Hit, Miss or Backfire? Controversial Ayres/Curtis 401k Fee Paper Claims Broad-Based Fiduciary Breach
They’re back! Last year, Ian Ayres, William K. Townsend Professor at the Yale Law School and University of Virginia School of Law associate professor Quinn Curtis teamed up and threatened to disclose 401k plan sponsors who breached their fiduciary duty. They promised to reveal a study to back up their contention. The industry struck back and Ayres and Curtis suddenly faded from the scene, refusing interviews from credible journalism organizations, including FiduciaryNews.com.
In late February of 2014, the two resurfaced with the paper they promised to deliver. The pair claim to “provide evidence that fees lead to an average loss of 86 basis points in excess of low cost index funds.” They further state “In 16% of analyzed plans, we find that, for a young worker, the fees charged in excess of an index fund entirely consume the tax benefit of investing in a 401k plan.” As a result of these and other specified “breaches,” Ayres and Curtis recommend “the requirements for default fund allocations be enhanced to assure that the default investment is reasonably low cost.” Furthermore, they recommend “the Department of Labor (DOL) designate certain plans as “high cost” and mandate that participants in these plans be given the option to execute in-service rollovers to low-cost plans.” Finally, the two law professors recommend “participants be required to demonstrate a minimum degree of sophistication by passing a DOL-approved test before being allowed to invest in any funds that would not satisfy the enhanced default requirement.”
The two touch on several issues that many within the industry agree need to be addressed. However, their tactics repeat many of the same errors Martin Smith made in his PBS Frontline show “Retirement Gamble.” Several industry thought leaders commented on the Ayres/Curtis paper and its, perhaps ironic, consequences.
“I think the arguments are measured even if they may not be overly dramatic. They are still eye popping,” says Kathleen M. McBride, one of the founders of The Institute for the Fiduciary Standard and the Found of FiduciaryPath, LLC in New York City. “The points about fees being a larger issue than diverse, but less optimal, platform offerings is a very important one and the policy implications are huge here,” continues McBride. She feels, from a bottom-line perspective, at the very least the paper “raises many very good questions and boosts the volume and direction of the conversation.”
Yet, for all its good intentions, like last year’s Frontline episode, the methodology of the paper leaves much to be desired and relies on tangential issues that leave the author’s open to credibility questions. For example, while they state their dataset includes 3500 plans, yet nowhere in the heavily footnoted (138 in all) does it explain the source of this data. Brooks Herman, Head of Data Research for BrightScope, Inc. in San Diego, California, who didn’t see any documentation about the source of the data says, “Understanding the scope and source of the data source is paramount. We can dive into some basic statistics. If Ayers is looking at sample data, one has to be concerned sample bias, etc. If Ayers is looking at survey data, it’s important to understand the scope/breadth of the survey data. Can that be applied to ‘all’ 401k plans?”
McBride points out “Curtis and Ayers were roundly criticized for revealing their data sources last time. I would think that they would discuss the sources one on one, though maybe not for attribution.” (Editor’s Note: FiduciaryNews.com did not have time prior to this article’s deadline to contact either Ayres or Curtis, although we would be happy to print their comments if they were to provide them.)
The type and source of data is critical to any analysis. Herman says, “I know the paper says it looks at over 3,000 plans, but we don’t know if these are small insurance plans, large plans dominated with collective trusts or somewhere in the middle. I raise this point because insurance products and collective trusts are not Investment Act of 1940 compliant, so obtaining fee information about them can be very difficult, if not impossible.”
“Research can be an incredibly valuable tool in helping not only understand the current state of things, but changes over time, and crafting possible solutions to perceived problems,” says Nevin Adams, Co-Director, EBRI Center for Research on Retirement Income at the Employee Benefit Research Institute in Washington, DC. “That said,” continues Adams, “someone (I think it was Mark Twain) once famously cautioned that ‘figures lie, and liars figure.’ If you don’t know the source of the data – you (or those you trust to do so) aren’t able to independently verify the accuracy of the basis for the conclusions based on that data. So, presented with conclusions based on unknown data means that you have to either accept or reject (or possibly ignore) them, based on something other than a complete understanding of the very foundation of the conclusions presented.”
The lack of sourcing the data may be one problem – and a problem that can be easily remedied – but the paper also makes the identical mistake The Retirement Gamble made by digressing into the notorious “passive versus active” debate. McBride says, “The issue they may take heat on is the comparison of managed vs. passive. They make the point that with managed funds with high Betas of course the participant is much better off in a low cost index fund. But how many participants are equipped to understand the difference between a managed fund with a low beta and one with a high beta?”
“The authors do not differentiate between plans with financial advisors and those without – which is a key component of determining the value added from additional fees,” says Greg Carpenter, president of Employee Fiduciary, LLC in Mobile, Alabama. “Ayres and Curtis begin with the assumption that actively managed funds add no value over time, and the higher fees charged result in a drag on earnings for all actively managed funds – with weak analysis. I disagree with that assumption. A skilled financial advisor can provide value using actively managed funds. However, those advisors may only be available for a fee that is excessive for smaller plans. In that circumstance, I believe that plan sponsors and participants are well served utilizing a diversified selection of low cost index funds. The authors argue that fees matter. I agree in the larger sense, but not as applied by the authors.”
Part of the problem with the Ayres/Curtis active vs. passive argument is their performance data appears only to include the years from 2010 through 2013. During this period, passively managed funds tended to perform better than actively managed funds. The opposite was true during the preceding decade. “There’s nothing inherently wrong with selecting certain specific pieces of data, or timeframe(s) of analysis, as long as you are clear about those details/choices,” says Adams. “Obviously, as any time/data choice is selective, there is the risk of a certain bias in selection that could, in turn, influence the results. But as long as the author is clear about what they have done, and perhaps even the rationale behind those decisions, the reader has an opportunity to make their own assessment.”
Herman brings up the much larger problem of trying to transform the active vs. passive debate into an fee argument. He says, “My first reaction is that the paper is harsh on active management. In the non-retirement space, the debate has been raging for decades about the benefits of active vs. index investing, and the same applies to retirement space too. There are a lot of reasons why an active manager might add value and warrant the fees over an index fund: They could offer uncorrelated returns to the index; They could provide protection from movements to the downside; and, They could have less/more beta than the index, etc. Again, all these arguments are well documented in the non-retirement space and there is not definitive conclusion as of yet.”
Indeed, the authors do cite a research paper by Choi et al (see “Yale/Harvard Study Reveals Disturbing 401k Fee Paradox,” FiduciaryNews.com, June 1, 2010) they claim supports their conclusion. Ayres and Curtis say, “For example investors mix their portfolios between two S&P 500 index funds with different fees. Since index funds that track the same index will have very similar returns, it is irrational to allocate funds to a high-cost index fund when a low-cost index fund is available… A similar effect has been identified by Choi, Laibson and Madrian. Choi et al. note that investors seem to sense that this investing strategy is a mistake, but cannot overcome the impulse to partake of all portfolio options presented to them.” Ayres and Curtis seem to conveniently miss Choi’s purposeful intent only to compare fees of like index funds, avoiding the “apples-to-oranges” comparison with actively managed funds all together.
The broader point of bringing in Choi’s (and other’s) research into their paper is to introduce the downside of naïve diversification – a well-documented problem among plan participants. But, with the advent of default options and tiered menu options (see “Adding Categories: A Sample of a New and Improved 401k Investment Option Menu,” FiduciaryNews.com, June 6, 2013), one may question if the naïve-diversification issue hasn’t already been solved. Regarding the extent of naïve-diversification in today’s plans, Carpenter says, “The Ayres/Curtis paper does not distinguish between plans with financial advisors and those without. For plans with financial advisors, the issue is moot. One can argue over whether the advisory fee charged is appropriate, but advisors uniformly provide a diversified choice of investments and either custom advice or portfolio models that provide participants with access to appropriate diversification. For plans without advisors, the DOL’s QDIA regs have made a huge difference. Prior to 2008, the options to diversify were there, but inexperienced investors typically chose the default option for 100% of contributions (and they still do), or chose a Benartzi/Thaler diversification. Since QDIA, plan sponsors have typically opted for target date funds as a default investment. Inexperienced participants understand the target date concept and embrace the default. This is by far the most common arrangement we see for smaller plans without financial advisors.”
While many plan sponsors might see their eyes glaze over when talking about such academic arcana as data sourcing and portfolio management theory, they will focus on anything that accuses them of breaching their fiduciary duty, and Ayers/Curtis are not shy in launching just such an accusation. Unfortunately – and many others have commented on this in the first Ayers/Curtis iteration last year – the two law professors seem to have their own sort of naïve sense when it comes to the DOL’s strategy on 401k fees. “The DOL doesn’t want to be placed in a corner in requiring plan sponsors’ using ‘low-cost’ funds,” says Ary Rosenbaum at the Rosenbaum Law Firm in Garden City, New York. “First off,” continues Rosenbaum, “DOL cares only about reasonable costs and not low cost, which is a huge difference since the DOL allows plan sponsors to pay more in fees if they are getting something more in service. In addition, the DOL is not going to waste political capital in picking a fight against mutual fund companies whose bread and butter is actively managed funds because ‘low-cost’ funds is a buzz word for index funds. Thirdly, like most retirement plan providers, I’m sure the DOL has cause for concern by plan sponsors being obsessed with only paying low fees, which may influence a race to the bottom.”
Marcia Wagner of The Wagner Law Group in Boston, Massachusetts says, “ERISA, from its very start, has required that fiduciaries, when utilizing plan assets, obtain services commensurate with their price, with the needs of the plan and its participants of primary importance in the calculation. There is a balancing required of fiduciaries, based on facts and circumstances, that cannot by law or common sense be driven by cost alone. To imply so is disrespectful to: The DOL tasked with regulating this industry; Plan sponsors tasked with being prudent fiduciaries; Plan vendors tasked with proffering necessary and desirable services so the private pension industry may function; and, An academic, tasked with understanding that of which he writes.”
So, will this new entry in the Ayres/Curtis lexicon hit, miss or maybe even backfire? Rosenbaum says, “While the article may have an important statement to make, the way the data was collected and the approach that Ayres had in contacting plan sponsors by using old data undercuts its credibility. You can have a great message. but sometimes it gets lost in how you developed and presented that message.”
“Of course, human beings have a certain natural tendency to embrace research/data/statistics that affirm what they already believe, and to cast aside (frequently without much thought) items that call those beliefs into question,” says Adams. “The tendency, certainly among busy professionals, is to focus on the headline, the three-sentence conclusion, or – in the case of a research paper, the ‘abstract.’ But sources matter – and data choices, and data assumptions are critical. My experience tells me that it’s important to know your source, and your source(s) source(s) – to watch out for what I call ‘spliced’ data – where two completely different and unrelated datasets are integrated in some way to create a third. And finally, and perhaps most important, just because it confirms what you think is ‘right’ doesn’t mean it’s accurate.”
Like Adams, Ron A. Rhoades, Asst. Professor, Business Department and Curriculum Coordinator, Financial Planning Program at Alfred State (SUNY) in Alfred, New York is quite familiar with the academic process and the art of writing research papers. He says, “Every study will have some limitations, in terms of the quality of the data which is surveyed. While many will criticize this study on those grounds, this does not deter from the truth which Professors Ayres and Curtis have discerned – that many, many plan sponsors appear ignorant of, or inattentive to, their duty under ERISA to ensure only reasonable fees and costs are paid in connection with the investments chosen and service providers employed. And, more importantly, that these plan sponsors are potentially liable for their failure.”
“At the same time,” continues Rhoades, “we can’t expect plan sponsors, who are often the owners of businesses of all sizes, to become experts in discerning the many disclosed and hidden fees and costs of pooled investment vehicles. This is why it is vitally important that plan sponsors – fiduciaries – rely only upon other fiduciary investment advisors, as experts. The biggest problems occur where plan sponsors work with non-fiduciary ‘retirement consultants.’ In such instances the plan sponsor remains liable, but often has no redress against the non-fiduciary consultant, who is protected by FINRA’s low, and essentially meaningless, suitability standard.”
So, like Martin Smith before them, Ayres and Curtis appear to be trying to address very real issues in today’s 401k environment. Unfortunately, their enthusiasm may have led to some undesirable incompleteness in their research as well as some off-road excursions that can direct the reader off topic. How those in the know regarding 401k and fiduciary issues respond and whether the sloppiness of their research (despite all the wonderful citations in their paper) causes them to face the same fate as Smith remains to be seen.
Interested in learning more about this and other important topics confronting 401k fiduciaries? Explore Mr. Carosa’s book 401(k) Fiduciary Solutions and discover how to solve those hidden traps that often pop up in 401k plans.