What to Know: Indexing With a Twist

The Idea Bloomberg's Matt Levine--who has resumed his daily column after four months of parental leave--professes fondness for an investment idea that he has dubbed "direct indexing." Explains Levine, "In direct indexing, you buy 1) all the stocks in the index, 2) except the ones that you don't want, and 3) plus any other ones that you do want." Think of it as indexing with a twist.

Levine continues, "In traditional active investing, you buy stocks where you have an investment thesis, the stocks that you understand and like and can make a case for. The implicit default is not buying; you have to overcome some burden of proof to decide to buy a stock. In direct indexing … the default is buying everything; there is a burden of proof to delete a stock."

In support of his notion, Levine offers two examples. The first is Norges Bank Investment Management, which is Norway's massive sovereign wealth fund. (Oil has been very, very good to Norway.) The fund begins with the global equity portfolio, then selectively deviates from that benchmark, sometimes by saving transaction costs by reducing the number of trades that it makes, and sometimes by shunning companies that it suspects of financial irregularities. Hedge fund manager Jim Chanos goes one step further than NBIM, offsetting a passive long positive in his flagship Kynikos Capital Partners with active short sales.

Levine's concept makes intuitive sense. Why do portfolio managers attempt to outguess the marketplace consensus on hundreds of companies? Better instead to preserve one's intellectual capital by taking market prices at their word, unless there is a strong basis to believe otherwise. When such occasions do arise, act. If the outlook for the company is positive, double or triple the stock's weighting. If, on the other hand, the outlook is negative, eliminate or short the position.

Adopting the direct-indexing mindset has the additional advantage of placing portfolio managers away from the crowd. Traditional active investors seek grounds to own stocks, while indexers don't bother with the issue at all. Only direct indexers (and short-sellers) take the third path, that of determining which stocks are to be shunned. With that task, they face little professional competition.

Enhanced Indexes We need not speculate about how such approaches might work, because they have been tested with mutual funds. One tactic has been enhanced indexing. Enhanced indexers don't short, but they do follow each of Levine's three rules, by 1) beginning with an investment index, 2) bypassing stocks that their managers don't want, and 3) adding securities that their managers do desire. With the third item, enhanced-index funds typically buy more of a security that exists within their index, rather than go outside the benchmark, but the point remains.

Once, I was enamored with enhanced indexers. In 1993, I recommended Fidelity Disciplined Equity FDEQX for an investment contest. I wanted to include an index fund in my recommended portfolio, to signal my support of passive as well as active strategies, but naturally I also wished to beat the market. Fidelity Disciplined Equity seemed to fit the bill. Using neural networks to adjust its S&P 500 portfolio, it had amply outperformed that benchmark consistently through its short history.

The fund surpassed the index again in 1994 before collapsing the next two years. Score one for the wisdom of the crowds; remove one for the wisdom of your columnist. Fidelity Disciplined Equity has since partially righted its ship but nonetheless lags the S&P 500 over the trailing 10-, 15-, and 20-year periods, as have most other enhanced-index funds. The group still exists, but with an aggregate $85 billion in assets, it is justifiably dwarfed by conventional index funds.

Long-Short Approaches Worse yet has been the fate of 130/30 funds, which in the mid-2000s were the talk of institutional investors, and which then were introduced to the general public. (So much for institutional excellence.) Such funds were modified versions of Chanos' hedge fund, investing 130% into equities, then shorting stocks with 30% of their assets, thereby creating a net position that was 100% long--the same as their indexed starting point, but presumably with superior performance.

That superior performance never arrived, leading to the species' near extinction. (Martha still survives, in the form of the single remaining fund that continues to bear the strategy's name, Vela Large Cap 130/30 VELAX.) That failure did not catch me unawares. Since my Fidelity Disciplined Equity experience, I had grown suspicious of "better than index" promises. In addition, other long-short equity strategies had disappointed. Nevertheless, I was surprised to see 130/30 funds fail so uniformly, and disappear so rapidly.

Selective Indexing A third form of direct indexing is Norges Bank's method, by 1) seeking to lower trading expenses, thereby increasing the tracking error between the fund and the index, and 2) omitting potentially troubled companies. That is how Dimensional Fund Advisors operates. Once considered an index provider, DFA now describes itself as an active manager because of the changes that it makes to its theoretical indexes when converting them into portfolios. To coin a phrase, I will call such a practice "selective indexing."

Over the trailing 20 years, DFA's biggest large-value fund has beaten Vanguard Value Index VVIAX by an annual 150 basis points, while its two small-cap value funds have outlegged Vanguard Small-Cap Value Index VSIAX by about half a percentage point per year. Within the tightly bunched world of index investing, those are sizable margins. Most of the discrepancy likely owes to differences in index construction--DFA uses its own proprietary benchmarks, rather than those of outside parties--than to the use of selective indexing. Nevertheless, the results are encouraging.

In summary: There's little reason to believe that index funds will improve by adding security selection. Funds that have favored stocks that their managers have judged to be attractive, or shorted those that they have disliked, have rarely excelled. But there may be something to be said for the practice of selective indexing, by permitting the portfolio to stray modestly from the index. Doing so increases the tracking error but gives management the opportunity to pick up some nickels along the way.

If one is to index with a twist, best that the twist be gently and modestly applied.

John Rekenthaler (john.rekenthaler@morningstar.com) 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.