15 11 14 17 10 19 22 18 13 20 12 21 16 4 5 8 6 9 7 2 3 Figure 7 a Change in Median Replacement Rates from 401(k) Accumulations Relative to Baseline Model Assumptions for Participants Reaching Age 65 Between 27 th 32 ____ VanDerhei, J ____. “H ack, and Craig Copeland. “ ow Would Target-Date Funds Likel Can Amery ica ImAfford Tom pact Future 401(k) Accum orrow's Retirees: Results Fr ulations?” (June 18, om the the highest-tenure category incr percen 401(k) plans, Figures 11 and 12 s 401(k) plan. However, the studies were all pe assets in th However, the evidence presented in Figures 14 reduction in 401(k) account balan independent of current eligibilit Based on unpublished E the additional savings for a longer period and m em positiv account balances from ployer contribution to a define The analysis for defined benefit sponsors The bottom tage o e 25– e case of a rev 37 percen f preretirem line of Figure 7 shows th t r the enhanced a BRI research, ent incom e nge, while the loss rsion (VanDerhei, 198 eases the differential substantia y. In both figures, the top panel how how long it m d contribution plan based on th ces at retirem e assum contributions (if any) than the average equity allocation for target-date funds e m ed to be replaced b es as efreezing the plan for all m di rform e an replacem and 15 suggests that the lowest-salary quartile a nt would decrea socia 9) ny of , the best available choice m e id prior to the 2008 m ght take for va t them ed with the 25 ent rates for a ty would be able to achieve this status y Social Security. lly (13.1 years in Figure 11 vs. they would have had under the se between 25 and 35 percent for of figures pertains to the VE e EBRI/Mercer surv rious 401(k) participants to percentile is always less emab rket crisis and the er pical 401 s is not as ay be to gradually The resulting ey described (k) 2.4 How much will automatic escalation help 401(k) participants? 2. Adequacy from 401(k) Plans: Pre-September 2008 2030 and 2039, by Income Quartile at Age 65 th th 17 An Evaluation of the 2009c). W EBRI-ERF Retirement Securit orking paper available at SSR y ProjectiAdequacy and S on Mo N: http://ssrn. del.” EBRI Issue Bri com/abstract=1422726. ef, no. 2 tructure of 63 (Employee Benefit participant by incom with as little as an additional 5 percent of third section of the testimony focus additional defined benefit accruals. This percenta recover the losses experienced in than 6 percent. Moreover, even the m previously. participants born between 1956 and 1970. straightforward. W will a plans and the next five panels focus on the au designed for individuals in the 56–65 age range reduce the relative value of the defined benefit 20.7 years in Figure 12). Results for the 70 replacem lways be at least as ent ratio can th hile 57 percent e quartile at age en be com well off 2008, as a function of tenure w under AE (at least up to th p of those that have frozen th e ared with any s on how this 65 for those retiring between 2030 and 2039 (thus edian gains com percentile in each case show to-escalation feature for AE pplan in the future by the imposition of a pension e w n ge drops to 6 percent for those between ages 55 sation per year. Unfortunately, the m one of a set of previously com a in this ran has im s 51.2 percent at year-end 2007. That would pacted 401(k) participants by looking at g e 95 e are in excess of 5 percent for all e plan in the ith the cu percentile) regardless of which plans under five what is likely to resu rrent plan spo last two yea puted thresholds re s ndian sor. lt References Figure 6 (percentage points) After the passage of PPA provided design specifications for (safe harbor) automatic 13 33 Research Institute, 2003). guaranteeing the potential availability of 401( groups. 2.1 Av set of auto-escalation assum the change in average 401(k) balances as well values f indicated they have already a different sets of assum freeze. for those with either very large equity allo and 59. to provide insight as to whethe imply that approxim or single f erage acco em aa tely 43 percent of the consis les in the lowes ptiunt balances ons: pdopted 401(k) autom tions proves to be r the individual has adequate re t incom cations at the end of 2007 or those with low ek) plans during their entire working career). as the expected tim quartile would require in correct. The sam tent sam atic enrollm ple participan s eources for retirem nte features, com e to recover from can be said of the second- ts in the age 56–65 age excess of 25 percent of pared with 45 ent security. 2008 losses The Income Quartile Current U.S. Voluntary Retirement Plans, W 401(k) Account Balances ith Special enrollment plans, further behavioral evidence wa th s required to parameterize the simulation m th odel Obviously, r The enhanced em ecovery times will b ployer contribution e a fs unction are accumulated b of what future m ased on the age-s arket returns p are assum ecific asset ed. T-162 repla percen contribution-to-account-balance ratios. In both com as a function of future investm lowest salary quartile through the 90 category would have had at least a 20 percenta While the results in Figure 23 are specific pcem ensation in additional savings regardless of the age cohort. t of those who have not, the ph ent rates vary from slightly more than ½ ent returns. enom percentile and the th enon is to the assum ge point reduction in e cases, under these assum a f reversed for those planning to freeze the plan or the lowe p ird quartile through the 75 tions m st incom entioned above, sim quities at year-end 2007 e quartile to slig ptions, the recovery percentile. htly m ilar o if re 1 2 3 4 401(k) Account Balances Among 401(k) Participants ________. “E The f While the previous figures illustrated that ta 1. Assum ollowing analys RISA At 30: The Decline ing 401(k) opt-outs, lim is in this sectio of Privat it of sa n focuses e-Sector Defined Benefit Pro fe harbor minim rget-date funds can ind on the percentage increase um, start over; eed m mises and Annuit a or decrease of those ke a substantial y with respect to how long 401(k) participants 2.1.1 Averages by age and tenur were likely to allow annual increases in e 28 Figures 11 and 12 differ in their assumptions allocations derived from Finally, Figure 22 shows the m year-end 2006 EBRI edian percentage of com for the non-equity com /ICI data (VanDerhei, Holden, Alonso and pensation required as an ponents (bonds, money 5.1.1 EBRI/Mercer survey of retirement program changes after PPA Emphasis on 401(k) Plans than 2/3 for the highest incom in the nex results are obtained (albeit with different pe Butrica, Barb Even the highest-salary quartile tim they were allocated 100 per es are so large as to effectively elim t two years. In ara A., Howard M. Iams, Karen E. S that case, o cent to target-date funds. e quartile under the baselines as does at least as well under AE nly 33 per inate the rcentage losses) under va m cent of those that plan to freeze their p itpossibility th h, and Eric J. It would appear that this situation Toder. “The at the p sum through the m ptions. a rious com rticipant will ev Disappearing Defined edian regardless of binations of ension have er recover b Payments: What Will It Mean?” EBRI Issue Brief, no. 269 (Employee Benefit Research Institute, balances m difference in balances at retirem Figure 2 pro The next section of the testim oving from vides a s participant-directed invest Present From Year-End 1999 Through Year-End 2007 imilar ana ent for som lysis ony reviews tw ; however, this tim e participants, another concern that was often ments to a type of investm o potential changes to e the retirement wealth goa the current system ent strategy that l requires a contributions to continue before opting out. As 21 24 part of the 2007 Retirement Confidence Survey m Copeland, 2008). All simulation results were base ENHANCED e arket, and stable-value investm mployer contribution for future year ents) of future market returns. Figure 11 assum d on annual returns data from s covered by a defined contribution plan in Ibbotson and es a nominal 2. Assum Benefit Pension And Its P ing no opt-outs, lim otential Im it of safe harbor pact Figure 11 On The Retire minimum m,ent Inco maintain contribution rates; mes Of Boomers.” NBER Assuming Always Have Contributions to 401(k) Plan adopted 401(k) autom the se changed m their 2008 losses. minim t of um guarantees an auto-esc arkedly by year-end 2008; however, it is ala In fact, using the lower rate of return tion atic enrollm d assum assum 34 p etions chosen. d asset ent, as opposed to allocations and rates of return. likely that m 46 percent of those that do not plan to freeze 9.1 in Figuo re 12 resu st of th 8.9 e change is due to lts in a situa 6.5 tion in 4.6 At year-end 2007, the average account balance among all 21.8 million 401(k) participants in 2004). Figure 12 m that cou sufficient amount to achieve a 90 p expressed (after QDIA regulations akes use of target-date funds. It is im ld o portant to note that under the baseli ptimize benefits and m Target-date funds are often chosen as the default investm w itigate risk. T ercen ere proposed) dealt with the t probability ne assum he fi that th rst f ptions, a worker ocu e basic retirem ses on the impact of potential im who is currently a ent expenses will b pact switch on participants ing ent e In April 2007, EBRI and Mercer fielded a survey designed to elicit information from a (RCS) information was collected that allowed behavioral estimates to be made (Helman, 9 annual rate of return on the non-equity portion of Associates (2009). Tim lieu of a frozen defined benefit e series data for the plan for financial indem years 1926 through 2008 were used for Large Cap the portfolio of 6.3 percent, while F nification. The majority of the em igure 12 ployees Working Paper. No. 2009-2. Chestnut Hill MA: Center for Retirement Research at Boston Time Needed to Recover From 2008 401(k) Losses, a Using Various Equity Return Assumptions the plan in the next two years. However, 42 perc which, m market fluctuations, as opposed to particip athematically, the participant would ant transfer activity. never recover (infinite ent of those planning to freeze their pension in The 2008 asset allocation recovery time). the EBRI/ICI 401(k) database was $65,454 (VanDerhei, Holden, Alonso and Copeland, 2008). Time Needed to Recover From 2008 401(k) Losses, Using Various Equity Return Assumptions 3. Assuming no opt-outs, limit of safe harbor maximum, maintain contribution rates; 401(k) participant is assum strategy for em 401(k) plans from m who were likely to cash out their 401(k) balances et for the entire retirement peri ployees who are autom voluntary to automatic enroll ed to continue to work for 401(k) plan sponsors each tim od. As expected, the m atically en ment for all US workers (not just 401(k) rolled in 401(k) pl at job ch edian a ange rath dditional savings requirem an er th s given that thes an roll theme over to an they e ents 4.2 Target-Date F Mercer’s retirement business contact list on retir unds ement program changes after the adoption of the b Copeland and VanDerhei, 2007). However, information on two other actions was not available: cuts that assum under age 30 can be financially indem Stocks and Long-Term ption in half, to Corporate B a nominal return of 3.15 percent. onds to sim nified with ulate an em the portfolios of all 401(k) participants. At ployer contributi Five different panels showing a on of only 6 percent of Assuming Loans Are Never Taken College, 2009. b and a 6.3 Percent Non-equity Return Assumption ________. “The Impact of PPA on Retirement Income for 401(k) Participants.” EBRI Issue Brief, no. 318 the next two years are currently consider estimates in Figure 16 suggest that only 15 percent of the oldest ing 401(k) automatic enrollm 401(k) participants (ages 56–65 ent features. However this average varies substantially with the participant’s age and tenure. Figure 5 shows and a 3.15 Percent Non-equity Return Assumption change jobs. individuals often do not exercise participants and eligible non-part increase for all groups not previous IRA or retain them If a m in a 401(k) plan. Figure 19 s uch less optim th icipants). The second summarize istic assum e effective co ly capped at the 25 percent lim ption hows the expected im ntrol required is made (i.e., each tim for e it. s the likely im mployers to benefit from pact e there is a jo on these individuals of pact of a m b change, ove Pension Protection Act o An Evaluation of the Adequacy and Structure of Current f 2006 (PPA) and the new FASB accounting rules (VanDerhei, 2007). Ffrom 401(k) Plan Aaccount range of retu com age 65, all accum Introduction A policy qu 4. pensation; however, the num Assum rns are p ing 401(k) opt-outs, lim estion that h ulated account bala resented in bo as repeated ber increases to th figures nces attributed to the en it of safe harbor ly surfaced , one for each of the following nearly 16 percent fo since the financial m maxim 0.4 hanc um,ed contributions are converted ma 0.3intain contribution rates; r those over age 60. arket cris equity retu 0.4is began rn 0.3 is (Employee Benefit Research Institute, June 2008). in 2007) had 90 percent or m Another concern is the vulnerabil ore of thity of 401(k) participants to vol eir 401(k) assets in equities. Another 5 percent had 80–90 atility in the equity markets, that for participants in th • When 401(k) participants changed jobs and eir 60s the average account balance for an indi began participation in a new 401(k) plan, vidual who has been with Panel A: Equity Rate of Return: –10 percent 18 Choi, James J., David I. Laibson, and Brigitte C. Panel A: Equity Rate of Return: Madrian. “Plan Design and 401(k) Savings O –10 percent utcomes.” the em ERISA Sec. 404(c) protection with respect to from moving from participant-directed participant-directed inve ployee has only a random stm chance of worki investm ents to target date funds. ents to target-dat potential liability exposure resulting from ng for an employer sponsoring a 401(k) plan), e funds, as a function of the Employers that sponsored defined benefit pension plans in the United States were asked to $137,430 assum to nom whether th pinal annuities (for consistency tions: –10 percent, –5 percent, e impact will be disproportiona with th 0, +5 percent, and +10 percent. tely e def born ined benefit accruals) u e by the lower-paid employees. Va In addition to showing sing gender-s nDerhei pecific The first two figures only show the additional savings required for the median worker in 5.1.2 Impact of freezing defi Percentile of 401(k) Participants ned benefit plan accruals for new and this d percent in equities, and 9 pe U.S. Voluntary Retirement Plans, With Special Emphasis on eals with extreme equity concentr rcent had 70–80 percent in equiti ations—especially for older employees. Figure 16 es. Aggregating these three the current employer at least 30 would they remember what contribution rate years (and hence the likelihood of they had in the old plan, or would they substantial portions of 401(k) 5. Assuming 401(k) opt-outs, limit of safe harbor Percentile of 401(k) Participants minimum, maintain contribution rates, 5.2 Should DC plans be amended to NBER Working Paper. No. W10486. Cambridge, MA: National Bureau of Econom permit sponsors to invest funds ic Research, Average Assuming Preretirement Withdrawals Are VanDerhei, J the replacem investm employee’s tenure on the job. The median im The Em ent losses. They put em ack, and Kelly Olsen. “Def ent rates drop to 23.2 percent for the ployee Benefit Research Ins ployees into asset ined Contri titute (EBRI) pact bution Plan Dominance Gro allocation lowest incom is extrem is pleased to assist the E ely sm s5 that are considered age ap e quartile all (1 percent or less); w and 27.8 percent for the s AcrosR s Se ISA Advisory ctors and propriate complete the survey. Although similar types of surveys had been conducted earlier, this survey Job Tenure 10th 20th 30th 40th Median 60th 70th 80th 90th annuity purchase prices. the es (2009b) presents results for each job tenure timated recovery time for the median individual in each tenure and equity return group for six different salary groupings: $20,000– each cohort. Finally, two appendices are incl Figure 3 uses the same model and assum uded to summarize resu ptions but assum lts from studies on (1) the impact es each worker will save $122,037 Job Tenure shows, for the year-end 2007 EBRI/ICI 401(k) databa categories together, the percenta employ 10thees 20th 14 ge of 401(k) participants ages 56–65 in 2007 with m 30th 40th Median se universe, the asset allocation distribution 60th 70th th 80th ore than 70 90th generated wealth being rolled over to an IRA or cas start over? hed out are de minimis) is slightly more than May 2004. Employer Sizes, While Mega Defined Benefit Plans Remain Strong: Where We Are and Where th th th th th th th Never Taken From 401(k) Plan Account (years) (years needed to recover) 6.7 6.0 6.0 3.8 highest inco and then gradually decrease the equity exposur Council in its evaluation of the contributed by participants to bett however, the interquartile ra me quartile. W nge increases with duration, as expected, and the 75 here a pa adequacy and structure of the rticipan 401(k) Plans t’s s e as the em er ensure financial security and can imulated rep ployees approach their target-date current U.S. voluntary retirem lacement rate will ac pe tually f rcentile f all ent or had the advantages of being distributed at a much later date and the greater likelihood that the where: com $30,000, $30,000–$40,000, $40,000–$50,000, $50,000–$60,000, $60,000–$90,000, and greater bination, a distributional analysis is included to show the 10 , 20 , 30 , 40 , 60 , 70 , 80 , (years) (years needed to recover) an additional 5 percent of com of freezes on future private defined benefit plan pensation from the accruals and time of the sim (2) transferring part or a ulation (2003) until retirem ll of the ent percent of their 401(k) portfolio in equities had decreased from 43 percent at year-end 2007 to 29 of 401(k) participant account balances to “equity” by age, as of year-end 2007 and with an 1–4$210,000. — — — — — — — 0.1 0.9 We Are Going?” In Dallas Salisbury, Retirement Prospects in a Defined Contribution World. th 22 th 1–4(typically th those with 1 within this ra eir exp 1nge is a f or m——— o ected retirem re years w unction of ith the em e the percen nt date). If ployer exceeds 6 percent. ta one a — ge of their working career they are em ssum— es a positiv —e equity prem — iumployed by a going 0.1 1.0 plans. My testim Unlike previous applications of the EBRI/ERF As is typically true in the pr ony will review the results of ivate retirement universe, plan many em model, in this cas pirical and s sponsors’ reaction to influences $103,751 e th imue m lation s odule used to tudies EBRI has plan sponsor would have sufficient information for a detailed cost/benefit analysis of what types such sponsors r • Would plans sponsors tend to choose the m eceive fiduciary protection? inimum or maximum limit for than $90,000. and 90 percentiles a These findings show that, at least for s well. For example, the value fothe m r the 7 edian results, lo 0 percentile represents a tim wer-paid employees e investm age. This illustrates the overall Choi, Jam ent risk inherent in es J., David I. Laibson, Brigitt defined contribution plans from percentage of tim e C. Madrian, and Andr es that indivi ew Metrick. “Optimal Defaults and the em duals in each cohort are assum ployee to another entity. ed 5–9 — — — 0.3 0.8 1.5 2.4 4.1 9.5 percent at year-end 2008. estimate for 2008. Equity in this figure is defined as the percentage of the participant’s 401(k) Assuming Do Not Always Have 401(k) Plan Coverage 5–9 • 401(k) opt-outs——— denote that individua0.3 ls will opt ou 0.9 -27.5 t of future in 1.6 -30.8 creas 2.6 es as d -34.7 escrib 4.3-39.4 ed in the 10.9 Washington, DC: Employee Benefit Research Institute, 1997, pp. 55–92. $93,841 sim forward, it is likely that em 401(k) plan sponsor. ulate cash-out vs. rollover behavior for defi ployees (especially young em ned contribution balances ployees) who otherwise would have at job change was of plan m such as PPA likely will b undertaken to determ Even though the annuitized version of th odifications and/or invest ine whether future cohorts of re e quite varied. C ment changes were being considered. opeland and VanDerhei (2009) u is later amount would provide a significant tirees in the US are likely to have retirem se the results ent contribution rates? 10–19 — — 0.7 1.7 3.0 5.1 9.1 24.4 infinity will have sh period long enough to include the rec orter recovery times than their highe overy times of 70 percent of t r-paid counterparts, and in m hose in the tenure and equity any cases there is a 10–19 Active De—— cisions.” NBER Working Paper. 0.8 1.9No. 1107 3.5 4. Cambridge, MA: National Bureau of 5.9 11.1 38.4 infinity to have a suf Proposals have been suggested since the Enron de ficient amount of income to cover basic expenses for the entirety of their retirem bacle that would attempt to transfer part or ent. funds held in equity funds, company stock, and the equity portion of balanced and/or target-date empirical findings presented in VanDerhei (2007); 39 5. Appendices 20–29 — 0.1 1.2 3.0 6.0 13.1 72.3 infinity infinity effectively turned off for chosen a relatively low equity allocation woul published in VanDerhei (2007) to m income adequacy and the extent to which the vol any new jobs. W odify the E For presen hile th d end up with larger 401(k BRI/ERF Retirem untary retirem is is like tation at: ly to oversta ent system is contributing to this ent Security Projection Model te the eventua ) accumulations at l balances 20–29 retirement incom — e when com 0.1 pared with Social Secu 1.5 3.7 $80,592 rity ben 7.8 efits, it is importan 20.7 infinity t to no infinity te that the infinity 1. Retirement income adequacy 25 significantly return com This section reports on the results obtained Econom bination cohort (in ot shorter recovery tim ic Research, January her words, at that e for the lowest-paid 2005. using the EBRI simulation m value only 30 percent of category of participan that cohort would ts than the highest- odel to determine VanDerhei, Jack L., Russell Galer, Carol Quick, and John D. Rea. "401(k) Plan Asset Allocation, The percentages vary from all of the investment risk inhe 30–35 percent for the cohor rent in defined contribution plan t on the verge of retirem s from the employee to another ent in the lowest Assuming Never Cash Out Balance at Job Change 13.3 9.1 6.8 4.7 funds. 2.3 How much will automatic enro Figure 9 co The figure shows that 27 percent of young 401( mpares the median replacement rate llment help 401(k) participants? s agains k) participants (those 35 or younger in t the baseline in each of the six 40 The EBRI/Mercer survey provide Panel B: Equity Rate of Return: s several useful statis –5 percent tics with respect to the overall type 23 Panel B: Equity Rate of Return: –5 percent attributed to the enhanced contributions, it doe and provide additional analysis retirem objective in its cu ent with a target-date f rrent form as well as possible und as opposed to participant-dir to inform public policy on th s allow a cons modifications that m isten e lik ected investm t com ely im ay increase its efficiency. parison to the def pact of continued trends ents. However, ined proposed regulations for 401(k) plans were not released until November 1981, and it took several • No opt-outs denotes that individuals will not opt out of future increases until they reach $66,660 $66,677 paid. have recovery tim how target-date funds would lik Account Balances, and Loan Activit es greater than that am $65,936 ely impact 401(k) participan ount). y," EBRI Issue Bri This additional detail is important, due to the ef (E ts who are assum mployee Benefit Research Institute, ed to be 5. 1 The impact of pension freezes on future private defined benefit incom entity. Although the party initia e quartile to more than 95 percent for thos lly exposed to said risk vari e in the youngest cohort in the highest incom es among the proposals, the likely e com 2007) have 90 percent or m binations of maintain contribu ore of their 401(k) assets in equities tion rate Percentile of 401(k) Participants /start over scenarios by whether the contribution is (broadly defined). Another 13 $61,341 Percentile of 401(k) Participants and frequency of defined benefit changes and the association between whether a plan sponsor Copeland, Craig, and Jack VanDerhei. “ 29 The Declining Role of Private Defined Benefit Pension Plans: benefit accruals th older em with respect to defined b Although the 2002 sim ployees who would exhibit at would have res enefit plan f ulation study allowed one u relatively ris rlted eezes. Th but for the new pen e variations to k-avers to project the 401(k) e tens d the bas ion freeze s encies if th ic model used in this ey inves cenario. accumutlations of ed the assets years before many employers implemented them. Therefore, even the oldest 401(k) participants an employer-induced constraint; large degree of diversity within each 1.1 Adequacy for all U.S. families automatically enrolled. January 1999). It is important to note that target equity return/job tenure com -date funds use in 401(k) plans is not bination. Assuming Preretirement Withdrawals Are Job Tenure 10th 20th 30th 40th Median 60th 70th 80th 90th targets would be the em quartile. The testimony begins with a review of a ployer, a government agency (perhaps the Pension Benefit Guaranty national retirement income adequacy model plan accruals Job Tenure constrained by the safe harbor percent of this cohort have 80–90 pe 10th 20th min 30th imu rcent of their assets allocated m or ma 40thximum, Median or whether the full RCS distribution can 60th in this fa 70th shion, and another 11 80th 90th closes its pension plan to new Who is Affected, and How.” Spring workers or freezes the ac 2009 Pension Research Council Conference: Restructuring crual of pension benefits for current 30 them 4. Potential Changes t selves may find the higher volatility of the target-date f o the Current System T und results in a sm hat Could aller account participants under various scenar analysis are similar to EBRI's recent analys ios, it was not able to do a pr is of the potential im oper analysis of the potential pact of the PPA's safe harbor for (years) are unlikely to have spent an entire working career in the 401(k) (years needed to recover) environment. A later portion of Never Taken From IRA Balances “Approaches for Retirement Secu 11.1 rity in th 12.8 e U.S.” 14.8 18.4 limited to those automatically enrolled; however, based on unpublished simulation results, it (years) (years needed to recover) Corporation), and/or a private insurance company. While the cost of the guarantees and/or EBRI constructed in 2003 to identify which groups of current workers are likely to have be used. Not surprisingly, the m percent have 70–80 percent alloca axi ted to equities. mum impact is seen when the full RCS distribution is used 35 • Retirem Retirement Risk Management in Safe harbor minimum ent income adequacy can be defined in denotes th a Defined Contributi at employers will lim a numon Worl ber of ways. Som it the a d. utomatic inc e have focused on reases to 6 Two additional assumptions were used for this analysis: (1) all defined benefit plans are 1–4VanDerhei, Jack, Sarah Holden, Luis A workers, and several employer and — — — pllonso, and Cr an-specific characteristics. — aig Copeland. 2008. “401( — — k) — Plan Asset 0.1Allocation, 0.8 1–4balance a significant percentage of For example, in panel D of Figure 11 (+ ——— the time (but less than 50 percent). 5 percent equity return assum — — — p— tion), the median 0.1 tim 0.8 e this paper describes the various sim 1. 2 Alternat autom benefits of autom atic enrollm aent and autom ive asses tic enrollment for 401(k) plans saments of tic escalation ulation models adequac on the likely account balances for 401(k) due to its inability to m that have been constructed to provide m y odel those workers who ore appears that 401(k) auto The dawn of the new year in 2006 began with a flood of news -enrollment will represent the majority of target-date fund use in the reports about the supposedly Optimize Benefits and Mitigate Risk financial uncertainty inherent in such an arrangement may be borne by the employer at least sufficient financial resources to meet basic expenses for their entire retirement and, for those that 5–9(without constraints) and the — — maintain contribution rate — 0.3 0.7 scenario is assu 1.1 m 1.7 ed. In that case, th 2.6 e 4.5 whether retirees will have suffici percent of compensation; ent financial resources to be ab le to generate a standard of th 5–9currently treated as thou ——— gh they were final average plans, an 0.3 0.7 d (2) only private-sector workers 1.2 1.8 2.7 4.8 Account Balances, and Loan Activity in 2007.” EBRI Issue Brief, no. 324 and ICI Perspectives Memo: to recovery for an individual in the highest participants (VanDerhei and C were eligible but chose not to participate unde opeland, 2008); however, there is one m job tenure category is 1.8 years. However, the 10 r a voluntary enrollment situation. The EBRI/ICI ajor change with respect information on the true wealth accumulation potential for 401(k) plans. 10–19 fut “new” trend am uAlthough many asset allocation m re. — ong private defined b — 0.6 eodels and/or fi nefit plan 1.3 sponsors of “freezing” th nancial advisors m 2.1 3.0 ay suggest that extrem 4.3 eir pension plans for 6.6 13.5 e 10–19 Copeland, Craig. “Use of Target-Date —— 0.6 Funds in 401( 1.4 k) Plans, 20 2.3 07.” EBRI Issue Brief, no. 327 3.3 4.7 7.3 15.8 do not, how much additional savings would be requi initially, it is Given the in unlike credib ly tha le rta , in th nge of e long asset a -term lloca , su tions ch a shift in risk-bearing would not som because of individual red to meet this objective either 75 or 90 participant investm ehow alter ent Survey respondents were asked to indicate what Although the EBRI/ERF Retirement Security Projection Model™ allows a variety of public changes they have made, or expect to make, median replacement rate for the lowest-income quartile increases by 28 percent, while the rate living or consumption stream similar to that available imm b ediately prior to retirement after were modeled (and if a worker is currently in (December 2008). the private sector, it was assumed they would Median Replacement Rates for Typical 401(k) Participant 4.1 The shift from voluntary enrollment to automatic enrollment and 20–29 percentile is zero (no recovery ti to cash-out/rollover behavior at job 401(k) Accumulation Projection m — 0.1 1.0 meo ), due to the f change that is explained below. del was m 2.1 odified act that at 3.3 to allow for the inclusion of synthetic 50.7 least 10 percent of the 401(k) 5.1 54.0 8.0 59.5 14.7 67.2 63.2 20–29 — 0.1 1.1 2.4 3.9 6.0 9.7 18.7 214.6 • Safe harbor maximum denotes that employers will limit the automatic increases to 10 current or new workers. In reality, these decisions have been quite U.S. Department of Labor prevalent in recent years, and concentrations in equities for (Employee Benefit Research the young cohorts would be accep Institute, March 2009). table, few would recommend it percen direction, it should not be surprisi the provisions of the existing t of the time. This is followed by the re defined contribution plans. ng that the adoption of target-d sults of another retirem ate funds has a large range of ent income adequacy to their defined contribution plans. One-third of policy scenarios to be modeled, it has only limited ab the defined benefit sponsors expect to m ility to assist individuals in terms ofa setting ke an for the highest-income quartile increases by 12 percent. Even for the scen 19 ario with the sm th allest adjusting f 2.1.2 Consistent participati or the differential impact of items such on numbers as taxes, savings, work-related expenses, and rem The sim ain there until age 65 ulation model sta ). The first assu rts with all workers, whether or not th mption reflects an upper bound on the expected ey are currently enrolled in a Panel C: Equity Rate of Return: 0 percent Panel C: Equity Rate of Return: 0 percent participants in this eligible nonparticipants category were estim (Holden and VanDerhei, 2005). The ated to have no losses in 2008 first set of colum and the 90 ns in Figure 8 percentile is the impact on 401(k) accumulations for all workers (including those percent of compensation; are part of the well-documented and long-term decline of “traditional” pension plans; what’s VanDerhei, Jack. The Role of Com for those approaching retirement. p Ne any vertheless, the 2007 asset allocation infor Stock in 401(k) Plans. Written Statement for the Hous mation in Figure e Ways analys different outcom is assuming that all cu es. Figures 17 (for participants rrent workers save an additional five percen younger than age 45) and 18 (for participants age t of compensation each increase in employer m targets for what multiple atching contributions, and s of final earnings they will n 20.9 percent expect to m eed to save to have sp ake an increase in ecific probabilities of expected impact (start over and limited by the safe harbor minimum), the lowest-income quartile age-specific expenditures. EBRI has taken a 401(k) plans, and tracks them through age 65 by st Percentile of 401(k) Participants somewhat different approach, focusing instead on ochastically assigning job change, whether the Source: Tabulations from EBRI/ICI 401(k) Accumulation Projection Model. reductions in future retirem For purposes of this analysis, the m ent wealth for m odel assum o Percentile of 401(k) Participants st cohorts of defined be es that no CURRENT e nefit participants (see mployees will be Advisory Council on Employee Welfare and Pension Benefit Plans DiCenzo, Jodi . “Behavioral Finance and Retirement Plan Contributions: How Participants Behave, and estim shows that the m ated to take 4.9 years before edian replacement rates d their 401(k) balances are expected rop substantially for the firs to be equal to their January t income quartile when the 6 Although the EBRI/ICI 401(k) database has been publishing average 401(k) balances since unusual is th and Means Committee e large size of som Hee aring on Retire of the employe ment Securi rs that ha ty and Defined Contrib ve recently announced pension freezes ution Pension Plans not currently eligible) 16 shows that almost a quarter (22 percent) of the oldest 401(k) participants (ages 56–65 in a 45 and older) show the m 1999 It is obviously im 2000 possible to m 2001 edians and interquartile odel the fi 2002 nancial consequences of such proposals until 2003 range for the percentage increase in balances 2004 2005 2006 2007 Job Tenure year until retirem 10th ent. The results of an addi 20th 30th 40th tional sim Median ulation model are presented from the 60th 70th 80th 90th non-m having sufficient income to cover basi atching employer contributions. A total of c expenses for their en 42.5 percent of the defined benefit sponsors tire retirement. Figure 4 show the Job Tenure still experiences an in • Start over 10th denotes that workers will start over fro co20th me replacem 30thent rate 40th increase of 11 percen Median m the default contribution when they 60th t and the highest-incom 70th 80th e 90th the probability of whether retirees will be able new employer sponsors a 401(k) plan, cash out beha to meet certain vior, and financial m minimum e axpenditures, including rket performance. In Change in median replacement rate for 401(k) accumulations relative to final five-year average VanDerhei (2006) for a detailed analysis of the impacted by a defined benefit plan freeze (in ess various defined benefit plan types). T ence, all freezes will impact NEW employees he second Prescriptive Solutions.” EBRI Issue Brief, no. 301 (Employee Benefit Research Institute, 2007). 1, 2008, level (in nominal terms). eligible nonparticipants are included (cf 50.7 percent in Figure 7 with 23 percent in Figure 8); (years) its inception in 1996 (VanDerhei, (February 26, 2002b). 16 Galer, Quick and Rea, 1999), it is (years needed to recover) important to note that year and the frequency of the announcements. 1 2007) had 90 percent or more of their 401(k) as (years) (years needed to recover) sets in equities. Another 10 percent had 80–90 moving from participant additional detail is provided; how direction to target-date funds, with ever, a highly stylized example the relative gains displayed as a of one method of achieving this standpoint of individual financia type of projected multiples that are p l planning to demonstrate how pos roduced from the Employee Benefit Research In t-retirement investm stitu ent risk, te indicated that they would in quartile increases by 5 percen crease employer contributions. t. medical expenditures that ar change jobs; and e not covered under Medicare and/or Medigap policies. salary. This is the first-order difference and does not take into account changes in participant assum addition, the EBRI/ICI 401(k) database is used ption was required as a result of the survey population used to collect the enhanced to statistically impute asset allocation under only). Each tim As it is far too soon to analyze what percenta e an employee is simulated to have ge of 401(k) sponsors with voluntary enrollm a job change, the probability that they would ent 1–4 — — — — — — — 0.10 0.64 1–4however, the im——— pact is much more muted for the high inco — — me quartile (56 percen — — t vs 0.1 . 67.2 0.7 to year differences in cro ® ss-sectional averages do not provide a true measure of the changes due percent in equities, and 11 percent had 70–80 percent in equities. function of the participant’s in objective can be readily sim longevity risk and the possibility of potenti ulate itial equity allocati d. Assumeally catastrophic nursi a proposal that w on. Obviously, the prim ould ng hom require the em e and other retiree ary advantage of ployer to ensure Fronstin, Paul, Dallas L. Salisbury, and Ballpark E$timate Monte Carlo for high incom Jack VanDere hei. “Savings males retiring at age 65 assum Needed for Health Expenses in ing 100 percent behavior that might occur as a result of changing the activity in question. 5–9 — — — 0.23 0.56 0.92 1.36 1.92 2.93 5–9contr (VE) will ad participant-directed baseline scenarios. be covered by a defined benefit plan is com The choice between the two non- ibution inf opt an auto ——— ormation in the EBRI matic enrollmequity return assum /Merc eSeptember 17, 2009 nt (AE) e put r su 0.2 approach, sim ed based on the assum rvey. 0.6ptions (nam ilar to VanDerhei and Copeland 1.0 ely, Figure 11 or Figure 12) ption that defined benefit 1.4 2.0 3.0 ________. “Pension Plan Surplus: Revert, Transfer, or Hold?’ 10 EBRI Issue Brief, no. 88 (1989). While it is obvious that pension plan freezes affect some workers negatively, it is not obvious percent). Beginning with a series of state-specific retirement incom 7 e adequacy studies funded by the to cash flows and investm • Maintain contribution rate ent returns. denotes that workers will r Figure 6 shows the ave erage tain th (and m e deferra edian values) of those l level rate from b The most important association tracked for those defined benefit sponsors increasing their 3. What Happened to 401(k) Participants After Last Year’s 10–19 10–19 Retirement: An Exam ——— — ination of Persons 0.490.5 1.03 Ages 55 and 1.1 1.57 1.765 in 200 2.15 2.3 9.” EBRI Notes 2.84 3.0 , no. 6 (Em 3.80 4.0 ploy 5.70 6.1 ee target-date funds when viewed in that participants receive an medical costs can be quantif account balance no less than what would have been obtained under a ied in an attem this context is the e pt to provide a dollar target depending on the level of xpected gains for those with an initial low equity allocation in retirement and no annuitization. This figure shows that for this individual, The ratio of the income generated in the first year of retirement from 401(k) accumlations to final (2004), this analysis m plans have not been frozen. The cumulative value appears to be of relatively m odels the scenario in which inor consequence as long as the equity rate of of all defined benefit accruals for N all VE sponsors switch to AE. The results return assumption is EW jobs is 11 which workers are affected, nor to what degree they are affected by a pension freeze. There are Target-date funds with automatic rebalancing and a “glide path” ensuring “age-appropriate” 20–29 20–29 Milbank Me — morial Fund — 0.06 0.1 in the early part of this d 0.820.9 1.561.7 ecade, EBRI expanded th 2.32 2.6 3.18 3.5 4.26 4.7 e simulation 5.85 6.4 model to 9.01 9.9 participants who have been w Although the m the previous job. odel produces several output m ith the same employer in th etrics, the one of m e EBRI/ICI universe since 1999. ost interest for this This employer contributions to a defined contribution plan is whether they recently closed their Benefit Research Institute, June 2009). five-year average salary (percentage) for the baseline model. equity allocation (of less than 30 percent). W m conf ________. “ ini Copeland and VanDerhei (2009) show the expected The 2005 study was perfor idence that is d mum rate of return. While som Measuring Retirement Income Adequacy, esired in rem tired a year prior to th eme ent. employers hilem som a Part One: Traditional Replacement Ratios and y choose to voluntarily assum e passage of e financial adviso reduction in nom the Pension Protection Act of rs m inal replacem ay argue that less e the additional ent rates if only 3.3 times final earnings is needed in addition to Social Security to provide a 50 percent Market Decline? determ allow the u non-negative (i.e., either 0, +5 or ined for each em sers to determ ployee under the assum ine the lik +10 percent). However, under a ne ely impact of these chan pti 31 on that no term ges by apply inated vested benefits are gative equity rate of return ing whatever relative many reasons for this, most importan Panel D: Equity Rate of Return: tly the unique characteris +5 percent tics and terms of each pension asset allocation are likely to becom Panel D: Equity Rate of Return: e much more common aft +5 percent er full implementation of the discussion is the ratio of provide a national retirem “401(k) accum ent income adequacy ulations” assessm divided by wage at the tim ent (EBRI/ERF Retirem e of retirem ent Security ent— “consistent sample” of participants had a year-end 1999 average balance of $66,660 – this had defined benefit plan to new hires in the last two years (78 percent of these sponsors indicated that than a 30 percent equity allocation m cost of this arrangem Median Results for Workers at Lar ent, others m ge Com ay wish to ay be optim panies. re-think the investm EBRI Notes, al for those very close to no. 9 (Sept ent e options provided to the mber 2004) retirement age, it is all p 2006 (PPA) and therefore autom rivate defined benefit plans were to freeze accruals for N atic escalation of contributions was not accounted for. EW employees immediately, by chance of covering basic expenses for the full reti rement period. However, as one moves to the growth in the percentage of commuted to lum assumption, somep interesting differences take plac -sum distributions AE participants they think is m prior to retirem Percentile of 401(k) Participants e in the right hand tail ent age (which is currently assum ost likely to o of the recovery tim ccur. ed to be e Percentile of 401(k) Participants The second section of the testimony focuses on defined contribution plans, specifically Gale, W., L. Papke, and J. plan and each freeze, and the age an VanDerhei. In William d characteris G. tics of the w Gale, John B. Shoven, orkers. Van and Derhei (200 Mark J. Warshawsky 6) provides Pension Protection Act of 2006 (PPA), with an expected increase in automatic enrollment for Projection Model™) in 2003 (VanDe or, for purposes of cash-out behavior discussed later, the tim rhei and Copeland, 2003). Th e of job change. The ratio of 401(k) is model simulates assumed grown to $137,430 by year-end 2007. they would increas 3.1 Balances e employer contributions to the defined contribution plan) or plan to do so in em likely that th gender and age. The averages are less than ployees and provide little or no participan is will not be the case for younger pa one percent for employees who are currently young t direction. In fact, an easy rticipants. As can be seen in Figure 17, the way of mitigating the 75 and 90 percent levels, investm Moreover, the future predominance of life-cycl ent and longevity e funds as part of the QDIA option was not risk as well as the possibility of extended Job Tenure 10th 20th 30th 40th Median 60th 70th 80th 90th Job Tenure 10th 20th 30th 40th Median 60th 70th 80th 26 90th distributions. For example, in Panel A (–10 percent equity rate of return) of Figure 11 (6.3 age 65 for all employees). (ed.), Understanding the Shift Toward Defined Contribution Plans in The Evolving Pension 401(k) plans. A brief snapshot of average account balances as of year-end 2007 is provided for ________.“Measuring Retirement Income a detailed an Even for the m alysis of ho ost conservative set of w pension freezes are lik Adequacy assump:e tions for auto-escalation (second panel of Calculating Realis ly to impact existic In ting employees as come Replacem a function ent Rates.” 401(k) plans and the attendant in retirement wealth from defined benefit pensions, terest in qualified default inve defined contribution plans, and IRAs as well as stment alternatives (QDIAs). accumulations divided by wage can be a convenient proxy for retirement security by dividing the The analysis in this section focuses on employees currently ages 25–29. This serves two (years) the next two years (80.9 percent). Similar but slightly smaller percentages were associated with (years) (years needed to recover) (years needed to recover) positiv new risk im (under 25) and old (over 55 or 60, known and scenarios were m e resu posed by the m lts of target-date funds in the inim odeled with both m um guarantee woul depending on gender). They peak lower equity allocation range are m one d be to force all contributions (whether y market funds and life-cycle funds as the at slightly over 2 percent uch more stays in a nursing home begin to move the multiple to higher ranges: a 75 percent probability percent non-equity rate of return Most of the policy concerns with respect to 2 ), the me 3dian participant with 401(k) participan 20–29 years of tenure is assum ts in the last few months of ed System Trends, Effects, and Proposals for Reform. Brookings Institution Press, Washington DC, the entire universe as well as for various age and tenure breakouts. The results of a so-called of plan type and em EBRI Issue Brief, ployee dem no. 297 ( ographics. Employee Benefit Research Institute, September 2006). 1–4Figures 14 and 15), the AE plans res ——— th ult in 401 —(k) accum —ulations at least as larg — — e as the VE plans 0.1 0.6 1–4Social Security — and net housing equity. — — Figure 1 provides the cohort- — — — specific m — edians of the 0.1 8 0.6 purposes: (1) it indicates what the ma ratio by an immediate (real) annuity purchase price at retirem ximum impact of a change from ent age and then adding it to the VE to AE is likely to be in defined benefit sponsors freezing their plans to all members: Of those that had frozen in the last 20 contributed by the em pronounced with the 75 for m default investm Under the (h ales between 30 and 34, and 1.75 pe eighly sty nt. ployee or the em percen lized) assum tiles fp oployer) into a tion that a r those with less th rcent for fe ll prre m ivate d lativ ales between 30 and 34. an a ee ly r fi 30 percen ne isk-f d benef ree investm it pla t alloc ns will be ation in ent. W the hile this is would require a multiple of 5.0 and a 90 percen t probability would require a multiple of 11.6. th 5–92008 have focused on those close to retirem to need 6.0 years to r 2005. ——— ecover their 2008 losses, e 0.2 nt whereas the sam age. Figure 10 shows estim 0.5 0.8 e individual in Figure 12 (3.15 1.1 ated changes in 1.6 2.2 5–9 — — — 0.2 0.5 0.8 1.1 1.5 2.2 “consistent sample” of 401(k) participants is then tracked to show how account balances have for all four salary quartiles through and including the medians. At the 75 percentile, the AE additional percentage of compensation that would need to be saved (beyond that already 2.2 How much will 401(k) p the future, and (2) it allows refinem 24 ent of the re articipan suts accumulate under alternative lts with respect to additional percentiles in the (next) two years, 61.9 percent (76.4 percent) indicated they would increase employer th 10–19 —— 0.5 0.9 1.3 1.8 2.2 2.8 3.7 unlikely to be popular with young employees and immediately amended in such a manner that any new e other participants desiring high long-term mployees will not be able to accrue 10–19 One pub — lic policy co — ncern often0.4 raised, espe0.9 cially as the private-sector 1.3 1.7retirement sy 2.1 stem cont 2.7 inues to evol 3.6 ve ________. “Defined Benefit Plan Freezes: Who's Affected, How Much, and Replacing Lost Accruals.” The literature documenting the evolution from defined benefit (pension) to defined average 401(k) account balances from Ja percent rate of return) would need 7.8 years. Moving to the 60 n. 1, 2008–September 1, 2009, broken down by age and percentile in the same panel for m atured over tim The third set of e without the biases introduced by columns in Figure 8 illustra Submitted by Jack VanDerhei tes th sim at, with a 3 p ply looking at year-to-year averages. ercent default contribution and plans have higher balances than the VE plans for all but the highest salary quartile (again 20–29 contributed to def — ined co 0.1 ntribution p 0.8lans and/or 1.4 IRAs) by current worke 1.9 2.5 rs each yea 3.1 r u 3.9ntil their 5.1 distributional analysis as well as These numbers may seem relatively sm the impact of salary and numbe all but they are diffused over a large segm r of years participating ent of the in a 20–29 contributions. — 0.1 0.7 1.3 1.8 2.3 2.9 3.7 4.9 Hel scenarios? from m dan, Ruth, Jack VanDerhei, and Craig Copeland. efined benefit (pension) to defined contribution (401“The Retirement Sy (k)-type) plans is the p stem robab in Transition: The 2007 ility that a worker will en $76,946 d expected returns, it would minimize the new risks shifted to the employer. pension benefits, any employee selected by the m EBRI Issue Brief, no. 291 (Employee Benefit Research Institute, M odel to otherwise have been eligible for a arch 2006). tenure. Focusing on those on the verge of reti contribution (primarily 401(k)-type) retirement plans in the la rement (ages 56–65) m st 20 years is replete with studies c akes it clear that the c a life-cycle fund default investment, the lowest income quartile would expect a median increase reflecting the often-demonstrat Panel E: Equity Rate of Return: ed empirical observation that +10 percent high-salary individuals do not 41 assumed retirement age to provide a 75 percen Panel E: Equity Rate of Return: t probability th +10 percent at they will be able to meet th $68,866 e population that is not expected to 401(k) plan on the final balances. up with no 401(k) accumulations at retirem have a defined benefit accrual ent age. While many would argue from that the future jobs (this is 401(k) accumulations 32 Retirement Confidence Survey.” EBRI Issue Brief, no. 304 (Employee Benefit Research Institu 36 te, Although some individuals attem Research Director, EBRI pt to gauge the overall success of the 401(k) plan by defined ben efit plan (in the absence of a freeze) is assigned a non-elective ENHANCED If the individual is assumed to have defined benefit accruals at retirement, this may be added to the previous total changes, to a large extent, depe analyzing the change in the relativ nd on the participant' e composition of plans s tenure with the plan sponsor. W and participants; however, very few ithin this As mentioned above, simply tabulating aver Percentile of 401(k) Participants age account balances provides an inadequate of 82 percent in their replacement rates while th Percentile of 401(k) Participants e highest income quartile would basically have p benefit as m resen VanDerhei (2007) found that at least am ted in Fig uu ch from res 14 an a higher participation rate d 15 provide more substantiv ong the e evidence under AE plans, and at least som defined benefit pension sponsors that have of the likely overall impact of PPA o e of them n retirem end ent basic retirement expenses for their entire retirement. For those close to retirement at the time of ________.“The Expected Impact of Automatic Escalation of 401(k) Contributi $58,071 ons on Retirement particula April 200 rly true of 7). the older employees). Therefore, Figure 20 shows the expected conditional looking at these averages, it is im be treated as exercising control over the ass poretant to note that very few ts in the account with respect to the am (if any) individuals on the verge of ount of contributions and after m Figure 23 shows the expected results of runni ultiplying by the ratio of an immediate (nominal) ann ng one such proposal through the EBRI/ERF uity purchase price at retirement age divided by 15 Job Tenure 17 Figures 14 and 15 provide a deta 10th 20th 30thiled distribution analysis of th 40th Median 60th e difference between V 70th 80th E plans 90th Job Tenure have focused on the sizeable num group, average account b 10th 20th alance chan 30th ber of large plan sponsors that have had ges varied b 40th etween a po Median $51,927 sitiv 60the 29.8 p 70th eboth rcen defined benefit t for th 80th e short- 90th measure of the wealth accumulation potential of Employee Benefit Research Institute 401(k) plans, even if one focuses on the oldest the sam income from e m 4 e0 dian rep 1(k) planls, acem VanDe ent rate. rhei and C opeland (2008) demonstrate the likely reduction of workers with no These losses are defined as the difference between actual year-end 2007 and estimated 2008 account balances. It closed their plan to new hires in up with a lower contribution rate for a tim Income” EBRI Notes, no. 9 the last two years or are planning (Employee Benefit Research Institute, S e due to the inertia of keeping the default contribution to do so in the next two years, eptember 2007). the study, very few groups (defined by income quartile, gender and/or family status) would have reduction in nominal replacement rates if all private defined benefit plans were to freeze accruals immediate (rea earnings whichl,) annuity purc in the absence o hase f aprice at n invest re mtire ent el ment age. ection by the participant, are invested by the plan in accordance retirement have had the opportunity to participate in a 401(k) plan for their entire working (years) (years needed to recover) (years) Retirement Income Projection Model. Instead (years needed to recover) of allowing employees to direct their own $43,127 and AE plans with automatic escalation by salary quartile. (Figure 13 provides the same analysis 401(k) accumulations as a result of switching from voluntary to automatic enrollment plans. Whether one assumes and defined contribution plans in pl tenure individuals (less than five years) to m ace, at leas ore than a 11 percent loss for those with tenure of t since the advent of the 401(k) plan in the early Holden, Sarah, and Jack VanDerhei. “ participants with the longest should be noted that this includes est tenure. The EBRI/ICI 401(k) A imated Can 401(k) Accumulations contribution activity (as welGenera l as ccum other u te Significant Income for Future lation Projection Model was cash flows) for 2008 and is not a relatively large percentage have already adopted automatic enrollment in their 401(k) plan, and 1–4with rate—even though in this case it is assum regulations prescri——— bed by the se cretary of labor. T ed to be — he th increasing annually to ree types — of funds s —pecifically enum a 6 percent contribution — erated 0.1 for safe 0.5 sufficient retirement wealth to achieve this status unless they contributed in excess of 25 percent 1–4for NE 33 W employees immediately. In essence, th — — — www.ebri.org — is filters out anyone — — without a new defined — 0.1 0.5 careers. The results of two sim 37 ulation studies conducted by EBRI and the Investment Company 12 contributions and perhaps those of the employer, assume employers are forced to guarantee a serial correlatio VanDerhei (2004) re n in elig views ibility o how re r not, t placem he red ent rates ha uction in th ve t is p rr aob ditionally bee ability is strik n use ingd , esp to establi ecially fo sh m r tih ni e l mo uwest-salary m targets for for AE plans without autom limited to investmeth nt losses. atic escalation of contributions). Figure 14 provides results under the ____ 1980s. ____. “ For these sponsors, the prim Retirement Inco $30,737 4 me Adequacy $30,727 ary decision After PPA and FAS 15 in many cases is not whether to retain 8: Part One—Plan Sponsors’ both 5–9more than 2 Retiree 0 years. s?”——— EBRI Issue Brief (Employee Benefit Research Institute, 2002). 0.2 0.4 0.7 1.0 1.3 1.8 created in an attempt to project the “401(k) accumulations” that would be generated under 5–9har bor treatme— nt in the reg— ulations a re:— lifecycle (target 0.2 -date) fu 0.4 nds, balanced 0.7funds, and 0.9 managed acc 1.3 ounts. 1.7 a considerable percentage of those who have not $28,106 are currently considering it. Of those that have rate). At the 90 39 percentile, the two lowest-salary quartiles have larger 401(k) accumulations of compensation per year. Younger cohorts would obviously have the benefit of contributing benefit plan from the results in the previous figure. Now the mean reduction in replacement rates $24,844 Institute are reviewed to show fut qu See a urtile. If re V ret ainD rees fu erth u by e re elig i an cald Cop cul iba ility i tin eg land (2 tsh assu e am 003 m ount the potentia e) d to be a fun fo nee r a detailed ded to c l f ptio r d o on of v e r re i scr de i tirem cu p th tio e sam rren n of th e t eli nt wea e am g e EBRI/ERF Retir ibount ility as p lth accum of afatram er-ta e u x te lemen i a rized n tion of thes com t Secu ie n i th n ret is sectio rity Pro ire plans ement n jection , t as t he h at 10–19 minimum rate of return of 5 pe —— rcent nom 0.4 inal and 0.8 they are able to find a GIC (or its sy 1.1 1.4 1.7 2.1 nthetic 2.7 18 10–19 assum ption of serial correlation, whereas th Reactions.” — EBRI Issue Bri — 0.4 ef, no. 307 (E 0.7 e valu mploy es in Figure 15 assum ee Benefit Institute, July 2007). 1.1 1.4 e 1.7 future elig2.1 ibility is th 2.6 forms of retirement plan, but the liabilities of each in terms of future accruals or contributions. alternative scenarios (H 13 27 olden and VanDerhei, 2002). Some may question why any 401(k) participant would choose to continue to invest in equities if the assumed rate already closed the plan to new hires, 59 per under AE plans, but the two highest-salary quar cent have already adopted tiles do better under VE autom plans. At the 95 atic enrollment 42 20–29 This is explained — in more d0.1 etail in Cope 0.7 land (2009a). 1.1 1.5 1.9 2.3 2.8 3.4 M are m receive pr The ob odel ab re onotonically increasing with age: starti ility o d placement rates a priofr t ho a v retirem ing no ent after adjus 40 re 1(k de)rive balan d by ce for t ting for dividi hng is diffe g the projecte rorence upng at approxim drops s in savi d fro 401(k) m ngs 41 , a acc p ga e e, a tely 1.5 percent for em rcen um nul d work-relate t to ations 24 p by the a ercen d e t b g xpe y e-s switch nses pecific ployees . ing Howe (nom from ve inal) VE r, a 20–29 ____ under continuous coverage as well ____. “— The Influence of Automa 0.0 0.6 tic Enrollment, as several alternatives. In 1.1 Catch-Up, an 1.4 d IRA Contri particular, the results of a 2005 1.8 2.2 butions on 4 2.7 01(k) 3.3 equivalent) that will provide that return in perpetuity. If all existing balances and future 401(k) 3.1 Recovery times While this m of return were na ey not be considered to be an opt gative. While this would certainly seem im ual choice for som nlikely if the long-term e sponsors, after recognizing assumptions were negative, this Source: Employee Benefit Research Institute. percentile, the VE plans have la features in the 401(k) plan as opposed to 42 pe k an to e AE. nu y weakn ity pu If serial rch ess ao correlation se price and f many retiremen is i div ginored, th dit in ng co by rger 401(k) accumulations for al m th e d ee si m ifferen om de uls is th lated ce is salary at retirem at ev rcent of those that ha they use a en greaterv , era dren opp gt ag e estim ine. g fr l but the lowest-salary quartile. a ote m ve not. Plan sponsors s for 40 plife expectancy ercent to 16 pe, and, rcent. 5 1 28 Source: Employee Benefit Research Institute. 40 ____ currently ages 20–24, and increa ____. “The Impact of the Financial Crisis on Workers’ sing to 8.3 percent for those Retirement Security.” Testi ages 60–64. The m m edia onyn before the s are a Accumulations at Retirement.” (Employee Benefit Research Institute, 2005). sim ulation study to explore the likely advantages of automatic enrollment (AE) features in a contributions were required to be Ad Th It Bu i is d s trica, ition will also possi aIam l assu ble t s in , Smith hm at clu p stio d oe m n , and th e s are of t e po Tod h m te eo s ne e pa d tially catastro r (2 eled rt 009 in ic invested in this single invest i p ) u Van ant ses th sD p were erh hie M c co e i i (2 n o st vest d 004 s el oed i of f). nIn un rsing co co m m e ho p in any m th e st e care (at least m o Near Term ce k nt option, the average expected via empl to un oy si til th er m mulate th e po atchi in ng c e i t wh m op ere th nact of an tribut e ions Losses are defined as the difference between year-end 2007 and 2008 account balances. This is NOT limited to investment loss. a analysis is attempting to conduct sensitivity analysis on the possible short-term consequences of various equity certain legal and/or financial constraints, such as the inability to terminate an underfunded Losses are defined as the difference between year-end 2007 and 2008 account balances. This is NOT limited to investment loss. consequently, provide workers with only a 50 percent chance of having adequate income in retirement. 14 9 25 b For the m indicating that they will ost generous set of assum close the plan to new hires in the ne ptions for auto-escalation (fourth pane xt two years have adopted autom l of Figures 14 and 15), atic 21 There has been considerable discussion as to what the current market downturn might do to significan House Education and Labor Committee (2008). tly lower than the means, as expected, and increase until they reach 4 percent at ages en th Th at tity is assu we e year-end re notmed a b 2l008 e to b to figu be e elig di res will b vers ible if fo ied. r Med e availab icaid). le in a joint EBRI/ICI publication in early October. "Non-equity" meaning a bond or other stable-value investment. 401(k) plan are discussed. This is followed by accelerated tra It sh See ro ouw ld b 4 ien no n F sition from ited gure th7. at th DB to e results in DC pe thnsi is fi ons on t gure are he no distribution t d a review of the results from a 2007 enhancem irectly co of retire mparabm leent in with com Figu er a e m 4 ong boom in VanDer eh rs in a ei (200 8) ent . In b 6 retu It s rnh assu ouldm be p n tio on te s. d that the participant and/or the employer may increase contributions to a higher percentage of "Non-equity" meaning a bond or other stable-value investment. This assumes no defined benefit payments are available in the form of an annuity. The full model adjusts for these c pension plan (with the exception of certain sponsors satisfying the bankruptcy conditions 34 Ibbotson enrollme Associates. nt features 61 percent of the tim Stocks, Bonds, Bills and Infl e, in c ation: 200 ontrast to only 39 percen 7 Yearbook: Market t for those that do not Results for 1926–2006. The historic equity rate of return on equities is about 10 percent per year. the AE plans result in 401(k) ac retirem 1999 ent ages. The decision-m 2000 2001 cumulations at least as larg aking process 2002 undertaken by individu 2003 2004 e as the VE plans f 2005 als or households to or all f 2006our salary 2007 c scenari th Due to s e earlier pub o in whi pace c licatio ch em onstn rployers aints, anal , equity co freeze all rem ysis in this sect ncentration asining were m ion is prieasu v li ate-se mited to the red ctor for t DB he co com plans nsisten paris and a thi on of “a t samp rd of a le o vera f particip ge ll state and l ” target-date ants d oe cfi a fund in l plans ned earlier. over 55–59 and then drop slightly. 10 29 2 15 compensation in the future. This contingency is not included in this analysis. of the m 19 odel to analyze the likely impact of automatic escalation of contributions in 401(k) plans The historic equity rate of return on equities is about 10 percent per year. The views expressed in this statement are s olely those of Jack VanDerhei and should not be attributed to the accordingly. Th See Th For e e figures e reason Van xam D ple, a 40-year-old e for th p rhe resen i (2 is fi ted in 009a nd ) t in fo hg is r c wi with a repo oth un resp trt assu er$1 fact ect to the h 00ual m ,0e 0 e th 0v acc at So ideiou n gh ce cial nt est i on bal Secu h na come q o nce w rity b 40 wh 1( ue o c a k n rtile is th )e acc h fits are ano gunt ed at th jo n bal o bs t m a ey an nc oh d d es a irol fied ve little to wl oul ed t . Fo d hhave e r add bal ga g a in from ition nce row t a no l an f a rth om nalysis e IR A 38 th th necessary to trigger pension insurance coverage This may be due to a number of factors, but in most cases it was by the Pension Benefit Guaranty Corporation, or either a function of a large contribution-to- ________. “F Chicago: Ibb rozen Pensions and otson Associates, 2007. Falling Stocks: What Will Happen to Retirees' Incomes?” Urban plan to close the plan in the next two years. t term By he ne ds o efin xtf equ fi ition ve y ity allo , p ears. articip catio ann ts wou ; howev ld n er, VanDerh eed to be ine i (20 the p0 lan 9 c) in at lclu east sev des sensitiv en years to ity an balysis fo e in the co r bo nsth isten the m t sam ost agg ple. Th ressi is v will e quartiles in every ca determine their retirem se, with the ex ent age(s) is extrem ception of the highest-salary quartiles for the 90 ely complicated and the actual impact of a sudden and 95 22 Em with AE features. 19 would show sho 9p w 9l– io n 2 y g t 0 ee Ben 0h 6e i , iup as a n m cl epact fit Research usiv of va 41-year-old with a e, had t rioh In uese bal s S stitu ocite (EBR al ances Sec mu u bch sm ri I), th een i ty ref e n aller balanc vest o EBRI Edu rme pr d i on pos th catio e the ree als, see di nfollowing year. an ffer Va d Research Fu ent nDe types rheiof anTD d C ndFs , any o opel (avera and f its prog ( ge, a 2003) ggr . ram essi s, o ve, a fficers, nd automatic enrollment features per se, given that their participation rates are so high in the voluntary enrollment 7 42 account-balanc Participants with e ratios salaries less th or a very conse an $20 rvative ,000 asset allocation. were excluded in an attempt to deal with part-time employees. PBGC) and the im Institute Presentation, Febr position of a 20 uary percent or 50 percent excise 3, 2009a. tax on the recoupment of excess pr an Resu The o d m vid com o lts for females as well as altern e si st co gni putations nfi ser cant vativ bi ass e targ as i un m tet-d e h a long-term e eq ate fun uity conc ativ ds as av e retire ent w erag e rat ll. i e retu m ons A en lth f t ag o rn ough r t of 1 es he , th equi y 1o e r p ue nge te rcen y su al st lts in lt for bo co ocat hth o ions ris p ts. th a a a d n pd e an ir all assu ven rsified uitizat m p ion e o b rtfo a perce selin lio an n e r ta d a ges ca an te of r n etur be n 41 Figure 21 shows the percentage of those with "lost" DB wealth due to a pension freeze who percentiles. drop in equity prices on retirement behavior will take years to analyze. However, as a convenient trustees, sponsors, other staff, or any other individual or organization. The Employee Benefit Research Institute is a situ conservative) instead. The atio tn erm . Howev "enhanced" er, unlik ine the lo cludes twest in hose wi co thme qu no adartile, ditional th em eir con ploy tribu er co tion ntri rates un butions t do e t r a hevolu defin ned c tary en ont ro rill bmen ution t p pl lan an 11 3 Source: Tabulations from EBRI/ICI Participant-Directed Retirement Plan Data Collection Project. 20 23 Park, Young. “Plan Demographics, Participants’ Saving Behavior, and Target-Date Fund Investments.” i assum fndi Th oun Th vi e figures d e add dual p in tions (s V itio st an oc D p n of sev r ee k er esen but hPar ei ( a ted in k 200 st e,ral 2 an 0 6) t 0 d dar h 9 a . is ta , fo drepo prov devi r details), res rt assu id aters i ion of m n e 1 19 th 999 ults .6 at hou pe m foa r rce k asing e ln ts it d etrnat f equ or t ifficu iv h ity is n e ret e flt to orm urn e start er co v ass er li u th m m qp u e con p ared wi iti dons a ated sisten . For r te h 6 pr t sam 5ovi ad perce d ded p ition le in n itn a f l an prev Va or tn alysis hD ie l oer ua shei t years ter. I 26 a Ev Then ere are sev th These sim ough th eral po ey are assu uten lation results suggest tial exp med to b lanation e sus ffering for th that a significant percentage is resu relativlt, ely h but th eave m y loo sses st lik oely is th n their equ at h of 401(k) participants are ity in igher-paid vestm ind ents, th ividu eir als h non-equ ave a ity are expected to have a larger total nominal replacement rate from the DC-enhanced contributions Account balances are participant account balances held in 401(k) plans at the participants' current employers and are net of pl nonprofit, nonpartisan, education an d research organization established in Washingtonan loans. Retirement savings held in plans at , DC in 1978. The testimony proxy for participants with a vast m were (ap The pr co o De xi nsi m part a dte er lm a yb 2 elnt y 1 hi o perce fg La her n b t t o h of t ra in ss t hu h e ed e defi defa final ned ul re t be rat gul an jority of e eat o fiif tons pl 3 perce an fo sp r their non-Social Security retirem qu o nn talified .s If ors t i hn e i d te h nfau e s ertu ilt invest a r i vm eypact were m pro e i nn t altern pose thisd c ativ a by te C go es (QDIAs) eh nt wealth in ry oi – , et in al ot ., her 20 on w 0 5o a rn dd s, 30 ________. “The Impact of the Recent Financial Crisis on 401(k) Account Balances.” (Employee Benefit sh have with (2 C 0 owing 09c o ou Other com ar pel ). t loss bi a EBRI Issue Brief, alternativ tn rd ari (o 2 lf y 0 sub assum 09a be assu inations of s) tan pr eo tial num d al v m id p les n tio no. 329 ( s one nib as s gni e , see VanDerh q rs o u sum fi i tca y f p i nt E nvest p a m det rticip tions for auto- ploy am ilan on e eee Benefit Research Institute, M nt i and ts. s earn the Cop di an ffe ere e lan ann sca nd ces. u l(2 ation resu al003 rate ). of retlt in interm urn of a 6 ype 2009) rcent ediate va .. See Van lues b Derhei etween (2002) higher 8 investm ratio of account balanc ents are assumed to be earn es to a ing nnua 6.3 l c percen ontributions tha t per year. Th n do their lowe is, coupled with r-paid esti com unte ated rparts. T contribu his m tiona activ y be the ity o res f th ult e previous employers or rolled over into IRAs are not included. d like raws ly to be able to repla heavily from research p ce at leas ublicationt 7 s of th 0-80 e Em percent of their preret ployee Benefit Researchirem Institu ent incom te, but any errors e fromor a (if any). As expected, young em 2 approxim Octob EB 006R . act e I r 24 has al u aa tely , 200 lly p so 4 out of 5 plan pe r 7, to pro evrf ails am orme vd si io dn eg , m s sponsors i th in uter alia, em le ath ions ighly com t pl hn at oyees have the highest percenta t p h fo le s o p cus s y en ers urvey sated p wh eci who o adop in fical a had froze bro lyt au on ad t t cross-sectio o hm e am n atic e or ou were n nt ro o llm n pla f m o ge, with nearly 40 percent of en f p nni oney t p lan ng lan s, tto freeze t hs a safe th ate relativ woulh d be a he rb eir redu re or defi qui fro ctio ned red m n in to 36 b The analysis is based on a sample of 2.4 million participants with account balances at the end of each year from 1999 through fo For a rev r more Research Institute, 20 detail. iew of this literature, s 09b). ee Gale, Papk e and VanDerhei (2005). 2007. 38 12 35 4 31 of these two extrem em co ployee and t nstraints im he em posed es. Again, it will be years befo ployer, is by IRC Sec. sufficient t 402(g) o, re plcoup t an-spoh ne so decre r rre researchers have enough em eact ase in t ions th oe esti potent mate ial AD d 20P/ 08 AC acP n couo nndi t bal pirical evidence sca ri nce minat byi t on t he e en std inof g, misinterpretations are those of the witnes s. cove com co fidu bene The Sponsors t Thi A si Thi ntciary risk ribu r ret fi b m s stm ination of 401(k) plans and Social Security t pl de a e i tio il xim rm r a a not ee h n r si n rates wo i s h u es n h tat wh m eal clude u a b d ei at pe already clos o en tih ton h rcent selectin ts cost t hu h e 40 (desc l r e d i 1( s (e a 40 reduce (v n ge c kr 1 r) x ig an e ( of b ed c ba k ae lsed em ) udi d com lthe pla bal a bel inces n nirtu v a g p o e nces n ensat w stm pl ally to wi u n )oy rsi in the wo wi e th i n er n o t tg hom ul tn h h for cont i zero) an e d ei s cu last two years be e cap p th ri a rre e er t b rticip xpect pe expe unt th y id at an ons ein cu an ns e crease in d t d ts 25 es); as long as they are continuously covered by a rre o f i wh o a f are e rm p nt em n e see Fr e em o fail to elect th rcen er em pl xx irep oy s cluded from pl t t, as it is un ion oy n ees cha pl lacemen gst oy er o d in e ers as fi , Sal r ne pre nge td c rates resultin eir ilik th v s wel d b iont e an o ely wo u jo o us em ry w lb as a ri n s alysis of tho a b a i n u n n d Van rkers tn pl vest d iy on oy cashe IR gm pl ers t fro wou D A e an nt e d rol m rh s . Sec. o hei l out at e in in d b rl i ove ha ( creased n t e t 2 ih ab tve h 0 r40 ie s a e “n 0 fr bee t 40 9 le to e 4) d a ne ( om 1 c) f ex ( o n k (5 r t two ) )( w A) those between ages 20 and 24 ending up with more retirement wealth from the annuitized or 37 16 the estim plan const ated recove raints fory r hi period. ghly compensated employees. to determine the relative likelihood that any of the five AE panels would be appropriate. of m pr p y retain account 4 one ears” g a 01( See For a o ER o rticip re d v ). i k) d ed as IS VanDerh e a p e n ation tail. A r balanc l a ou an dn p di wel p. s. alysis that rto for iv onal es withi lie as an des t i (200 th sa is group h vi y at did l 7) n n IRA balan , a g ffor add s i ofew r . o p n ok excess uyears of rat the cas pitio oses ces th no aof fl t at are attrib their entry date. d Sec. h h fl eitail o s a ow m 40 ount nim 4 th (c) p u e . licatio ( tresu ab 1) le to olts. f E ns R in 401(k) ISA stead, see , a ro pa llovers. rtiVan cipant D erh in a ei and n indi Olsen (1 vidual acc 997 ou ). nt plan shall J J J J J J J J J J J J J J J J J J J J Ja a a a a a a a a a a a a a a a a a a a ac c c c c c c c c c c c c c c c c c c c ck k k k k k k k k k k k k k k k k k k k k V V V V V V V V V V V V V V V V V V V V Va a a a a a a a a a a a a a a a a a a a anDerhei nDerhei nDerhei nDerhei nDerhei nDerhei nDerhei nDerhei nDerhei nDerhei nDerhei nDerhei nDerhei nDerhei nDerhei nDerhei nDerhei nDerhei nDerhei nDerhei nDerhei,,,,,,,,,,,,,,,,,,,,, EB EB EB EB EB EB EB EB EB EB EB EB EB EB EB EB EB EB EB EB EBRI, RI, RI, RI, RI, RI, RI, RI, RI, RI, RI, RI, RI, RI, RI, RI, RI, RI, RI, RI, RI, DOL DOL DOL DOL DOL DOL DOL DOL DOL DOL DOL DOL DOL DOL DOL DOL DOL DOL DOL DOL DOL Ad Ad Ad Ad Ad Ad Ad Ad Ad Ad Ad Ad Ad Ad Ad Ad Ad Ad Ad Ad Advisory Counci visory Counci visory Counci visory Counci visory Counci visory Counci visory Counci visory Counci visory Counci visory Counci visory Counci visory Counci visory Counci visory Counci visory Counci visory Counci visory Counci visory Counci visory Counci visory Counci visory Councilllllllllllllllllllll,,,,,,,,,,,,,,,,,,,,, Sept. 17, 2009 Sept. 17, 2009 Sept. 17, 2009 Sept. 17, 2009 Sept. 17, 2009 Sept. 17, 2009 Sept. 17, 2009 Sept. 17, 2009 Sept. 17, 2009 Sept. 17, 2009 Sept. 17, 2009 Sept. 17, 2009 Sept. 17, 2009 Sept. 17, 2009 Sept. 17, 2009 Sept. 17, 2009 Sept. 17, 2009 Sept. 17, 2009 Sept. 17, 2009 Sept. 17, 2009 Sept. 17, 2009 Electronic copy available at: http://ssrn.com/abstract=1474865 Electronic copy available at: http://ssrn.com/abstract=1474865 Figure Figure 14 15 Figure 8 Figure 9 Auto-Enrollment With Auto-Escalation vs. Voluntary Enrollment: Auto-Enrollment With Auto-Escalation vs. Voluntary Enrollment: Figure 10: 1 2 Median Replacement Rates From 401(k) Accumulations for All Eligible Workers Post-PPA 401(k) “Accumulations” as a Multiple of Final Earnings for Those Currently Age 25–29 Post-PPA 401(k) “Accumulations” as a Multiple of Final Earnings for Those Currently Age 25–29 Impact of Automatic Escalation: Figure Figure 22: Median percentage of compensation required as an ENHANCED employer contribution for future years covered 21: Percentage of those with "lost" DB wealth due to a pension freeze who are expected to have a larger total Figure Figure Figure Figure 2 1 34 (assuming future eligibility is a function of current eligibility) Figure Figure Figure Figure Figure 20: Expected CONDITIONAL 23 17 18 19 : : Increase in balances (401(k) + rollover IRA) at retirement age as a function of initial : Increase in balances (401(k) + rollover IRA) at retirement age as a function of initial Expected change in average 401(k) account balances if all participants were to prospectively : Increase in balances for those assumed to cash out when they change jobs as a (assuming future eligibility is NOT a function of current eligibility) percentage point reduction in nominal replacement ratios if all private Change In Average Account Balances (by Age and Tenure) From Turning 65 Between 2030 and 2039, by Income Quartile at Age 65 Figure 16 a by a defined contribution plan in lieu of a frozen defined benefit plan for financial indemnification Percentage of Added Compensation That Must Be Saved Annually Until Percentage of Added Compensation That Must Be Saved Annually Until nominal replacement rate from the DC enhanced contributions (if any) Percentage Increase in Median Replacement Rates Percentage of Retirees Estimated to Have Sufficient Retirement Income/Wealth defined benefit plans were to freeze accruals for NEW employees immediately, by gender and equity allocation equity allocation Impact of Final Earnings Multiple on the Probability of function of tenure in in average target date vs participant direction: Participants ages 45 and older average target date vs participant direction: Participants younger than 45 in average target date vs participant directed age change to a guaranteed investment yielding 5 percent nominal, by gender and year of birth Voluntary Enrollment Voluntary Enrollment (percentage of final five-year average salary) Asset Allocation Distribution of 401(k) Participant Account Retirement For a 90% Chance of Covering Basic Retirement Expenses Retirement For a 75% Chance of Covering Basic Retirement Expenses January 1, 2008 – by Saving 5% of Compensation Each Year From 2003 Until Retirement September 1, 2009 Among 401(k) Participants Salary Quartile 5th Percentile 10th Percentile 25th Percentile Median 75th Percentile 90th Percentile 95th Percentile Salary Quartile 5th Percentile 10th Percentile 25th Percentile Median 75th Percentile 90th Percentile 95th Percentile From 401(k) Accumulations for Workers Turning 65 (see text for assumptions of asset allocation under status quo) Retirement Income "Adequacy," by Retirement Income Category (assumes current Social Security and housing equity is never liquidated) (assumes current Social Security and housing equity is never liquidated) (assumes current Social Security benefits) Figure 13 1 0.0 0.0 0.0 0.1 1.8 5.9 9.3 Balances to “Equity,” by Age: Year-end 2007 and 2008 1 0 0.0 0.0 0.1 1.8 5.9 9.2 45% 18% with Account Balances as of Dec. 31, 2007 7% 40% 20% 2 0.0 0.0 0.1 1.5 4.6 9.9 13.5 2 0 (Assumes 100% Equity Allocation and No Annuitization) Auto-Enrollment Without Auto-Escalation vs. Voluntary Enrollment: 0.0 0.1 1.4 4.9 10.5 16.0 Between 2030–2039, by Income Quartile at Age 65 Quartile 1 63% 3 0.0 0.0 0.3 2.5 7.0 13.7 17.9 3 0 0.0 0.2 2.2 7.1 14.2 18.9 (“Equity” is defined as equity funds + company stock + the relevant portion of balanced and target date funds) Post-PPA 401(k) “Accumulations” as a Multiple of Final Earnings for Those Currently Age 25–29 5.00% 9% For: Males Retiring at Age 65 4 0.0 0.0 1.8 6.0 12.0 19.8 25.7 4 0 0.0 1.3 5.7 12.0 19.8 26.0 (compared to baseline of PPA matching without automatic escalation) 40% 16% 6% Quartile 2 80% Automatic Enrollment (assuming 401(k) opt-outs, limit of safe harbor minimum, start over)* 58% 60% Automatic Enrollment Voluntary Enrollment (assuming (assuming future eligibility 401(k) opt-outs, limit of safe harbor IS a function of current eligibility) minimum, start over)* 57% 57% 30% | family family single female single female single male single male Option: Building Block 3 (investment income, longevity, and health care expenses stochastic) family single female single male Figure 5 56% Salary Quartile 5th Percentile 10th Percentile 25th Percentile Median 75th Percentile 90th Percentile 95th Percentile 28% 30% Salary Quartile 5th Percentile 10th Percentile 25th Percentile >90% 80–90% Median 70–80% 75th Percentile 90th Percentile 95th Percentile Quartile 3 Salary Quartile 8% 5th Percentile 10th Percentile 25th Percentile Median 75th Percentile 90th Percentile 95th Percentile 54% 15% | 5% 0.00% a 1 0.0 0.0 0.6 2.5 5.9 9.8 12.4 a 1 1 0 0.0 0.0 0.0 52% 0.2 0.0 0.1 2.5 1.8 6.5 5.910.3 9.2 12.9 35% 14% Account Balances Increase With Age and Tenure 52% 25% 25% Income Quartile 1 Income Quartile 2 Income Quartile 3 52% Income Quartile 4 100% | 2 0.0 0.0 1.5 3.9 7.6 10.6 12.3 60% Quartile 4 2 2 0 0.0 0.0 0.0 1.0 0.1 1.4 4.0 4.9 7.6 10.5 10.9 16.0 12.5 1936-40 1941-45 1946-50 1951-55 1956-60 1961-65 1966-70 100% 50% 50% 3 0.0 24% 0.3 2.1 4.6 8.2 11.7 13.3 | 4% 3 0 0.0 1.7 4.7 9.048% 12.4 13.9 3 0.0 0.0 0.2 2.2 7.1 14.2 18.9 7% 25% Average 401(k) account balance, by age and tenure, 2007 4 0.3 1.0 3.3 6.1 9.2 12.0 13.3 90% 4 4 0 0.0 0.5 0.0 3.0 1.3 5.7 6.2 12.0 9.5 19.8 12.6 26.0 13.9 | 45% 30% 12% 45% 23% 45% 20% -5.00% 90% 10% Automatic Enrollment (assuming no opt-outs, limit of safe harbor minimum, maintain contribution rates)* Automatic Enrollment (assuming no 401(k) opt-outs, limit of safe harbor minimum, maintain contribution rates )* Automatic Enrollment 43% (assuming future eligibility IS a function of current eligibility) | Tenure (years) 3% 40% 42% Salary Quartile 5th Percentile 10th Percentile 25th Percentile Median 75th Percentile 90th Percentile 95th Percentile 20% 20% 6% 80% Salary Quartile Salary Quartile 5th Percentile 5th Percentile 10th Percentile 10th Percentile 25th Percentile 25th Percentile Median Median 75th Percentile 75th Percentile 90th Percentile 90th Percentile95th Percentile 95th Percentile 40% 40% 20% | 1 0.0 0.0 0.9 2.8 6.8 10.4 12.5 Age Group 80% 0–2 >2–5 >5–10 >10–20 >20–30 >30 10% 25% 19%1 1 0 0.0 0.0 0.0 0.3 0.1 1.2 2.8 3.3 7.1 5.711.2 6.8 13.2 20% 37% | 2% 2 0.0 18% 0.5 1.9 5.0 8.6 12.6 13.8 -10.00% 2 2 0 0.0 0.1 0.0 1.8 0.7 2.3 4.8 4.6 9.1 7.113.0 7.6 14.2 20s 70% $4,491 $10,748 $18,564 10% 20% 3 0.1 0.5 2.6 5.5 9.9 12.6 17% 14.6 | 5% 70% 5% 3 0.0 0.0 1.1 2.8 5.1 7.1 8.0 3 16% 0 0.4 2.6 5.9 10.2 13.2 14.6 33% 25th percentile 25th percentile 4 0.3 1.5 4.1 7.6 11.0 13.5 14.6 30% 1% 8% 4 4 | 0 0.0 0.3 0.0 3.3 1.5 3.7 7.3 6.2 11.3 7.613.8 8.2 15.1 20% 30s $11,502 $23,024 $42,861 15% $62,207 25th percentile 15% Automatic Enrollment 15% 15% (assuming no opt-outs, limit of safe harbor maximum, maintain contribution rates)* 60% median median median Automatic Enrollment Voluntary Erollment (assuming (assuming future eligibility is no 401(k) opt-outs, limit of safe harbor NOT a function of current eligibility) 14% maximum, maintain contribution rates )* 15% | 60% -15.00% Salary Quartile 5th Percentile 10th Percentile 25th Percentile Median 75th Percentile 90th Percentile 95th Percentile median 40s 4% $16,672 $31,055 $58,262 $100,856 $151,193 0% Salary Quartile Salary Quartile 5th Percentile 5th Percentile 10th Percentile 10th Percentile 25th Percentile 25th Percentile Median Median 75th Percentile 75th Percentile 90th Percentile 90th Percentile95th Percentile 95th Percentile 75th percentile 75th percentile | mean 0% 12% 12% 12% male 1 0.0 0.0 1.7 4.5 9.4 14.7 17.0 6% 0% 75th percentile 50% 15% 1 1 0 0.0 0.0 0.0 0.4 0.0 0.1 4.5 1.8 10.4 11% 5.915.5 9.3 18.4 11% 23% 0% 1-5 6-10 11+ 20% 50% | 2 0.0 0.3 3.1 6.6 12.0 16.4 17.6 50s $20,603 $34,882 $63,783 $111,840 $194,385 $191,225 female 2 2 0 0.0 0.0 0.0 2.4 0.1 1.5 6.6 4.6 12.1 9.916.6 13.5 18.2 10% 10% 3 0.1 0.8 3.6 9% 7.7 12.8 17.2 19.9 -1% | -20.00% 3% 3 3 0 0.0 0.0 0.0 3.1 0.3 2.5 8.0 7.0 14.0 13.7 17.7 17.9 20.5 10% 10% 10% 40% 60s-20% $24,544 $35,399 $60,525 $105,504 $172,584 $210,457 4 0.7 2.1 5.3 9.8 13.9 17.5 19.1 8% 40% 4% | 4 0.0 0.0 1.8 6.0 12.0 19.8 25.7 10% 4 0 0.9 4.9 10.0 14.7 8% 18.1 19.8 7% 25-34 35-44 45-54 55-64 Automatic Enrollment Source: Tabulations from EBRI/ICI Participant-Directed Retirement Plan Data Collection Project. (assuming 401(k) opt-outs, limit of safe harbor maximum, maintain contribution rates)* Automatic Enrollment (assuming future eligibility is NOT a function of current eligibility) -2% Automatic Enrollment (assuming 401(k) | opt-outs, limit of safe harbor maximum, maintain contribution rates )* -10% 10% 6% Salary Quartile 30% 5th Percentile 10th Percentile 25th Percentile Median 75th Percentile Low Income90th Percentile 95th Percentile Note: At year-end 2007, the average account balance among all 21.8 million 401(k) particiants was $65,454; the median account balance was $18,942. -5% 2% 30% Salary Quartile 5th Percentile 10th Percentile 25th Percentile Median 75th Percentile 90th Percentile 5% 95th Percentile Salary Quartile 5th Percentile | 10th Percentile 25th Percentile Median 75th Percentile 90th Percentile 95th Percentile -25.00% 73.2% 49.9% 35.7% 29.8% 1-4 1 0.0 0.0 1.1 3.5 7.8 12.7 16.4 2% 5%5% HIgh Income 1 0.0 0.0 0.3 1.2 2.9 5.5 6.5 1 0 0.0 0.3 3.4 8.4 13.5 17.0 -3% 2 | 0.0 0.0 2.0 5.1 10.0 14.5 16.4 5% 5% 20% 2 0.0 0.1 0.8 2.0 4.4 6.7 7.6 2 0 0.0 1.5 5.2 10.0 14.0 16.6 35.7% 8.3% 2.8% 2.4% 20%5-9 3 0.0 0.4 2.9 6.1 11.0 15.7 17.9 | 1% 0% 3 0.0 0.2 1.2 2.5 5.0 7.0 8.0 3 0 0.0 2.3 6.0 12.2 16.6 18.5 4 0.5 1.4 4.1 7.6 12.1 15.5 18.3 -20% 0% -4% -30.00% | 0% 4 0.2 0.8 -0.9% 1.9 -7. 3.8 2% 6.1 -7.2% 7.6 8.2 10- 4 19 0 0.6 3.8 7.9 12.6 16.4 18.5 10% Up to Age 35 Up to Age 35 Age 36–45 Age 36–45 Age 46–55 Age 46–55 Age 56–65 Age 56–65 -10% Automatic Enrollment (assuming 401(k) opt-outs, limit of safe harbor minimum, maintain contribution rates)* 10% 0 1 - 9 10 - 19 20 - 29 30 - 39 40 - 49 50 - 59 60 - 69 70 - 79 80 - 89 90 - 99 100 Source: Author's simulations. 20 - 24 25 - 29 30 - 34 35 - 39 40 - 44 45 - 49 50 - 54 55 - 59 60 - 64 0% Automatic Enrollment (assuming 401(k) opt-outs, limit of safe harbor minimum, maintain contribution rates Tenure )* 20 - 24 25 - 29 30 - 34 35 - 39 40 - 44 45 - 49 50 - 54 55 - 59 60 - 64 0 1 - 9 10 - 19 20 - 29 30 - 39 40 - 49 50 - 59 60 - 69 70 - 79 80 - 89 90 - 99 100 Salary Quartile 5th Percentile 10th Percentile 25th Percentile Median 75th Percentile 90th Percentile 95th Percentile Note: Post-PPA 401(k) accumulations denote retirement money at age 65 in either a 401(k) plan or IRA rollover that originated w 0% ith contributions made on or after January 1, 2008 -10.9% -11.3% 20- 2007 29 2008 2007 2008 2007 2008 2007 2008 11 Initial equity allocation (percent) Salary Quartile 5th Percentile 10th Percentile 25th Percentile Age Cohort Median 75th Percentile 90th Percentile 95th Percentile 0% 0%Full RCS, Start Over 0% Full RCS, Maintain Maintain Contribution Maintain Contribution Start Over, Limit of Start Over, Limit of 1 0.0 0.0 0.7 Age cohort 2.8 6.3 10.1 12.7 The percentile columns represent the levels below which a certain percentage of observations fall. For example, the 75th percentile indicates the 401(k) accumulation multiple value below 0% Initial equity allocation (percent) 20 - 24 25 - 29 30 - 34 35 - 39 40 - 44 45 - 49 50 - 54 55 - 59 60 - 64 All Eligible Workers (Without Automatic Enrollment (3% Automatic Enrollment (3% Automatic Enrollment (6% Automatic Enrollment (6% 1 0 0.0 0.3 2.7 7.1 11.0 13.6 which fall 75 percent of all the values for an individual in that age and cohort -35.00% Source: Author's simulations based on June 16, 2009 modifications to the EBRI/ERF Retirement Security Projection Model. For ad 2 0.0 0.0 1.7 4.2 8.1 11.3 ditional 12.7 Contribution Rate Rate, Limit of Safe Rate, Limit of Safe Safe Harbor Min. Safe Harbor Max. 123412341234 123412341234 123412341234123412341234 123412341234 123412341234 Source: Author's simulations based on June 16, 2009 modifications to the EBRI/ERF Retirement Security Projection Model. For additional Automatic Enrollment) 2 0 Contribution Rate; Money 0.0 Contribution Rate; Life-Cycle 1.2 4.4 Contribution Rate; Money 8.2 11.7 Contribution Rate; Life-Cycle 13.2 Sources: Age Cohort 0 5 10 15 20 25 30 35 40 3 0.0 0.4 2.3 5.2 8.8 12.5 14.2 detail on the model, see VanDerhei and Copeland, "The Impact of PPA on Retirements Savings for 401(k) Participants," EBRI Issue Brief, Source: Author's simulations based on June 16, 2009 modifications to the EBRI/ERF Retirement Security Projection Model. For additional Harbor Min. Harbor Max. detail on the model, see VanDerhei and Copeland, "The Impact of PPA on Retirements Savings for 401(k) Participants," EBRI Issue Brief, 3 0 0.0 1.9 5.2 9.7 13.3 15.6 Source: Author’s simulations based on April 2009 version of EBRI/ERF Retirement Security Projection Model™ Market Fund) Fund) Market Fund) Fund) 2007: Tabulations from year-end 2007 data from EBRI/ICI Participant-Directed Retirement Plan Data Collection Project. The analysis is based on active 4 1936–1940 1936–1940 1936–1940 0.5 1941–1945 1941–1945 1941–1945 1.1 1946–1950 1946–1950 1946–1950 3.5 6.5 1951–1955 1951–1955 1951–1955 9.8 1956–1960 1956–1960 1956–196012.6 1961–1965 1961–1965 1961–1965 14.5 detail on the model, see VanDerhei and Copeland, "The Impact of PPA on Retirements Savings for 401(k) Participants," EBRI Issue June 2008 Brief, Final Earnings Multiple Source: Author’s simulations based on April 2009 version of EBRI/ERF Retirement Security Projection Model™ Source: Author’s simulations based on April 2009 version of EBRI/ERF Retirement Security Projection Model™ June 2008 4 0 0.6 3.2 6.6 10.3 13.2 14.6 Source: Author's simulations from the EBRI/ICI 401(k) Accumulation Projection Model. Source: Author's simulations. participants with account balances at the end of 2007. Sources: 2007 Account Balances: Tabulations from EBRI/ICI Par® ticipant-Directed Retirement Plan Data Colle a ction June 2008 1 Source: Author's simulations. -40.00% Birth Cohort/Income Quartiles Birth Cohort/Income Quartile Birth Cohort/Income Quartile Terms: 401(k) opt-outs denotes that individuals will opt out of future increases as described in the empirical findings presented in VanDerhei (2007a). No opt-outs denotes that individuals will not opt Source: EBRI/ICI 401(k) Accumulation Projection Model. * a 25% = 25% or more. Source: EBRI-ERF Retirement Security Projection Model. 2008: Author's projections based on year-end 2007 data from EBRI/ICI Participant-Directed Retirement Plan Data Collection Proje Retirement Confidence Survey. Source: Employee Benefit Research Institute, Ballpark E$timate Monte Carlo, August 2006 version. ct. 25% = 25% or more. Source: EBRI-ERF Retirement Security Projection Model. Terms: 401(k) opt-outs denotes that individuals will opt out of future increases as described in the empirical findings presented in VanDerhei (2007a). No opt-outs denotes that individuals will not 1 * out of future increases until they reach an employer-induced constraint. Safe harbor minimum denotes that employers will limit the automatic increases to 6 percent of compensation. Safe harbor Project; 2008 and 2009 Account Balances: EBRI estimates. The analysis is based on all participants with account The 401(k) accumulation includes 401(k) balances at employer(s) and rollover IRA balances. opt out of future increases until they reach an employer-induced constraint. Safe harbor minimum denotes that employers will limit the automatic increases to 6 percent of compensation. Safe harbor Source: EBRI-ERF Retirement Security Projection Model. Assumes current Social Security, and that housing equity is never liqui maximum denotes that employers will limit the automatic increases to 10 percent of compensation. Start over denotes that workers will start over from the default contribution when they change jobs. dated. The model includes the possibility of chronic long-term 2 balances at the end of 2007 and contribution information for that year. All eligible workers includes 401(k) plan participants with account balances at year-end 2000 and eligible nonparticipants. maximum denotes that employers will limit the automatic increases to 10 percent of compensation. Start over denotes that workers will start over from the default contribution when they change jobs. Maintain contribution rate denotes that workers will retain the deferral level rate from the previous job. home health care and nursing home expenses. Maintain contribution rate denotes that workers will retain the deferral level rate from the previous job. Note: Post-PPA 401(k) accumulations denote retirement money at age 65 in either a 401(k) plan Note: Post-PPA 401(k) accumulations denote retirement money at age 65 in either a 401(k) plan or IRA rollover that originated with contributions made on or after January 1, 2008. or IRA rollover that originated with contributions made on or after January 1, 2008. Probability of Adequacy Reduction in nominal replacement ratios for those with "lost" DB wealth due to a freeze

Volume T-162

Pages 45

EBRI Testimony

Sept 17, 2009

Jack VanDerhei