Use of Target-Date Funds in 401(k) Plans, 2007

Transcription

1 March 2009 No. 327 Date Funds in 401(k) Plans, 2007 By Craig Copeland, EBRI E X E C U T I V E S U M M A R Y WHAT THEY ARE: Target-date funds (also called life-cycle funds) are a type of mutual fund that automatically rebalances its asset allocation following a predetermined pattern over time. They typically rebalance to more conservative and income-producing assets as the participant s target date of retirement approaches. WHY THEY RE IMPORTANT AND GROWING: Of the 401(k) plan participants in the EBRI/ICI 401(k) database who were found to be in plans that offered target-date funds, 37 percent had at least some fraction of their account in targetdate funds in Target-date funds held about 7 percent of total assets in 401(k) plans and the use of these funds is expected to increase in the future. The Pension Protection Act of 2006 made it easier for plan sponsors to automatically enroll new workers in a 401(k) plan, and target-date funds were one of the types of approved funds specified for a default investment if the participant does not elect a choice. EBRI/ICI 401(K) DATABASE: This study uses the unique richness of the data in the EBRI/ICI Participant-Directed Retirement Plan Data Collection Project, which has almost 22 million participants, to examine the choices and characteristics of participants whose plans offer target-date funds. EFFECT OF AGE, SALARY, JOB TENURE, AND ACCOUNT BALANCE: Younger workers are significantly more likely to invest in target-date funds than are older workers: Almost 44 percent of participants under age 30 had assets in a target-date fund, compared with 27 percent of those 60 or older. Target-date funds appeal to those with lower incomes, little time on the job, and with few assets. On average, target-date fund investors are about 2.5 years younger than those who do not invest in target-date funds, have about 3.5 years less tenure, make about $11,000 less in salary, have $25,000 less in their account, and are in smaller plans. EFFECT OF AUTOMATIC ENROLLMENT: While the EBRI/ICI database does not contain specific information on whether a 401(k) plan had automatic enrollment, this analysis was able to proxy for those who could be identified as automatically enrolled. The data show that workers who were considered to be automatically enrolled in their employer s 401(k) plan are significantly more likely to invest all their assets in a target-date fund than those who voluntarily joined, and were also less likely to have extreme all-or-nothing asset allocations to equities. EQUITY ALLOCATIONS AND FUND FAMILIES: One of the major questions surrounding target-date funds is the equity allocations that these funds use over time (the so-called glide path ) as a participant s retirement target date approaches. The glide paths of different target-date funds have significantly different shapes and starting/ending equity allocations. As of 2007, the equity allocation ranges from about percent for 2040 funds (for workers about 30 years away from retirement), and from percent for 2010 funds (for workers one year away from retirement) a 40 percentage-point difference. Moreover, the fund families change their relative rank in equity allocation within the different fund years. This analysis finds that the relative rank of the equity allocation within a target-date fund does not appear to affect the percentage of participants investing all their account into that fund. Nevertheless, investors in specific fund families are more likely to invest all their assets in a single target-date fund from that family. A research report from the EBRI Education and Research Fund 2009 Employee Benefit Research Institute

2 Craig Copeland is senior research associate at EBRI. This Issue Brief was written with assistance from the Institute s research and editorial staffs. Any views expressed in this report are those of the author and should not be ascribed to the officers, trustees, or other sponsors of EBRI, EBRI-ERF, or their staffs. Neither EBRI nor EBRI-ERF lobbies or takes positions on specific policy proposals. EBRI invites comment on this research. Copyright Information: This report is copyrighted by the Employee Benefit Research Institute (EBRI). It may be used without permission but citation of the source is required. Recommended Citation: Craig Copeland, The Date Funds in 401(k) Plans, 2007, EBRI Issue Brief, no. 327, March Report availability: This report is available on the Internet at Table of Contents Introduction... 4 Prior Research... 4 Target-Date Fund Use... 5 Automatic Enrollment Target-Date Funds... 5 Tenure and Number of Target-Date Funds Used... 6 Year of Target Date... 6 and the Level of Target-Date Fund Use Glide Paths Conclusion References Endnotes Figures Figure 1, Percentage of 401(k) Plan Participants in a Plan That Offers a Target-Date Fund Who Use Them, by Various Factors, and the Distribution of Use Across Those Factors, Figure 2, Distribution of Target-Date Fund Users, by Auto-enrollment and Non-auto-enrollment Status and Various Demographic and Plan Characteristics and Participant Choices, Figure 3, Percentage of Target-Date Fund Users Having All of Their Assets in Target-Date Funds, by Plan Size and Automatic Enrollment Status, Figure 4, Percentage of Target-Date Fund Users Having All of Their Assets in Target-Date Funds, by Account Balance and Automatic Enrollment Status, Figure 5, Distribution of the Number of Target-Date Funds Used by Those Having Them and the Percentage Who Allocated All to Target-Date Funds, by Tenure, Figure 6, Distribution of Target-Date Fund Use, by Date of the Fund and Age of the 401(k) Plan Participants Using Target-Date Funds, Figure 7, Distribution of Target-Date Fund Users, by Year of Target-Date and Plan Size, Auto-enrollees and Nonauto-enrollees, ebri.org Issue Brief March 2009 No

3 Figure 8, Distribution of Target-Date Years for Target-Date Fund Users by Company Stock Ownership and Autoenrollment Status, and Having All Assets in the Target-Date Fund, Figure 9, Distribution of Equity and Company Stock Allocation for Those With a Target-Date Fund and Other Asset Allocations, by Date of the Target-Date Fund and Auto-enrollment Status, Figure 10, Distribution of, by Target-Date Fund Use and Age, Figure 11, Distribution of, by Target-Date Fund Use and Salary, Figure 12, Distribution of by Target-Date Fund Use and Plan Size, Figure 13, Distribution of, by Target-Date Fund Use and Account Balance, Figure 14, Equity Allocation Glide Paths of 10 Target-Date Fund Families, End of Year Figure 15, Target-Date Fund Equity Allocation and the Relative Rank of This Equity Allocation Within a Target- Date Year For Various Fund Families, Average Total Outside of the Target-Date Fund Equity Allocation, and Average Share This Allocation Represents of Participant's Using Each Target-Date Fund, by Company Stock Ownership, Figure 16, Percentage of Target-Date Participants Who Have All Their 401(k) Plan Assets in the Target-Date Fund, by Rank (Lowest to Highest) of Equity Allocation Within the Target-Date Fund, and by Year of the Target-Date Fund, Figure 17, Percentage of Target-Date Participants Who Have All Their 401(k) Plan Assets in the Target-Date Fund, by Rank (Lowest to Highest) of Equity Allocation Within the Target-Date Fund, Year of the Target-Date Fund, and Low or High Account Balance, ebri.org Issue Brief March 2009 No

4 Date Funds in 401(k) Plans, 2007 By Craig Copeland, EBRI Introduction Target-date funds (also called life-cycle funds ) are a type of mutual fund that is increasingly important in 401(k) retirement plans. These funds automatically rebalance their asset allocation following a predetermined pattern over time to a specified target retirement date, typically rebalancing to more conservative and income-producing assets as the participant s retirement target date approaches. These funds have proliferated and assets in target-date funds represented about 7 percent of total assets in 401(k) plans in 2007 (VanDerhei, Holden, Alonso, and Copeland 2008). The use of these funds is expected to continue to grow in the future, as the push from plan sponsors and policymakers for better diversification of 401(k) participants assets intensifies. The Pension Protection Act of 2006 (PPA) made automatic enrollment of new workers in their employers 401(k) plans easier for plan sponsors to implement: Target-date funds were one of the types of approved funds specified under PPA regulations for a default investment (funds in which contributions are placed in an automatic enrollment arrangement if the participant does not elect a choice). However, as some policymakers in Congress have focused on, the allocations to the various asset classes in the targetdate funds vary widely, especially to equity investments. 1 Consequently, there is a growing debate over what the "appropriate" allocation to equities is over the life-cycle of a worker s career even though, in many cases, the appropriate allocation would vary for different workers, both across plans and even within a plan. 2 As reflected by PPA s endorsement of them as a default investment, target-date funds have proliferated, in general, as a way to make diversified 401(k) investing more automatic in 401(k) accounts. However, despite all these developments surrounding target-date funds, there is limited research on how individuals use these funds, which individuals use these funds, and what participants will do if they are automatically enrolled in them. While some plan administers have looked at the use of target-date funds in the plans they administer, 3 no analysis has been done across a broad sample of the 401(k) plan participants. Therefore, this study uses the unique richness of the data in the EBRI/ICI Participant-Directed Retirement Plan Data Collection Project, which has 21.8 million participants from 56,232 plans across a spectrum of plan administrators. 4 In this database, 67.3 percent of the plans offered target-date funds as an investment option, and 37 percent had at least some fraction of their account in target-date funds in This study focuses on only those 401(k) participants identified as being in plans that offered target-date funds, as the goal is to examine participants choices when target-date funds are offered. This Issue Brief investigates the percentage of participants investing in target-date funds in 2007 across various demographic characteristics. Furthermore, it determines whether the entire account balance is allocated to target-date funds for those who invest in them. To control for the potential differences that are likely to result from those defaulted into the funds by automatic enrollment versus those actively choosing the funds, those who can be proxied as being auto-enrolled are compared with those who are not proxied as being auto-enrolled. Moreover, equity allocations outside of the target-date funds are studied both at the aggregate level of the target-date year and at the individual target-date fund level. Overall asset allocation to equities is also investigated, by comparing the equity for those who did not use target dates, those who have only some of their assets in target-date funds, and those who have all of their assets in target-date funds. Prior Research The available research on target-date funds has been rather limited. Using data from Vanguard the use of target-date funds in their plans has been studied. In the most recent study, 59 percent of the plans were found to offer targetdate funds and 27 percent of participants were invested in target-date funds as of year-end 2007 (Nessmith and Utkus, ebri.org Issue Brief March 2009 No

5 2008). Furthermore, they found investors in target-date funds on average to be younger, to have lower account balances, and to have shorter tenure. Fidelity analyzed the percentage of plans they administer that offer automatic enrollment and use target-date funds as the default option (Fidelity 2009). They found that automatic enrollment increased from 2 percent of plans in 2006 to 16 percent in 2008, and the percentage using life-cycle (target-date) default fund options increased from 14 percent in 2006 to 60 percent in 2008 (of those offering automatic enrollment) (Fidelity, 2009). Ibbotson released a study of target-date fund performance in the fourth quarter of 2008 and for all of 2008, and found that all the funds had negative returns for the year (only one had positive returns during the fourth quarter). In addition, the dispersion of the returns was significantly large due to the considerable differences in the asset allocations within the funds (Ibbotson 2008). A study by Cliff and Lucas (2009) compares the equity exposure of target-date fund families across target-date years and found significant differences across the families. By examining the equity exposure and performance of the funds within the families, they discuss key tools for establishing benchmarks for target-date funds. They conclude there is still a great deal to learn about what path the equity allocation should take as the target date nears. Consequently, as target-date funds become more popular choices for 401(k) participants, both participants and professionals need greater understanding of the funds and how they are used. This Issue Brief aims to fill some of these holes, but is not able to address longitudinal issues of participant choices (over time) in the use of target-date funds, either from automatic enrollment or from participants own voluntary selections. Target-Date Fund Use Of the 401(k) plan participants from the EBRI/ICI database who were found to be in plans that offered target-date funds, 36.9 percent had at least some fraction of their account in target-date funds in 2007 (Figure 1). The likelihood of a participant investing in target-date funds decreased as the age of the participant increased: 43.7 percent of participants under age 30 had assets in a target-date fund, compared with 27.0 percent of those ages 60 or older. Those with salaries less than $40,000 were more likely to use target-date funds than those with salaries larger than this amount. 5 Furthermore, as tenure and account balance increase, the likelihood of the participant using target-date funds declines. Consequently, those who use target-date funds compared with those who do not are more likely to be younger; to have lower salaries, less tenure, and smaller account balances; and/or to be in plans with a smaller number of participants. The average target-date fund investor is about 2.5 years younger than those who do not invest in targetdate funds. They make about $11,000 less on average in salary, have about 3.5 years on average less in tenure, have $25,000 on average less in their account, and are in plans with an average of 1,200 fewer participants. Automatic Enrollment Target-Date Funds One of the areas in which target-date funds are expected to dramatically increase is in plans that offer automatic enrollment and use target-date funds as the default investment option. 6 The EBRI/ICI database does not contain specific information on whether a plan had automatic enrollment. However, given the importance of the effects of automatic enrollment on the use of target-date funds, this analysis undertook an algorithm to create a proxy for those who could be identified as automatically enrolled. Auto-enrollees in this study were determined by identifying those participants who had less than two years of tenure, had 90 percent or more of their account balance in target-date funds or target-date funds plus company stock, and were in a plan where at least 60 percent of the participants who had less than two years of tenure were 90 percent or more invested in a target-date fund. While these criteria are likely to miss some of those who were automatically enrolled (especially those who changed their allocations significantly after being automatically enrolled or those who ebri.org Issue Brief March 2009 No

6 were automatically enrolled with a different default investment than a target-date fund), they are the most likely to identify those who were automatically enrolled into target-date funds and had made no changes in their investments during the year. For this study, this definition is defensible, as the focus is on the use of target-date funds in 2007, not over time; the individual must be invested in a target-date fund to be investigated further in this study; and automatic enrollment with the use of target-date funds as the default option did not accelerate significantly until after , 8 Among the participants who invested in target-date funds that could be completely identified within the database (name of fund, target-date year, and asset allocation within the fund by target-date year), 7.2 percent were determined to be auto-enrollees under this methodology (Figure 2). 9 In general, auto-enrollees were younger, had lower salaries, and were in later-dated target-date funds. 10 Furthermore, auto-enrollees were also more likely to be in the largest plans, to have all their account balance in target-date funds, to use only one target-date fund, and to have 75 percent to 89 percent of their assets in equities 11 (Figure 2). In particular, 33.3 percent of those determined to be autoenrollees were younger than age 30, while only 13.7 percent of those determined not to be auto-enrollees were younger than age 30. Approximately 50 percent of those determined to be auto-enrollees had salaries less than $20,000, compared with just over 15 percent of those using target-date funds but were not determined to be autoenrollees. About 55.5 percent of auto-enrollees were in plans with more than 5,000 participants, compared with 46.5 percent who were not auto-enrollees. Furthermore, 73.8 percent of auto-enrollees had a total equity allocation (inside the target-date fund plus any equity outside the target-date funds) of 75 percent to 89 percent. The non-auto-enrollees had a more diverse distribution, as only 40.2 percent had a total equity allocation in this range. A larger percentage of these non-auto-enrollees had allocations of 90 percent or more to equities or allocations of less than 75 percent to equities than the auto-enrollees had. Another characteristic of automatic enrollment is the likelihood of being invested only in target-date funds. Those identified as auto-enrollees were significantly more likely to have all their assets invested in the target-date funds. Except for participants in the largest plans (more than 10,000 participants), more than 90 percent of those automatically enrolled into target-date funds had all of their allocation in target-date funds (Figure 3). However, for those who appeared to select target-date funds on their own, 50 percent of those in the smallest plans (declining to 30 percent of those in the largest plans) had all of their assets in target-date funds. A similar result held true across account balance size. Among auto-enrollees, except for those with balances less than $5,000 (where 90 percent were invested all in target-date funds), approximately 80 percent of those investing in target-date funds had all of their assets in target-date funds (Figure 4). However, among those who were not autoenrolled, the likelihood of being completely invested in target-date funds decreased significantly as the account balance increased. More than 60 percent of target-date investors not auto-enrolled with account balances less than $5,000 had all their assets in target-date funds, compared with just over 10 percent of those with balances of $200,000 or more. Tenure and Number of Target-Date Funds Used As the participant's tenure increases, it is less likely that they are invested entirely in target-date funds (Figure 5). 12 For non-auto-enrollees, the percentage not completely invested in target-date funds increases from 48.4 percent for those with less than two years of tenure to 76 percent for those with 20 or more years of tenure. (By definition of the proxy used in this study, auto-enrollees have tenure of less than two years.) The portion of those with all their assets in a target-date fund, using only one fund, stays relatively constant across tenure levels at about percent (calculated from Figure 5). Year of Target Date The theory behind a target-date fund year is to match an investor to a fund that has the closet date to the investor's expected year of retirement. Therefore, a fund with a date closer to the current year is designed for an older individual about to retire, while funds with dates far into the future are designed for younger individuals whose retirement is still ebri.org Issue Brief March 2009 No

7 Figure 1 Percentage of 401(k) Plan Participants in a Plan That Offers a Target-Date Fund Who Use Them, by Various Factors, and the Distribution of Use Across Those Factors, 2007 Percentage Distribution of All Distribution of Distribution of With Target- Participants With Those Not Those Investing Date Fund Access to Target- Investing In in Target- When Offered Date Funds Target-Date Funds Date Funds Average/Percentage Age All 36.9% 100% 100% 100% Under or Older Salary a $59,757 $64,375 $52,986 <$20, % 9.6% 19.0% $20,000 $39, $40,000 $59, $60,000 $79, $80,000 $99, $100,000 or more Tenure b Less than 2 years % 11.4% 22.9% 2 to less than 5 years to less than 10 years to less than 20 years to less than 30 years or more years Total Assets $62,165 $71,426 $46,333 <$5, % 26.4% 37.2% $5,000 $9, $10,000 $19, $20,000 $39, $40,000 $59, $60,000 $99, $100,000 $199, $200,000 or more Plan Size (number of participants) 11,971 12,401 11, % 0.2% 0.4% , ,001 2, ,501 5, ,001 10, > 10, Source: EBRI tabulations from the EBRI/ICI Participant-Directed Retirement Plan Data Collection Project. a This category includes only the observations with salary information. b This category includes only those observations with tenure data. ebri.org Issue Brief March 2009 No

8 Figure 2 Distribution of Target-Date Fund Users, by Auto-enrollment and Non-auto-enrollment Status and Various Demographic and Plan Characteristics and Participant Choices, 2007 All With All With Target Date Auto- Non-auto- Target Date Auto- Non-auto- Funds enrollees enrollees Funds enrollees enrollees All 100.0% 7.2% 92.8% Number of Target Date Funds Used Age % 97.2% 86.8% Under or more or Older Total Equity Allocation Salary a 1-9% <$20, $20,000 $39, $40,000 $59, $60,000 $79, $80,000 $99, $100,000 or more Average Target Plan Size Date Year (number of participants) 2000 b , ,001 2, ,501 5, ,001 10, > 10, All Assets Allocated to Target-Date Funds? No Yes, One Fund Yes, More than One Fund Yes, One Fund and Company Stock Yes, More than One Fund and Company Stock Source: EBRI tabulations from the EBRI/ICI Participant-Directed Retirement Plan Data Collection Project. a This breakdown only includes those participants with complete salary data. ebri.org Issue Brief March 2009 No

9 distant. When examining participants by age and year of target-date fund, approximately three-fourths of target-date investors use a fund that has a target date near an expected retirement year for their age cohort. For example, 76.8 percent of participants investing in target-date funds who were younger than 30 years of age use a target fund dated from 2040 to 2050, while 83.6 percent of the participants 60 years old or older use a target-date fund of 2015 or earlier (Figure 6). This finding is even stronger for those who were identified as auto-enrollees, with approximately 90 percent of those in a target-date fund with a year that correlates to the expected retirement year of each age cohort. Among those under age 30, 93.4 percent are in a fund with a target date of 2040 or later, and 86.8 percent of those ages 60 or older used a target-date fund with a year of 2010 or earlier. However, for those who were not identified as auto-enrollees, the distribution is not as concentrated at the years of expected retirement of the various cohorts; nevertheless, still a significant majority within each age cohort is in funds that match the expected year of their retirement. These findings suggest that most (although not all) 401(k) participants using these funds understand the retirementdate concept and are choosing age-appropriate funds. Those automatically enrolled in a target-date fund are especially likely to be an age-appropriate fund. Larger plans typically have more investment options in their 401(k) plans than do small plans. 13 These differences in the number of options could influence how target-date funds are used among participants of plans of different sizes. When examining the distribution of target-date fund participants by year of the target date, the distribution of participants across these years was very similar across the plan sizes (Figure 7). One difference that does emerge is that smaller-plan participants were more likely to be in 2050 funds than were participants in the largest plans. However, the participants in the larger plans were more likely to be in 2040 funds, so if the participants are grouped into early-, middle-, and late-dated fund years ( , , and 2040 or later, respectively), the distributions are almost identical across plan sizes. Those who were not identified as auto-enrollees had distributions that were similar to that for all the target-date investors. In contrast, among participants who were identified as auto-enrollees, participants in smaller plans appear were less likely to use target-date funds further into the future, while those using target-date funds with more immediate dates are almost equal across plan sizes (Figure 7). Among target-date fund auto-enrollee investors in plans with one to 10 participants, 35.0 percent were in target-date funds dated 2040 or later. This gradually increased as plan size became larger, reaching approximately 50 percent for both participants of plans with 5,001 10,000 participants and with more than 10,000 participants. The percentage of target-date fund investors in funds dated 2015 or earlier remained relatively constant across plan sizes, hovering around 10 percent. 14 Therefore, a higher percentage of participants in the smaller plans were in target-date funds dated from The existence of company stock in a plan could also affect the use of target-date funds, but there is little difference in the distribution of participants across target-date years with and without company stock as an option. What difference there is appears to be a somewhat lower likelihood of investing in a later-year target-date fund when a participant has a company stock option (Figure 8). This appeared to hold true for both those identified as auto-enrollees and for those who were not. For example, among non-auto-enrollees who had company stock, 18.0 percent were invested in a fund with a target date of 2040 or later, compared with 23.9 percent for those without company stock. 15 Furthermore, among those who were not auto-enrollees, the likelihood of being completely invested in target-date funds increases as the year of the target date goes further into the future: 29.5 percent for those in 2000 funds, compared with 66.4 percent for those in 2050 funds (Figure 8). Among auto-enrollees, the likelihood of being completely invested in target-date funds remained just below 90 percent for those invested in each target-date year, until jumping to 96 percent for those in 2050 funds. A standard feature of target-date funds is that the further the fund year is into the future, the greater the allocation to equities. Consequently, investors in target-date funds who also invest in other options in their 401(k) plans may invest differently outside of the target-date fund, depending on the year of the target-date fund used. In Figure 9, the ebri.org Issue Brief March 2009 No

10 Figure 3 Percentage of Target-Date Fund Users Having All of Their Assets in Target-Date Funds, by Plan Size and Automatic Enrollment Status, 2007 Percentage of All Assets in Target-Date Funds Percentage Auto-enrolled Figure 4 Percentage of Target-Date Fund Users Having All of Their Assets in Target-Date Funds, by Acount Balance and Automatic Enrollment Status, 2007 Plan Size Non- (number of participants) All Auto-enrollees Auto-enrollees All % 96.8% 50.8% 8.9% , ,001 2, ,501 5, ,001 10, > 10, Source: EBRI tabulations from the EBRI/ICI Participant-Directed Retirement Plan Data Collection Project. Percentage All Assets in Taget-Date Funds Percentage Auto-enrolled Non- Total Assets All Auto-enrollees auto-enrollees All <$5, % 90.2% 62.5% 16.3% $5,000 $9, $10,000 $19, $20,000 $39, $40,000 $59, $60,000 $99, $100,000 $199, $200,000 or more Source: EBRI tabulations from the EBRI/ICI Participant-Directed Retirement Plan Data Collection Project. Figure 5 Distribution of the Number of Target-Date Funds Used by Those Having Them and the Percentage Who Allocated All to Target-Date Funds, by Tenure, 2007 All Assets Allocated to Target-Date Funds No company stock With company stock yes, more yes, more yes, than yes, than Tenure No one fund one fund one fund one fund All 54.7% 39.5% 2.2% 3.3% 0.3% Less than 2 years to less than years to less than 10 years to less than 20 years to less than 30 years or more years Auto-enrollees Non-auto-enrollees Less than 2 years Source: EBRI tabulations from the EBRI/ICI Participant-Directed Retirement Plan Data Collection Project. ebri.org Issue Brief March 2009 No

11 Figure 6 Distribution of Target-Date Fund Use, by Date of the Fund and Age of the 401(k) Plan Participants Using Target-Date Funds, 2007 Average Year of All Target Date Fund(s) Used Under or older 2000 a 3.0% 2.7% 3.0% 3.7% 13.5% Auto-enrollees Average Year of Target Date Fund(s) Used Under Age or older 2000 a 1.6% 2.0% 2.5% 3.3% 13.3% Non-auto-enrollees Average Year of Target Date Fund(s) Used Under Age or older 2000 a 3.3% 2.7% 3.0% 3.7% 13.5% Source: EBRI tabulations from the EBRI/ICI Participant-Directed Retirement Plan Data Collection Project. a 2000 represents target date funds labeled as "Retirement Income," "Target Now," "Target Today." Age ebri.org Issue Brief March 2009 No

12 All Figure 7 Distribution of Target-Date Fund Users, by Year of Target-Date and Plan Size, Auto-enrollees and Non-auto-enrollees, 2007 Average Year of Target-Date Fund(s) Used Plan Size (number of participants) 2000 a % 0.7% 8.4% 3.1% 20.0% 5.8% 24.3% 5.0% 19.0% 1.0% 8.5% , ,001 2, ,501 5, ,001 10, > 10, Auto-enrollees Average Year of Target-Date Fund(s) Used Plan Size (number of participants) 2000 a % 0.1% 6.4% 2.5% 21.1% 3.7% 26.1% 3.7% 24.5% 1.4% 9.2% , ,001 2, ,501 5, ,001 10, > 10, Non-auto-enrollees Average Year of Target-Date Fund(s) Used Plan Size (number of participants) 2000 a % 0.7% 8.6% 3.2% 19.9% 6.0% 24.2% 5.1% 18.5% 1.0% 8.4% , ,001 2, ,501 5, ,001 10, > 10, Source: EBRI tabulations from the EBRI/ICI Participant-Directed Retirement Plan Data Collection Project. a 2000 represents target-date funds labeled as "Retirement Income," "Target Now," "Target Today." ebri.org Issue Brief March 2009 No

13 distribution of participants' equity allocations outside of the target-date funds is examined across the year of the targetdate funds. In the top panel, which includes all participants, the distribution of outside equity allocations remains fairly constant across each of the target-date funds. Results are also very similar (ranging from percent) for target-date participants with 25 percent to 49 percent of their outside assets in equities (inclusive of company stock). Furthermore, about 20 percent of investors across the target-date years have less than 10 percent allocated to equities outside of the target-date fund, and percent of target-date investors have outside allocations of 75 percent or more. However, a significant variation emerges among auto-enrollees. Those in 2050 funds are much more likely to have very small (less than 10 percent) outside allocations to equities than investors in other target-date funds: 70 percent compared with no more than 39 percent (Figure 9). No major differences occurred among the non-auto-enrollees. and the Level of Target-Date Fund Use 401(k) participants who choose to invest in target-date funds are selecting a specific allocation of assets for their investments that is diversified between equities and bonds an allocation that becomes more conservative as the year of the target date becomes closer. Therefore, participants in target-date funds would be expected to not have extreme concentrations of assets (either all in equities or all in bonds). In fact, this is the case, as shown in Figure 10: Only 1.2 percent of participants younger than 30 who are completely invested in target-date funds have less than 25 percent of their account balance in equities. At the same time, only 1.8 percent have 90 percent or more of their account in equities. In contrast, for young participants (under the age of 30) who have no assets in target-date funds, just over 42 percent have very low equity allocations (less than 25 percent) while 23 percent have very high equity allocations (90 percent or more). These data show that by having at least some assets in a target-date fund, 401(k) participants avoid extreme (all or nothing) concentrations of equities that is more commonly found among those who do not invest in target-date funds. For those with only some of their assets invested in target-date funds and under age 30, 7.2 percent have less than 25 percent in equities and 22.5 percent have 90 percent or more in equities but none have 100 percent of their assets in equities. Also, as those invested completely in target-date funds get older, the percentage with high concentrations of equities declines. By age 60 or older, 89.1 percent of the all-target-date participants have an equity allocation between 25 percent and 74 percent (59.0 percent with percent in equities). By comparison, those in this age group who have nothing in target-date funds, 25.1 percent have 90 percent or more of their assets invested in equities; among those who invest only some of their assets in target-date funds, 9.2 percent have 90 percent or more in equities. Other factors besides age could influence the investment patterns of 401(k) plan participants, for those both with and without target-date funds: Three prominent potential factors would be the participant's salary, size of the participant's 401(k) plan, and the participant's account balance. Salary: Participants in each salary group that have all or some of their assets in target-date funds are less likely to have all or almost all of their assets in equities than those not using target-date funds (Figure 11). This even occurs for those with the highest salaries ($100,000 or more). Almost 30 percent of those having nothing invested in targetdate funds and with salaries $100,000 or more have a equity allocation of 90 percent or more, while 15 percent have less than 10 percent of their assets in equities. In contrast, among participants in this salary group who are completely invested in target-date funds, just over 90 percent have an equity allocation of percent, less than 1 percent have more than 90 percent in equities, and less than 2 percent have less than 25 percent in equities. Plan size: Overall, participants from plans of different sizes had relatively similar distributions of equity allocations, regardless of target-date fund use whether all or only some invested in target-date funds (Figure 12). What little difference there was resulted from the higher concentrations of participants with percent invested in ebri.org Issue Brief March 2009 No

14 Figure 8 Distribution of Target-Date Years for Target-Date Fund Users by Company Stock Ownership and Auto-enrollment Status, and Having All Assets in the Target-Date Fund, 2007 All Auto-enrollees Non-auto-enrollees Average Year of No company With No company With No company With Target-Date Fund(s) Used stock company stock stock company stock stock company stock 2000 a 3.6% 4.4% 2.5% 1.9% 3.7% 4.5% Percentage of All Assets in Taget-Date Funds Average Year of Non- Target-Date Fund(s) Used All Auto-enrollees auto-enrollees 2000 a 29.5% 91.7% 26.4% Source: EBRI tabulations from the EBRI/ICI Participant-Directed Retirement Plan Data Collection Project a 2000 represents target-date funds labeled as "Retirement Income," "Target Now," "Target Today." ebri.org Issue Brief March 2009 No

15 Figure 9 Distribution of Equity and Company Stock Allocation for Those With a Target-Date Fund and Other Asset Allocations, by Date of the Target-Date Fund and Auto-enrollment Status, 2007 Average Year of Equity/Company Stock Allocation Outside of the Target Date Fund Target Date Fund(s) Used Zero 1 9% 10% 24% 25% 49% 50% 74% 75% 89% 90% 100% All 14.6% 6.6% 10.6% 22.4% 27.5% 13.1% 5.2% 2000 a Auto-enrollees All a Non-auto-enrollees All a Source: EBRI tabulations from the EBRI/ICI Participant-Directed Retirement Plan Data Collection Project. a 2000 represents target date funds labeled as "Retirement Income," "Target Now," "Target Today." ebri.org Issue Brief March 2009 No

16 equities (for those with all of their assets in target-date funds and in large plans). However, in general, no major differences or trends emerge across the plan sizes. Account balance: As a participant's account balance increases, the likelihood of having a specific level of equity allocation changes depending on whether the participants is invested all, none, or some in target-date funds. For participants who have no assets in target-date funds, the probability of those with nothing in equities decreases sharply, as the participant's account balance increases. Among participants not invested in target-date funds with an account balance of less than $5,000, 45.4 percent have nothing in equities (Figure 13). This declines to 5.8 percent for those with a balance of $200,000 or more. Those completely invested in target-date funds across all account balances had equity allocations that are mostly between percent, but as the account balance increased there was a clear decrease in the likelihood of having an equity allocation of 75 percent 89 percent. Among those completely invested in target-date funds with low balances (less than $5,000), 66 percent had a total equity allocation of percent. However, this decreased as the account balance of these all-target-date investors increased: For those with an account of $200,000 or more, 29.8 percent had an equity allocation of between percent. Across all account balances, those with at least some of their 401(k) account invested in target-date funds were less likely to have very high or very low concentrations of equities. Furthermore, those who invested at least some of their assets in target-date funds, there was little change in the distribution of their equity allocations regardless of the size of the account balance. Glide Paths One of the major questions surrounding target-date funds is the equity allocations that these funds use over time. The academic literature has adopted the phrase glide path to refer to the downward direction that the equity allocation takes within a target-date fund as the target date approaches. The glide paths of different target-date funds have significantly different shapes and starting and ending equity allocations. Figure 14 illustrates this variation in glide paths of 10 families of target-date funds identified in the database. 16 The percentages of equity are lowest for funds aiming at producing income for a current retirement (to the left). Funds for retirement further in the future are to the right, such as the 2040 or 2050 funds. For 2040 funds, the percentage in equities (all types domestic, large, small, and foreign) ranges from about 80 percent to 97 percent. As the fund dates approach the current year, the percentage in equities in these funds decreases at different rates and to different terminal values. For the 2010 funds, the percentage in equities ranges from 26 percent to about 66 percent a 40 percentage point difference. Moreover, the fund families change their relative rank in equity allocation within the different fund years. Given the considerable differences in the glide paths of these plans and the equity allocations in a specific year, the identification of any allocation outside of the target-date fund that leads to systematic increases or decreases in diversification would be a significant finding. Furthermore, a systematic result could lead to a better understanding of the risk preferences of investors in 401(k) plans. In an attempt to see if participants are taking steps to lower or increase their allocations to equities given the relative equity aggressiveness of the target-date funds they are offered, the equity allocation across each target-date fund family for each target-date multiple of 10 is ranked. The rank is from the lowest percentage of equities in the funds to the highest for each target-date fund year. Furthermore, the participant's average equity allocation outside of the target-date funds and percentage of the outside allocation that these equities represent are compared across these target-date funds in Figure 15. ebri.org Issue Brief March 2009 No

17 Under Age 30 Figure 10 Distribution of, by Target-Date Fund Use and Age, 2007 All 0.0% 0.0% 1.2% 1.8% 3.1% 92.0% 1.8% 0.0% Auto Non-auto Some Auto Non-auto None Age All 0.0% 0.0% 1.2% 1.8% 7.6% 88.6% 0.9% 0.0% Auto Non-auto Some Auto Non-auto None Age All 0.0% 0.0% 1.5% 2.4% 40.5% 55.5% 0.2% 0.0% Auto Non-auto Some Auto Non-auto None Age All 0.0% 0.0% 1.7% 10.6% 73.3% 14.4% 0.1% 0.0% Auto Non-auto Some Auto Non-auto None Age 60 or older All 0.0% 0.0% 6.6% 60.3% 29.8% 3.3% 0.1% 0.0% Auto Non-auto Some Auto Non-auto None Source: EBRI tabulations from the EBRI/ICI Participant-Directed Retirement Plan Data Collection Project. ebri.org Issue Brief March 2009 No