Reassessing TDFs

In May, the heads of two key congressional retirement policy committees—Sen. Patty Murray (D-WA), Chair of the Senate Health, Education, Labor & Pensions (HELP) Committee, and Rep. Bobby Scott (D-VA), Chair of the House Committee on Education & Labor Committee—sent a letter to the U.S. Government Accountability Office, requesting that the GAO review target date funds (TDFs).

That’s not a bad idea. Over nearly 15 years, TDFs have emerged as “the” 401(k) investment, with 27% of 401(k) assets now invested in TDFs. It’s time for an assessment and some thinking about how they can be improved.

TDFs—the Auto-Enrollment Investment

The TDF era began in 2007, with DOL’s publication of its qualified default investment alternative (QDIA) regulation. Until then, most participants were encouraged to make an affirmative election of an appropriate portfolio mix from a fund menu that (in many cases) contained a large and heterogeneous set of funds/investment objectives/styles. 

Participants who did not make an affirmative investment election presented a problem under ERISA Section 404(c)—the sponsor was on the fiduciary hook for whatever investment they were defaulted into. The solution—no doubt dreamed up by a lawyer—was to default them into a principal preservation vehicle, a GIC or money market fund, on the theory that if a participant didn’t lose money, they wouldn’t sue. (How naïve that all seems from the rear-view-mirror of 2021, when a few basis points “underperformance” can provide the basis for a major, multi-million dollar class action.)

Nobody, however, thought that “principal preservation” was a very good investment strategy, especially for younger participants. In the early 2000s this problem became acute as the use of auto-enrollment expanded, with participants being defaulted into 401(k) plans without making any investment election.

DOL’s QDIA regulation provided the solution—an age-based, balanced fund default investment that relieved sponsors of fiduciary responsibility (provided the QDIA was itself prudently selected). And the QDIA of choice for most sponsors was the target date fund—a fund asset allocation that tilted toward higher return/higher risk investments for younger participants and became more conservative as a participant aged.

Since 2007, TDFs have evolved into the master 401(k) asset allocation solution for nearly all participants. Indeed, some sponsors now regularly re-enroll participants who have already made an affirmative asset allocation election, defaulting them out of their own election and into a TDF (unless they make another affirmative election of a different investment).

Time for a Review

After nearly 15 years of TDF practice, a review is (as I said) a good idea.

And I think that, in their letter to GAO, Sen. Murray and Rep. Scott ask some excellent questions about the effectiveness of TDFs as a comprehensive 401(k) investment solution. Critically, about how TDFs differ in glidepath and asset allocation structure, how those different TDF designs have performed, and how resilient they might be to negative market events.

Considering these issues, I’d like to point at some features of TDFs and 401(k) investing generally that need some thoughtful scrutiny.

Benchmarking Complexity

A significant problem that TDFs present, and that the GAO will have to deal with if its evaluation of TDF performance is to have any utility at all, is complexity. Setting aside issues of fees and active vs. passive strategies, TDF performance is affected by two different and hugely significant variables: glidepath and asset allocation strategy. 

These variables are not easily accounted for in TDF comparisons and benchmarking. In many respects they make evaluation of TDF performance difficult if not impossible.
There is a (simplifying) binary argument about glidepaths—whether they should be designed “to” retirement (with retirement taken as the goal/investment horizon) or “through” retirement (with death (life expectancy) taken as the goal/investment horizon). But in truth the variability of glidepaths of either flavor is wide.

Asset allocation is, if anything, even more variable, with (nearly) as many asset and portfolio theories as there are investment management companies.
When you put these two variables together, it’s hard to think how you could usefully compare the performance of two different TDFs.

For instance, in a year in which the stock market does better than expected, “through” glidepaths may do better than “to” glidepaths. If most of the better-than-expected performance has been in U.S. large cap (as a sector), S&P 500 strategies will do better than more diversified strategies. The performance of any one TDF will depend on how “through” its glidepath is and how un-diversified its equity strategy is. 

So what? Would the mess that is such a year’s TDF “performance scatter” be any different than the mess it would be if, say, all the value stocks did well and the growth stocks tanked?

How, in this context, do you “benchmark” a TDF’s performance? What are you trying to accomplish by doing so? The GAO needs to look at that question, hard.

Investing Against Uncertainty

Just focusing on the “risky,” return-seeking part of the TDF asset allocation pie: While we may know a lot about how different strategies have performed in the past, we don’t know how they’ll do in the future. That’s why we have a market and prices—they are just the sum of the bets investors are placing on future performance. A lot of people—including a lot of policymakers, lawyers, and courts—seem to have forgotten this. The “equity premium”—a return above the risk-free rate of return – only exists because of this uncertainty. 

And the fundamentally unpredictable, in the form of, say, the global financial crisis of 2008 or the COVID-19 “event” we just went through, will have a massive effect on equity returns, in all sorts of ways.

Probably the elephant in the room in this regard is the performance of U.S. large caps over the last decade. The S&P 500 has seen a bull market more or less since 2009. It has significantly outperformed “the world” (with returns nearly three times the EAFE Index) and, e.g., value investment strategies. Funds with a U.S. large cap focus generally did better than more diversified equity funds. And the S&P 500 index beat almost everything else.

That performance is an outlier—it’s not a function of modern portfolio theory and it couldn’t be anticipated by a “prudent” investor.

But (again), so what? There is nothing to say that at current valuation levels, U.S. large caps will continue to outperform most other strategies.
In this regard, Sen. Murray and Rep. Scott ask a really interesting question: “what, if any, is the cost of a passive investment stance in a tumultuous market? Are TDFs properly structured to withstand major stock turbulence?”

I’d really like to know the answer to that question. Should TDFs be increasing their allocation to defensive strategies, even in the face of the U.S. large cap bull market? Is there a theory for that? Or is it just how we feel right now?

Certainty, Interest Rates, and Income

Focusing on the other side of the TDF pie (the non-risky allocation): One thing I have learned in a long arc in this business is that as individuals approach retirement, they begin to care about how much income they will have when they stop working, not the size of their account balance.
And for individuals concerned about income, the critical question is interest rates, not asset returns.  Because (among other things) interest rates (on investment grade instruments, at least) represent the price of income.

The importance of interest rates has only grown as, over the last decade or three, interest rates have declined. Just ask any DB plan sponsor.

The Effect of Interest Rates on Retirement Income

For 401(k) participants, changes in interest rates will ultimately show up as the cost of turning their account balance into retirement income. Which is why Department of Labor’s new lifetime income disclosure rule is so important.

To demonstrate, I did a quick analysis, applying the new lifetime income disclosure rules (backwards) to the period 2017-2020. Those rules compute lifetime income from an account balance based on the year-end yield on 10-year Treasury securities. 

Here’s the monthly annuity a participant could buy (and that would have been posted to the participant’s annual statement), starting with a balance of $100,000 at the end of 2017, crediting earnings based on a typical 2020 TDF (that is, a TDF for participants turning 65 in or around 2020), and tracking changes over that period in the 10-year Treasury rate.

Growth in Lifetime Income 2017-2020

 

Year	Assets (2020 TDF)	Interest rate (10-year Treasury)	Monthly Annuity	Cumulative return on income
2017	$100,000	2.37%	$528.02
2018	$95,760	2.98%	$535.39	1.40%
2019	$112,642	1.83%	$564.70	6.95%
2020	$126,205	0.92%	$578.13	9.49%

 

The point here is that while this participant’s account balance grew by more than 26% over this period, because of declines in interest rates, his retirement income only grew by 9.49%.

I would like to see GAO look harder at whether current TDF designs are hedging this interest rate risk adequately. And, in that regard, they should be looking at how DB plan sponsors are addressing a similar risk. And learning from that DB practice.

More Conservative TDF Strategies Are Not Necessarily the Answer

Reading between the lines, it appears that in addition to perennial concerns about fees, Sen. Murray and Rep. Scott are concerned that TDFs may be overallocating to stocks, putting older participants’ retirement adequacy at risk. Especially in view of possible “major stock turbulence.”

I have an open mind on all of that. But consider that the 3-year return of that 2020 TDF we just looked at was around 26%. The cumulative return of the same firm’s investment grade total bond market index fund did significantly worse over the same period, notwithstanding the decline in interest rates (including a nearly 150 basis point decline in the yield on 10-year Treasuries).

It may be that as currently designed, equities remain the best hedge 401(k) participants have against declining interest rates this side of a duration-matched (LDI) bond strategy. The latter, of course, has not been implemented in the 401(k) space. More’s the pity.

It is my most profound hope that GAO’s review will be thorough and thoughtful and that it will help frame consideration of rules—or simply stimulate consideration of alternative TDF designs—that will improve TDF results for participants.

Michael P. Barry is a senior consultant at October Three and President of O3 Plan Advisory Services LLC, which provides retirement plan regulatory analysis targeted at plan sponsors and those who provide services to them.

Opinions expressed are those of the author, and do not necessarily reflect the views of ASPPA or its members.