Tue, 3 Nov 2015
Shifts toward target-date funds, low-cost index investments, and auto-enrollment features have largely benefited investors in the form of increased savings and higher returns, says Vanguard's Kevin Jestice.
Christine Benz: Hi, I'm Christine Benz for Morningstar.com. With the ebbing away of pensions, more and more workers are relying on defined-contribution plans like 401(k)s to fund their retirements. I recently sat down with Kevin Jestice--who heads up Vanguard's institutional investor services--to discuss what's working in the defined-contribution marketplace.
Kevin, thank you so much for being here.
Kevin Jestice: Thanks for having me.
Benz: Your group oversees institutional investments at Vanguard, including defined-contribution plans. I'd like to get your take on the defined-contribution space at large. When you look at features of 401(k) plans, what seems to be working in terms of getting participants to save appropriately within their plans and to invest appropriately within those plans?
Jestice: We've seen a lot of changes, I'd say, in the last 10 years in the defined-contribution landscape that's benefiting participants. The first is getting participants into the plans, getting them saving. We've seen things like auto-enrollment--defaulting people in immediately when they start employment. Even more recently in the last few years, there's been a spike in re-enrollment, where we take all the participants in a plan and re-enroll them all in the plan to make sure everyone has access to the plan.
Benz: So, if I haven't opted in before and the next enrollment period comes around, I get re-enrolled on a default basis?
Jestice: Absolutely. Just automatically enroll people. Inertia is really powerful in the defined-contribution space. So, can we use that inertia to generate a good retirement-planning outcome? So, auto-enrolling or re-enrolling participants into the plan gets them saving, and then making sure we enroll them at the right savings level and amount that fully utilizes the match from the plan sponsor.
And then we've seen features like auto-escalation, where you may increase your contribution 1% per year, timed with when your employer payroll increases each year when you get a raise. So, you don't necessarily feel the economic hit to your paycheck. The paycheck goes up; it just goes up less because you are increasing your defined-contribution plan deferral. That auto-escalation feature is helping people get to the right savings level from 4% to 5% to 6% to 7%--and more. Then, the last change we've seen is the use of target-date funds in defined-contribution plans, and that has had a huge impact on the ability for investors to have a well-diversified portfolio--a prudently diversified portfolio--inside the plan.
Benz: How about automatic rebalancing? I know that's a feature that investors can often switch on as well.
Jestice: Yes, participants do use that, but it tends to be used more for the self-directed type of investor who is selecting their own funds and wants to maintain that asset allocation. We haven't seen a lot of movement there. The real movement over the last several years has been toward use of the target-date funds. Over 90% of plans are now offering target-date funds inside the plan. For participants who have access to target-date funds, we see that two thirds of them this past year were actually utilizing the target-date funds, so that's a huge adoption of target-date funds from inside the plans.
Benz: In part, because of the default?
Jestice: Exactly--because of the default. So, people are being auto-enrolled into the target-date fund; they are being re-enrolled into the target-date fund. When plans are consolidating, when a merger and acquisition occurs and the plans merge together, we often see a default into the target-date funds. We've seen a real increase in the use of target-date funds. So, we have more people in plans today than ever before, we have more people saving at higher rates because of the escalation, and we have more participants exposed to prudently diversified portfolios.
The statistic, there, we watch is how many participants have extreme asset allocations--think 100% stocks or 100% bonds. And over the last decade, we've seen a real decline in the number of participants who have those extreme asset allocations. It's not that that's wrong for everyone, but for most participants, that's probably not the right allocation. We've seen a real decline in those extreme allocations as we've seen an increase in the use of target-date funds.
Benz: How about when you look at the choices offered on 401(k) menus? I know that in the '90s, for example, in some plans there was choice proliferation; people wanted maybe the technology-sector fund--or whatever it might be. Have you seen plans starting to skinny down the number of options that they are offering participants?
Jestice: We are. It's a gradual decline, but the numbers are in decline. As you mentioned before, if we went back 10 or 20 years, we've seen the numbers increasing year over year. Now they are gradually decreasing. If you think of the target-date suite as one fund--because there's one that's appropriate for any individual and not as a suite of 12 or 13 funds--then the number of plan options today is around 18 investment options in the average plan. That's still quite a few options, but that's coming down from having been in the 20s in years past. So, the trend is toward fewer options.
That choice overload is difficult for autoparticipants to navigate. When you think about trying to select the funds for your own retirement, not everyone has the time to invest in researching potentially dozens of funds in their plan and understanding them all. So, that choice overload leads a lot of people to potentially make bad decisions for their retirement--for instance, putting a little bit in every fund,"1/N diversification" is what we call it in the industry. We see most participants actually using three funds. The average participant uses about three funds in their portfolio. A lot of participants are only using one with the target dates. So, we see that most participants are using very few funds, but plans are still offering, let's say, around 18 funds today on average.
Benz: I would imagine it also depends on the data points that you show participants about each of those options. Maybe the untrained investor might naturally gravitate to the one with, say, the best five-year return or something like that. Do you work with plans to help improve the data that participants are looking at when they are making those choices?
Jestice: Participation education is a critical part of working in the defined-contribution space--ensuring that the participants understand and have access to the information they need to make good choices. It includes things like whether you are listing the investment options alphabetically or by expense ratio or by their volatility or the asset class--money market, stocks, bonds, and so on. How are you aggregating them? It has an impact because, as participants are looking at funds, they start at the top of the page and run their finger down. So, how you communicate that information is important.
What we have gravitated toward in the last several years has been a tiered model, which is that you recommend Tier 1 or Tier A for participants, and that's the target-date series. That's the single-fund solution that can take care of the asset allocation for you. Tier B, then, would be core index funds, just offering three, four, or five funds--total stock market, total international stock market, total bond market, total international bond market, maybe a TIPS fund. It's just the core components of the target-date fund; offer those as index options in the plan for those who want to control the asset allocation of the options. But we'd still use low-cost highly diversified index funds. Then, we'd use the third tier, Tier C, which then can be the active funds--think your large cap, small cap, value, growth, active international, and those kinds of things. This tier has more of the active funds for the folks who really do want to select funds and potentially have overweights or style and capitalization factors built in to their portfolio.
Communicating it in three tiers, you can still offer several funds, but it helps screen the choice. Your first choice is which tier you are in; then, within that tier, the answer becomes much easier for the participant to then figure out. So, we've definitely moved toward a tiered structure, and the tiering is really helping because we see, again, more than 90% of plans offer the target dates and 52% of plans were offering an index core in the lineup. So, that's offering participants a lot of good choices to build low-cost portfolios, and we think that that communication strategy simplifies the decision for the participant.
Benz: My last question for you relates to this idea of investor outcomes. Morningstar has a data point called "investor return," in which we look at dollar-weighted returns based on our fund-flow data into a fund. So, if heavy flows have arrived more recently, they get a greater weight in the calculation. Vanguard has data on participants and their own outcomes. How do the participants who direct their own investments compare with the target-date funds when you look at returns there?
Jestice: The data is pretty intriguing. If we think of the target-date funds as a bit of an "efficient frontier," if you will, we can plot folks who are self-managing their accounts on a risk/return basis and see how they are performing against the target dates. What we generally see is that the group, by and large, underperforms the target-date funds, and there are two key reasons for that. One is that the investments that they are utilizing are higher cost. Lower cost--all else equal--leads to better returns. We see them underperforming that efficient frontier of target dates by amounts similar to cost.
The second contributing factor, though, is also the buy-high-sell-low phenomenon of people somewhat chasing returns. People like to invest in funds that have high historical returns. They sell funds that have low historical returns. And past performance is not a great predictor of future outcomes. We print millions of statements every quarter that, at the bottom, say, "Past performance does not predict future results." And yet, we see participant behavior time and time again buying funds with strong records--buying high--and then selling funds with weak record--selling low--when what we really want to do is buy low and sell high in the investment markets.
Because the target-date funds are a consistent buy-and-hold strategy, we tend to see much more consistent returns for participants at a very low cost because they are index-based. And because the participants are sticking with them long term--it's very low churn inside the funds--we don't see this buy-high-sell-low phenomenon. And because the costs are typically much lower than what they are choosing on their own, we see much less of a headwind from cost. (I don't know if it's a tailwind, but it's much less of a headwind from cost relative to the other options available.) So, the participants in the target-date funds are meaningfully outperforming the participants who are self-managing their own accounts, by and large.
Benz: Kevin, this is such an important topic. Thank you so much for being here to share you insights.
Jestice: Thanks for having me.