Vanguard Research February 2016

Transcription

1 The Reshaping buck stops participant here: Vanguard outcomes money through market funds reenrollment Vanguard Research February 2016 Cynthia A. Pagliaro, Stephen P. Utkus Executive summary. Reenrollment into low-cost target-date funds is a highly effective strategy to improve portfolio diversification and reduce fund fees paid by participants. Most participants remain invested in the default fund after the reenrollment event. A small group of participants opt out and make their own portfolio choices but use the default fund as an additional investment option. Opt-out behavior. Six months after the reenrollment event, only 16% of all participants fully or partially opted out of the default target-date fund. Among active plan participants (those currently contributing), the opt-out rate was 20%; among inactive participants, who are less attentive, the opt-out rate was 4%. Those participants who opted out were older, wealthier males with a preference for more complex investment portfolios. Asset allocation. After reenrollment, 94% of all participants held a position in the default target-date series, and that series captured 74% of. Participant risk profiles became more age-appropriate, reflecting the target-date glide path. The percentage of participants with extreme equity allocations was reduced from 23% to 7%. Impact on fund fees. Reenrollment into a low-cost target-date default fund resulted in a dramatic decrease in the annual fund expenses paid by participants. The average expense ratio decreased from 0.41% to 0.10% (0.37% and 0.07% at the median). Over the long term, this reduction saves participants thousands of dollars in fees. Implications. Reenrollment into a low-cost qualified default investment alternative (QDIA), such as a low-cost target-date series, can rapidly improve diversification and reduce fees for participants, potentially leading to higher retirement wealth accumulations in the future.

2 Background Over the past decade, there has been a material improvement in participant portfolios among defined contribution (DC) plans. An increasing number of participants, nearly 4 in 10, are invested in a single target-date fund due to the growing prevalence of target-date options in plan menus and the growth of automatic enrollment plans designating a target-date default fund. As a result of these trends and the expanded use of managed account advice, we anticipate that nearly two-thirds of all participants will be in professionally managed investment allocations within five years. 1 At the same time, a meaningful fraction of participants, particularly the long-tenured in a plan, maintain portfolios that are not fully diversified. As our research has shown, their portfolios often reflect decisions made many years ago as well as the investment menus in effect at the time of enrollment into the plan. 2 Reenrollment is one way to quickly improve plan diversification and address these concerns. Reenrollment is a plan design strategy whereby participants account balances are transferred automatically to the plan s qualified default investment alternative (QDIA). are given advance notice of their right to opt out of the transfer. Sponsors also receive fiduciary protection under the Department of Labor s QDIA regulations. 3 Reenrollment can be highly effective in changing participant behaviors for several reasons. Inertia often dominates financial decision-making. Similar to the effect of inertia in automatic enrollment and saving decisions, many participants will choose not to take any action in response to the reenrollment event and their portfolio holdings. A second reason is that many participants may lack strong convictions about their original portfolio choices. They may be unsure about the appropriate level of risk to take or how to adequately diversify their portfolios. In this paper, we assess the effects of reenrollment on one large plan in terms of participant opt-out behavior, the impact on asset allocation, and the effect on fees paid. Reenrollment case study The reenrollment event analyzed in this paper occurred in December 2014, at the time a large DC plan transferred recordkeeping services to Vanguard. At the time of conversion, the plan had nearly 18,000 participants and more than $1.2 billion in. The conversion involved both a change in recordkeeping administration and a change in the plan investment menu. Before the conversion, the plan offered a total of 33 funds well distributed among various asset classes (Figure 1). The majority of funds offered were actively managed, including the entire suite of target-date funds, designated as the plan s QDIA. In addition, the plan offered five risk-based model portfolios, constructed from funds available in the lineup. Figure 1. Plan menu composition Individual funds Preconversion Postconversion Stable value 1 1 Bond Actively managed 2 2 Index 1 1 Domestic equity Actively managed 6 6 Index 3 3 International Actively managed 2 2 Index 1 1 Professionally managed options Target-date funds Actively managed 11 NA Index 0 12 Model portfolios 5 NA Active traditional balanced fund 1 1 Total menu options Percent of actively managed funds in menu 85% 41% 1 See How America Saves See Pagliaro and Utkus, For a detailed discussion of the fiduciary and plan design aspects of a reenrollment, see Improving plan diversification through reenrollment in a QDIA. Vanguard, forthcoming. 2

3 After conversion, the plan menu remained essentially the same in basic design with two key differences. First, an actively managed target-date series was replaced with a passively managed series with much lower fees. Second, the risk-based model portfolios were no longer offered. However, most of the funds used to construct the model portfolios were still offered in the plan. The prior menu also included a stable value investment fund, which required advance notification to the insurer of the plan reenrollment event in order to maintain par value. As a result, the reenrollment occurred in two phases. The first phase occurred in December 2014 at the time of the plan s administrative conversion, when all funds except stable value were reenrolled into the new target-date series. The second phase took place in June 2015, when assets in the stable value option were reenrolled into the then-current contribution allocations of the participants. Consistent with the requirements governing a reenrollment event, all participants received timely written communication and notification of the event. During this communication period, participants were able to view their new contribution allocations and account balances on a new participant website and had the opportunity to choose their own investments from the new menu of investment offerings. In our analysis of the reenrollment, we compared participant portfolios just before and then six months after the event labeled as PREevent and POSTevent in this report. We also studied the participants who opted out of the default fund to better understand the factors influencing their decisions. To accurately reflect the changes in opt-out rates over time, our sample contains only those participants present before and after reenrollment. In other words, participants hired or terminated during the interim six-month period were excluded from our analysis. 4 Opt-out rates The percentage of participants who chose to opt out of the default fund, either in part or in full, was measured after both the first and second phases of the reenrollment event. Immediately after the first phase, at the end of December 2014, 10% of participants partially or fully opted out of the default fund and elected their own portfolios (Figure 2). Ninety percent remained in the default. Because of the stable value option, 54% of participants were wholly invested in the new default fund and an additional 36% held both the default fund and the stable value option. Figure 2. Opt-out rates All plan participants Immediately after reenrollment Six months after reenrollment 100% default fund 54% 84% Default and stable value funds 36 0 Subtotal Partial opt out 2% 10% Full opt out 8 6 Subtotal Total 100% 100% 4 Ninety-five percent of participants were present during both first and second phases of reenrollment. 3

4 After the second phase, the opt-out rate increased from 10% to 16%. There was an increase in the fraction of individuals who partially opted out of the default target-date series. Only 6% of participants chose a portfolio without any target-date holding. Opt-out rates varied based upon a participant s status within the plan (Figure 3). Twenty percent of active participants (those currently contributing to the plan) either partially or fully opted out compared with 4% of inactive participants. 5 This difference most likely reflects a higher level of inattention among inactive participants. Figure 3. Opt-out rates by participant status All plan participants 100% balanced strategies. More than one-third of participants also chose to invest in international options as well as the most conservative stable value fund. In terms of plan assets, half of total were invested in equity funds, both domestic and international. One-quarter of were investments in conservative options like bond and stable value and slightly less in traditional balanced funds. The plan had previously introduced automatic enrollment for new hires into a target-date series. As a result, one-quarter of all participants held target-date funds and 5% of were invested in these funds. Figure 4. Changes in portfolio All plan participants Panel A: Before reenrollment (PREevent) 84% 80% 96% Professionally managed options Target-date 25% 5% Traditional balanced % 6% 8% 10% 12% All Active Inactive 1% 3% Individual funds Stable value 42% 15% Bond U.S. equity % default Full opt-out Partial opt-out International equity 39 9 Changes in portfolio allocation One of the primary objectives of reenrollment into a target-date series is to address both extreme portfolio allocation and non-age-appropriate allocations. Before the plan conversion, a significant percentage of participants invested across the broad array of investment offerings (Figure 4). More than one-half of all participants were invested in individual domestic equity, bond, or traditional Panel B: 6 months after reenrollment (POSTevent) Professionally managed options Target-date 94% 74% Traditional balanced 5 3 Individual funds U.S. equity 13% 16% International equity 8 3 Bond 8 2 Stable value Active participants are currently employed participants that defer a percentage of current compensation to the DC plan. Inactive participants are term-deferred (terminated employees who left their account balances in the plan) or suspended participants. See Appendix 1 for detailed statistics for each group. 4

5 After the reenrollment, the distribution of participant and shifted dramatically. As a result of the reenrollment, 94% of participants held a position in the target-date series and these funds captured 74% of all. The remaining assets were distributed among the other asset classes with domestic equity funds holding 16%, the largest share among remaining categories. In terms of equity allocations, before the reenrollment, 23% of all participants held extreme equity positions (Figure 5); 14% had equity allocations exceeding 90% of their portfolios and 9% had equity allocations of 20% or less. After reenrollment, the occurrence of extreme allocations was significantly reduced and overall portfolios became more age-appropriate. Figure 5. Distribution of equity allocation All plan participants 50% 47% 23% of participants had extreme equity allocations before reenrollment 28% 9% 14% 0% 0% 1 10% 11 20% 21 30% 31 40% 41 50% 51 60% 61 70% 71 80% 81 90% 91 99% 100% PREevent POSTevent 5

6 In terms of age-related equity allocations, before the reenrollment, the distribution of participants average equity allocations varied widely and the median value differed from the equity allocations of the target-date funds (Figure 6, Panel A). After the event, the distributions narrowed significantly and the trajectory of the median equity allocation aligned more closely to the target-date series (Figure 6, Panel B). The distribution widens somewhat after age 40, reflecting the later-than-normal retirement ages anticipated by some older investors. This expectation leads to somewhat riskier portfolios. Figure 6. Reenrollment and age-related equity allocation All plan participants The shaded area is bounded by the 5th and 95th percentile of the equity distributions within each age group. Panel A: PREevent 100% 0 30 or younger or older Median equity holding Target date Panel B: POSTevent 100% 0 30 or younger or older Median equity holding Target date 6

7 Impact of participant fees One of the important results of this particular reenrollment event was the dramatic reduction in investment fees paid by participants. Before reenrollment, a participant paid, on average, 41 basis points in fees each year, reflecting the greater use of active strategies in the plan menu (Figure 7). Figure 7. Impact on fund fees All plan participants The cumulative reduction in fees will be significant over time. On an annual basis, the median participant will save about $100 per year in fund fees (Figure 8). After ten years, this results in a projected cost savings of about $2,400. After 20 years, the projected savings is more than $9,000. Figure 8: Fund expenses paid by median participant All plan participants 0.50% $12,000 $11, % 0.37% $9,020 0% 0.10% Average Median 0.07% $0 $2,965 $2,171 $2,397 $568 $128 $24 Year 1 Year 10 Year 20 PREevent POSTevent PREevent POSTevent After reenrollment, participants experienced a 75% decrease in fund fees paid, paying, on average, 10 basis points. The median fees paid were 0.37% before reenrollment and 0.07% after reenrollment. The lower fees reflect the fact that the new target-date series used low-cost passive options. Note: Assumes the median participant saves 9% annually (includes matching contributions) and has 5% annual rate of return. This hypothetical illustration does not represent the return on any particular investment, and does not reflect any taxes or penalties that may be due upon distribution. Withdrawals from a tax-deferred account before age 59½ are subject to a 10% federal tax penalty unless an exception applies. 7

8 who opt out We sought to understand the reasons why nearly 1 in 6 participants chose to opt out of the default and construct their own portfolios. Because the majority of participants who opt out are active participants (those currently making contributions), our analysis considered only this group. Among active participants, as noted above, 20% opt out fully or partially from the target-date default. Broadly, participants who opt out tend to be older and longer-tenured men who have higher income and account balances (Figure 9). Full opt-out participants are the oldest and wealthiest. who opt out are also more likely to be men. This finding is consistent with previous academic findings on male overconfidence and trading. These findings are confirmed by a logistic regression controlling on a range of factors simultaneously (Figure 10). who opt out tend to be older, wealthier males with specific portfolio preferences. They are more likely to hold five or more funds and have strong preferences for equity investments in both domestic and international funds. Using a statistical clustering technique to identify common portfolio preferences, we identified three distinct groups among the 20% of participants who Figure 9. Participant characteristics Active participants only A. Demographics A. Remained in default (n=10,588; 80% of sample) B. Partial opt out (n=1,619; 12% of sample) C. Full opt out (n=1,028; 8% of sample) Average Percent Average Percent Average Percent Age Tenure Balance $63,390 $122,849 $134,437 Compensation $87,952 $105,751 $119,438 Male 45% 54% 60% Female 42% 33% 29% Missing gender 13% 13% 12% Figure 10. Factors influencing opt-out behavior Opt out versus participants who remain in default 15% 21% 19% 23% 21% 22% 0 7% 9% 7% 7% 13% 2% 5 5% 5% Age Age Age Age 60 or older Male 4% Balance $30 $59K Balance $60 $89K Balance $90K or higher 3% Income $50 $74K Income $75 $100K Income 5 or $100K more or more funds 20% or more in active domestic 20% or more in active international 5% 20% or more in TDF 20% or more in stable value Note: The blue bars are statistically significant at the 95% level. The grey bars are statistically insignificant. 8

9 partially or fully opted out of the default fund (Figure 11, Panels A and B). The largest group, representing twothirds of these participants, are status quo investors. For the most part, they actively chose to maintain their existing portfolios, making no substantive changes to their elections. They allocated, on average, 13% to target-date funds. An additional 7% are cost-conscious participants who exhibited a similar preference for their existing portfolios but made two distinct changes; they reduced the total number of funds held by an average of 2.5 funds and traded more expensive active funds for lower-cost index funds. Similar to the status quo group, they also allocated a small percentage (12%) to target-date funds. The third group, core-satellite investors, represents 28% of these participants. They reduced the number of funds they held from 11 to 4, keeping the default target-date fund as a core portfolio holding. In addition to the targetdate, they added other funds, predominately passively managed domestic equity funds to their portfolios. Figure 11. Active participants who opt out Panel A. Behavioral traits by cluster Percentage allocated to target-date funds 100% 0 3 Core-sateliite 65% allocated to target-date 28% 4 Cost-conscious 12% allocated to target-date Simplified portfolio, different management strategy Number of funds 7% 65% 5 Status quo 13% allocated to target date No significant changes, predominantly active 6 Panel B. Portfolio characteristics by cluster Average asset allocation Status quo (n=1,718; 65% of sample) Cost-conscious (n=181; 7% of sample) Core-satellite (n=748; 28% of sample) PREevent POSTevent PREevent POSTevent PREevent POSTevent Target-date 11% 13% 7% 12% 4% 65% U.S. equity active U.S. equity index International equity active International equity index Bond Stable value Traditional balanced Number of funds

10 Implications This reenrollment case study highlights the benefit on portfolio diversification that this strategy provides. This benefit is particularly valuable for longer-tenured participants who made portfolio choices under prior investment menus on their own, or who did not benefit from automatic enrollment into a diversified default fund. One of the most significant benefits of this particular case study was a dramatic reduction in investment fees paid by participants. The selection of a low-cost, passively managed target-date series as the default fund will result in considerable fee savings over time for the typical participant, which, all things equal, can improve wealth outcomes. Some plan sponsors are reluctant to override the individual investment choices of their participants and regard reenrollment as an intrusive plan design strategy. Yet participants retain the right to opt out of the event and confirm their own investment elections. The evidence that most participants remain in the default option confirms the notion that many participants lack strong views on portfolio construction and prefer to cede these detailed choices to the professional portfolio manager designing and managing the default option. References Pagliaro, Cynthia A. and Stephen P. Utkus, Behavioral effects and indexing in DC participant accounts Vanguard Center for Retirement Research. institutional.vanguard.com Vanguard, How America Saves Vanguard Center for Retirement Research. institutional.vanguard.com Vanguard, (forthcoming). Improving plan diversification through reenrollment into a QDIA. Vanguard Strategic Retirement Consulting. institutional.vanguard.com Lamancusa, John and Jean A. Young, Target-date fund adoption in Vanguard Center for Retirement Research. institutional.vanguard.com 10

11 Appendix 1: A. All participants (n=17,028) Opt-out rates Before reenrollment 9/30/2014 Immediately after reenrollment 12/10/ months after reenrollment 6/30/ % default fund 54% 84% Default fund and stable value 36% 0% Subtotal 90% 84% Partial opt-out 2% 10% Full opt-out 8% 6% Total 100% 100% Switched target-date fund 2% 3% Portfolio allocation Target-date 25% 5% 93% 73% 94% 74% U.S. equity 54% 41% 7% 9% 13% 16% International equity 39% 9% 5% 2% 8% 3% Bond 53% 10% 4% 1% 8% 2% Stable value 42% 15% 41% 14% 2% 2% Traditional balanced 57% 20% 3% 2% 5% 3% Other portfolio statistics Average equity allocation 67% 67% 77% 100% stable value 3% 2% 0% Hold multiple target-date funds 2% 1% 2% Expenses Average index allocation 18% 82% 91% Average expense ratio 0.41% 0.12% 0.10% 11

12 B. Active participants (n=13,235; 78%) Opt-out rates Before reenrollment 9/30/2014 Immediately after reenrollment 12/10/ months after reenrollment 6/30/ % default fund 54% 80% Default fund and stable value 33% 0% Subtotal 87% 80% Partial opt out 3% 12% Full opt out 10% 8% Total 100% 100% Switched target-date fund 2% 3% Portfolio allocation Target-date 28% 6% 92% 73% 92% 72% U.S. equity 55% 42% 9% 10% 16% 17% International equity 40% 9% 6% 2% 8% 3% Bond 56% 10% 5% 1% 9% 3% Stable value 40% 14% 39% 13% 3% 2% Traditional balanced 55% 19% 4% 2% 6% 3% Other portfolio statistics active participants Average equity allocation 69% 67% 77% 100% stable value 1% 1% 0% Hold multiple target-date funds 2% 1% 2% Expenses Average index allocation 19% 82% 91% Average expense ratio 0.42% 0.11% 0.10% 12

13 Before reenrollment 9/30/2014 Immediately after reenrollment 12/10/ months after reenrollment 6/30/2015 C. Inactive participants (n=3,793; 22%) Opt-out rates 100% default fund 50% 96% Default fund and stable value 44% Subtotal 95% 96% Partial opt-out 1% 3% Full opt-out 5% 1% Total 100% 100% Switched target-date fund 0% 0% Portfolio allocation Target-date 14% 4% 95% 76% 99% 90% U.S. equity 50% 37% 1% 3% 3% 7% International equity 35% 8% 0% 0% 1% 1% Bond 46% 9% 0% 0% 1% 1% Stable value 49% 21% 49% 20% 0% 1% Traditional balanced 63% 22% 0% 1% 1% 1% Other portfolio statistics inactive participants Average equity allocation 62% 64% 77% 100% stable value 2% 2% 0% Hold multiple target-date funds 0% 0% 0% Expenses Average index allocation 17% 82% 99% Average expense ratio 0.37% 0.13% 0.08% 13

14 14

15 15

16 Connect with Vanguard > institutional.vanguard.com Vanguard research > Vanguard Center for Retirement Research Vanguard Investment Strategy Group For more information about Vanguard funds, visit institutional.vanguard.com or call to obtain a prospectus or, if available, a summary prospectus. Investment objectives, risks, charges, expenses, and other important information about a fund are contained in the prospectus; read and consider it carefully before investing. Investments in target-date funds are subject to the risks of their underlying funds. The year in the fund name refers to the approximate year (the target date) when an investor in the fund would retire and leave the workforce. The fund will gradually shift its emphasis from more aggressive investments (stocks) to more conservative ones (bonds and short-term reserves) based on its target date. An investment in a target-date fund is not guaranteed at any time, including on or after the target date. Diversification does not ensure a profit or protect against a loss. All investing is subject to risk, including the possible loss of the money you invest. Vanguard Research P.O. Box 2900 Valley Forge, PA The Vanguard Group, Inc. All rights reserved. Vanguard Marketing Corporation, Distributor. REEROLLP