The new report – a collaboration between Hewitt Associates and Financial Engines – titled “Help in Defined Contribution Plans: Is It Working and for Whom?” – comes to the unsurprising conclusion that participants who get (investment) “help” – defined in the report as target-date funds, managed accounts, or online advice – are better off than those who do not, in all but the most extreme circumstances (in 2008, the most conservative allocations, no matter how undiversified or age-inappropriate, did better than more diversified portfolios, according to the report). On average, the median annual return for Help Participants was almost 2% (186 basis points) higher than for Non-Help Participants, net of fees, according to the report.
Additionally, those participants using “Help” have portfolios with risk levels that the survey’s authors suggest are both “more appropriate for their retirement horizons and more efficiently allocated among the options in their plan”. Factors contributing to that risk “gap” were the tendency of “non-helped” participants to make no adjustments in their portfolio (or risk level), and a gravitation toward larger holdings in company stock over time. In fact, the survey noted a particular concern – in view of a more limited time to recover from mistakes – that the greatest variability in observed portfolio risk levels was found among retirees and near-retirees not using “Help”.
The report looked at participant behavior, portfolio risks and returns during a particularly volatile period in the markets – January 1, 2006 and December 31, 2008 – across a dataset of seven large plans representing more than 400,000 individual participants and over $20 billion in plan assets.
Who’s Getting HELP?
On average, across the more than 400,000 plan participants represented in the report, about a quarter use at least one of the types of Help offered within their 401(k) plans. However, average usage of Help overall varied across the seven plans in the sampling, from a low of 15% to a high of more than 35%.
Of the quarter of participants using Help, 9.8% are invested appropriately in target-date funds (e.g. 95% or more of their balance in that offering – as an interesting side note, of the 75% not using “Help”, 43% have allocated some money to target-dates, but less than 95%. The average portfolio allocation among those participants was 36%), while 9.7% were enrolled in managed accounts, and 5.8% use online advice.
Similarly, 13% of participants in the Non-Help group have used online advice but not within the last 12 months. It was assumed that all participants using a managed account were taking full advantage, since the managed accounts provider had discretionary control over the participant’s full account balance. The study’s authors note that, while just 25% have used Help “appropriately, an additional 41.7% of Non-Help Participants show partial Help usage.
How it Matters
The study notes that so-called “Non-Help Participants” with appropriate risk levels underperformed in only one case—when compared to Non-Help Participants with too low risk levels in the severe bear market in 2008. Even so, across all three years in the study, Non-Help Participants with appropriate risk levels outperformed those with too low or too high risk levels by an average of 138 basis points, according to the report.
The report notes that, across all three market conditions included in the study range (a bull market, a mixed market, and a bear market), Help participants outperformed Non-Help Participants 88% of the time, and that - across the three years in the study, the extra efficiency provided by Help portfolios provides an average annual return benefit of 67 basis points, even when adjusted for similar risk levels.
What Kind of Help?
The survey’s authors note that plan design (specifically, automatic enrollment coupled with a qualified default investment alternative, or QDIA), the length of time investment “help” has been in place in a plan, and participant demographics all impact Help usage. However, one of the stronger influences appears to be age. For example, the study found that target-date fund users tend to be younger, with lower tenures and with significantly lower account balances, salary, and contribution rates (the study’s authors estimate that approximately half of the participants using Help through target-date funds have been defaulted into the funds by their employers).
The study also noted that online advice users tended to be younger, but with higher account balances (as compared with target-date fund users), salaries, and contribution rates, while managed account users tend to be older and with longer tenures – both compared with participants using target-date funds or online advice, and with those receiving no “Help” at all.
The study also noted that plan design and possibly what they described as the “one-size-fits-all nature” of target-date funds make them the most popular choice or default for younger workers (those under age 35). In contrast, the study noted that baby boomers (participants age 45–64) and retirees (age 65+) are far more likely to prefer managed accounts , while Generation X participants (age 30–44) tend to be more likely to enroll in managed accounts than Generation Y, but less likely than members of Generation Y to use target-date funds or online advice. That said, the study notes that there are two ages in particular where the probabilities shift significantly; starting at age 50, participants begin to be far more likely to enroll in managed accounts versus the other two types of Help, and participants tend to be more likely to use target-date funds over the other two types of Help up to age 35. There was, however, no clear shift tied to age for online advice users.
Account size also appeared to have an influence (though age, and an increased opportunity to accumulate savings, can be correlated with larger balances). In the study sampling, the higher the balance, the more likely participants were to use online advice. In fact, nearly 12% of participants with account balances over $250,000 use online advice, according to the report. Managed accounts appeal to participants across several balance categories (though with higher concentrations of participants with balances between $15,000 and $100,000).
The study’s authors note “When we look at the relationship between account balances and where the probabilities for selecting a certain type of Help shift, we can see that $5,000 is a key threshold for target-date funds in our sample,” with those having balances less than $5,000 being 80% more likely to use target-date funds than those with account balances between $5,000 and $10,000. “This is not surprising,” the study’s authors noted, “since roughly half of the target-date group were new employees defaulted into the funds”. That said, even though they noted a correlation between balance and use of managed accounts and online advice, the study said there “are no specific account balance levels where there are significant shifts in usage probabilities”.
Which Matters More?
So, all other things being equal, which matters more – age or account balance? According to the researchers, age is the dominant driver of the type of Help used. However, they went on to note that “These findings suggest that no single type of Help is necessarily ideal for a participant’s entire work life, and that certain types of Help correlate with various life stages. At 25 years of age, for example, most participant risk preferences and life situations are relatively similar, making it easier to combine them into a cohort based only on age, as is the case with a product solution like target-date funds. As participants age, however, their life situations grow more heterogeneous, requiring greater flexibility and personalization.”
As for why different kinds of help were made available had such different usage patterns, the study’s authors note that “Help” was introduced at different times and in different ways at the various employers in the study. However, they note that, in general, the longer Help has been available in a plan, the more participants tend to use it.
Hewitt and Financial Engines say they have additional research in progress for phase two of this study, which will attempt to identify why participants use or do not use certain types of investment help.
The report is available here. A free registration is required.
« MOSERS Board Votes to Drop Staff Bonuses