Good Female 80 8% % a the Changes to the Current Population Survey Annual Social and Economic Supplement on Estimates of Income,” As participatio Trad n in itional em In ployment-based reti come Measure rement plans in the private s (in billioe nctor shift s) ed from defined benefit (DB) pension With the exception o af 2007, the survey has never found differences in self-rated health status between HDHP Traditional 19 20 18 19 17 17 16 13 14 $34.7 15 the lar Individuals gest dif we Tradit fre as erence ional sign un ed der to the the n CDHP ew inco anm d HDHP e measur group e—25.4 s if th pe ey h rcent ad o af de total ducti retirem ble ofe at leas nt income, c t $1,25 o0 mpared with for individua 6.5 l 45 35 38 34 32 34 34 32 33 33 In 2001, a handfu a l of employers started offering health reimbursement arrangements (HRAs)—a then-new type of $35 b 51 Traditiona 51 l Income Measure New Income Measure Traditional b 50 52 50 50 50 50 52 52 Total Income $1,451.32 $90.09 $176.48 $310.64 $874.11 Proceedings of the 2015 Allied Social Scie HDHP nce Association (ASSA) Research Conference, available at 22 30 20 15 16 14 16 15 22* 17 plans to enrollees an defin d i HDHP en d co divintributio duals wit nh (DC) plans, 42 traditional 34 cthe sources overage. 35 In co of 37 inc ntrast, in ome in 32 reti ninerement out of 33 10 have year 34 become a s of th 35 e surve n incr 70.0% 34 easingly y (2009 37 impor was th tant e Characteristics of the b Population With Consumer-Driven and perce coverage or nt under $2,500 the traditio for fami nal ly coverage. measure (FiTo be assign gure 5). The ed fourth-i to the ncome CDHP g quartile roup, th had th ey mu es smallest cha t also have b neen ge in t elig hib e le to fracti on c 49 52 health plan. 70 7% % The mos c t prevalent HRA-plan design then had a deductible of at least $1,000 for employee-only coverage HDHP 52 50 54 56 52 54 55 56 Social Security 531.24 76.15 144.04 161.41 149.64 CDHP CDHP 35 23 33 28 34* 16 30 12 28 14 30* 13 30* 13* 28* 11 32* 13 30 12 http://www.census.gov/content/dam/Census/library/working-papers/ 19.7% 2015/DEMO/ASSA-Income-CPSASEC-Red.pdf policy topic. Historically, annuity payments have been predominantly made from DB plans at retirement, but the shift exception), it c was found that CDHP enrollees were more likely than traditional-plan enrollees to report excellent or 20% 41 49 of total contribut retire e to ment in an HSA or come had a h coming ealth from I reim Rbursement A and 401( arr k)-type ange pla ment n in (HR come un A) with der a rollo the ne ver provision w measure (6.9 that they percent coul d use CDHP 44 48 48 53* 55* 54* 55 51 and a tax-preferred account that could be tapped by workers and their families to pay out-of-pocket health care $50, Fair 000–$99, /Poor Total R 999 etirement 253.57 2.32 10.07 63.22 177.96 $30 High-Deductible Health Plans, 2005–2014, p. 2 a EBRI Employee a Benefit Research Institute Notes (ISSN 1085 ?4452) is published monthly by the Employee Benefit Research to D very goo C pladn h 60 s ha 6% e %als r th e (Fi suglture 4). ed in th Furthermore, i e increasing un s s e of even lu of mp- the sum distri 10 years butio of th ns. Consequ e survey (2 ently, the 005, 200 tradit 7, 200 ional ma 8, 2011, nner 2012, used Ages 21–34 Tradit ional 13 1 2 13 10 9799 11 12 Traditional traditional, to pay for m co edical mpared Com ex pany penses with 1 or th 434 .7 perc e aent ne 107. b38 ility 44to take th w). 36 eir 1.08 a 36 ccount with t 38 5.57 hem shoul 38 d th 37 33. ey 37 chan 36ge jobs. 67. 38 Individua 41 35ls with th expenses. The Medicare Prescription Drug, Improvement, and Modernization Act of 2003 included a provision to allow b Institute, 1100 13 St. NW, Suite 878, Washington, DC 20005-4051, at $300 pe16.2% r year or is included as part of a membership b a HDHP 17 13 11 12 12 10 11 10 9 10 to measure the sources of income in retirement became obsolete. 5.1% 2013, a Traditional nHDHP d 2014), CDHP enrollees were less 28 likely 33 to repor 34 t bein33 g in fair28 or poor 31 health, though th 27 24 e actual 26 differe 25 nces State 36 76.20 31 35 0.55 40 42 2.71 48* 36 15.80 37 57. 39 13 1 40* only a flexible spending account (FSA) were not included in the CDHP group. subscription. Pe criodicals postage rate paid in Washington, DC, and additional mailing offices. POSTMASTER: Send address Examining the New Income Measures in the Current individuals with certain high-deductible health plans to contribute to a health savings account (HSA). HRA- and HSA- 14.7% $25 5% b CDHP c 50% 9 12* 9* 8* 10 7 8* 7* 7* 7* 15% th HDHP CDHP 33 4314 42 25 40 23 21 45 26 51* 21* 35 19* 36 19* 36 23* 38 18* Federal 32.00 0.30 0.88 7.14 23.68 were small. changes to: EBRI Notes, 1100 13 St. NW, Suite 878, Washington, DC 20005-4051. Copyright 2015 by Employee Benefit Obesec eligible plans are today collectively referred to as consumer-driven health plans (CDHPs). Social Security Population Survey, $100, CDHP 000–$149, 999 p. 8 20 23 19 22 25 20* 18* 18 20 19* One of the most cited sources Research Institute. All rights reserved, Vol. 36, no. a of income data for 5. those whose ages are associated with being retired is the Current IRA, 401(k) 14.08 0.15 0.21 1.42 12.30 Because the Tradit base sample ional (national sam 36 30ple) inclu 27 ded onl 26y 294 in 31 dividuals in a 29 CDH 29P and 23 287 indivi 27 duals with 25 an a 3.8% 40 4% % Traditional Ages 35–44 b 14 14 14 14 17 15 17 20 19 19 $20 Annuities 1.82 0.02 0.02 0.45 1.34 Population Survey (CPS), conducted by the U.S. Census Bureau. The Annual Social and Economic Supplement to the Whil CDHP e e thne rolle HDHP new es income exhibite meas d mo ure i re he 35 n CP althS 31 -cons founcd si ious be 30 gnificant havior 26 in than creases in in 27dividu retirement als with 26 traditi inco 26 moe as nal covera 28definege. d in t 26 In a his artic ll ye 23 ars of le, the b HDHP, an oversample o f individuals with a CDHP or HDHP was added. The oversample included 1,162 individuals with a By 2014, 27 percent of employers with 10–499 workers and 48 percent of employers with 500 or more workers HDHPc 11 5* 13 18* 17 19* 17 16 15 17 Traditional 10% 26 23 22 23 23 23 24 24 23 25 CDHPInterest 75.09 1.37 3.00 9.88 60.83 26* 25 30 28 27 26* 28 23 25 23 c income survey exc CPS (fiefrom S lded 30 3% ept in % o 201 March o cial S 3, C ecurity DfHP eac eh year still provi nrollees ) hwer a ds b ede t e lh en th ee over ss likely th e sourc whelan mi e of the muc thos ng majority o e withh tra -ci f income ted ditional numbers for covera for those in t ge to retirem report th he e low nt income. est three at they smoked. However, income 2 b CDHP The Employee Benefit Research Institute (EBRI) was founded in 1978. Its mission is to a CDHP and 741 individuals with an 11 HD 7*HP, resultin 16* g in 23* a total sample 19* (base 17 plus over 20* sample) of 21* 1,4 22* 59 for th 21e CDHP offered HDHP either an HRA- or HSA-eligible plan. 21 As a resu 23 lt, t 24 hese plans co 25 24 vered 24.5% about 30* 26 mill 21 ion 24 people in 21 2014, 26 Smokes c Digar ividend ettes 57.83 0.93 2.50 7.91 46.48 6.9% a research $15has s c hown that this survey has contribute to misclassified c , to encourag erte, an ain types of d to enhanc income an e the developmen d has get of sound empl nerally under-reoy porte ee ben d efit quartil Similarly, es when in all years e combin 6.5% x ed wit cept 201 h retir 0, e 201 ment 1, a in nd come, as A T 2014, Cwel DHA P l as enro G L A N C almost llees were less l half of this i E ikely t ncom oe report in the that four t th h quart ey did not ile (Figure $150,000 or More Traditional 21 24 24 20 18 15 15 14 16 14 group an CDHP d 978 for the HDHP group. After 30 factoring 30 out t 25 he base 27 sample 25 —the 29 30* 4 individuals 27* with 26 a CDHP 25 and th 25 e representing a New bout Inco15 perc me Meas ent of the priva ure tely insured market (Fronstin and Elmlinger, 2014). a $12.3 Who we are 20 2% % b 1 $12.1 programs and sound public policy through objective research and education. EBRI is the only T HDHP raditional 7779 10 10 12 16 13 18 13* 18* 13* 16* 12* 12 11* 10* 11* 9* income—an 6). In fact, regularly exer u d i ncise. Howev d ner both particular, inco e sources of r, m there were generally e measur retir es, e 97. ment 0 perce income non di t of ffer . the enc retir es in e obesity rates ment and Social bet Secur weeni CDH ty inco P a mne rec d trae dived by those itional-plan Ages 45–54 1.4% 237 individual 5% To bs with tal Inco an H me DHP—there $1,584. were 03 1,450 individuals in $89.27 the$181. sample 56 with tra $328. ditional 87 health $984. coverage. 32 c private, nonprofit, nonpartisan, Washington, DC-based organization committed exclusively to HDHP CDHP a 16* 2 16* 3* 15* 9 14* 9* 14* 8 12* 7* 10* 13* 11* 13 13* 13 11* 11* 10.1% Characteristics of the Population With Consumer-Driven and High-Deductible $1 Traditional 0 29 26 27 26 28 27 27 29 26 26 in th enrolle e lo es. west-income quartile was from Social Security. The percentage from Social Security in the fourth quartile was S c ocial Security 551.12 74.98 149.14 161.99 165.02 This article 10 1% ex % amines the po 0.8% pulation with a CDHP and how it has differed from the population with traditional health public policy research and education on economic security and employee benefit issues. No regular exercise 0.6% 2.2% CDHP 2.1% b 6* 3* 10* 11* 10 9 20* 20* 20* 22* 4.4% In response to this research, the Census Bureau redesigned the income questions in the CPS to reduce the HDHP a 32 29 30 27 25 28 34* 28 29* 30 Health Plans, 2005–2014, Total Retirement 324. by Paul Fronstin, 42 2.29 Ph.D., EB11. RI, and 08 Anne 74. Elmlinger, Gr 91 236.14 eenwald In ad 45.7 percent dition to ubein nderg s the trati traditional m fied, the b EBRI’s member aese sa asure mple compare was a ship includes a d lso we with 4 ight 1cross-section of .1 ed by percent gen un der, der t pension funds; businesses; trade associations age, he e ne dw meas ucation,ur re e. gi on, income, and ; Traditional 0.2% n/a 252 1.5%525212324202018 coverage. Data from t 1.1% he 2005– 1.2%2007 Employee B 0.1% enefit Research Institute (EBRI)/Commonwealth Fund Consumerism Source: EBRI/Commonwealth Fund Consumerism in Health Care Survey, 2005 ?2007; EBRI/Greenwald & Associates Consumer Engagement in c b CDHP 32 29 31 30 28 28 31 31 30 32* 3 discrepa With respect t ncies HDHP 0% from the oC HDHP ompany and tr prior aditional design. In 141. -plan 270 01 enrolle 4, researc es, ther 0. hers 73 e w at th ere e C no stat ensus Bure 4.60istically sign au co33. ndu ifi41 cant diff cted a test o erenc 102.96 fe s in t the nh ee o w set of bese 0% n/a labor un 26 ions; h 20* ealth care prov 22 iders and insur 21 e 18 rs; government org 20 a 19 nizations; 19 and service firms. 16 & Associates Health Care S urvey, 2008 ?2014. 0% race/ethnicity $5 to reflect the actual proportions in the population ages 21–64 with private health insurance coverage. in Health Care c Survey and the 2008–2014 EBRI/Greenwald & Associates Consumer Engagement in Health Care 1s 1s t Q t Q uu ar ar til til ee 2n 2nd Q d Qu ua arrttiille e 3r 3rd d Qu Quar arti tile le 4t 4th Q h Qu ua arrtitile le Ages 55–64 a CDHP 1st Quartile 2nd Quartile 3rd Quartile 4th Quartile n/a 22* 18* 19* 14* 20 21 15* 16* 15 State 56.24 0.64 1.48 12.87 41.26 CPS-income Traditional questions by = health plan w do ith no deducti ing a spibl lt-pan e or <$1,250 (i el design, ndi$2 vidual w .2 h ), < ere 3 $2,500 (fam /8 of the ily) in sample 2014. received the redesigned questions while Whe percentage in n adding Social S any years o ecurity f th income e survey to r aned no r tirement ecent incom differences e, the p in ercent exea rcise. Howev ge of income er, in coming from all years of the IRA an survey d 401(k)- The CDHP and HDHP ov a ersamples were weighted by gender, age, inc $1 ome, an .4 d race/ethnicity. Survey (CEHCS) were used for the analysis. Differences between the populations with traditional coverage and those b Source: EBRI/Commonwealth Fund Consumerism in Health Care Survey, 2005 ?2007; EBRI/Greenwald & Associates Consumer Engagement in Health Traditional 17 18 18 19 21 19 22 24 24 24 $0.6 2 Federal 28.46 0.13 0.77 5.48 22.09 HDHP = High-deductible health pl Source: Employ Source: Employ $0.2 ee ee Benefit R Benefit R esearch Institut esearch Institut an with deductible $1,250+ e e estimates estimates $0.2 of the M of the M arch (individual), $2,500+ arch 2014 2014 C C urrent Population urrent Population (fam Surv Surv ilyey ), not HS ey (T (T raditional raditional A-eligible in 2014. and N and N ew ew Income IncomeM M ee asures). asures). EBRI’s work advances knowledge and understanding of employee benefits and their Source: Employee Benefit Research Institute estimates of the March 2014 Current Population Survey (Traditional and New Income Measures). type the rem except 2 plans a0 inin 10 looks diff , g HDHP e 5/8 rec ere n erolle ive nt du de tshe tra e to th wered less l e itiona highe ikely than l questions. r relativ tra e amo ditiunt onal-plan of Socia enro l Se llees to curity i report ncome that they going to thos smoked. e in the lower- Care Su brvey, 2008 ?2014. c Note: The income quartiles are only for those individuals ages 65 or older. 401(k) income includes income from any similar defined contribution plan. $0 Note: The income quartiles are only for those individuals ages 65 or older. 401(k) income includes income from any similar defined contribution plan. HDHP 33 23 24 27 24 22 27* 30* 26 25 with high-deN ductibl ote: The income e health quartiles plans are only (HDHPs for those indiv ) wer iduals ages 65 e also or exa older. 401(k) mine income includes d. Differences discussed in income from any similar defin th ed contribution plan. e remainder of this article ? The CDHP populatio = Consum IRA, 401( ns of a er-driven health pl k) dults within consum an w49.56 ith deductible $1, er-drive 250+ (individual 0. n 58 health ), $2,500+ (fam plans 2.18 (CDH ilyPs), high-de ), with HRA12. , HS 12 A ducti , or HS bA le -eligible in 2 healt 34.h 67 pla 014. ns (HDHPs) a importance to the nation’s economy among policymakers, the news media, and the public. It Traditional = Health plan with no deductible or <$1,250 (individual), <$2,500 (family) in 2014. c 1st Quartile 2nd Quartile 3rd Quartile 4th Quartile income While pane quarti l In le ternet surv s. Under this eys are definition nonran ofdo inm, studies come (inclu hav dine g de Soci monstrated that al Security), the such share surveys, when c of income from areful IRA a ly design nd ed, CDHP 17 18 25 22 22 21 24 25 24* 24 * Difference betw b Annui een HDHP ties /CDHP and traditional 20.91 is statistically 0. significant at p 15 = 0.05 or better. 1.28 6.59 12.88 were statistically significant. (More does this b information abo y conducting ut theand p dataublishing policy re can be found in tsear he appen ch, analysis, dix.) and special reports on and traditional health plans were each split about 50–50 between men and women in 2014. HDHP = High-deductible health plan with deductible $1,250+ (individual), $2,500+ (family), not HSA-eligible in 2014. What we do This Notes article provides a comparison of the income levels from the redesigned questions with those from the Source: Employee Benefit Research Institute estimates of the March 2014 Current Population Survey (Traditional and New Income Measures). 401(k)-type obtain results comparabl White, non-Hispa plan incom nic e i en wi th th e ran first quartil dom-digit-dia e was l 0.8 pe telepho rcn ent u e surveys. I nder th n e ne a study w m e that asur e of provides income the r (Figure esults from 7). This a c employee benefits issues; holding educational briefings for EBRI members, congressional and Interest 84.22 1.26 3.41 10.81 68.75 CDHP = Consumer-driven health plan with deductible $1,250+ (individual), $2,500+ (family), with HRA, HSA, or HSA-eligible in 2014. Employer Size Differences Note: The income quartiles are only for those individuals ages 65 or older. 401(k) income includes income from any similar defined contribution plan. a traditional questions. The focus in t his article is on the income of those ages 65 or older and on the income categories Traditional federal agen 68 cy st 71 aff, and th 71 e news 72 media; and sponsoring public opinion survey 70 70 69 71 68 s on emplo 75 yee percentage was larger number of surve Diviy dend s tha for t we each hi re cond gh ucer i t 49.84 edn at co t me quartile, he same0. ti78 me reachin using th g 8.6 percent e sa 1.63 me questionn for the 5.56 a fourth-income quart ires both via41. telephone 88 ile. and * Difference between HDHP/CDHP and traditional is statistically significant at p = 0.05 or better. ? CDHP enrollees were less likely than those with traditional coverage to be between the ages of 21 and 34 in b In the earlier years of the survey (2005–2009), the CDHP population was more likely than the population with associated HDHP wit h retiree income to se benefit issues. e th92* e impact o EBRI’s Ed 84* f the 76* ch ucanges in ation and Re 75 the qu se 71 ar estions on ch Fund 73 (EBRI-ERF) performs sources of 75* 71 income in 75* the char retir83* it eable ment. , Demographic Differences in the CDHP, HD Figure 3 HP, and Traditional-Plan Populations Percentage Change New Measure Relative to Traditional Measure online (Taylor, 2003), the researcher found that the use of demographic weighting alone was sufficient to bring 2014, a cnd more likely to be ages 45 ?54. educational, and scientific functions of the Institute. EBRI-ERF is a tax-exempt organization traditional CDHP cov erage to have that covera92* ge through 81 sma Figure 8 78 ll employers 77 (be 74 tween two 75 and 77* 49 employe 78* es)77* (Figure 5). 82* Figure 6 Education, by Type o Figure 4 f Health Plan, 2005–2014 Particular Even with em thT e o phasis is giv tal high Ince or amou me en ntto the s of inincome come i 9.1%d from in entified un divi -0dua .der 9% lthe n retire ement w inco 2.9% ac mcounts e measur (IRAs) and 401(k e, 5.9% Social Security )12. -ty remains 6% pe plans, as by fatr the his Gender—Generally, regardless of plan type, the population of adults with private health insurance has been split 50– almost all of the results from the online survey close to the replies from the parallel telephone survey. He also found supported by contributions and grants. Minority Figure 5 Distribution of Percentage Retirement of Income Plus Individuals A Social ges 65 or Older Security Income, by With Income Income Quartile More recently (2010–2014), there were few statistically significant differences by employer size between the CDHP Social Percentage Security of Indiv 3.7 iduals Ages -1.5 65 or Older W 3.5 ith Income W 0.4 ho Have IRA 10.3 , appears to be the income type with the most underreporting, given the lump-sum nature of the payments typically most important source of inc 2005 ome for thos 2006 e ages 2007 65 or o 2008 lder who 2009 have incomes in 2010 2011 the lowe 2012 r half. I 2013 n fact, 68. 2014 3 percent 50 between mean and women. Throughout 2005Types of Income –2014, about 50 percent of traditional-plan enrollees were male and that in ? CDHP enrollees were more li some cases, propensity wei kely than ghting (meanin traditional-plan g the propen enro sity llees to for a be certain in househol type ofds person to with $15 be 0,00 onlin 0 or e) re more duc in ed Firm Size, by Type of Health Plan, 2005–2014 Traditional 32 29 29 28 30 30 31 29 32 25 and Source of Retirement Who Receive More Than 90% Income, Tradit of Their Total ional and New Income Measures of Income From Social in the 401(k) Income, by Income Quartile, Traditional and New Measures Total Retirement 27.9 -1.1 10.0 18.5 32.7 populaHigh School G tion and the raduat popue or Less lation with traditional coverage. found from of the indivi th du b ese plans, inst als who received ead o income f regu in lar 201 annu 3 a itn y payments traditiona d were in the lowest-income lly received fr quartile r om pe eceive nsions. d more than 90 percent about 50 percent were fema2005 le (Figur2006 e 1). No statistically 2007 2008 signif2009 icant diff2010 erences hav 2011e been 2012 found be 2013 tween2014 HDHP income in eve a ry year except 2006, 2009 and 2010. They were also more likely to be in households with the rem HDHP aining gaps, but in other cases EBRI Issue B it 8* did not riefs 16* re are period duce 24* the rem icals providing exp 25 aining gaps. P 29 ert evaluati e 27 rhaps the ons of emplo 25* most striking 29 yee benefit issues and 25* difference 17* in Security, by Income Quartile, Traditional and New Measures of Current Population Survey, Individuals Ages 65 or Older, 2013 Traditional Company of Income 31.9 in the Current -32.3 Population -17.5 Survey, 2013 0.1 52.7 Retirement Income—This 32% income com 38%es from 42%pensio 33% ns or retir 35% ement38% plans from 34% private- 29%and/or 33% public-sector 28% Self-Em c ployed With No Employees 20% of their total in b come from Social Strends, as well ecurity (Figur as cr e 8itical analy ). Those in t ses of he sec emplo ond-i yee benefit po ncome quartile licies an had a d proposals. similar l EBRI Notes evel, at is a CDHP 8* 19 22 23 26 25 23* 22* 23* 18* enrollees and traditional-plan enrollees. Statistically significant differences in gender have been found between CDHP demogra Our $10p 0,0 hics 00– betw $1 a 49 een ,99 tel 9 in ephone a income in Income in nd onli most ye ne surveys the ars. They Current Population wa we s the u re rou nder-r ghly t epresentatio w Surv ice as lik ey, 2013 ely as n of minor indiviiduals ties inwit online h traditional samples. HDHP 100% State -26.2 15.0 -45.5 -18.5 -27.8 9* 18* 14* 15* 15* 11* 14* 12* 16* 14* When comparing HDHP enrollees with traditional-plan enrollees, it was found that, in all years of the survey except employers or Tradit ional unions and includes tr 2% aditio 3% nal defin 3% ed ben 2%efit plans 3% and defi 3% ned contr 3% ibution 3% plans (401 5% (k)-type 3% plans). Source: EBRI/Commonwealth Fund Consumeri monthly sm in period Heaical providing current informati lth Care Survey, 2005 ?2007; EBRI/Greon on a variety of employ enwald & Associates Consumer Engag ee benefit topics. ement in c 80% 61.4 percent. b enrollees an CDHP d those wit Federal h traditional 7* covera -14* 11.0 ge in a 12* few -57. ye 10* 3ars ?2007, 10* - 12.2 2008, 20 9* 10, 20 9* -23.3 11, and 20 9* 12 ? -10* 6. though not m 7 7* ore coverage HDHP to have college or postgraduate educations in nearly all years of the survey. Overall Comparisons 12* 10* 11* 8* 7* 6* 9* 9* 17.8% 6 7* Health Care Survey, 2008 ?2014. 2007, HDH 18P % enrollees were less likely th Tan radi traditio tional Income nal-pl Measuan re enrolle Ne ew Inc s too m w eo Mreas k fur oe r large employers (500 or more It also includes income from individu EBRIef al reis a weekly tirement accou roundup of nts and Keo EBRI resear ghs, as wel ch and l as insights, as payments from well as updates on survey annuities and paid- s, c publications 90% a Some College, CDHP Trade or Business School IRA, 401(k) 9* 252.07* 6* 284.86* 656653 934.7 754.5 181.9 References Traditional = Health plan with no deductible or <$1,250 (individual), <$2,500 (family) in 2014. recently, and when they 71.0% were found the differences were not very large. studies, litigation, legislation and regulation affecting employee benefit plans, while EBRI’s a Compared employees). wit Th h the estimated amou ey were more likely to nt of i worn kcome u for small nder t emh ployers i e traditiona n all y l in ears of t come questio he surn vey exc s for 20 ept 13, t forh 2010. e redesi gned up life insurance. b 2–49 E Tradit m ional ployees 68.3% 31 29 29 31 31 29 30 29 31 27 70% Annuities 1,046.4 752.7 6,485.7 1,362.9 864.2 HDHP = High-deductible health pla n with deductible $1,250+ (individual), $2,500+ (family), not HSA-eligible in 2014. Blog supplements our regular publications, offering commentary on questions received from 80 16% % a b Conclusio Traditional n questions resulted in an estimated tota 15 19l annual 19income 9 16.1 percent 15 larger for 16 those 16 ages 6 17 5 or olde 16r, an ag 16 gregate Fronstin, Paul, HDHP and Anne Elmlinger. "Fin 43 35* dings From the 2014 EBRI/Green 28 29* 27 wa27 ld & A 15.1% ssociates Cons 28 27 umer Engagem 28 23 ent in c Interest 12.2 -8.5 13.6 9.4 13.0 Examining the New Income Measure in the Current Population Survey, by Craig Age— CDHP = In 2014, Cons b CDHP umer-driv enrol en healt leeh s wer plan w news reporters, e it l h dedu ess li ctkely than ible $1,policy 250+ thos (indiv make w idual), ers, ith tra and others. The $2,500+ ditio (family nal covera ), wiEBRI Databook th HRA ge , HSA to be , or H betwe SA-eligible on Employee en in the a 2014. Benefits ges of 21 is a and c 61.4% HDHP Traditional Income Measure 44* 37* 60.7% 30* 30* 31* 28 30* 27* 32* 31* CDHP Company—Income from a private 31 - 37* sector de 30 fined benef 25* it (DB 25* ) pension 25plan. 25* 24* 24* 22* amount of The new me Healt alm asure o h Do Care S ivst an a idend f in udditio come rvey."n in th EBRI Issue Brief no. 4 al $1e CP 33 -billio 13. S wa 8 n (Fi s able gure to i 0 1 7 - (Empl d 16. ). ent Furthermore, 2 ify oyee significa Benefit Res -34.9 n a tlnnual y more i ear retir ch Institute), Decemb ncom eme -29.8 e nt incom cominge from was 27. -9. er 20 IRAs an 9 9 14. perce d 401(k)- nt 70% c statistical reference work on employee benefit programs and work force-related issues. Copeland * Differ14 60 enc , Ph % %e be .D., tween HDHP/ EBRI CDHP and traditional is statistically significant at p = 0.05 or better. Appendix—About the 2014 EBRI/Greenwald & Associates Consumer Engagement CDHP 29* 26* 25* 22* 20* 23 21 16 16 16 34, and they were more likely to be between the ages of 45 and 54. This pattern was consistent in nearly all years of College Graduate or Some Graduate Work Source: Employee Benefit Research Institute estimates of the M arch 201 4 Current P opulation Survey (Traditional and New larger, an aggregate difference of almost $71 billion annually. Income Ne from w Inco IRAs a me Measurn ed 401(k)-type plans was an type plans, which researchers have noted was being missed by the survey. This income was also shown to be a much 50–199 Employ a ees in Health Care Survey the survey, Taylor, Hump Tradit though the diffe h ional rey. "Does Interences rnet Rese 24 were arch ‘ 22 noW t often ork’? Comp 20 statistically signific 24 aring Onl 23 ine Sur anv t. e 23 y Results W 24 ith Te 26 lephone Surve 25 ys." 27 60% State—Inc Inco ome from me a M easur a st es). ate- or local-government employer DB pension plan. ? TheT U.S. Ce raditionalnsus Bureau’s Current Population Survey (CPS) is a primary source of income data for those whose b 8 10111211 8 13 8 1211 12% Contact EBRI Publications, (202) 659-0670; fax publication orders to (202) 775-6312. important larger percent component age of r of etirem this ehigh nt ine come r amou (ent xcluding Socia of annual ret l Se irement curity) tha incom n ha e fou d bee nd in n the previously fou new question nd in t s, which he CPS. overa It was ll HDHP 50% b 33 36* 41* 40* 41* 47* 41* 43* 37* 45* Note: The income quartiles are only for those individuals ages 65 or older. 401 (k) income includes income from any similar defined International Journal of Market Research 42, no. 1 (August 2003). The Employe HDHPe Benefit Research Institute (EBRI) and Greenwald & Associates created the EBRI/Greenwald & 11 14 16 15 17* 14* 14 15* 15 14 c Subscriptions to EBRI Issue Briefs are included as part of EBRI membership, or as part of a ages are associated with being retired. In response to research showing that the survey has misclassified and 50% was more tha CDHP con cnt 250 perc ribution planent . higher than that found by the traditional measure. a particularly large sh 97.0%are for those i 44* n th 37*e lowest-incom 39* e 44* quartile, reach 45* ing just ov 43* 45* er 25 percent. 46* 46* 49* Orders/ 97.0% Similar results were found in comparing the HDHP population with traditional-coverage enrollees. Between 2010 and CDHP 7 12101213121211111 3 Federal—Income from a federal-emplo 93.5% yer DB pensio 93.1% n plan. Associates Consumer Engagement $199 annual sub in Health Care scription Surve to y (CEHCS EBRI Not ) to es examin and EBRI Issue e issuesBrie surroundi fs. Change of ng consumer-dire Address: EBRI, cted Income Differences 10% under Graduat -ree Degree ported certain types of income, the 2014 CPS included a redesigned set of questions aimed at better 200–499 Employees Endnotes 2014, HDH 40P % enrollees were less likely than those with traditional coverage to be ages 21–34. In addition, in all years a 1100 13th St. NW, Suite 878, Washington, DC, 20005-4051, (202) 659-0670; fax number, 40% a health care, Tradit in ional cluding the cost of insurance, the cost of care, satisfaction with health care, satisfaction with a health 1311 9 12111012161218 CDHP e Subscriptions nrolle Tradite ional s have been more likely than traditional-plan enrollees to be in higher-income households in most years However, Social Security remains the9898 overwhelmingly predomina11nt source o8998 f income for those ages 65 or ol11der. Over capturing income from individual retirement accounts (IRAs) an 71.9%d 401(k)-type plans, among other goals. 1 IRA, 401(k b)—Income (distributions or (202) 775-6312 withdr; e-mail: awals) from indiv subscriptions@ebr idual retirem i.orge nt Me accounts, Keo mbership Information: ghs, and employ Inquiries ment- b 68.4% of the survey, See Fronstin a 8% HDHP enrollees were more li nd Elmlinger, 2014, for more information about HRAs and H kely than those with traditional covera SAs. ge to be ages 45–54, though the IRA and 4 HDHP 01(k) Type Plan Income HDHP 156767 12 18* 17* 17*7* 15*779 17* 18 1 20*0 1182 care plan, reasons for choosing a plan, and sources of health information. The 2014 CEHCS was comparable with 30% of the survey. In fact, CDHP enrollees were more likely than traditional-plan enrollees to be in households with quartil 60 perc es. In ent o the h f indivi igh duals i est-inc n t ome quartile, he t wregarding o lowest the EBRI -income quarti percentage of membership an les re IRA d/or con ceived m and 4 tributio 01(k ore than )-ty ns to EBRI-ER pe 90 plapercent n incom F should be dir of t e decr heir eas total i eed cted from ncome to 87 EBRI .3 from c c 30% based defined contribution plans (such 401(k) plans). dif 2 ferences CDHP CDHP were not often st18* atistically signi 6* 10 13 ficant. 18* 8 20* 7 20* 8* 23* 8 21* 9 10 21* 21* 8 7* 22* 6.2% Whe See n http://w examinww.mercer ing the IRA an .comd 4 /content/mer 01(k)-type cer/global/al plan income l/e o n/news f thosroom/m e ages 6 odest-health-b 5 or older byenefit the income -cost-grqu owth-continues-as- artiles of these findings from ? This artictle he 200 finds the 5–20 ne 07w m EBRI/Commonwealt e President Dallas asure of income in Salisbur h Fun the y d at th CPS Consumerism e identif above address, ( ies sign in Healt ificantly 202) 659-0670; h Car more e surveys, and e-mail: income salisbur (an the 2 d a y@ebri.org much 008 - larger 2013 $150,000 or more in income in every year except 2006, 2009 and 2010 (Figure 2). CDHP e 45.7% nrollees were also more Social Security. percent un6% der the traditional measure of income to 70.0 percent for the new income measure (Figure 3). The third S500 or ource: E MB or R e E I/Cm om ploy monw eesealth Fund Consumerism in Health Care Survey, 2005 ?2007; EBRI/Greenwald & Associates Consumer Engagement in 41.1% 20% a consumerism-kicks-into -high-gear.html indivi CEHCS. duals, each income quartile showed higher levels of IRA and 401(k)-plan type income under the new measure. percentage of Traditional income) coming from IRAs and 401(k)-type plans. Compared with the estimated amount under Health Care Survey, 2008 ?2014. 54 46 43 50 48 52 49 54 46 50 likely than traditional-plan enrollees to be in households with $100,000–$149,999 in income since 2007 (with 2010 quartil Annuiti e hae d t s—income he largest from perc annu entage-point ities and pai increase o d-up liffe i IR nA and 4 surance 0purch 1(k)-ta ype sed outsi plan inco de o m f an e relat emiploym ve to the tra ent-based pl ditiona al n. Race—With t 20%he exception of 2005, CDHP enrollees have only recently (2011 ?2014) been more likely than traditional- b 4.0% Editorial Board: Dallas L. Salisbury, publisher; Stephen Blakely, editor. Any views expressed in this publication and those of the authors should a HDHP 21* 24* 28 31* 31* 38* 32* 34* 28* 28* 3 The first Tradi (lo 10 4% % tiw onal es = t) income Health plan w qu ith no deducti artile sho blw e or < ed $0.4 $1,250 (i bindi llion vidual m ), < ore, or a $2,500 (fam 20 ily0 ) i n perce 2014.nt increase over the traditional measure Yet, incom the tra e dfr itiona om retirem l-income ent quest sources such ions for 20 as IRAs an 13, the red 401( designe k)d -ty questions pe plans and DB have resulte pensions from d in an estimate private- d to and public- tal annual and In theory, a r 2014 beina g e ndom sample xceptions). Sinc of 2,000 yields a statistical prec e 2010, traditional-plan enrolle ision of plus or minus 2.2 per es have been more likely than centage points (with 95 percent CDHP enrollees to be in not be ascribed to the officers, c trustees, members, or other sponsors of the Em13.5% ployee Benefit Research Institute, the EBRI Education and measure—its share rose from 10.1 percent to 24.5 percent. The lowest-income quartile had a small increase in its plan enrollees to be non-Hispanic white. The 2005 difference may have been due to a small sample size of minorities, b CDHP 43* 37* 44 45 50 47 11.9% 46 51 53* 54 The 2014 HDHP surv = High-deductible health pl ey was conducted an w withi ith deductible $1,250+ n the United State (individual), $2,500+ s between Au (fam g. 7 ily), not HS and Au A-eligible in 2014. g. 27, 2014, through an 11-minute (Fi sector employers plays a gResear ure 2ch Fund, ). This or amount i their staffs. significant ncr Nothin eased g her for role ein is to be co eac in h suppleme successively nstrued as an attem ntin higher g Soci pal S incom t to aid or ecuri e quartile hi ty. Ther nder the adoption to efore, $2.0 it is im of a bill ny ion pending le portant (1,000 gislat to point perc ion, ent), re out t gulation, hat In additio income n, th 9.1 ere ar percent la e payme rger 2.1% nts from oth for those a erg speci es 65f or o ic sourc lder, e an s (e. aggr g., U.S. egate ra am ilroad retir ount of almost ement) an an a d u dditional ndefine d or $13 unkno 3 billion. wn confidence) of what the results would be if the entire population ages 21–64 with private health insurance coverage was households c Source wi : EBRI th incom /Commones les wealth Fu sn than d Consu$3 mer0,00 ism in Heal 0. th Care Survey, 2005 ?2007; EBRI/Greenwald & Associates Consumer Engagement in Health share, from 1.1 percent to 1.2 percent. 10 0% % which was addressed in 2006. When comparing HDHP enrollees and tr 1.7%aditional-plan enrollees, it was found that in CDHP = Consumer-driven health plan with deductible $1,250+ (individual), $2,500+ (family), with HRA, HSA, or HSA-eligible in 2014. 2% or interpretative rule, or as legal, accounting, actuarial, or other such professional advice. Internet survey. The national or base sample was drawn from Ipsos’s online panel of Internet users who have agreed Care Survey, 2008 ?2014. $10.7 billion (764 Tr adi pe tiona rcent l ), and $2 New 2.4 bil Trli adi on tiona (18 l 2 perce New nt), respectiv Traditiona elly. New Traditional New sources of r those wit Retireme h th eti ent i rement se income ncome is 2 incom soue re 7 rces are .9 percent porte more d in larger, thlikely to e survey an aggre be (th in eg total of th ate e up difpe fere all r- th in nce o cese other ome f a qu lmost $7 ar sour tiles be 1 billi ces is less th cau on. se th eya ha n 1ve 0 th perce ese sour nt of the ces surveyed with complete accuracy. There are also other possible sources of error in all surveys that may be more serious * Difference between HDHP/CDHP and traditional is statistically significant at p = 0.05 or better. 0.5% 2005, 200 a 6, 2007, 2011, 0.4% 2013 and 2014, a higher percentage of HDHP enrollees were non-Hispanic white. Again, the Traditional = Health plan with no deductible or <$1,250 (individual), <$2,500 (family) in 2014. 1.2% to participate in researc 1st Q h sur uartileveys. Nearly 2,000 adults ages 21 2nd Quartile -64 wh 3ro d had he Quartile alth insurance t 0.5% 4th Q hrou uartig leh an employer, In general, there have been few income differences between HDHP enrollees and traditional-plan enrollees. of income. Moreover, these individuals’ incomes before they were in retirement cannot be determined from a single than theoretical calculations of In additio total retir b e nme to nt incom the higher e). amount sampling e of IRA a rror. These include nd 401(k)-type plan refusals to be income, a lar interv ger iewed an percend o tage ther fo of inrms o dividuals r f nonrespo eceivinse, ng t the his EBRI Notes 0%is registered in the U.S. Patent and Trademark Office. ISSN: 1085 ?4452 1085 ?4452/90 $ .50+.50 HDHP = 0% High-deductible health plan with deductible $1,250+ (individual), $2,500+ (family), not HSA-eligible in 2014. 2005 finding may also have been due to a small sample size. purchased Despitec thedir ine creasing leve ctly from a cls of t arrier, or he higher purchased t amounts of hrough thi a sgovernment income type exc for eac hange h iwer ncome quartile, e drawn ran t do hmly from t e distributih on e I of psos ? However, Social Sec 1st Qu uar rity r tile e So mains the cial Security ove Co 2n mpa d Q rwhelm n uy artileStingly ate Fe pred deralomina IRA3r , 4 d nt so 01(k Quar ) tiurce o leAnnuities f incom Other e fo 4trh Q those uartile ages 65 or older. 1st Quartile 2nd Quartile 3rd Quartile 4th Quartile effects of ques income year’s snapsh CDHP = was alot of i Con tion wording a so foun sumn ercome. -d d. Amo riven heal Conseque nn d question thg t planh w ose in ith dedu ntly, order, and scree th cti those ble low e $1,250+ e who st-inco (indivha idun ad acc m ing. While attempts are l), $2, e qu 500+ eartile, ss and t (family 0. ), wook advanta 4 percent ith HRA, HSA, mad w or HSA- e gere of these pl to minimize these factors, it is e fou eligibn le id to h n 2014.aan ve rec s are t eh ived inc e ones with ome this income * Differ ty encpe ac e between ross the HDHP/CDHP an income d tradi quartil tional is ses shifte tatistically sid to gnifican at at larger p = 0.05 or percentage o better. f the income going to the lower income The redesigned CPS still finds that over 60 percent of individuals in the two lowest-income quartiles receive more impossible to q higher amouSource: Employ nt Source: Employ u s of this antify the ee incom ee Benefit R err Benefit R ors that may re e—not n esearch Institut esearch Institut ee cessarily those e estimates estimates sult from them. of the M of the M arch arch wh 2014 2014 o CC urrent Population urrent Population had the hiSurv gh Surv ey est overall ey (T (T raditional raditional and N and N inc ew ew omes prior Income Income M M easures). easures). to retirement. Source: Employee Benefit Research Institute estimates of the March 2014 Current Population Survey (Traditional and New Income Measures). Education Differences Note: The income quartiles are only for those individuals ages 65 or older. 401(k) income includes income from any similar defined contribution plan. N Note: T ote: The he income income quartiles quartiles are only are only for those for those indiv indiviiduals duals ages 65 ages 65 or or older. 401(k) older. 401(k)income includes income includes income from any income from any similar defin similar defined contribution plan. ed contribution plan. than 90 percent of their total income from Social Security. © 2015, Employee Benefit Research Institute ?Education and Research Fund. All rights reserved. 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 Nottttttttte e e e e e e e ess ss s s s s s • Ma • Ma • Ma • Ma • Ma • Ma • Ma • Ma • May y y y y y y y y 201 201 201 201 201 201 201 201 2015 • Vol. 36, No. 5 5 • Vol. 36, No. 5 5 • Vol. 36, No. 5 5 • Vol. 36, No. 5 5 • Vol. 36, No. 5 5 • Vol. 36, No. 5 5 • Vol. 36, No. 5 5 • Vol. 36, No. 5 5 • Vol. 36, No. 5 10 11 15 4 3 8 2 7 9 A monthl y newsletter from the EBRI Education and Research Fund © 2015 Employee Benefit Research Institute ebri.org ebri.org ebri.org ebri.org ebri.org Notes Notes Notes Notes Notes • • • • • May May May May May 2015 2015 2015 2015 2015 • • • • • Vol. Vol. Vol. Vol. Vol. 36, 36, 36, 36, 36, No. No. No. No. No. 5 5 5 5 5 14 12 13 5 6