ETFs And 401(k) Plans

August 25, 2008


Darwin Abrahamson   David Blanchett Gregory Kasten


Editor's Note: David Blanchett and Gregory Kasten's article, "Why ETFs And 401(k)s Will Never Match," which appeared in the July/August 2008 issue of the Journal of Indexes, elicited a lot of discussion in the exchange-traded fund industry. Blanchett and Kasten argued that the extra costs associated with incorporating ETFs inside a 401(k) program far outweighed any cost savings from the funds' lower expense ratios.

Darwin Abrahamson took exception to that argument. Abrahamson is the CEO of Invest n Retire, the leading provider of software that allows ETFs to function in an ETF platform. His position is not just that ETFs can work inside 401(k)s, but that they represent a dramatically better solution for the 401(k) market.

Abrahamson submitted a letter to the editor disputing Blanchett and Kasten's article, which follows below. Because it is such an important discussion, we invited Blanchett and Kasten to respond.

Debunking The Myth that ETFs Have No Place In 401(k)s
by Darwin Abrahamson

The authors of the white paper, "Why ETFs And 401(k)s Will Never Match," David Blanchett, internal consultant, Unified Trust Co.; and Gregory Kasten, president, Unified Trust Co., N.A. purport to explore the benefits and costs associated with using exchange-traded funds in 401(k)s. The authors also give a veiled attempt at providing guidance as to whether ETFs represent a better indexing option than traditional index mutual funds.


Unfortunately, instead of basing their analysis on facts, the paper relies on the myth that the technology required to add ETFs to 401(k) plans has not been developed.

For the past decade, the entire financial industry has lamented that ETFs cannot be added to 401(k) plans due to the prohibitive cost of trading ETFs on the open market throughout the day, and the industry's inability to record-keep ETFs in 401(k) plans. If the authors had fully researched these challenges, they would have discovered that Invest n Retire, LLC, (INR), a recordkeeping firm in Portland, Ore., has been trading and record-keeping ETFs in 401(k) plans, on a cost-effective basis, for over five years.

Prior to the publication of their white paper, neither David nor Gregory called me to discuss our platform. In fact, INR has a patent pending on its technology for trading and record-keeping ETFs in 401(k) plans.

In their conclusion, David and Gregory state: "Given current technology, the cost savings from ETFs in 401(k) plans appear to be minimal. …"

Debunking The Myth

Debunking the myth that the technology necessary to trade and record-keep ETFs has not been developed requires a re-examination of Blanchett and Kasten's white paper. In their conclusion, they further give the impression that even if the technology were available, the cost of trading ETFs would remain as an insurmountable obstacle to adding ETFs to 401(k) plans.

"…While the expense ratios for ETFs may be less than their respective indexed mutual fund peers, this lower cost is materially eroded by the explicit and implicit costs associated with making the ETFs ‘401(k)-ready.' In fact, it is likely that an ETF 401(k) strategy would end up being more expensive than a mutual fund strategy after all the costs are considered."

Transaction Costs

To support their conclusion, they present an argument that ETFs should be avoided due to transaction costs in the form of a bid/ask spread and commissions: "There are two primary transaction costs associated with purchasing an ETF. … The first cost is the bid/ask spread (or spread) and the second is commissions."

True, ETFs do incur a cost in the form of a spread when an ETF is bought or sold on the open market. What David and Gregory fail to mention is that the stocks and bonds, which are the investment components of mutual funds, are also bought and sold on the open market.

This oversight demonstrates that their theories are flawed by their preconceived beliefs. In case you miss my point, let me be very clear: The white paper includes misleading information regarding the cost of purchasing and selling mutual funds: "Commissions, similar to the bid/ask spread, are a cost paid each time an ETF is bought or sold, since unlike mutual funds, ETFs cannot be redeemed at NAV and must be purchased on the open market."

Invisible Costs

Mutual funds buy and sell thousands of shares a day on the open market. Each mutual fund transaction also includes commissions, as well as spreads. In a white paper commissioned by The Zero Alpha Group (ZAG), "Mutual Fund Brokerage Commissions," the researchers examined the magnitude of brokerage commissions paid by mutual funds and concluded that brokerage commissions are significant, averaging 27 basis points.

In 1994, John Bogle, founder of The Vanguard Group, coined the term, "invisible costs," in reference to the high costs of trading shares in a mutual fund.1 Bogle was referring to the fund's expenditures for trading the securities in the portfolio. In contrast to fund fees, which are reported as the expense ratio, trading costs within a mutual fund are invisible because they are not included in the expense ratio, which makes them difficult to assess.


In the white paper, "An Analysis of Mutual Fund Trading Costs," by John Chalmers, Roger Edelen and Gregory Kadlec, the researchers directly estimated the annual trading costs for a sample of equity mutual funds and found that trading costs are large and exhibit substantial cross-sectional variation. Abstract: "Trading costs average 0.78% of fund assets per year and have an inter-quartile range of 0.59%. Trading costs, like expense ratios, are negatively related to fund returns and we find no evidence that on average trading costs are recovered in higher gross fund returns. ..."2

Using data from Morningstar Principia, the researchers of the paper, "Mutual Fund Brokerage Commissions," illustrate that the Fidelity Growth fund had an estimated total cost of 1.2259 bps after adding the implicit trading and commission costs (explicit expense ratio 0.85 bps plus the implicit trading cost which includes the trading cost for the securities within the fund, based on turnover with an expense ratio of 0.2484 bps and commission cost of 0.1275 bps).3

Empirically, the researchers make it clear that mutual funds pay trading costs for the securities traded within the fund, and those costs are substantial.

Index Funds Vs. ETFs

Blanchett and Kasten also attempt to make a case that choosing a more costly index fund makes more sense than choosing a lower-cost ETF since they wish their readers to conclude, from their research, that index funds are a lower-cost option once you add the expense of trading the ETF.

"In the aggregate, since the expense ratio differences between the ETF and mutual fund strategies were so small (at most, 14 bps for Investor-share classes), it is unlikely that any material benefits are going to be obtained from unitizing an ETF, once considering all the costs (both explicit and implicit)."

Again, Blanchett and Kasten fail to mention the implicit trading and commission costs within the index mutual fund. However, the white paper, "Mutual Fund Brokerage Commissions," debunks the myth that index funds are free of trading and commissions costs.

Using the Vanguard 500 Index as its example, the paper illustrates that the fund's estimated total cost is 0.2155 bps (explicit expense ratio of 0.18, plus the implicit trading cost, which includes the trading cost for the securities within the fund based on turnover, with an expense ratio of 0.0324 and commission cost of 0.0031).4

When you consider the fact that an ETF receives shares in-kind—that is, the securities within the ETF are not traded on the open market, unlike traditional mutual funds—the implicit trading costs of ETFs are actually lower than traditional mutual funds. For example, the Vanguard Large Cap ETF has an expense ratio of .07 bps and .000 bps in trading cost for receipt of shares in-kind within the ETF.

Revenue Sharing

Blanchett and Kasten even defend the practice of revenue sharing and attempt to illustrate that by paying a higher fee for a traditional mutual fund, a plan will actually reduce its costs if the revenue-sharing fee is used to pay plan expenses.

Blanchett and Kasten rationalize, "If the revenue share monies from mutual funds are returned to the plan to offset fees … revenue share can actually decrease the total net cost of the mutual fund. In some cases, this can make an index mutual fund that has a higher expense ratio than an ETF actually be less expensive than the ETF."

In my view, that is the most preposterous justification for including investment options that pay revenue sharing. In my opinion, relying on revenue-sharing fees to pay plan expenses only encourages a lack of scrutiny of the fees being charged and who is getting paid. Case in point, how many times have you heard a plan sponsor say the plan doesn't cost us anything—it's free?

It is interesting to note that Unified Trust discloses on its Website (60/40 Asset Allocation Newsletter 10/31/06) that Unified charges a fee for revenue [sharing] collection.5

In an article, "The Legality of Kickbacks: How 401(k) Vendors are Paid on the Eve of SEC and DOL Investigations," another employee of Unified Trust, Pete Swisher, vice president and senior institutional consultant, stated: "Kickbacks is perhaps an unfair term since it suggests that revenue sharing payments are somehow illegal or unethical when in fact they are a perfectly reasonable practice. … payments typically take the form of rebates of fund expense ratios in amounts ranging from 0.25% to 0.50%, though the full range is more like 0 to 1.00%."

Mr. Swisher does admit that revenue sharing can be used for good or evil. "In a perfect world, revenue sharing would simply be a number subtracted from expense ratios to determine the true cost of money management. A fund costing 1.00% but refunding 0.50% is still more expensive than a fund costing 0.40% and refunding none. …net cost is the number to target when evaluating fund and plan expenses."6


In another white paper, "Revenue Sharing Aspects of Qualified Retirement Plan Management," Kasten states, "The practice of revenue sharing in the retirement and investment management businesses is here to stay."

I guess this helps us understand why both Blanchett and Kasten believe high-cost mutual funds are appropriate investment vehicles for 401(k) plan participants.


Darwin Abrahamson is the CEO of Invest n Retire LLC; Portland, Ore. He can be reached at [email protected].


The Rebuttal: Why ETFs Just Don't Make Sense On A 401(k) Platform
By David Blanchett and Gregory Kasten

Abrahamson's critique largely misses the entire point of our article: that low-cost, high-quality, passive open-ended mutual funds are a better solution than ETFs for 401(k)s when the total costs of making ETFs "401(k) ready" are considered.

It is standard knowledge that ETFs cost less than the average mutual fund. Most mutual funds are actively managed and ETFs are passively managed, and active management tends to cost more than passive management. Therefore, the issue is not whether the average ETF costs less than the average mutual fund (they do), but whether ETFs are a better solution than mutual funds to obtain a passive exposure to the markets. This issue is much less clear and was the underlying purpose for the original research that resulted in a paper published by these authors titled "Why ETFs And 401(k)s Will Never Match," which appeared in the July/August 2008 issue of the Journal of Indexes.

While Abrahamson rebuts our point on transaction costs, he ignores the "total cost" issue of making ETFs 401(k)-ready. ETFs in their traditional form cannot readily be used in 401(k) plans. Instead, a variety of implicit and explicit "costs" must be incurred in order to make an ETF 401(k)-ready. For example, each time an ETF is purchased, the buyer incurs the bid/ask spread and pays a commission—two fees that are not paid when purchasing a mutual fund. While mutual fund portfolio managers incur the bid/ask spread and commissions when they buy and sell stocks, the most common passive indexes have very little turnover, and such turnover tends to have a very small impact on the net performance of the fund.

Abrahamson also asserts that we "fail to mention" the "invisible costs" associated with portfolio turnover. This assertion is incorrect at both a practical and realistic level. Both of the authors (Blanchett and Kasten) have independently written and published research on the costs associated with turnover (one was an empirical study published in the Journal of Indexes "Tracking Error And The Efficient Frontier, November/December 2007); the second was a review of past literature published in the Journal of Pension Benefits). We are both incredibly familiar with the issues surrounding turnover. Also, in reality, the returns of the larger, well-run mutual funds tend to precisely mirror the performance of the benchmark index minus the explicit fee (i.e., expense ratio).

Just because an ETF has a lower expense ratio doesn't mean an investor will realize a higher return. Let's compare the largest ETF and the largest index fund based on assets as of June 30, 2008, which are both based on the same underlying benchmark, the S&P 500. The largest ETF is the SPDR Trust Series 1 (AMEX: SPY), with $71.7 billion in assets; and the largest index mutual fund is the Vanguard S&P 500 Investor Share Class mutual fund (VFINX), with $54.8 billion in assets. Since the ETF is cheaper, we would expect the ETF to have the better performance, right?

As of June 30, 2008, the net asset value (NAV) performance (according to Morningstar) of SPY was -13.58%, 4.14% and 7.36% over the preceding 1-year, 3-year and 5-year periods, respectively, while the NAV performance of VFINX was -13.19%, 4.28% and 7.45% over the same respective periods. The S&P 500 Index returned -13.12%, 4.41% and 7.58% over the respective periods. VFINX outperformed SPY by 39 bps, 14 bps and 9 bps, over the preceding 1-year, 3-year and 5-year periods, respectively, despite the fact the expense ratio of VFINX was 7 bps higher (than SPY as of June 30, 2008).

The relative outperformance of VFINX over SPY would increase further if the additional costs associated with making SPY 401(k)-ready were incorporated into the analysis, not to mention the fact that Vanguard has cheaper share classes available for larger investors. Unlike mutual funds, though, ETFs cannot be purchased in fractional shares unless they are unitized. The unitization process is not free, and the costs associated with unitization reduce the net return realized by the investor. The vast majority of platforms that offer ETFs utilize them in a unitized fashion, a point wholly ignored in Abrahamson's critique.


Abrahamson also comments that we "defend the practice of revenue sharing" in our paper. We don't defend the practice; we simply note the practical reality that revenue isn't going anywhere anytime soon. The most prudent investments through a sound fiduciary process should be selected for a plan, and if such an investment happens to pay revenue share, it should be collected and returned to the plan to reduce total plan costs. This is a common duty of any true fiduciary. Abrahamson clearly does not understand Unified Trust's business model and has made little, if any, attempt to do so.

Unified Trust's fiduciary approach in dealing with revenue-sharing issues has been carefully examined by auditors, banking regulators, the DOL and CEFEX (Centre for Fiduciary Excellence). As a discretionary ERISA fiduciary, Unified Trust prudently follows the DOL "Frost" Opinion Letter (DOL 97-15A). In the letter, the DOL sets out how an ERISA fiduciary should deal with revenue-sharing payments. The fiduciary's compensation should not vary because of any particular investment selection. Revenue-sharing funds are used to offset the fiduciary fees on a dollar-for-dollar basis. Any excess revenue above the fiduciary's fees is returned to the plan. This is the fiduciary approach Unified Trust has utilized ever since we received our first revenue-sharing payment.

Abrahamson's additional references to previous research by the authors and other Unified Trust employees are not pertinent to the underlying purpose of our research—to determine the potential benefits of ETFs in 401(k)s—and appear to be ad hominem or red herrings if little else. While he does note that his company has created a patent-pending technology for "trading and record-keeping ETFs in 401(k) plans," he provides no information on the total cost of these services or the general availability and acceptance of this approach. Without providing information on the costs of the services, or the resulting net performance actually earned by investors, Abrahamson cannot contend our argument is without merit, and that his ETF approach—or really any ETF approach for that matter—is better than a mutual fund approach.

We determined that "ETFs and 401(k)s will never match" because when all the costs are factored in, it just doesn't make sense to try to make an ETF 401(k)-ready when such a large number of low-cost, high-quality mutual funds already exist. If we had found that ETFs were a better investment option than low-cost, high-quality, passive mutual funds, that would have been the position taken in our paper; however, this was not the case. Our analysis was independent and our results should be viewed as such. We had no underlying agenda when writing our piece; we were merely trying to uncover the truth.

We recognize there may come a day when ETFs are a better solution than mutual funds for 401(k) plans, and if/when this day comes, we will gladly make the switch. Today, though, a plan sponsor desiring a passive approach is better served investing in low-cost, high-quality mutual funds than attempting to utilize an ETF approach. While there may be a few, if any, exceptions to the rule, mutual funds are clearly the rule, and ETFs are clearly the exception.



1 An interview with John Bogle in "Volume-weighted Trading Costs and Mutual Fund Performance," Roger M. Edelen, Echo Investment Advisors; Richard Evans, Boston College; and Gregory B. Kadlec, Pamplin College of Business, Virginia Tech, Nov. 16, 2006.

2 "An Analysis Of Mutual Fund Trading Costs," John M. R. Chalmers, University of Oregon; Roger M. Edelen, University of California, Davis - Graduate School of Management; Gregory B. Kadlec, Virginia Polytechnic Institute & State University - Pamplin College of Business, November 1999.

3 "Mutual Fund Brokerage Commissions," Jason Karceski, University of Florida; Miles Livingston, University of Florida; Edward S. O'Neal, Babcock Graduate School of Management - Wake Forest University, January 2004.

4 Ibid.

5 Unified Trust passes through 100 percent of all revenue-sharing payments received to the Plan, less a trading fee cost for fund trading and revenue collection, but cannot guarantee what amounts, if any, will be collected. The trading fee is a charge levied by Sungard STN and passed through to the plan by Unified Trust to offset trading and revenue collection costs.

6 "The Legality of Kickbacks: How 401(k) Vendors are Paid on the Eve of SEC and DOL Investigations," Pete Swisher, CFP, vice president and senior institutional consultant, Unified Trust Company, N.A.

Find your next ETF

Reset All