Jan. October 2001 Feb. EBRI Issue Brief (ISSN 0887-137X) is published monthly at $300 per year or is included as part of a membership subscription by the Employee Benefit Research Institute, 2121 K Street, NW, Suite 600, Washington, DC 20037-1896. Periodicals postage EBRI Table 4 Table 5 rate paid in Washington, DC. POSTMASTER: Send address changes to: Table 1 EBRI Issue Brief, 2121 K Street, NW, Suite 600, Mar. Chart 3 Table 2 Chart 1 Table 3 Table 6 a Chart 8 Chart 4 a Average Total Contribution Rates, by Age and EMPLOYEE Washington, DC 20037-1896. Copyright 2001 by Employee Benefit Research Institute. All rights reserved, No. 238. Distribution of Participants by Plan Match Level and Plan Match Rate, Average Participant Before-Tax Contribution 1999 Sarah Holden is senior economist, Research Department, at the Investment Company Institute (ICI), and Distribution of Participants by Total Contribution Rate, 1999 Average Participant Before-Tax Contribution Rates, Distribution of Participants by Tax Status of Employee Average Participant Before-Tax Contribution 401(k) Participant Contribution Data, 1999 Participants Contributing a Among Participants Making Contributions, t Employer Contribution Contribution Behavior of 401(k) Employer Contribution Status, 1999 Rates, by Age and Salary, 1999 (Percentage of Participants) BENEFIT Contribution, 1999 a by Age and Tenure, 1999 Apr. Jack VanDerhei, Temple University, is research director of the EBRI Fellows Program. Special thanks to Luis Rates, by Age and Employer Contribution Percent Match Level, age of Particip by Salants a ary, 1999 t 402(g) Limit (Percentage of Salary Contributed by Both Employer and Employee) (Percentage of Salary Contributed) (Percentage of All Participants) (Percentage of Salary Contributed) a a RESEARCH Item Median Mean Alonso at EBRI, who managed the database, and to Janet Thompson-Conley at ICI, who prepared the graph- 10% Status, 1999 by Age, 1999 b The Employee Benefit Research Institute (EBRI) was founded in 1978. Its mission is Match Rate Made No 20% May INSTITUTE (Percentage of Salary Contributed) ho we are ® Tenure (Years) Non-Highly Compensated Employees ics. This document is being published simultaneously by the Investment Company Institute as 9% to contribute to, to encourage, and to enhance the development of sound employee benefit Plan Has ICI Plan Does Not Have W Salary Range Contribution Participant Contribution Variables 18% 18% b b $0.25 $0.33 $0.50 $0.67 $0.75 $1.00 Other Total Age Employer Contributions Highly Compensated Emplo Employer Contributions yees Plan Par programs and sound public policy through objective research and education. EBRI is the only private, nonprofit, ticipants 8% 15% 17% 8% Perspective Before-Tax Participant Contribution as a Percentage of Salary (the . Any views expressed in this report are those of the authors and should not be ascribed to the Before-Tax Contribution Rate). 6.0% 6.8% Plan Has Age Plan Does Not Have 0–2>2–5>5–10 >10–20 >20–30 >30 $20,000 >$40,000 >$60,000 >$80,000 Jun. 16% 16% 16% Made Both Bef 15% ore- nonpartisan, Washington, DC-based organization committed exclusively to public policy research and education on b b After-Tax Participant Contribution as a Percentage of Salary (the After-Tax Contribution Rate). 3.6 4.0 Age Employer Contributions Employer Contributions 15% 15% 7% 20s 8.5% 6.9% officers, trustees, or other sponsors of EBRI, EBRI-ERF, or their staffs. Neither EBRI nor EBRI-ERF lobbies 2% 0% 0% 2% 0% 0% 3% 2% 8% Age –$40,000 –$60,000 –$80,000 –$100,000 >$100,000 and After-Tax 20s 5.7% 6.4% 6.0% Total Participant Contribution as a Percentage of Salary (the economic security and employee benefit issues. EBRI’s membership includes a cross-section of pension funds, Total Participant Contribution Rate 11% ). 11% 6.0 6.9 15% 30s 9.6 7.2 3% 12% Contributions 4110051 12 10% or takes positions on specific policy proposals. EBRI invites comment on this research. by Sarah Holden, ICI, and Jack V 6% anDerhei, Temple University 30s 6.4 6.8 6.7 6.3% Jul. Sum of Employer and Participant Contribution as a Percentage of Salary (the Total Contribution Rate). 9.0 9.7 20s businesses, trade associations, labor unions, health care providers and insurers, government organizations, and 5.7% 6.9% 20s 5% 5.3% 6.8% 7.4% 6.8% 4.8% 4% 10401229 40s 10.1 7.3 40s 6.8 7.2 7.1 6.7 6.6% 5% 9% 30s service firms. 6.5 30s 7.1 6.2 6.8 7.2 6.9 5.1 50s 11.1 8.3 5% 1010255 13 Personal Participant Characteristic Variables Made an After- 50s 7.5 7.9 8.0 7.6 7.4 and EBRI Fellow 40s 6.8 7.3 40s 6.7 7.1 7.3 6.8 5.0 Aug. 5% 60s 11.8 9.5 4% 6% Tax Contribution 24 27 5345 49 Annual Participant Salary $32,730 $44,187 60s 6% 8.5 8.7 8.7 8.4 8.1 8.0% 50s 7.6 8.2 50s Onl 7.6 y 8.3 8.2 7.3 5.1 10%7% 00400005 Participant Age (Years) 41 41 3% All 10.0 7.4 7 2% 60s 8.4 60s EBRI’s work advances knowledge and understanding of employee benefits and their 9.5 8.5 9.3 9.0 7.9 5.1 For example, U.S. GAO (October 1997) and Papke (1995) analyzed data counted in the test (everything else equal, the more that can be counted, the 1999) reported that, in 1997, 98 percent of full-time employees of medium and participant contribution activity. Several samples of participants were tested 8% 00100002 Participant Tenure (Years) 3% 58 Source: Tabulations From the EBRI/ICI Participant-Directed Retirement Plan Data Sep. 41 55 hat we do 60 These estimates are based on regression of participants earning $80,000 or For example, Even and Macpherson (May 2001), Even and Macpherson All 2% 6.8 7.4 new employees were automatically enrolled in the 401(k) plan and must have VanDerhei and Copeland (January 2001) also found a positive effect of compiled from the Internal Revenue Service (IRS)/Department of Labor lower their aggregate contribution rate will be). W large private establishments (with savings and thrift plans) were allowed to for the same reason. More details of the regression analyses are presented in 9% or more 00100012 importance to the nation’s economy among policymakers, the news media and the public. Made a Before- Plan Characteristics— although many plans do not fall thereafter. For Economic Perspectives nut Hill, MA: Center for Retirement Research at . Vol. 15. no. 3 (Summer 2001): Holden, Sarah, and Jack VanDerhei. “Regression Analy- 401(k) plans than for those who did not. The average VanDerhei, Jack, and Craig Copeland. “A Behavioral employee contributions to 401(k) plans in 1999. their 60s contributed 9.3 participants in their 50s and participants in their 50s Collection Project. For many Ameri- Table 3, Average Participant Before-Tax Source: Tabulations From the EBRI/ICI Participant-Directed Retirement Plan Influence of Personal Par • This ticipant Issue Brief examines the 1999 contribution behavior of 1.7 million 401(k) plan Employer Influence of the IRC Source: Tabulations From the EBRI/ICI Participant-Directed Retirement Plan 0 less per year and making before-tax contributions to the 401(k) plan. The (April 1999), Bassett, Fleming, and Rodrigues (June 1998), and Andrews Plan Characteristics Offered to Participant Variables opted out if they did not want to participate) on participant contribution match level on participant contribution rates. (DOL)/Pension Benefit Guaranty Corporation (PBGC) Form 5500. The IRS/ contribute on a before-tax basis, and 54 percent were allowed to contribute Holden and VanDerhei (forthcoming). Tax Contribution 1% It does this by conducting and publishing policy research, analysis, and special reports on employee benefits issues; 15 25 Note: Average calculated among all participants making before-tax contributions. Data Collection Project. The regression analysis 3–22. example, contributing allow both options. PSCA (2000) reported that 11.5 percent of plans had excess contributions Boston College, December 2000. Among sis of the Influence of Participant and Plan Sponsor total contribution rate was 10 percent of salary for Eighty-five percent of participants in the sample only Model for Predicting Employee Contributions to percent of salary (table 1). 18 percent of participants in with two or fewer years’ Total 8 can workers, 5 41 5 6 20 15 100 regression model included age, tenure, salary, plan loan provision (yes/no), (1992) found that the presence of an employer match (or contribution) Employer Contribution to Participant Data Collection Project. ’s Account as a Percentage of Participant 20s 30s 40s ’s Salary. behavior. They found that new participants tended to stay with the default Contribution Rates, by Age and Employer 50s 60s Onl All y 2.8% 3.3% Oct. DOL/PBGC Form 5500 data also are tabulated by the DOL Pension and participants drawn from the EBRI/ICI Participant-Directed Retirement Plan Data only on a before-tax basis. U.S. DOL (April 1999) found that, in 1996, Source: Tabulations From the EBRI/ICI Participant-Directed Retirement Plan a Characteristics 5% 61 Contributions 34 §402(g) Limit holding educational briefings for EBRI members, congressional and federal agency staff, and the news media; and Average calculated among participants with any contribution. Includes For example, Madrian and Shea (May 2000) analyzed participants in one T that were returned to participants after the plan year ended; 1.5 percent of able of 0 Gender information was not available for a sufficiently large and 85% and employer contribution (yes/no) variables to examine their effects on increased participation in the 401(k) plan. Note: Average calculated among all participants making before-tax contributions. Age Group contribution rate of 3 percent of salary. Analysis of the effect of automatic Welfare Benefits Administration and summarized in an annual report (for the Employer Match Rate–the Percentage of Each Dollar Contributed by the Employee That the Employer Will Match (Cents on the Dollar). 88 percent of full-time employees of small private establishments (with savings $0.50 $0.51 tenure contributed Clark, Robert L., and Sylvester J. Schieber. “Factors found that participants Data Collection Project. Papke, Leslie E. “Participation in and Contributions to the participants studied here, participants in their 401(k) plan Characteristics on 401(k) Plan Participants’ Contribu- employees in plans offering an employer contribution, made before-tax contributions to their plans, and 401(k) Plans: Preliminary Results.” their 60s were at the limit The regression analysis North American $20,000– >$40,000– >$60,000– >$80,000 Contribution Status, 1999 – >$100,000 All ........................................ 10 Collection Project. The findings in this paper build on previous academic research large 401(k) plan and found that prior to the introduction of automatic before- and/or after-tax employee contributions, and/or employer plans deposited excess contributions into a non-qualified account; and >0–sponsoring public opinion surveys on employee benefit issues. 1% Source: Tabulations From the EBRI/ICI Participant-Directed Retirement Plan Data Collection Project. >2%–3% >4%–5% >6%–7% >8%–9% >10%–11%representative sample of participants. However, other research has found >12%–13% EBRI’s Education and Research Fund >14%–15% >16%–17% >18%–19% >20% (EBRI-ERF) participant before-tax contribution rates. The stylized estimates reported EBRI enrollment default options was beyond the scope of this study. However, PSCA most recent report, see U.S. DOL (Winter 2001)). a Employer Match Level–the Percentage of Salary Up to Which the Employer Will Provide A Matching Contribution. and thrift plans) were allowed to contribute on a before-tax basis, and 6.0% 4.2% Source: Tabulations from the EBRI/ICI Participant-Directed Retirement $40,000 $60,000 $80,000 $100,000 56 Average calculated among participants making before-tax contributions. Nov. a An algorithm comparing the employee contributions and the employer enrollment with a 3 percent of salary default contribution rate, more than one- 7.5 percent of were more likely to 85 percent contributed only on Affecting Participation Rates and Contribution Levels 40s and earning 12.3 percent of plans limited contributions of highly paid participants within 401(k) Pension Plans: Evidence from Plan Data.” accounts have The Table 4, Average Total Contribution Rates, by Age tion Activity.” compared with 7.4 percent for those in plans not mixed results for the influence of gender on participant contribution rates. For contributions. 97 percent of all dollars contributed by employees Actuarial Journal. Working Paper, Vol. 5. No. 1 (January 2001): also confirmed that the (chart 4). However, this forthcoming. • Older participants tended to contribute a higher Sample of nearly 1 million participants for whom employer matching contribution information was provided or derived. Source: Tabulations From the EBRI/ICI Participant-Directed Retirement Plan Approximately 91 percent of assumed that the participants were offered a loan provision and provided an This section examines the performs the charitable, educational, and scientific functions of the Institute. EBRI-ERF is a tax-exempt organization examining the contribution activity of 401(k) participants, by using a large sample (2000) reported that only 4.2 percent of plans had automatic enrollment of In addition, Levinson (Winter 2000) analyzed plan-level data from a 1998 Plan Data Collection Project. 41 percent were allowed to contribute only on a before-tax basis. b Employer Effective Match Rate–the Employer Match Level Multiplied by the Employer Match Rate (Percentage of Participant’s Salary). 3.0 2.4 Percentage of Salary Contributed by Employee and Employer Plans may or may not permit participants to take out loans. Salary Range Overview b contributions for each participant in a given plan looking for a pattern b Chart 6 third of participants clustered at the 6 percent match level. In addition, the plan year once the maximum allowed by the test was reached. Data Collection Project. example, VanDerhei and Copeland (January 2001) and Even and Macpherson Plans may or may not permit participants to take out loans. employer contribution. Contents Match rate is the percentage of each dollar contributed by the employee for which the employer makes a matching contribution participants. In addition, larger plans (with 5,000 or more participants) were survey of 1,292 employers; and Papke and Poterba (1995) analyzed plan-level contribute at the 402(g) limit if the plan offered a loan a before-tax basis, another in 401(k) Plans.” In Olivia S. Mitchell and Sylvester J. $20,000–$40,000 Journal of Human Resources. Vol. XXX. No. 2 (1995): Note: Excludes participants earning less th Holden, Sarah, and Jack VanDerhei. “401(k) Plan Asset percentage of salary contributed by the participant rose offering an employer contribution. an $40,000 a y were contributed on a before-tax basis. On average, 80–94. ear because positive correlation between salary, while supported by contributions and grants. The American Savings Education Council (ASEC) and the Consumer Health become an percentage of their salaries to plans than younger Maximum Contribution Allowed by the Employer as a Percentage of Participant’s Salary. and Employer Contribution Status, 1999 all participants in the sample 16.0................ 17.0 10 participant and plan charac- EMPLOYEE of participants in a wide range of plan sizes and by examining in detail the factors 16 established both the (initial) match rate as well as the maximum amount of Kusko, Poterba, and Wilcox (1998) analyzed data for one 401(k) plan and Estimated Participant Before-Tax Contribution Ra By design, the IRC limits, in 1999, prohibited highly compensated (May 1997) found (in the latter case, in some, but not all, of their regression tes by Tenure The relationship between salary and a participant’s contribution rate was (shown as cents on the dollar). another IRC limit prevented them from reaching the IRC Sec. 402(g) limit. more likely to have automatic enrollment than smaller plans. Dec. survey data (for 1986 and 1990) from 43 plans. 26 Source: Tabulations From the EBRI/ICI P Source: Tabulations From the EBRI/ICI P articipant-Directed Retirement Plan Data Collection Project. articipant-Directed Retirement Plan Data Collection Project. 35 Although there have been recent modifications (e.g., prior-year testing and Education Council (CHEC) are programs of EBRI-ERF. They are coalitions of private- and public-sector institutions similarly aged provision. This result is consistent with prior research 5 percent made both before- Schieber, eds., per year contrib- compensation matched at that rate. 311–325. Living with Defined Contribution important part of with participant age. Allocation, Account Balances, and Loan Activity in found that 37 percent of employees contributed at the match level. Further- Xiao, Jing J. “Saving Motives and 401(k) Contributions.” participants contributed 6.8 percent of their salaries For example, after controlling for age and contributions at the participants, even after factoring out differences in c Table 5, Distribution of Participants by Plan employees from contributing (on a before-tax basis) more than 12.5 percent of models) that being female had a positive effect on the participant’s contribu- were in plans offering em- teristics that influence that influence contribution activity. Tabulated from a sample of 0.8 million participants. not linear. A $1,000 increase in salary produced a larger increase in the Match level is the percentage of salary up to which employee contributions will be matched by the employer. • Employee before-tax contribution rates tended to be Note: Sample of participNote: Sample of nearl ants with before- and/or after y 1 million p -tax emplo articipants (whether contributing or not) f yee contributions, and/or employer contributions. or whom employer matching BENEFIT safe harbor approaches) that allow plan sponsors to reduce the uncertainty Source: Tabulations From the EBRI/ICI Participant-Directed Retirement Plan Data Collection Project. 46 8 with the goals of public education on saving, retirement planning, health insurance, and health quality. 7.7% Results from the broader 1999 EBRI/ICI 401(k) database showed that more, Yakoboski and VanDerhei (June 1996) analyzed three large 401(k) For example, see Madrian and Shea (May 2000); Kusko, Poterba, and salary ($10,000/$80,000). tion rate, while Clark, Goodfellow, Schieber, and Warwick (2000) found that participant before-tax contribution rate at higher salary levels than at lower tenure and other factors, the regression estimated that a participants with and the results presented below, which found, in general, Papke, Leslie E., and James M. Poterba. “Survey Evi- and after-tax contributions, Pensions uted an average of 57 their retirement planning. The income that these . Philadelphia, PA: The Pension Research contribution information was provided or derived. 1999.” Match Level and Plan Match Rate, 1999 on a before-tax basis. Financial Counseling and Planning Perspective. Vol. 7, no. 1; and 402(g) limit may have EBRI Issue Brief . Vol. 8 (2) (1997). ................. 11 salary and job tenure. A 60-year-old participant with a ployer contributions. For whether a participant U.S. DOL (September 1999) reported similar results. Their report found inherent in this annual testing procedure, many sponsors place limits on the Medians and means calculated among those participants with positive values of the variable in question. Means are not dollar-w In the EBRI/ICI database sample of 1.7 million eighted. slightly lower for participants whose employer made a 2001 Overview .......................................................................... 3 82 percent of participants were in plans offering loans. See Holden and plans and found that between one-fifth and one-third of participants Wilcox (1998); and Yakoboski and VanDerhei (June 1996). Influence of Loan Provision being female had a negative effect in their OLS regression analysis. In salary levels. 17 that 47 percent of all full-time employees of medium and large private firms RESEARCH that offering a loan provision increased participant and another 2 percent made only an after-tax contribu- Council, The Wharton School of the University of 6.7 percent of amounts that highly compensated employees may contribute to a plan in an dence on Employer Match Rates and Employee Saving • Eighty-five percent of participants in the sample only made before-tax contributions 25-year-old participant with five years of job tenure and The clustering of participants in the more than 5 percent–6 percent of no. 230. Washington, DC: Investment Company Yakoboski, Paul, and Jack VanDerhei. “Contribution resulted from other factors five to 10 years’ accounts are expected to provide in retirement depends, the same salary and job tenure as a 30-year-old Table 6, 401(k) Participant Contribution Data, 1999 VanDerhei (January/February 2001). contributed at the plan match level. employees receiving employer contributions, the em- .. 13 contributed at the 402(g) addition, Ippolito (1997) and U.S. GAO (October 1997) found that contribu- contribution activity. When designing a matching Contributions to 401(k) Plans .................................... 3 participants, 91 percent of participants were in 401(k) • level of 6 percent of salary would provide a participant contribution. The average participant contribution Before-tax contribution activity varied among partici- 9 were offered a 50 cents on the dollar match rate in 1997, 48 percent were 42 For example, see Fidelity Investments (2001), which tabulated information attempt to improve their chances of compliance without the need for year-end salary range occurred because 49 percent of participants were in plans offering Even and Macpherson (May 1997) found similar results: Contribution EBRI Issue Briefs tion rates were about the same between men and women (the variable was not are monthly periodicals providing expert evaluations of 7.6% to their plans, and 97 percent of all dollars contributed by employees were contrib- tenure contributed contributions to 401(k) plans. tion (chart 1). On average, 97 percent of all dollars Pennsylvania, and University of Pennsylvania Press, salary (table 1). The in part, on the contributions that plan participants make Behavior in 401(k) Plans.” 7.5% Economics Letters 49 $40,000 in salary per year would contribute 5.8 percent Institute, January 2001, and Employee Benefit tied to age, such as salary or tenure. To control for these Rates and Plan Features: An Analysis of Large 401(k) 51 Contribution rates tended to be higher in 401(k) plans participant is predicted to contribute nearly 2 percent 47 62 ployer contributed, on average, 3.3 percent of participant limit imposed on individual before-tax contributions. INSTITUTE contribution policy, the employer must determine: (1) the offered matching employer contributions on up to 6 percent of salary ® plans offering an employer contribution. This result is consistent with previous research using the Federal Reserve with an effective match rate of 3 percent (0.50 x 6 per- Participants are individuals who were currently employed and holding In plans with on defined contribution plans; Clark, Goodfellow, Schieber, and Warwick modifications to contributions. For a listing of the alternative approaches to Influence of Internal Revenue Code Limits ur publications ............... 3 employer matching contributions on up to 6 percent of salary (table 5). rate in plans offering an employer contribution was pants. About 61 percent of participants contributed rates increased with tenure up to about 17 years and then were pulled down sated employees achieved a total contribution in excess O statistically significant). jobs and associated 401(k) accounts maintained by employee benefit issues and trends, as well as critical analyses of employee 30 contributed, and 25 percent were offered the combination of 50 cents on the uted on a before-tax basis. On average, participants contributed 6.8 percent of their Board’s Survey of Consumer Finances (SCF) data. For example, Munnell, of his or her salary to a 401(k) plan. Meanwhile, a 401(k) account balances—whether they contributed in 1999 or not. 8.0 percent of (2000), which analyzed 1995 participant-level data for 87 plans; Hewitt contributed by employees were contributed on a before- 1998, pp. 69–97. same age group maximize the likelihood of compliance, see Allen, Melone, Rosenbloom, and during their wage-earning years. Contributions by plan (1995): 313–317. Another plan characteristic, the presence of an Chart 1, Distribution of Participants by Tax Status Research Institute, February 2001. other factors, a statistical regression model was used. Plan Data.” EBRI Issue Brief no. 174 (Washington, more of his or her salary to a 401(k) plan. salary. The total contribution to a participant’s account that permitted loans to participants, a feature available by longer tenures. On the other hand, Clark, Goodfellow, Schieber, and Because participants earning less than $40,000 a year percentage of each dollar contributed by the employee employer contributions, participants contributed an cent) of salary. About one-quarter of participants were in Other research has studied the effect of the participant’s level of education 18 18 6.8 percent of salary. Contributing participants in more than 5 percent of their salaries on a before-tax or derived. Participants were then identified as in a plan Influence of Personal Participant Characteristics .... 4 Allen, Jr., Everitt The clustering of participants in the more than 8 percent–9 percent of participant before-tax contribution rates fell minimally benefit policies and proposals. Each issue, ranging in length from 16–28 pages, thoroughly explores one topic. companies, insurance companies, and consulting firms. EBRI of 15 percent of salary (chart 3). dollar on contributions up to 6 percent of salary. U.S. DOL (April 1999) found the individual. Sundén, and Taylor (December 2000), using 1998 SCF data, found that Associates (1999); and Clark and Schieber (1998), which analyzed 1994 VanDerhei (1997), pp. 191–192. Warwick (2000) found a negative correlation between contribution rates and 46 Clark, Robert L., Gordon P. Goodfellow, Sylvester J. employer contribution, lowered the likelihood of a Poterba, James M.; Steven F. Venti; and David A. Wise. tax basis. contributed an salaries on a before-tax basis. 27 Investment Company Institute. “401(k) Plan Partici- 65-year-old participant with otherwise similar character- 63 on contribution rates. For example, Ippolito (1997), Even and Macpherson This model showed that even after controlling for these DC: Employee Benefit Research Institute, June 1996). salary, and those participants depend on a variety of factors, including the 7.4% to 84 percent of participants analyzed in this study. of Employee Contribution, 1999 (Percentage of was the sum of the employee (before- and after-tax) and The universe of investment plan administrators varied from year to year, cannot reach the $10,000 limit, these individuals were salary range occurred because 27 percent of participants were in plans offering with a formal specified limit on before-tax contribution that the employer will match—i.e., the Influence of Plan Characteristics that 36 percent of all full-time employees of small private firms were offered a 7.3% 58, 59 ............................... match rate; and 4 average of 6.8 percent of salary to their plans on a participants in plans offering loans contributed 1 percentage point more of plans offering an effective match rate of up to 2 percent plans not offering an employer contribution directed basis and about 21 percent set aside more than T.; Joseph J. participant-level data for 19 plans. PSCA (2000) reported that 7.7 percent of plans indicated that the before-tax Notes is a monthly periodical providing current information on a variety of employee benefit topics. EBRI’s Washington • Participants tended to increase the share of their as the employer match rate rose. all levels of tenure. The variation in contribution rates raises However, the Records were encrypted to conceal the identity of em- (May 1997) and Bassett (August 1995) found that contribution rates rose with IRC §415(c) imposes limits on the total contribu- and thus, aggregate figures should not be used to estimate time trends. employer contributions of 50 cents on the dollar on up to 6 percent of salary with more than participant being at the 402(g) limit. However, the Schieber, and Drew Warwick. “Making the Most of average of 50 cents on the dollar match rate in 1996, 38 percent were offered matching regulatory framework under which 401(k) plans operate, “Implications of Rising Personal Retirement Saving.” Participants making before-tax contributions set their salaries; and U.S. GAO (October 1997), using 1992 SCF data, found istics would contribute 8.5 percent of his or her salary to pants: Characteristics, Contributions, and Account other factors, older participants were more likely to be at Statistical regression analysis found that participant contributions of highly paid participants were limited by plan design. In 67 All Participants) employer contributions. Among participants whose .......................................................... 5 not included in the analysis of the influence of the (2) the percentage of the employee’s salary up to which Bulletin provides sponsors with short, timely updates on major federal developments in employee benefits. before-tax basis (table 3). On the other hand, partici- of salary, and 41 percent of participants were offered an EBRI’s Background rates or in a plan with no plan sponsor limit. ...................................................................... In plans 4 an average of 7.4 percent of salary to their plans. participant education levels. On the other hand, Munnell, Sundén, and Taylor 10 percent of their salaries on a before-tax basis. Melone; Jerry S. 10 salary (and amounts) contributed to their 401(k) plan regression analysis found that as the match level chosen questions about the factors that lead to the differences in 43 For example, see U.S. GAO (September 2001); Munnell, Sundén, and ployers and employees, but were coded so that both could (table 5). 1 • Before-tax contribution activity varied among participants. About 61 percent of tions that may be made to a participant’s account in a These estimates are based on regression of participants making before-tax employer contributions on up to 6 percent of salary contributed, and 19 per- contribution rates of participants in plans with borrowing privileges were 36 In this analysis, addition, larger plans were more likely to have limits on the before-tax 64 the plan. the 402(g) contribution limit. The median age of participants in the sample (41 years old) was similar to The regression estimated that, all other 20 years’ tenure average total (employee and employer) contribution rate aside, on average, 6.8 percent of their salaries in their 401(k) Plans: Who’s Choosing What and Why?” In 7.3 percent of personal participant characteristics, and the plan’s NBER Working Paper Fundamentals of Employee Benefit Programs . No. 6295. Cambridge, MA: offers a straightforward, basic explanation of employee benefit programs Chart 2, Distribution of Participants by Before-Tax Activity.” (December 2000) found that education level was insignificant when a variable ICI Research Series. Washington, DC: accounts received any contribution—from the employer, contribution rates were higher when plans allowed 402(g) limit. In the sample of about 0.9 million partici- Taylor (December 2000); Bassett, Fleming, and Rodrigues (June 1998); U.S. contributions will be matched—i.e., the match level. pants in plans with no employer contributions directed effective match rate of more than 2–3 percent of salary contributions to the 401(k) plan. The regression model included age, tenure, Contributing on a Before- or After-Tax Basis with employer contributions and for which no match cent were offered the combination of 50 cents on the dollar on contributions up 7.2% ........... 4 3 percentage points higher than those of participants in plans without a loan participants are Rosenbloom, and contributions of highly paid participants than smaller plans. as their salaries rose until salaries reached $80,000. by the employer rose, participant contribution rates 19 be tracked over multiple years. Complete employee and contribution activity among participants. The remaining participants contributed more than 5 percent of their salaries on a before-tax basis given 401(k) plan. Under this provision, the sum of the median age among the 10.3 million participants in the entire 1999 EBRI/ 7.1% IRC rules are discussed in this • representing the participant’s planning horizon was included (people with Approximately 91 percent of the participants in this Issue Brief in the context of analyzing 47 in the private and public sectors. The EBRI Databook on Employee Benefits Bibliography is a statistical reference volume on employee in plans with employer contributions was higher than in GAO (October 1997); Even and Macpherson (May 1997); Poterba, Venti, and 401(k) accounts. Among those participants who made Olivia S. Mitchell, P. Brett Hammond, and Anna M. salary when National Bureau of Economic Research, November factors being equal, participants in their mid-40s or Investment Company Institute, Spring 2000. contributed overall design. borrowing. salary, plan loan provision (yes/no), and employer contribution (yes/no) to 6 percent of salary. Indeed, a participant in a plan offering provision. Participant Contribution Rate, 1999 (Percentage the employee, or both—the average total contribution individuals who were pants earning more than $40,000 a year, 11 percent In addition, employee after-tax and employer matching contributions are These two plan components can have a significant Employer Contributions formula information was provided, an algorithm was 60 ............................................. 5 an average of 7.4 percent of salary into their plans. (chart 7). Furthermore, about 9 percent of participants ICI database (42 years old). The median tenure of participants in the sample This Jack L. Issue Brief participant contribution activity in 1999. EBRI and ICI do not provide legal, short planning horizons contributed less). And Bernheim and Garrett (July For individuals with salaries above $80,000, before- rose. sections of this Issue Brief examine the influence that employer contribution and salary information was Wise (November 1997); Ippolito (1997); Xiao (1997); Bernheim and Garrett and about 21 percent set aside more than 10 percent of their salaries on a before-tax employee before- and after-tax contributions and em- benefit programs and work force related issues. sample were in plans that featured an employer variables to examine their effects on participant before-tax contribution rates. currently employed and subject to a similar nondiscrimination test (the Actual Contribution 28 7.4 percent of plans without employer contributions, as discussed in before-tax contributions, however, 61 percent contributed Rappaport, eds., earning $60,000– 1997.This Issue Brief Forecasting Retirement Needs and examines the 1999 contribution Ippolito, Richard A. younger increased their before-tax contribution rate by was five years on the job, compared with a median job tenure of seven years in Participant Salary— Pension Plans and Employee Perfor- The likelihood that a participant loans was expected to contribute 0.6 percentage point 58 48 investment, or employee benefit plan design advice or advice concerning 1996) found that a participant’s education level was not significant when a of Participants) was 9.7 percent of the participant’s salary. ............................................................ 5 contributed at the $10,000 limit. However, it appears impact on participants’ contribution rates. For example, This result is from a regression on a sample of all participants (whether Although participants themselves contributed less on This estimate is based on regression of all participants, whether contribut- were offered employer contributions of more than 5 per- Tax Code Limits (July 1996); Bassett (August 1995); and Andrews (1992). In addition, ICI developed, which examined employer and employee ............................................................... 6 examines the VanDerhei. The stylized estimates reported assumed that the participants were offered a tax contribution rates (though not the amounts who had 401(k) account IRC limits, personal participant characteristics, and plan Percentage [ACP] test). Previous research on a few large 401(k) plans available for a sample of 1.7 million participants from basis. ployer contributions for a participant to a given plan rd 1 the entire 1999 EBRI/ICI database. (See Holden and VanDerhei (January/ contribution. compliance with federal regulations, and this variable indicating whether an employer provides educational materials was Issue Brief should not be used detail below. Profit Sharing/401(k) Council of America. more than 5 percent of their salaries and 21 percent Retirement Wealth $80,000 per year, contributing or not) for whom match rate and match level information was . Philadelphia, PA: The Pension 43 Annual ing or not. The regression model included age, tenure, salary, and employer about 0.06 percentage point for each additional year of mance, Evidence, Analysis, and Policy contributed at the $10,000 limit rose with salary. Thirty- salary (table 2). . Chicago, IL: (Spring 2000) included some results on contribution activity among 401(k) behavior of 1.7 million 401(k) plan participants. 6.9% This Chart 3, Distribution of Participants by Total Forty-three percent of participants had a total loan provision and provided an employer contribution. more of his or her salary to the 401(k) plan than a that among participants not contributing at the 402(g) balances—whether or not a participant contributing up to a 6-percent-of-salary average in plans with an employer contribution, the total cent of their salaries if they contributed at least at the variation in IRC Rules Governing Contributions to contributions for all participants in any given plan, to February 2001).) Pension Planning: Pensions, Profit-Sharing, and contributed) tended to fall as salaries rose because for such purposes. considered (participant contribution rates were higher in plans with more found that participants tended to cluster at match 27 the 1999 EBRI/ICI database. characteristics have on participants’ before-tax contribu- provided or derived. The regression model included age, tenure, salary, plan could not exceed, in 1999, the lesser of $30,000 or contribution (yes/no) variables, in addition to the loan provision (yes/no) participant households surveyed. This is because $10,000 represented more than 25 percent of the partici- Endnotes 1 3 485 7 9 111315 171921 23 252729 they made a contribution contributed more than 10 percent of their salaries. Research Council, The Wharton School of the Univer- and an average of Survey of Profit Sharing and 401(k) Plans: Reflecting Contact EBRI Publications, (202) 659-0670; fax publication orders to age. Older participants, those between their mid-40s and The University of Chicago Press, 1997. nine percent of participants with annual salaries greater Regression analysis sample of participants was drawn from data that the participant with no borrowing privileges.68 The avail- Contribution Rate, 1999 (Percentage of contribution of more than 10 percent of salary and 44 limit, 52 percent could not have contributed at the IRC detailed educational materials). At this time, information was not available for the defined benefit plan match level in a plan offering a 50 percent match rate derive the match formula for the plan. loan provision (yes/no), employer match rate, and employer match level 401(k) Plans 61 ............................................................ Two types of 6 amount going into participants’ accounts was higher in variable, to examine their effects on participant before-tax contribution rates. match level. Other Deferred Compensation Plans. 21, 22 contribution Eighth edition. pant’s salary, and therefore, the participant would have reached the 415(c) ubscriptions/orders • Eleven percent of participants analyzed in this study earning more than $40,000 a 65 IRC, and possibly plan sponsor, contribution limits 20 tion behavior. The contribution behavior of participants levels. In the EBRI/ICI database sample of nearly Plan-specific information on loan provision was available for the majority For this contribution activity analysis, the in 1999. See the Appendix S 25 percent of total compensation. In 2000 and 2001, the limit on employee elective deferrals (before-tax 11 2 Years of Job Tenure (202) 775-6312. Subscriptions to Background EBRI Issue Briefs are included as part of Joulfaian and Richardson (September 2001) used IRS Statistics of Income Influence of Internal Revenue Code 29 found that tenure coverage of participants in the EBRI/ICI database. However, other research Before-tax contribution rates varied widely among both sity of Pennsylvania and, University of Pennsylvania 5.0 percent when variables to examine their effects on participant before-tax contribution rates. Employee Benefit Research Institute (EBRI) 1999 Plan Year Experience. Chicago, IL: Profit Shar- Although the and the Joulfaian, David, and David Richardson. “Who Takes mid-60s, increased their before-tax contribution rate by than $100,000 contributed at the limit, while 7 percent ability of plan assets in the event of financial need seems limit first. Participants) 19 percent had a total contribution of more than 15 per- ................................................................ 6 limit because the plan-imposed limit was lower. This 35 receives the same amount in employer matching contri- those plans. Indeed, the total (employer and employee) of the plans in the sample. Some plans without this information were A regression analysis of the influence of the for a detailed description match variables were created for each plan: (1) the Influence of the IRC §402(g) Limit............................. 7 49 contributions) was $10,500. EGTRRA raised the limit further (see endnote 12). Clark, Goodfellow, Schieber, and Warwick (2000); Clark and Schieber Homewood, IL: Richard D. Irwin, Inc., 1997. activity of 401(k) year contributed at the $10,000 before-tax IRC limit in 1999. Thirteen percent of became binding for some participants. On average, See Holden and VanDerhei (January/February 2001). grouped by age, salary, and/or tenure is also examined. 1 million participants, there was some evidence of individual tax return information combined with contribution information EBRI membership, or as part of a $199 annual subscription to following data were determined for each participant: EBRI Notes and EBRI Issue Briefs. Individual copies are Furthermore, an individual employee’s contribu- has found mixed results for the effect of defined benefit plan coverage on Source: Regression Anal 16 3 ysis Using Data From the EBRI/ICI Participant-Directed Retirement Plan Data 37 highly compensated employees Press, 2000, pp. 95–138. salary exceeds ing/401(k) Council of America, 2000. and non-highly com- average before- 0.07 percentage point for each additional year of age. Advantage of Tax-Deferred Savings Programs? classified as having a loan provision if any participant in the plan had an earning between $60,000–$70,000 a year did so (chart was positively of the data. 59 Investment Company Institute (ICI) have collected as Chart 4, Among Participants Making (1998); Even and Macpherson (May 1997); and Andrews (1992) also found 28 cent of salary. Total contribution rates varied widely to encourage participants to contribute more to their analysis could not identify the participants who, because Other research results (using participant-level data) on the effect of the Limits from W-2 forms to tabulate average contribution and participation rates in bution as a participant contributing up to a 3-percent- Six percent of participants made no contributions and the remainder made contribution rate among participants in plans with match rate on participants’ contribution rates found that Collection Project. 21 plan participants Participant Characteristics participant contribution rates to their 401(k) plans. Par match rate ticipant —the percentage of each dollar contributed by available with prepayment for $25 each (for printed copies) or for $7.50 (as an e-mailed electronic file) by calling ............................................. participants with salaries between $70,000 and $80,000 contributed at the cap, and 8 Andrews, Emily S. “The Growth and Distribution of individuals making more than $100,000 per year 50 Technically, 415(c) limits total “additions” for a given participant’s account clustering at the match level. Indeed, 16 percent of all date of birth, which determined an age and age cohort; In addition, statistical regression analysis is used to tion may be limited as a result of the contribution These nonmatching employer contributions may be a fixed percentage of outstanding loan balance. This may have understated the number of plans 17 that percentage of salary contributed rose with age. 31 match rate on the participant’s contribution rate have been mixed. For Engen, Eric M.; William G. Gale; and Cori E. Uccello. individual retirement accounts (IRAs), defined contribution plans, and pensated employees (chart 2). U.S. Department of Labor. Bureau of Labor Statistics. $100,000 per year. contributions of less than $10,000. tax contribution Evidence from Federal Income Tax Data.” 25). After controlling for other factors, the regression This positive correlation between contribution correlated with National © 2001. part of the EBRI/ICI Participant-Directed Retirement Note: Estimated before-tax contribution rate based on regression results for participants making before- accounts. For example, Clark, Goodfellow, Schieber, and Warwick (2000) found the Contributions, Percentage of Participants at 402(g) among participants, although more non-highly compen- EBRI is a nonprofit, nonpartisan, public policy research organization that of nondiscrimination testing required by the IRC, were (the sum of employer contributions, employee contributions, and any EBRI or from www.ebri.org. Change of Address: EBRI, 2121 K Street, NW, Suite 600, Washington, DC 20037, (202) of-salary match level where the plan sponsor has a the employee that is contributed by the employer, and (2) Influence of Participant Age 62 ....................................... 8 during their wage-earning years. Contributions to 401(k) participants’ salaries or a fixed dollar amount per employee. These contribu- employer contributions averaged 10 percent of salary, offering loans (or participants eligible for loans) because some plans may have 401(k) Plans.” In John A. Turner and Daniel J. Beller, 18 percent of those with salaries between $80,000 and $90,000 were at the limit. tended to contribute a much smaller share of their isolate the effect of individual factors on participant participants example, VanDerhei and Copeland (January 2001), using a variation of a contributed at the match level offered in participant date of hire, which assigned a job tenure and Simplified Employee Pension (SEP) and Keogh plans, for individuals and tax contributions. Values are for participants with $50,000 annual salary activity of other participants in the plan. In order to , 50 years old, and years of tenure 4, 5 36 projected value (replacement rate) of the defined benefit plan had a positive does not lobby or take positions on legislative proposals. forfeitures allocated to the employee’s account) for any given year. In 2000, the • These estimates are based on regression of participants making before-tax Eleven percent of participants analyzed in this study Employee 29 participant contri- “The Adequacy of Household Saving.” Plan Data Collection Project. Employee Benefits in Small Private Establishments, IRC and rate by 401(k) Brookings tions are called nonelective contributions (NECs) and, in certain circum- rates and age was expected for two reasons. First, Tax Journal offered, but had no participants take out, a plan loan. It is likely that this Limit by Age, 1999 (Percentage of Participants analysis also found that participants with higher salaries (September 2001): 669–688. Plan-imposed limits were reported or derived for a sample of 0.9 million Despite the availability of plan loans to partici- 775-9132; fax number, (202) 775-6312; e-mail: Publications Subscriptions@ebri.org. Membership Information: Inquir- prevented from reaching the 402(g) limit. 100 percent match rate. However, the participant in the sequential response regression model (in which the marginal incentive and indicated. Participants are in plans with an employer contribution and a loan provision. compared with 7.4 percent among participants in plans households in 1996. the Influence of Participant Salary match level—the percentage of salary contributed by .................................. 8 accounts play a crucial role in determining how much eds., Trends in Pensions 1992. Washington, DC: U.S. However, it appears that among participants not contributing at the IRC limit, Chart 7 effect on the 401(k) contribution rate, which they noted was inconsistent with salaries to their 401(k) plans than did individuals contribution behavior. their plans (chart 8). Remarkably, there was only slight 415(c) limit was the same as in 1999. In 2001, 415(c) limited additions to the tenure range; salary, which determined a salary range; contributions to the 401(k) plan. The regression model included age, tenure, Characteristics ensure that employees of all income ranges attain the stances, may be used to assist the plan sponsor in complying with ADP tests. omission is small, as the U.S. GAO (October 1997) found that more than participants (see the Appendix for details). 3 Benefit earning more than $40,000 a year contributed at the Papers on Economic Activity plan sponsor limits opportunity cost of contributing at each sequential percentage of compensation 1996. Bulletin 2507. Washington, DC: U.S. Depart- ies regarding EBRI membership, and/or contributions to EBRI-ERF should be directed to EBRI President Dallas . Vol. 2. Washington, DC: plan participants Kusko, Andrea L., James M. Poterba, and David W. economic theory suggests that individuals experience a were more likely to be at the limit. However, the positive ICI is the national association of the American investment company bution rates for The findings in this paper build on previous in Age Group) .............................................................. a 7 pants, only 18 percent of the participants offered the target savings hypothesis (which would argue that participants wouldn’t second example obtains a match equivalent to the first, lesser of $35,000 or 25 percent of salary. EGTRRA raised the dollar amount in with no employer contributions (table 4). salary, plan loan provision (yes/no), and employer contribution (yes/no) Distribution of Effective Match Rates 12 income individuals can expect to receive at retirement. the employee up to which the employer applies the Influence of Participant Job Tenure 52 percent could not have done so because of formal plan-imposed contribution limits .......................... 9 See Chapter 11 of Allen, Melone, Rosenbloom, and VanDerhei (1997) for 95 percent of 401(k) plans that offered loans had a least one plan participant Department of Labor, Pension and Welfare Benefits The IRC §402(g) limit in 1999 was $10,000. In 2000 and 2001, the limit with salaries between $20,000 and $40,000. variation in meeting match levels across salary. For dollar amount of employee contribution; and dollar was measured), found a positive effect for match rates. Clark and Schieber benefits of the 401(k) plan, several provisions of the IRC industry. Its membership includes 8,638 open-end investment companies Salisbury at the above address, (202) 659-0670; e-mail: salisbury@ebri.org 38 Research 30 need to save as much on their own because the defined benefit plan would do $10,000 before-tax Internal Revenue Code (IRC) limit Brookings Institution, December 1999, pp. 65–188. complicate the analysis of the effect of salary on contri- Because so many factors simultaneously influence whether a given ment of Labor, April 1999. was 6.8 percent 415(c) to $40,000 and the percentage to 100 percent; both new limits are life-cycle pattern of saving. participants with up to about 18 years’ tenure; a nega- Wilcox. “Employee Decisions with Respect to 401(k) variables to examine their effects on participant before-tax contribution rates. In other words, younger relationship academic research examining the contribution activity of borrowing privileges in the 1999 EBRI/ICI database additional detail. Chart 5, Among Participants Making Contributions, with an outstanding loan. Among Plans Offering Matching Employer was $10,500. The “Economic Growth and Tax Relief Reconciliation Act of Participant Plan Design despite a lower personal contribution rate. match rate. An (1998) also found that higher match rates produced higher participant ...................................................................... effective match rate was derived by 9 Even small contributions early in an individual’s career, (“mutual funds”), 498 closed-end investment companies, and seven sponsors of Administration, 1992, pp. 149–176. After controlling for all other factors, the regres- below the IRC limit. Chart 2 example, 15 percent of participants earning between some saving for them). They noted that the positive effect was consistent with amount of employer contribution for the individual 12 participant is at the limit or not, regression analysis was used to identify the provide for a framework by which firms must carry out effective Jan. 1, 2002. The dollar amount is then indexed for inflation in The stylized estimates presented assumed that the participants were offered a Institute- b • 2001” (EGTRRA) raises the 402(g) limit to $11,000 in 2002, $12,000 in 2003, Contribution rates increased or decreased as job The Internal in 1999. Thirteen percent of participants with Contributions, 1999 tive correlation began for participants with more than Even, William E., and David A. Macpherson. “Determi- of salary in 1999, individual contribution rates varied ________. bution rates. Because these constraints primarily limited contribution rates. In addition, Even and Macpherson (May 1997) found (in 401(k) participants, by using a large sample of partici- Employee Benefits in Medium and Large Chart 5 people may save less because school expenses, the costs Plans.” In Olivia S. Mitchell and Sylvester J. Schieber, unit investment trusts. Its mutual fund members manage assets of approxi- Percentage of Participants at 402(g) Limit by was not the actually had a loan outstanding. In addition, for partici- 51 66 Age—Older the savings preference hypothesis put forth by Ippolito (1997), which argued Match rates and match levels in this study were sion analysis estimated that a participant in a plan with Even and Macpherson (May 1997), using 1993 Current Population Survey If a single participant in a plan received an employer contribution, the plan effects of each individual factor, while simultaneously taking each factor into multiplying the match rate by the match level. Table 6 Influence of Loan Provision Distribution of P ...................................... articipants by Before-T 10 through compounding, can become sizable balances by $1,000 increments on an annual basis. ax P Bassett, William. “Defined Contribution Plans: The Role loan provision and provided an employer contribution. articipant Contribution Rate, 1999 $20,000–$40,000 a year contributed at the plan match $13,000 in 2004, $14,000 in 2005, and $15,000 in 2006. The limit will be participant. Using contribution and salary information, Education nondiscrimination testing. Actual Deferral Percentage the 1988 CPS sample) that the presence of a matching employer contribution mately $7 trillion, accounting for approximately 95 percent of total industry 42 (Percentage of Participants) tenure rose, depending on whether a participant was Editorial Board: Among P Dallas L. Salisbury, publisher; Steve Blakely, managing editor; Cindy O’Connor, production and distribution. Any articip Revenue Code ants Making Contributions, Percent salaries between $70,000 and $80,000 contributed at age of Participants widely around that figure. This section analyzes the that participants with a taste for savings would want both a defined benefit nants and Effects of Employer Matching Contri- the contribution behavior of highly compensated employ- Private Establishments, 1997. Bulletin 2517. Wash- (CPS) data, found that 81 percent of employees offered a 401(k) plan indicated of acquiring a home, and the expenses of raising a family 18 years’ tenure. eds., was identified as offering an employer contribution and all participants in the Living with Defined Contribution Pensions For example, the regression esti- result of highly . account. Each participant was identified as at the limit or not (with a yes/no pants in a wide range of plan sizes and by examining in 56 Salary, 1999 ................................................................. 7 pants with loans outstanding, loan balances amounted to 401(k) plan • Older participants tended to contribute a higher percentage of their salaries to plans indexed for inflation in $500 increments after 2006. In addition, starting in provided or derived for a subset of nearly 1 million an employer contribution would contribute 0.4 percent- retirement. Participants themselves, as well as plan 22 37 T presents the median and average (mean) values of these Influence of Employer Contributions increased participant contribution rates. ax Code ....................... 10 assets, and represent more than 88 million individual shareholders. of Income, Age, and Match Rates.” Federal Reserve Given the IRC limits on before-tax contributions and typical plan designs, The relationship between age and participant contribution rates was not and Research level, and 17 percent of participants earning between each participant was identified as at the IRC §402(g) and defined contribution plan and would be attracted to employers offering views expressed in this publication and those of the authors should not be ascribed to the officers, trustees, members, or othe (ADP) nondiscrimination tests essentially require that their employers offered matching contributions. plan were coded as having been offered an employer contribution. r variable). A probit regression model was used to estimate the impact of plan at 402(g) Limit by Salary, 1999 16% early or late in his or her tenure. Holding other factors (IRC) applies two the cap, and 18 percent of those with salaries between variations, focusing on how personal participant charac- 2002, participants who are age 50 or older will be allowed to make additional butions in 401(k) Plans.” ees, the regression analysis in this section focused on ington, DC: U.S. Department of Labor, September Working Paper, Oxford, OH: may be of more immediate concern. Older individuals mated that, while a 50-year-old participant earning Philadelphia, PA: The Pension Research Council, The compensated detail the factors that influence contribution activity. >6% of Salary only 14 percent of total account balances (net of the Clark, Goodfellow, Schieber, and Warwick (2000) found that participant than did younger participants, even after factoring out differences in salary and job Chart 6, Estimated Participant Before-Tax participants it was difficult for participants to reach this limit. Indeed, only about 1 linear. A one-year increase in age produced a larger increase in the participant both. In addition, U.S. GAO (October 1997) found that coverage by another plan variables. participants. Among plans offering a matching employer Influence of Employer Match Rate and sponsors of the Employee Benefit Research Institute, the EBRI Education and Research Fund, or their staffs. Nothing herein is sponsors (or employers), can contribute to their 401(k) age point less of his or her salary to the plan than would Bank of New York Research Paper. No. 9517. New to and participant characteristics on the likelihood of a participant being at the 4 Fund. $80,000–$100,000 a year contributed at the plan match limit or not and at the IRC §415(c) limit or not. Informa- Papers summarizing the 401(k) plan asset allocation, account balances, and > 5%–6% “catch-up” contributions. 19 before-tax contributions by highly compensated employ- 52 67 4% 6 contribution rates fell as the match rate rose, and Andrews (1992) found that This estimate is based on regression of participants making participant Formal plan limit information was provided for some participants and like age and salary constant, a participant’s contribu- 49 broad limits on percent of all participants (regardless of income grouping) hit the 415(c) limit. before-tax contribution rate at higher ages than at younger ages. $80,000 and $90,000 were at the limit. However, it $50,000 a year with only one year of tenure would teristics influenced participant contributions. pension had a positive effect on the household’s contribution rate. Miami University and Tallahassee, FL: Florida State non-highly compensated employees making contribu- Most previous research 1999. into contribution activity has usually can save more because their living expenses Wharton School of the University of Pennsylvania, Contribution Rates by Tenure .................................... individuals 52 9 unpaid loan balance). 402(g) limit. The probit regression model included variables for participant be construed as an attempt to aid or hinder the adoption of any pending legislation, regulation, or interpretative rule, or as 13 legal, were more tenure. Participants tended to increase the share of their salary (and amounts) of Salary contribution, the most common match rate offered was 14% a participant in a plan with no employer contributions. loan activities of participants at year-end 1996, 1997, 1998, and 1999 have Non-Highly Compensated Employees Participants earning $40,000 a year or less were excluded because another Match Level In order to ensure typical results that accurately .......................................................... 10 accounts. 23 York: Federal Reserve Bank of New York, August According to data on the 1.7 million 401(k) 24 All rights the presence of an employer contribution reduced participant contribution before-tax contributions to the 401(k) plan. The regression model included age, derived for others. When derived, if any participant in the plan in question 39% level. tion on before- and after-tax participant contributions Limits ees (as a percentage of their eligible compensation ) On the other hand, Even and Macpherson (May 1997) found that if the 7 age, salary, and tenure, as well as plan variables (i.e., loan provision (yes/no), 5% Up to 2% tion rate rose slowly for the first 18 years of tenure accounting, actuarial, or other such professional advice. participants’ 23 38 appears that among participants not contributing at Specifically, it examines the average behavior of several U.S. Department of Labor, Pension and Welfare Benefits University, May 2001. tions. Among contributing non-highly compensated typically decrease relative to income. Second, another contribute about 7.2 percent of salary to a 401(k) plan, and University of Pennsylvania Press, 1998, been published as previous issues of EBRI Issue Briefs and contributing a ICI Perspective, IRC limit prevented them from reaching the $10,000 limit. See endnote 27. used either highly aggregated plan data, 40% individual Chart 7, Distribution of Effective Match Rates Under IRC §414(q), an individual in 1999 who, for the preceding year had An augmented version of the life-cycle theory predicts that the optimal Highly Compensated Employees likely to reach rates. In other research, Munnell, Sundén, and Taylor (December 2000) found contributed to their 401(k) plan as their salaries rose until salaries reached $80,000. tenure, salary, plan loan provision (yes/no), and employer contribution (yes/ contributed the 25 percent of salary permitted by IRC 415(c) in 1999, then the 50 cents on the dollar: 41 percent of participants were However, the regression analysis estimated that total reserved. participants drawn from the 1999 EBRI/ICI database, 401(k) plan was the “main” retirement plan that covered the participant, Appendix represent 401(k) participant contribution activity, ........................................................................ Non-Highly 12 1995. Highly Compensated of Salary and employer contribution (yes/no)). More details of the regression analyses 12% was available for about three-quarters of the partici- do not exceed the contributions of non-highly compen- available through EBRI’s Web site at www.ebri.org and ICI’s Web site at 33 compensation in excess of $80,000, was considered a “highly compensated savings pattern increases with age. For a summary discussion of life-cycle and then declined. that offering a match (a yes/no variable) increased participant contribution contributions to no) variables to examine their effects on participant before-tax contribution plan was assumed to have no plan sponsor limit. On the other hand, if no the IRC limit, 52 percent could not have done so ________. “Employee Participation in 401(k) Plans.” groupings of participants and uses regression analysis employees, the percentage of salary contributed by Administration. Private Pension Plan Bulletin, theory indicates that many individuals find it difficult to an otherwise identical person with 18 years’ tenure pp. 98–112. higher percent- 13 firm and participant data from one or a few select Compensated Employees >4%Emplo –5% yees 24% participant contribution rates were higher. For individuals with salaries above $80,000, before-tax contribution rates (though Among Plans Offering Matching Employer If the employer chooses to make a contribution, it may be a matching the 402(g) are presented in Holden and VanDerhei (forthcoming). EBRI Issue Brief is registered in the U.S. Patent and Trademark Office. ISSN: 0887-137X 0887-137X/90 $ .50+.50 Influence of Employer Contributions several regression models were tested using several offered a matching contribution of 50 cents on the dollar The EBRI/ICI Participant-Directed Retirement employee dollars accounted for two-thirds of the total contribution rates for participants in plans with em- Bassett, William F., Michael J. Fleming, and Anthony P. www.ici.org/economy/perspective.html employee.” This limit is indexed for inflation in $5,000 increments; thus, in models, see Browning and Crossley (Summer 2001). For a more extensive rates, while increasing the match rate decreased participant contribution pants. Table 6 presents contribution activity and rates. The result held true when all participants, whether contributing or not, participant reached the IRC maximum, the plan was assumed to have a plan sated employees (as a percentage of their eligible of Salary 8 40 The EBRI/ 39 43 Two other studies found that other pension coverage was not a significant 401(k) plans: individual participant-based and 401(k) contribution that is contingent on the employee making a contribution, and/or 10% because of formal plan-imposed contribution limits 30% would contribute 7.6 percent of salary (chart 6). to isolate the effect of each personal participant charac- Working Paper participants rose as salary rose. firms, Abstract of 1997, Form 5500 Annual Reports, administrative data on participants from one . Oxford, OH: Miami University and For example, the Wash- Levinson, Gerald M. “KPMG Benefits Survey Reveals focus on long-term concerns. Contributions, 1999 .................................................. Younger workers are age of their 53 11 2000 and 2001, participants earning in excess of $85,000 in salary were discussion, see Engen, Gale, and Uccello (December 1999). limit than 31 not the amounts contributed) tended to fall as salaries rose because IRC, and rates. However, both the yes/no variable and the continuous match rate were included in the regression model. contribution limit, and the contribution rate of the participant with the 13% higher groupings of participants. The four participant sample (table 5). The most common match level was 6 percent of A $10,000 contribution for an individual earning $60,000 a year is a Plan Data Collection Project ................................ 12 dollars contributed to these retirement plans. Thus, ployer contributions were 2.8 percentage points 5 Rodrigues. “How Workers Use 401(k) Plans: The factor in participants’ contribution activity. Clark and Schieber (1998) found The EBRI/ICI database covers only 401(k) participants and does not a nonmatching contribution that is deposited into the participants’ accounts participant characteristic variables for the “typical” compensation) by more than a specified amount. In cases 9 10, 11 32 counted as highly compensated employees. 22% plan-based. below the IRC limit. This analysis could not identify The typical 401(k) plan allows an employee to choose teristic. Both the “average” and regression analyses Tallahassee, FL: Florida State University, April 1999. regression analysis estimated that a 40-year-old partici- variable were included in the same specification and the influence of the ington, DC: U.S. Department of Labor, Winter 2001. more likely to consider retirement a vague and far-off However, an otherwise identical person with 30 years of Low 401(k) Contribution Rate highest contribution rate in the plan was assumed to be the maximum allowed .” Journal of Pension salaries: Influence of Plan Characteristics 16.7 percent contribution rate. For an individual earning $70,000 a year, it is record keeper, or household survey data. ICI Using Chart 8, Participants Contributing at Employer 39 that the defined benefit replacement rate had a small, but statistically younger possibly plan sponsor, contribution limits became binding for some participants. contain information on employees who do not participate in 401(k) plans. regardless of whether they contribute. salary: 49 percent of participants were offered matching 8% 53 than total contribution rates for participants in plans Research by Munnell, Sundén, and Taylor (December 2000) supported this 64 given the importance of employee decision-making, this Bibliography groups predominantly used were: (1) all participants in ................................................................... 13 This suggests that 2.8 percentage points of the 2.6 percentage point Participation, Contribution, and Withdrawal Deci- At the employer’s discretion, employees that are in the top-paid group of 401(k) plan participant. 18% match level was not taken into account. Bassett (August 1995) concluded that where a plan may not meet nondiscrimination testing by the plan. Although this method may have overstated the number of a 14.3 percent contribution rate. insignificant, negative effect on participant contribution rates. And, Bassett 20% Thus, it is not possible to study why employees choose to participate. For the participants who were prevented from reaching ________. “Factors Influencing Participation and Contri- whether to contribute a portion of his or her salary to a found that participant contribution rates varied with U.S. General Accounting Office. “Private Pensions: pant with 10 years’ tenure and earning $20,000 annually event. Retirement and its requisite planning typically job tenure would contribute about 7.4 percent of salary. Planning & Compliance view. They found that a short planning horizon reduced a participant’s . Vol. 25. No. 4 (Winter 2000): Even with participant-level data from the 1999 EBRI/ICI database, Did you read this as a pass-along? Stay ahead of employee benefit 54 difference in the average total contribution rates presented for participants in Contribution Match Level, by Salary, 1999 participants. employees for the preceding year may also be considered highly compensated 14 Endnotes the sample whether contributing or not; (2) participants employer contributions of up to 6 percent of salary the effect of the match rate on participant contribution rates was positive until Could we send a friend or colleague a complimentary ........................................................................ Participant- 15 paper focuses primarily on what factors influence the without employer contributions. participants with formal plan contribution limits, this method did not identify sions.” National Tax Journal. Vol. LI. no. 2 (June Furthermore, other For research examining possible plan sponsor motivations for employer 6% 13% (August 1995) found that the effect of having only a defined contribution plan research on the factors that influence an employee’s decision to participate, see Several descriptive plan variables also were used requirements, highly compensated employees who contribution rate by almost 1 percentage point. However, their specification of • 32 Giving employees the option of borrowing from their 34 plans with and without employer contributions (in table 4) was explained by EBRI Issue Briefs for only $49/ Because the 402(g) limit is a dollar-amount limit, by design, the percentage • Giving employees the option of borrowing from their 401(k) accounts increased employees. An employee is in the top-paid group of employees for any year if issues with your own subscription to the IRC $10,000 limit because of nondiscrimination age, salary, and tenure. IRC Rules Gover 401(k) plan account. Plan sponsors usually make a bution Levels in 401(k) Plans.” would contribute 6.6 percent of salary. An otherwise the match rate exceeded 70 cents on the dollar. this Issues of Coverage and Increasing Contribution Issue Brief examines the roles that contribution ning Contributions to Final Report submitted become more prominent concerns as people age. 50–62. participants limited in ad hoc ways. (Percentage of Participants in Salary Range) lower contribu- ......... 12 contribution activities, see Ippolito (1997), which argued that employers offer >3%–4% For example, was barely positive and statistically insignificant, and concluded that people Appendix copy of EBRI Issue Brief? research has shown that offering an employer contribu- Joulfaian and Richardson (September 2001); Even and Macpherson (May making before-tax contributions; (3) all non-highly contributed. A little more than one-quarter of partici- percentage of income that 401(k) participants contribute factors affecting household contribution rates did not include age, which 1998): 263–289. the employer contribution. The remaining difference resulted from varying of salary contributed at the 402(g) limit falls as salary rises ($10,000 is a in the analysis. All participants were identified by such employee is in the group consisting of the top 20 percent of the employees Directed originally made higher contributions (possibly at the IRC year electronically e-mailed to you or $99/year printed and mailed. 401(k) accounts increased participant contribution Evidence (using participant-level data) on whether increasing the match 7% Among plans with formal contribution limits, the most common maximum of Salary matching contributions to attract and retain workers who are “savers” because 10% 4% participant contribution rates. On average, a participant in a plan offering loans testing—IRC rules that are designed to ensure that whose defined contribution plans were supplemental saved as much in them to Department of Labor Limits for Defined Contribution Plans.” , Pension and Welfare Benefits Letter Report. Madrian, Brigitte C., and Dennis F. Shea. “The Power of 2001); Munnell, Sundén, and Taylor (December 2000); Madrian and Shea Existing research contribution matching a portion of the participant’s identical participant earning $70,000 a year would limits, participant characteristics, and detailed plan might be expected to influence a person’s planning horizon.tion rates, they 55 14 while 5 age, salary, tenure, and loan provision status (among other things) of the 20 percent contribution rate for a participant with a $50,000 salary, a pants were offered a combination of 50 cents on the when ranked on the basis of compensation paid during such year. In addition, tion increased participation in the 401(k) plan. 65 401(k) Plans annually. 13% compensated employees, whether contributing or not; rate increased participation or not is mixed. Clark, Goodfellow, Schieber, and Bernheim, B. Douglas, and Daniel M. Garrett. “The contribution rate applied by plan sponsors was 15 percent of salary. Thirty- savers tend to be more productive; and Even and Macpherson (April 1999), For more information about subscriptions, visit our Web site at Tables and Charts whether or not their plan offered a loan provision. as people whose defined contribution plan was their sole plan. limits) may have a portion of their contributions re- (May 2000); Clark, Goodfellow, Schieber, and Warwick (2000); Even and rates. On average, a participant in a plan offering 41 2% Influence of Par Send an issue to ticipant Age appeared to contribute 0.6 percentage point more of his or her salary to the plan than Retirement participants in the two columns in table 4. Influence of Par employees of all income ranges attain the benefits of ticipant Salary contribution. However, the plan sponsor is not required Administration, May 1997. 10 percent contribution rate for a participant with a $100,000 salary). contribute 7.9 percent of salary. 09/17/01, GAO-01-846. Washington, DC: U.S. General any individual employee with 5 percent ownership in the firm is also counted Suggestion: Inertia in 401(k) Participation and has indicated were more design features play in the contribution decisions made 57 40 Warwick (2000), and Clark and Schieber (1998) concluded that increasing the five percent of participants were estimated to be in plans with this 15 percent which summarized Ippolito’s hypothesis, and an alternative view that percent of 2% U.S. GAO (October 1997) and Even and Macpherson (May 1997) found a and (4) non-highly compensated employees making dollar on contributions up to 6 percent of salary. Macpherson (April 1999); Bassett, Fleming, and Rodrigues (June 1998); Clark www.ebri.org or complete the form below and return it to EBRI. Determinants and Consequences of Financial Educa- >2%–3% Furthermore, all participants were also identified by 45 turned or recharacterized after year-end to move the as a highly compensated employee. loans appeared to contribute 0.6 percentage point 0 Other employer actions also may influence participant contribution match rate increased participation. Bassett, Fleming, and Rodrigues (June limit. Seven percent of participants faced a plan contribution limit of 10 suggested that employers offer matching contributions to help meet nondis- a participant in a plan with no loan provision. 54 positive relationship between salary and contribution rates. Munnell, Sundén, the 401(k) plan. 33 that, among Fidelity Investments. to offer contributions to employees, and some offer no The key IRC provision that applies to participants is IRC by participants and employers in 401(k) plans. Organization Accounting Office, September 2001. Building Futures, Volume II: This estimate is based on regression of participants with any contribution Savings Behavior.” and Scheiber (1998); Kusko, Poterba, and Wilcox (1998); Poterba, Venti, and NBER Working Paper. No. 7682. of Salary likely to reach Ordinary Least Squares (OLS) regression analysis was used to estimate the 69 Plan Data Collection Pr participants in oject Table 1, Average Participant Before-Tax It is the combination of the match rate and the before-tax contributions. For additional detail on the tion in the Workplace: Evidence from a Survey of 25 0 activity. For example, Clark and Schieber (1998) found that participants at 1998) concluded that increasing the match rate did not necessarily increase $60,000 >$60,000– >$70,000Contributing on a Befor Influence of Employer Match Rate and – percent of salary or lower. >$90,000– >$100,000 e- or After-Tax crimination testing requirements. Average contribution rates tended to rise with age whether or not their plan offered an employer contribu- Average before-tax contribution rates tended to rise and Taylor (December 2000) admittedly did not account for IRC limits or plan >$80,000– plan into compliance. 24 As a result, some participants in 41% more of his or her salary to the plan than a partici- (before-tax, and/or after-tax, and/or employer) to the 401(k) plan. The Wise (November 1997); U.S. General Accounting Office (GAO) (October 1997); effects of personal characteristics on the percentage of salary a participant IRC §401(a)(17) defines the annual compensation limit (to be included in Influence of Participant Job Tenure Opportunities and Challenges for Workplace Savings ________. “401(k) Pension Plans: Loan Provisions En- Name Cambridge, MA: National Bureau of Economic Re- other things, two §402(g), which in 1999 capped an individual’s annual contribution or a contribution whether the employee The principal findings of the analysis are as follows. the limit or less $70,000 $80,000 $100,000 >15% firms that distributed specifically tailored information on the firm’s 401(k) their 20s Contribution Rates, by Age and Salary, 1999 participation. Kusko, Poterba, and Wilcox (1998) tracked employees through Address >0–1% >2%–3% >4%–5% ........... >6%–7% 8 >8%–9% $90,000 >10%–11% >12%–13% >14%–15% match level that determines the effective match rate— • constraints and found a negative relationship between salary and participant Older workers were more likely to meet the before-tax 66 regression analyses, see Holden and VanDerhei (forth- regression model included age, tenure, salary, plan loan provision (yes/no), Bernheim and Garrett (July 1996); Bassett (August 1995); Papke (1995); Households.” NBER Working Paper. No. 5667. Cam- contributed on a before-tax basis to the 401(k) plan. The regression models nondiscrimination tests) to be $160,000 in 1999. In 2000 and 2001, the 68This document is available electronically in pdf among participants making before-tax contributions, 15 tion of any kind. among higher-salary groups of 401(k) plan participants, • Total contributions—the sum of employee and employer contributions—were higher these plans were unable to contribute at the IRC §402(g) This exercise produced the first tier of any given match formula, but may Plan Design The Profit Sharing/401(k) Council of America (PSCA) (2000) reported that pant in a plan with no loan provision. 50 Match Level 20 Basis Plan administrators that are either EBRI or ICI mem- plan had higher contribution rates, and Bernheim and Garrett (July 1996) match rate changes occurring over time in one 401(k) plan and found that contributes or not. in America, A Report on Corporate Defined Contribu- hance Participation But May Affect Income Security This section examines how em- Salary Range search, May 2000. contribution rates. Clark, Goodfellow, Schieber, and Warwick (2000) also 401(k) plan because of before-tax contributions to all 401(k) plans at $10,000. and employer contribution (yes/no) variables to examine their effects on the Organization Papke and Poterba (1995); and Andrews (1992). Table 2, Average Participant Before-Tax contributed at included participant age, salary, and tenure as well as variables reflecting 401(a)(17) annual compensation limit was set at $170,000. EGTRRA Source: Tabulations From the EBRI/ICI Participant-Directed Retirement IRC contribution limit than younger workers. Eigh- that is, the percentage of salary that the employer would coming). City/State/ZIP have missed second or multiple tiers in the matching calculation. bridge, MA: National Bureau of Economic Research, 70 percent of all plans in their survey allowed participants to contribute on a within any given salary grouping. For example, contrib- found that higher contributions prevailed among participants in plans where Up to a point, plan participants tended to increase their only when the match rate was dramatically increased did non-contributors but only until IRC, and possibly plan-sponsor, contribu- format. To order online, visit ww.ebri.org/store/ Additional detail for plan sponsor contribution for participants who received an employer contribution as part of their 401(k) plan limit. In addition, some plans impose formal found a negative relationship between salary and contribution rates. When total contribution rates achieved by participants. A similar estimate was Plan Data Collection Project. bers provided records on participants in 401(k) plans plan characteristics (i.e., loan provision (yes/no), and employer contribution Source: Tabulations From the EBRI/ICI Participant-Directed Retirement Plan Data Collection Project. increased the limit to $200,000 in 2002 and then indexed it for inflation in design character- ployer contributions affected employee contributions, This cap applies to the aggregate before-tax contribution tion Plans for Some.” . Boston, MA: Fidelity Investments, 2001. Source: Letter Report, Tabulations From the EBRI/ICI P 10/01/97 GAO-HEHS-98-5. articipant-Directed Retirement Plan Data Collection Project. Munnell, Alicia H.; Annika Sundén; and Catherine their higher • before-tax basis only; about 2 percent of plans allowed participants to Total contributions—the sum of employee and em- 6 the limit, Contributions to 401(k) Plans a the employer provided financial educational information. contribute for the participant, if the participant contrib- Contribution Rates, by Age and Tenure, 1999 decide to start participating in the plan. Otherwise, they observed stability in .......... 9 Address The notable exception is VanDerhei and Copeland (January 2001), which teen percent of the participants in their 60s and 69 highly compensated employees were included in the regression sample, a July 1996. This section examines the employer contribution feature Your Name obtained when all participants, whether contributing or not, were included in Employees may be able to choose whether to make In addition, regressions analyzing total contributions (not just participant ebriib.htm or call EBRI Publications at (202) The effective match rate is the match level multiplied by the match rate. uting participants in their 20s earning between $40,000 (yes/no) variables). Several combinations of variables were tested to ensure contribution rates the longer they stayed in a job. Note: Excludes participants earning less than $40,000 a year because another IRC limit prevented them from reaching limits (for about 0.9 million participants) and matching $5,000 increments. This limit is important in the calculation of nondiscrimi- tion limits restrained some participants’ contributions. than for those who did not. The average total contribution rate was 10 percent of contribution limits that fall below the IRC limits for contribute on an after-tax basis only; and 17 percent of plans allowed both Note: Sample of participants with before-tax participant contributions. administered by these organizations in 1996, 1997, 1998, b Hewitt Associates, LLC. without considering how the contribution amount was In addition, using data from one large 401(k) plan, Madrian and Shea participation status of participants tracked over time. Washington, DC: U.S. General Accounting Office, Trends & Experience in 401(k) istics—plan loan features and employer contributions— Taylor. “What Determines 401(k) Participation and analyzed a sample of 163,346 participants drawn from the 1998 EBRI/ICI of an individual participant regardless of the number of negative relationship between salary and participant contribution rates salaries. ployer contributions—were higher for participants Sample of nearly 1 million participants for whom employer matching 15 percent of the IRC Sec. 402(g) limit. Tabulated from a sample of 0.8 million p the regression. before-tax contributions) were also run. articipants. uted at least up to the match level. For example, a plan that the results presented were typical and accurately characterize 401(k) nation tests because it has an impact on the contribution rate of highly 15 percent of those in their 50s met this limit, while 26 in more detail, considering the design of employer City/State/ZIP Browning, Martin, and Thomas F. Crossley. “The Life- before- and after-tax contributions. Eleven percent of plans in the PSCA before- and/or after-tax contributions to the 401(k) plan, 659-0670. and $60,000 a year contributed an average of 6.8 per- However, contribution rates tended to drop off among formulas (for nearly 1 million participants) was reported Within any given age group, the average contribution (May 2000) examined the effect of introducing automatic enrollment (where 63 salary for employees in plans offering an employer contribution, compared with some participants. database to examine the effect of employer match formulas on a participant’s resulted here as well. contribution information was provided or derived.44, 45 and 1999. • Before-tax contributions accounted for most of the These administrators included mutual fund determined. Plans October 1997. . Lincolnshire, IL: Hewitt Associates, 1999. compensated employees by influencing how much of their salaries may be influenced participant contributions. Contributions?” CRR Working Paper 2000-12. Chest- that received an employer contribution as part of their survey had no participant contributions. In addition, U.S. DOL (September only 5 percent of those in their 20s did so. with a match rate of 50 cents on the dollar and a match decision to contribute an additional percentage point of salary. Cycle Model of Consumption and Saving.” Journal of matching contributions and its effect on participant cent of salary, while similarly salaried participants in long-tenured employees. For example, contributing Mail to: EBRI, 2121 K Street, NW, Suite 600, Washington, DC 20037 rate tended to rise for salaries up to $80,000 a year, and 7.4 percent for those in plans not offering an employer contribution. Mail to: EBRI, 2121 K Street, NW, Suite 600, Washington, DC 20037 or Fax to: (202) 775-6312 or Fax to: (202) 775-6312 EBRI Issue Brief Number 238 • October 2001 • © 2001. EBRI 10 20 14 16 12 2 6 18 4 8 October 2001 • EBRI Issue Brief October 2001 • EBRI Issue Brief October 2001 • EBRI Issue Brief October 2001 • EBRI Issue Brief October 2001 • EBRI Issue Brief October 2001 • EBRI Issue Brief October 2001 • EBRI Issue Brief October 2001 • EBRI Issue Brief October 2001 • EBRI Issue Brief October 2001 • EBRI Issue Brief October 2001 • EBRI Issue Brief October 2001 • EBRI Issue Brief October 2001 • EBRI Issue Brief October 2001 • EBRI Issue Brief October 2001 • EBRI Issue Brief October 2001 • EBRI Issue Brief October 2001 • EBRI Issue Brief October 2001 • EBRI Issue Brief October 2001 • EBRI Issue Brief October 2001 • EBRI Issue Brief 19 17 15 11 13 3 7 1 9 5 Percentage of Participants Issue Brief Percentage of Participants Issue Brief c Match Level Percentage of Participants in Salary Range Percentage of Participants in Salary Range Percentage of Salary Contributed by Employee (Percentage of Participants in Age Group)

Volume 238

Pages 20

EBRI Issue Brief

Sarah Holden

Jack VanDerhei

Retirement