Managed accounts have been praised, criticized, and litigated — often on the theory that they’re little more than expensive target-date funds. However, a recent report actually quantifies their impact — and turns out, it’s not an investment story, it’s behavioral.
That report — inauspiciously titled “The 2026 Managed Accounts Research Series: Analyzing the Value of Managed Accounts” — was published in mid-January by Morningstar. Of course, Morningstar has a fair amount of “skin” in the managed account space — a reason, if you will, to find a favorable outcome for the design.
At this point, my natural cynicism kicked in — though based on my personal and significant experience working with Jack VanDerhei, one of the coauthors, over a period of decades during his long-standing tenure at the Employee Benefit Research Institute (EBRI) — well, let’s just say if Jack says something is “so,” I tend to believe him.
Following the report publication, I had an opportunity to talk with Jack (and Spencer Look, his collaborator in this effort) to better understand their model. For those who haven’t stumbled across the report, they take a significant database of actual 401(k) plan balances and activity and apply sophisticated statistical behavioral modeling techniques to project long-term outcomes based on various assumptions. While it’s not unusual for researchers to deploy statistical modelling, most suffer from a lack of actual data, not only as a baseline, but as a behavioral predictor. Which, I should add, explains (to me, anyway) why the results often don’t match up with how real people respond/react in the real world.
All that said, this kind of modelling is also dependent on the quality of the underlying assumptions — and here none is perhaps more focused on than cost. Here the assumptions are 40 basis points cost for managed accounts (plus another 31 basis points in fund fees), 30 basis points for target-date funds, and 73 basis points for the DIY group. Don’t like those assumptions? VanDerhei is willing to plug in different numbers.
I also questioned whether it makes sense to model “managed accounts” generically, given the wide variation in personalization, design, and cost. Look explained that their review of roughly half a dozen managed account structures — including but not limited to Morningstar’s — showed sufficient similarity to support a generalized model. The same held true for target-date funds.
But here is the part that matters.
As important as factors like cost and asset allocation (not to mention the cost of asset allocation) are to outcomes, the report acknowledges that “…higher contribution rates are the primary driver.” The researchers further note that, “based on our analysis of the empirical data, MA users consistently save more than TDF or DIY investors, even after controlling for age, wage, tenure, and plan design features” — a pattern they say “…suggests that personalized savings-rate recommendations[i] embedded within MAs play a key role in encouraging higher savings rates.”
While you have to go to page seven of the 19-page report to find that,[ii] to my eyes, it is the most important sentence in it.
Now, I’ve long said that while we talk about “managed accounts” as though they are a monolithic concept, they are not. There are different — in some cases, widely different — levels of personalization deployed — variations in cost and construct. Anyone who ignores these potential underlying differences in application is missing the point.
I will admit that in considering the value of managed accounts, I had — perhaps as many of you — tended to focus more on the differences in asset allocation that might be possible if we knew more than projected retirement date — not to mention variation in the underlying costs. What I had not factored in was what the Morningstar researchers have — the application of personalization to influence and impact savings rates.
If managed accounts meaningfully increase savings rates — not just tweak allocations — that changes the fiduciary conversation entirely.
Because better investing matters. But saving more matters more.
- Nevin E. Adams, JD
[i] It’s worth noting here that the impact is smaller, but still positive, for AE with escalation plans, with TDF investors seeing an increase of 2.7% and DIY investors seeing an increase of 7.8%. Moreover, approximately 92% of AE plans with auto-escalation show an improvement in the median projected retirement wealth for TDF investors under the MA scenario. In other words, while automated increases in salary deferrals help, those timed and personalized via the managed account platforms provide a superior result.
[ii] It IS, however, right there with the first findings. Apparently, the folks who created the executive summary didn’t view it as being as significant as I did.
No comments:
Post a Comment