Does It Ever Make Sense to Index Portfolios?
Not for mutual funds but perhaps for hedge funds.
An Unpopular Idea
Almost nobody thinks that indexing investment portfolios makes sense.
There are a few exceptions. Certain versions of hedge fund indexes exist. Clearly, some people will defend the proposition. However, I have never met any of them, except for Andy Lo in passing. When Bloomberg's Matt Levine scoffed at indexing hedge funds, he spoke for nearly all investment researchers. Morningstar's team, for example, is universally skeptical.
Elsewhere, of course, indexing is a huge success. By Morningstar's estimates, index mutual funds and exchange-traded funds now possess more U.S. equity assets than do conventional mutual funds. The U.S. equity indexers won't be caught; that race is over. Other contests have yet to be settled. Eventually, though, indexers will likely dominate the remaining asset classes, too.
However, they haven't made any traction with indexing portfolios. Hedge fund indexes have few assets, if there are indexed private equity or venture capital funds I have not heard of them, and mutual funds do not index investment portfolios. (Some funds are structured as funds of funds, meaning that they buy other funds, but they are actively run.) The providers are also unconvinced.
Such suspicion doesn't necessarily mean that the idea is unsound. After all, stock indexing itself was initially scorned. Vanguard 500 remained the only index mutual fund for half a decade after its 1976 launch, partially because the company's competitors didn't wish to cannibalize their existing fund lineup, but also because the Vanguard fund wasn't selling well. The marketplace is an unreliable judge.
In that spirit, let's examine whether indexing portfolios might make sense. Understanding the reasons for indexing individual securities is a good starting point. Why do most of today's U.S. equity fund investors choose funds that hold every stock, rather than invest in funds that buy only the good ones?
You know why: The past does not reliably repeat. Investors can easily determine what has occurred, but they struggle to use that information effectively. History gives valuable clues. Unfortunately, the marketplace contains thousands of highly informed shareholders, who each use those clues when setting a security's price. Consequently, reading the evidence accurately isn't enough. To be successful, investors must also perceive where the consensus is wrong.
Also, a diversified portfolio, created by indexing, doesn't face an individual security's "event risk." The portfolio will not go to zero, but the stock (or less commonly, the bond) might do so. This is not an argument for indexing, per se, because diversification also occurs from active portfolio management, but it does offer additional support for owning many securities, rather than selecting the handful that look the most attractive.
These attributes are obvious. Less appreciated is the investment math for securities, such as stocks, that do not follow a normal distribution. Over the long term, most stocks stink. The typical U.S. equity breaks roughly even in nominal terms during its lifetime and outright loses money after inflation. Therefore, a diversified portfolio that keeps roughly apace with the median will have a higher (much higher) expected return than a single stock.
Stocks and Bonds
In summary, equity investors have three reasons to index:
1) They can't identify winners ahead of time.
2) Their investment won't go to zero.
3) Their median performance will improve.
The logic for fixed-income investing is not as strong. The first item persists, as bond investors also have difficulty predicting the future, but the next two attributes lose importance. Absent a sharp increase in interest rates, against which diversification cannot protect, investment-grade bonds are unlikely to sustain severe losses. And their returns are not skewed as with equities. Thus, the median performance for investment-grade bond portfolios lies near the average.
The final consideration when evaluating the attractiveness of indexing is whether the strategy faces competition. Are there existing investments that accomplish essentially the same thing? The answer was negative for U.S. equities, because while there were plenty of diversified U.S. stock mutual funds, none could match Vanguard 500's price. Also, for those investors who cared greatly about relative performance, no rivals could reliably hug the S&P 500 so closely.
From Stocks to Portfolios
Let's score the tactic of indexing portfolios by these measures.
First, registered funds--meaning publicly available funds subject to the Investment Company Act of 1940, such as mutual funds, ETFs, and closed-end funds. Once again, it is difficult to identify future winners. One can modestly improve the odds of selecting an above-average fund by screening on positive attributes, such as relatively low expenses and turnover ratios (or, alternatively, on high Morningstar Analyst Ratings), but such actions aren't remotely close to being guarantees. For the first principle of indexing, the logic for doing so with registered funds is just as sound as with stocks or bonds.
From there, though, the appeal declines. As with high-grade bonds, registered funds rarely implode (when disasters do strike, they hit highly specialized funds like leveraged ETFs, rather than mainstream choices), and their performances pretty much follow a normal distribution. Consequently, the tactic of indexing registered funds doesn't help with the investment math in the same way that occurs with indexing individual stocks.
Also, fierce competition already exists. A stock fund that bought all actively run large-blend funds that passed its quality screens would be a good investment, assuming it kept its own fees very low. Unfortunately, that index of large-blend portfolios would be compared against a conventional stock index fund, which would have much the same holdings, but would charge less. There's no practical need for such funds.
Considering Hedge Funds
The case for indexing hedge funds, ironically, is stronger than with registered funds. Although many seem to believe that identifying future winners is easier for hedge funds than for stocks, bonds, or mutual funds, such is not the case. The typical hedge fund of funds performs like the underlying hedge fund indexes, minus their additional expenses. In other words, the professionals who are paid to select the hedge funds that have the best "alphas" don't on average beat dart-throwing monkeys.
So far, the justification for indexing hedge funds is no greater than that for indexing registered funds. However, hedge funds derive a greater benefit from the investment math. They disintegrate more often than do registered funds, and their returns are more skewed, too. Those tendencies aren't as pronounced as with equities, so the argument for indexing stocks is stronger yet than with hedge funds, but they are valid reasons to consider indexing.
And there aren't true rivals for an indexed portfolio of hedge funds. There are several hundred hedge fund of funds, but they are actively managed, thereby adding a steep second layer of fees. A hedge fund index portfolio would be significantly cheaper than any of them. The cost difference would make an indexed hedge fund a different category of fund (assuming, of course, that the indexed fund kept its own fees low).
In summary, the argument for indexing hedge funds is better than is commonly acknowledged. It's not overwhelming, but neither is it silly. On the other hand, indexing mutual funds, ETFs, or closed-end funds is pointless. The practice would not lead to something that investors cannot already buy.
John Rekenthaler has been researching the fund industry since 1988. He is now a columnist for Morningstar.com and a member of Morningstar's investment research department. John is quick to point out that while Morningstar typically agrees with the views of the Rekenthaler Report, his views are his own.