New Innovation, Just Not in the ETF Space
By Noah Hamman
First, congratulations to Eaton Vance on the SEC approval of their exemptive relief on the exchange-traded managed fund (ETMF). Interestingly, that distinction is very important as Eaton Vance has not created a new ETF. Instead, they have created an entirely new product category call the ETMF. The SEC makes this categorization clear by listing the Eaton Vance exemptive relief under “other,” and not under ETF. This truly marks the start of a new product type. I know Eaton Vance and other firms have spent a significant amount of time and resources to reach this point, and they should be congratulated on their persistence and success.
After reading through the regulatory filings, here is what I understand to be key functionalities of the new structure and how it compares to both mutual funds and ETFs.
Difference Between Mutual Funds
First, let us look at the comparison against a traditional mutual fund. On the positive side, ETMFs should be both more cost-efficient and tax-efficient, due to the exchange listing (reduced transfer agent costs) and the basket in-kind process (tax efficiency) to create and redeem new shares.
On the negative side, as far as I can tell, an investor will never actually buy at net asset value (NAV). It appears there will always be a premium to buy and a discount to sell (which is why they call it “NAV-based”).
At a high level, those are really the only differences. Eaton Vance seems to position the new structure as a better alternative to a mutual fund, but does not position it as a better alternative to an ETF.
Difference Between ETFs
ETFs are transparent. ETMFs are not. I am a big believer in transparency, and based on the feedback I hear, investors and financial advisors are demanding it. With their assets, investors have a right to lift the hood, which does not mean they want to run out and compete with, or be a mechanic. The window dressing that can occur in non-transparent products and funds is a real source of frustration for investors. It is not illegal, but investors do not want to feel they are being misled about the driver of performance for their investments.
It is a fact that outperformance can occur with transparency. Today, there are five-star Morningstar rated transparent active ETFs in both bond and equity strategies. I understand that some managers do not want to be transparent. Given the statistics that show how many portfolio managers underperform their index benchmarks, the lack of transparency is certainly not helping most of them.
Intraday execution is the other big difference between ETFs and ETMFs. The whole point of the exchange is to bring two parties together to negotiate the price of a security and conduct a transaction. ETMFs do not do this. One of the great features of ETFs is limit orders – the ability to implement similar risk control as done for stocks. Trading control and intraday liquidity are important features that also make paying for spreads and commissions worthwhile.
Other Items to Consider: Trading
With an ETMF, investors are able to transact during the day, but are just locking in their cost (spread). I will be curious to see how the optics will work for the investor with the intraday quote. On the asking price, it might just display $0.50, but I am not sure what it will show on the bid price. I would guess a negative number, say -$0.50, although I have never previously seen a quote for a negative price. It feels like adding a new color to a traffic light, we can be trained on what it means, but it will look weird for a while until we get used to it. It would seem the exchanges and the brokers will require programming updates for their systems to allow negative quotes. That is not a difficult thing to do, but in this case, the question is both a return on the cost and competition. Certainly if this will generate significant income for the broker-dealers, then of course they will invest the time and money for software/systems development. But competition might also be a factor. Schwab has their own funds and ETFs, Fidelity has their own funds and ETFs, are they going to change their systems for a patented process owned by a significant asset management competitor? Maybe, but certainly that consideration will come into play.
An investor will not manage execution price – that will be determined later in the day after market close. Buyers will only manage their execution costs with trades. It would be possible to see an ask price at $0.25, but later in the day watch it at $0.20. A change in the value of the underlying assets would not cause that difference; instead the difference is just based on market maker perception/estimate of the cost of conducting the transaction at that size and time of the day. With a basket in-kind process, the market maker will need to buy the underlying assets to deliver to the fund. Therefore, it is likely the less time the market maker has, the more likely a higher cost will be charged, and the larger the size, may increase the cost/spread as well. What reactions would surface if a submitted order executed at $0.25 per share is then “on-sale” an hour later in the day at $0.20 per share? There will certainly be an answer for that, but the process of explaining it to a client or investor will be challenging for a while (until we get use to that new traffic light!).
Right now, it’s unclear that limit orders will be available, but I could see that being a good option to help control the spread costs, and possibly use more people/orders to drive the cost down. Consider a scenario where several investors put in limit orders to buy at $0.01 over NAV. When enough orders cue up at that price to create a full creation unit (CU), then the market maker will take all those orders. Who knows how many hours or days that might take, as well what would happen to the price of the investment (which honestly is what an investor really cares about) while waiting. I am not sure this would work, but also not sure “cost” limit orders will even be an option for ETMF.
Extra Market Maker Costs (Spreads)
I believe the ETMF will bring more risk to market makers that could lead to higher transaction/spread costs. A statement in the SEC summary about market maker risk notes: 1
“According to the applicants, in NAV-based trading, by contrast, Market Makers do not engage in arbitrage and assume no intraday market risk in their share inventory positions because all trading prices are linked to NAV. 21 Applicants state that no intraday market risk means no need for Market Makers to engage in intraday hedging activity, and therefore no associated requirement for current portfolio holdings disclosure to maintain a tight relationship between Share trading prices and NAV.”
I interpret that as no need to arbitrage intraday prices because the ETMFs do not trade intraday, but it doesn’t mean the market maker will not take on risk. I see three areas of risk to market makers which investors should expect in turn will be reflected in the spread costs.
One is multi-day inventory risk, but within the footnotes of the SEC summary says, “Applicants believe that Market Makers will generally seek to minimize their exposure to price risk in Shares by holding little or no overnight inventory. ETMFs also will have smaller creation unit sizes than ETFs”2. This will help, but a smaller CU size alone will not reduce the costs. In fact due to the CU fees, it is likely the market maker will have a higher cost per share they redeem back to the fund. I like the smaller size CU, it really does help, but unless Eaton Vance has innovated the cost the ETF service providers charge for CUs, then there will definitely be a cost to the market maker that is likely passed on to investors in the spread.
Second, the market maker has risk with purchasing the underlying portfolio. For example, an investor comes in to buy $20M of the ETMF. The market maker goes out to buy the underlying basket for delivery, but because it’s not the actual basket, there is risk in the difference in the value of the made-up basket versus how the actual NAV moves that day. That slippage is a cost to the market maker that again is passed on via spreads, and more importantly, unlike a traditional transparent ETF that can be perfectly hedged, the larger the purchase size of the ETMF, the larger the risk to the potential market maker.
Third, the SEC summary also says, “Any purchasers or redeemers of Creation Units are expected to incur a transaction fee to cover the estimated cost to the ETMF of processing the transaction, including the costs of clearance and settlement charged to it by NSCC or DTC, and the estimated trading costs incurred in converting the Basket to the desired Portfolio Positions.3 I italicized for emphasis, but this basically explains the market maker or authorized participant (AP) pays a new fee that is not a fee that is part of the ETF CU process. This is the added cost of non-transparency, because the basket is not the actual underlying holdings, or at least all of the actual underlying holdings. The market maker will pay the cost of any trading adjustment needed to acquire the actual shares of securities in the portfolio. So the market maker has no idea what this will be, how big or how small (it’s also unclear if this is a positive number, does it go the market maker or investors?) – but again the reality is not only are there several costs to the market maker – this is a third impact to spreads of which two are unique to ETMFs.
Taking a Better Approach
If creating a brand new structure, implementing other improvements would have been great too. For example, it appears that the premium and discount calculation will be the same – using only end of day prices to determine the average. In today’s data-filled day and age, it should be very easy to take execution prices and the indicative value at the time of execution, and then calculate a real average of what an investor should expect for a premium or discount. It is too bad a little more innovation was not added to this new approach. The end of the day is probably the worst time to trade an exchange-traded security. Measuring efficiency at that time of the day is an area that should have been improved in a new product, and hopefully, eventually will be improved for ETFs.
Why an Exchange?
I am still stuck on the fact that the SEC is changing what occurs on an exchange. Investors are “trading fees”— which are the only items priced and valued immediately as part of the trade. Then at the end of the day, the exchange provides a “service” to process orders at NAV, similar to how a mutual fund transfer agent operates, except in this case a fee is paid to the market maker instead of the transfer agent.
Imagine a scenario where a Registered Investment Advisor (RIA) manages his or her clients’ assets at TD Ameritrade. Let us say that this advisor has two Eaton Vance fund choices, one, likely an institutional class share mutual fund, and the other an ETMF. Both are easily accessible for the advisor and the client. The mutual fund can trade at NAV and be guaranteed without the additional fee. The advisor could buy the ETMF, be guaranteed a higher fee and not have a reassurance that the tax efficiency will be there. Which choice would a fiduciary make for a client, and how would a compliance department react?
The Loss of Crowd-Based Pricing
I wrote about this previously, but one of the great features of ETFs is the crowd-based fair value pricing. See the article here: http://www.thinkadvisor.com/2013/08/26/crowdsourced-etf-pricing
In short, an investor relies upon the fund sponsor to make the decision when fair valuation is needed for both mutual funds and ETMFs. I would argue the ETF has a much more market efficient crowd-based approach in that regard.
Be Careful in Cash or Retirement Accounts
Here is a cautionary note for those potentially investing in an ETMF in a retirement account (and cash accounts too). Many advisors invest in mutual funds for their client accounts, of which many have brokerage accounts that are either IRAs, or rollover IRAs. A great feature about investing in mutual funds at NAV is the ability to both buy a share or a dollar amount. If only $5,000 in cash sits in an IRA brokerage account, an advisor submits an order for $5,000 to ensure that IRA account is not overbought. It’s a very easy process. For the ETMF, an advisor or investor does not really know what exactly to expect, and potentially runs the risk of overbuying in an IRA account, which as most financial professionals know is not a great scenario for a retirement account.
Much More Work Ahead
It would seem between the exchanges and brokerage trading systems, a lot of software programming will need completion to support ETMFs – I think that would be the case with any new product that operates differently. There is probably a lot of work that will be done among compliance departments as well to ensure they understand how these products will work for clients. It will be interesting to see how everything develops. Finally, I cannot say that I understand everything about these products, and being in the active ETF space, I will continue to learn more and will update my thoughts accordingly. If you spot any inaccuracies in my analysis of this new product structure, please do not hesitate to let me know.
Looking at the differences between ETFs, mutual funds (MF) and exchange-traded managed funds (ETMFs), it would seem ETMFs have a few more features than mutual funds, but not nearly as many as ETFs.
1 Source: Security and Exchange Commission, Release No. 31333, November 6, 2014, p. 10 (www.sec.gov).
2 Source: Security and Exchange Commission, Release No. 31333, November 6, 2014, p. 10 (www.sec.gov).
3 Source: Security and Exchange Commission, Release No. 31333, November 6, 2014, p. 9 (www.sec.gov).