Will ETFs Replace Open-End Mutual Funds?

By Heather Bell

A recent article on the SSRN Web site by Ilan Guedj and Jennifer Huang at the University of Texas offers a fresh look at the great exchange-traded funds versus open-end (mutual) funds debate (ETFs vs. OEFs).

The title, "Are ETFs Replacing Index Mutual Funds?", pretty much sums up the point of the article, which looks at how ETFs and traditional index mutual funds size up against each other using hard data. The authors, by the way, are affiliated with the university's McCombs School of Business.

We all know what the accepted advantages of the two different fund types are, but with ETFs having been in existence for about 15 years and having seen exponential growth in the last few years, a revisit of the hard numbers behind those precepts seems like a good idea.

However, I found the article's organization fairly chaotic and the authors' grasp of ETFs a bit weak (unit investment trusts, for one, are not the most common form of ETF, as the authors assert). So, be warned. On the upside, it is chock-full of some good insights.

As the study's authors ask, are ETFs then going to replace index mutual funds? I know the suspense is killing you. Brace yourselves: The answer is no, at least in this piece of research.

Guedj and Huang approach the issue from the perspective of the small investor's trading needs, examining how the two types of investment vehicles meet those needs, and they draw heavily from existing research on OEFs in constructing their report.

As we know, a mutual fund spreads the transaction costs incurred among all of its shareholders when shares are bought and sold, while the individual ETF shareholder alone bears the cost of his or her own transactions, with small investors simply trading existing shares on the open market rather than creating or redeeming them. However, the article finds that the actual costs of both types of funds are approximately the same.

Reckless Trading Encouraged?

In the ETF industry, much is made about the stock-like trading capabilities of ETFs and the fact that shareholders are not affected by the trades of other shareholders. The authors point out that although share redemptions in OEFs affect all shareholders and can encourage reckless trading, thereby increasing tracking error, risk-averse investors may find the OEF structure appealing because of the likelihood of getting a better price in the event of a large liquidity shock. OEFs offer a sort of "liquidity insurance," and some investors are willing to pay a premium for it in the form of lower overall returns.

However, this liquidity insurance is something more associated with smaller funds, which do not have much of an effect on the market, and the benefits diminish with larger funds, which can often see greater tracking error and price distortion.

The fact that trading expenses are shared among all investors, not just those buying and selling shares, causes what the article's authors term a "moral hazard." Since costs are shared across the entire shareholder base, there is little reason for shareholders to take them into account when buying and selling shares, which basically can lead to willy-nilly trading, which in turn drives the costs up and increases tracking error. Basically, the risk-averse investor who prefers small OEFs will switch to ETFs when a fund becomes too big, or "moral hazard" trading becomes excessive.

Of course, a sort of corollary to this is that in the case of very large OEFs, trading by shareholders can actually affect the underlying stocks, which means that ETFs are also affected. Further, the more correlated investors' liquidity shocks become, the less attractive the OEF begins to seem since the impact on stock prices and tracking error is greater, and the ETF becomes more attractive as a result.

The article also observes that the number of index OEFs and ETFs is about the same, but that ETFs cover five times as many indexes. Moreover, the article points out that newer ETFs tend to track indexes that are less liquid, more volatile and more heavily concentrated in a single industry. Such indexes are more appropriate for an ETF structure than they are for an OEF structure, which would likely experience high tracking error.

This would seem to be part of the reason for the growth of sector and emerging markets investing. Although the article is focused on OEF and ETF structures, it can be extrapolated that ETFs have opened up investors' opportunity sets significantly.

In an interesting aside, the authors suggest that actively managed ETFs might be an ill-conceived idea, citing evidence that the OEF structure allows investors to reward manager ability by voting with their feet. However, I fail to see why a similar trend would not be possible with actively managed ETFs. After all, if active ETFs are managed poorly, investors will simply trade them at discounts to their actual worth or not trade them at all. And, in theory, new shares will not be created either. Mainly, the great virtue of the article is that it offers some quantitative proof for something we already kind of knew—that ETFs appeal to a great many people, but that OEFs are more suited to certain specific groups of investors, and primarily the risk-averse small investor.

Costs Less Important

Usually the example that is cited involves costs, not liquidity shocks: Trading costs associated with ETFs may wipe out their advantages for smaller investors who may be dollar-cost averaging their investments or using a similar strategy. For those investors, OEFs are by far the better choice. But with the advent of no-commission ETF trading, costs are slowly becoming less of a factor for small individual investors.

In light of these more recent market developments, the Guedj and Huang article provides another dimension. They highlight the fact that ETFs are growing in popularity not just because of their "stock-like" characteristics but because they can more efficiently offer exposure to more asset classes and sectors than OEFs.

Extrapolating from Guedj and Huang's conclusions, a person familiar with the ETF industry would conclude—as many already have—that while index-tracking OEFs will not face extinction, they are unlikely to grow any further as an industry, and it is only a matter of time before they are dwarfed by the ETF industry.

Comments

[Briggsy]

What feature of ETFs allow them to trade very close to their net asset value, unlike typical closed-end funds? Anyone? Anyone? Bueller?

[rainman]

If one carefully config his own porfolio of ETF, he can be a super fund manager of his own, forget mutual fund, slow and inefficient and it may be hurting those retirement money in general.

[B. Ray]

The arbitrage character of ETFs will make ETFs very close to their NAVs.

[huangthomas]

Closed-end funds are subjected market premium or discount, which can sometime deviate 20 to 30% from NAV.