Active on the Active Front

October 13, 2005

Actively managed funds are the next frontier for the ETF industry; a look at how the battle’s shaping up in the United States.

For years, actively managed funds have topped the "what's coming next" list for the exchange-traded fund (ETF) industry.  Fund issuers have called active ETFs "the Holy Grail of the ETF industry" and "the greatest thing since the money-market fund."

Why all the excitement? Simple: The size of the market.  The active mutual fund industry dwarfs the passive/index business.  If ETFs can tap into active investors, the opportunity for growth is enormous.

Unfortunately, the challenges are just as big. 

One of the key features of ETFs is that their share price closely tracks their net asset value (NAV).  The reason is that shares of an ETF can always be converted back and forth into shares of the underlying securities of the fund.  If you hold 50,000 shares of most ETFs, you can return them to the fund issuer and demand the equivalent value in the underlying stocks; it's called "redeeming" shares.  Similarly, if you want to "create" 50,000 shares of an ETF, you simply buy the underlying securities and deliver them to the ETF provider - who, in turns, gives you the ETF shares.

This interchangeability means that arbitrageurs ("arbs") are always on the lookout for discrepancies between the NAV of the fund and the share price of the ETF.  If the two move apart, the arbs buy the ETF and sell short the underlying shares (or vice-versa), and profit from the difference. 

That's easy for index ETFs, where the portfolio holdings are freely disclosed and the NAV is published every fifteen seconds.  But active fund managers are notoriously shy about disclosing their holdings on a regular basis.  They worry that traders will trade ahead of the fund - driving up the price of stocks they're buying, and driving down the price of stocks they're selling. In fact, active fund managers are so paranoid that they don't even want to publish the true NAV of their funds on a regular basis - for fear that savvy traders will "reverse engineer" the actual holdings of the fund.

In designing an active ETF, then, the core challenge is finding a way to allow the arbitrage traders to do their work - without tipping the hand of active managers. 

Two groups -  the American Stock Exchange and Managed ETFs, a start-up company founded by ETF pioneer Gary Gastineau - have been working on this problem for a number of years.  Both sides say that they're getting closer to a solution.  In fact, the first actively managed fund could be just around the corner.

The Amex's Tracking Portfolio

The AMEX's solution revolves around the creation of something called the "tracking portfolio," a kind of alternate portfolio that can be used as a stand-in for the real thing.  The "tracking portfolio" for an active ETF will hold different securities than the ETF, but will share the same risk/reward profile and will "track" the ETF's performance closely.  The Amex will create this tracking portfolio based on what it calls a "complicated risk analysis" of the actual ETF holdings. Then, instead of publishing the actual holdings and NAV of the ETF every fifteen seconds (as the SEC requires for most ETFs), the Amex will publish information about this tracking portfolio.

"The tracking portfolio will be another portfolio composed of the same kind of securities as the ETF, and constructed in such a way as to closely mimic the intraday value of the ETF," said Tony Baker, who's been heading up the active effort at the Amex..  "It will be constructed so that it cannot be used to reverse engineer the actual holdings of the fund." 

Using back-testing, Baker says that his tracking portfolios have performed well - running within a standard deviation of 5 to 38 basis points of the underlying fund, depending on the specific nature of the fund.  (By standard deviation, Baker means that the tracking portfolio will stay within X basis points of the true NAV most of the time - 68% of the time, to be exact.)

The question with these tracking portfolios is whether the specialists and market makers will be able to deliver tight enough spreads to entice investors to make the products attractive. After all, the market makers will bear the risks of any tracking error - and that risk is sure to be captured through the spread.  The Amex will need to convince the market-makers that the tracking mechanism is robust, and that it can be counted on to perform well ... all the time.

"What's important here is that all of this technology and process is intended for a simple purpose - to put enough information in the hands of market participants for them to quote fair and efficient markets," said Baker.  "Investors don't really need to know how this works - they just need to know that it does work."

Baker said that market makers are confident they can provide a reasonably efficient market based on the information in the tracking portfolio.  At the end of the day, it will publish the true NAV and holdings of the fund.

Amex has no plans to develop funds on its own; it's working only on the platform.  It will be up to the fund managers to partner with the Amex and actually move funds forward with the Securities and Exchange Commission (SEC). A number of fund companies have expressed interest, but there are no funds in registration today.

Gastineau's Random Offset

Gastineau's efforts differ from the Amex's in many ways. For one, Gastineau's outfit - Managed ETFs - plans to both license its technology AND manage funds of its own. But the bigger difference lies in how Managed ETFs will distribute information about its funds. Instead of creating a tracking portfolio, Managed ETFs will publish a value based on the actual holdings of the underlying fund.  The twist - and this is important - is that it won't be the true value of the fund; it'll be a close approximation.

"You can't give the exact price every 15 seconds, or traders could back-calculate and figure out what's in the portfolio and what is changing," said Gastineau.  "Instead, we will take the exact value and increment or decrement it by a number drawn from a random distribution."


In other words, it will tweak the NAV - just enough, it hopes, to throw traders off the scent.  For instance, the published value may be off by 10 basis points at 10:00AM, and then 30 basis points fifteen seconds later, and then minus two basis points fifteen seconds later, and so on. 


Once an hour, the group will distribute the true NAV of the fund.


The advantage of Gastineau's plan is that the publicly-disclosed value has a definitive relationship with the NAV; the downside is that you don't know what that relationship is.


Building a Basket


The second major problem for active ETF managers lies with the creation/redemption process.  The trouble should be obvious: To create a basket, you would have to know what's in the fund.   But how can you know what's in the basket without tipping the manager's hand?


The answer is that you can't. Which is why Gastineau's funds comes with a few interesting twists.


Gastineau says that fund managers are perfectly happy for other people to know what they already own, but they don't want people to know what they're in the process of buying or selling.  As a result, the creation/redemption basket for the Managed ETFs won't reflect changes in stocks that are being bought or sold … until the total position is created (or liquidated).


"When you're selling a stock - say, a five percent position in the portfolio - you'll say nothing about it in the creation/redemption basket until the five percent position is all gone," says Gastineau.  "And when you're adding a stock, you won't show anything until you get your full position established.  Then, it will appear in the basket."


The trouble with this is that during creations, the ETF manager will not receive shares that match his new fund composition - the delivered basket will overweight shares that the fund manager is selling, and underweight shares that he's buying.  To deal with this problem, Gastineau says that all orders to create new shares must be submitted to the fund issuer by 2:30.  The fund manager will use the remaining 90-minutes of the trading day to adjust the supplied shares - say, by selling part of one holding and using the proceeds to buy another, to make it match up with the real portfolio.


The ETF will continue to trade after 2:30, but there will be no additional creations.  If a market maker needs additional shares to sell, he can still create a basket after 2:30 - but he will have to deliver it the next day, and it will have to reflect the next day's basket.  He must take the risk that there will be changes. Chances are, this risk will be reflected in a wider spread for these shares.


Amex Offers Options


The Amex is taking a much simpler approach. If the active fund managers don't want the creation basket to reflect the active portfolio, why force it?  Why not let the active fund managers decide what they'd like to receive instead?


Under the Amex system, the creation basket would take one of three forms. It'd be up to the individual fund designer to choose what they'd like to receive:


1) They could receive cash - plain and simple.  It would be the fund manager's job to decide how to deploy that cash.  This would be a straightforward process, but perhaps less attractive to the specialists and market makers.


2) They could receive shares in one or more passive ETFs.  This may sound strange, but mutual funds buy and sell passive ETFs all the time as a way of quickly "equitizing cash."  This way, the cash would already be "equitized" for them.


3) They could receive a named basket of securities.  Fund managers would choose the basket as they see fit.


The concept is borrowed (in part) from the way traditional mutual funds receive inflows.  Traditional mutual funds create new shares all the time - they do a "creation" whenever an investor sends in cash.  It's the manager's job to deploy that cash, either by buying stocks, ETFs or bonds - or by keeping the cash as cash, and reducing the fund's exposure.


"The one we think will be the most popular is the benchmark ETF approach, but it's the manager's choice," said Baker.  "Each manager will weigh what is most advantageous for them, and will work with the specialists to see what works most efficiently.  (But) fund managers equitize cash in their portfolios all the time. This is the same thing. They're just getting the cash already equitized for them."


The leap is that this doesn't exactly mirror how other ETFs work - the ETF fund manager doesn't just watch their holdings grow. They have to play an active role in the process. But then again, that's the point, isn't it?


The downside is that the inflows will skew the exposure of the fund - a problem for traditional active funds as well.


The Issue With Spreads


Higher spreads are one of chief concerns of active ETFs critics.  Given the imprecise information about the value of the fund, arbs will not be able to keep the spreads on the fund as tight as they can with some indexed ETFs. 


But both Baker and Gastineau argue that high spreads won't be a major problem, saying that they will be similar to the less popular ETFs on the market today.  


"The reason the spread on the SPDRs and the QQQQs is a penny is due to two things," says Gastineau.  "One is the high degree of predictability of the value of the fund at a given time, and the second is the number of orders and the amount of trading. We don't expect actively managed ETFs to have the kind of trading volume that the SPDR has, so on that basis alone, you would expect a slightly larger spread.  But I don't think it's going to be a major problem."


Where's The Demand? And Who's Interested In Issuing Funds?


Which brings us to the big Field of Dreams question: If they build it, will investors come?  Are investors actually hungry for active ETFs? 


Both Gastineau and Baker say "yes," pointing to the huge amount of money tied up in active funds.


"On the mutual fund side, people are willing to pay for the skill of an active manager," said Baker.  "There are an awful lot of assets in active funds.   Using the ETF structure doesn't change that equation."


Nor, according to Gastineau, does active management change the basic benefit of ETFs: convenience, tax efficiency, and lower expenses.  While it remains to be seen how much the active ETFs will charge, they should be able to pass on some of the inherent administrative cost savings along to investors.


To help attract assets, Gastineau's group plans on creating different types of shares for its ETFs - front-end load, level-load, etc. - to match the fund and distribution structure of existing active funds.


Baker has no specific plans for multiple share classes, but then again, the Amex isn't creating its own funds - it's merely creating the technology. It will be up to each individual fund manager to decide on the share class structure for the funds.


"Amex is an enabler, not an issuer," he says.


Baker says that the Amex is working with fund managers of all sizes, although most of the interested parties are large scale clients. Some of these managers may follow the Vanguard model and create ETF share classes of existing active funds, while others may launch entirely new products.


"We think the share class approach could be very attractive. But for some managers, this will be an opportunity to expand their product line with new funds."


And for everyone, it will be an opportunity to grow the ETF industry.

Find your next ETF

Reset All