William J. Bernstein
The Big Lie
As Joseph Goebbels knew when he drafted the Nazi campaign blaming the Jews for Germanyís failure to rise out of the ashes of the Great War, if you tell a lie often enough, it eventually becomes accepted as the truth.
Year after year, investment pundits repeat a fundamental falsehood about indexing that has somehow acquired the ring of truthóthat indexing works for large-cap stocks, but not for foreign or small-cap stocks. The origins of this monstrosity are lost to time, but Charles Schwab deserves a major portion of the credit for popularizing this notion in its "core-and-explore" strategy, where one indexes more of a portfolioís large-cap segment than its small and foreign.
I had hoped that core-and-exploreís manifest failure over the past three years would put this old chestnut to rest, but sadly, it hasnít. In the Mutual Funds supplement to the May 6, 2002 Wall Street Journal, former Fidelity chief Robert Pozen, now firmly ensconced at Harvard, opined: "Active managers beat the relevant indexes on a regular basis for things like international funds, small-cap funds, etc."
Well, I really didnít expect a balanced view of indexing from a Fido bigwig, but I had hoped that Harvard profs were a bit more data-driven. Iíve written about this one until my fingers ache, but no one seems to listen. So, Professor Pozen, this oneís for you.
Letís start by dividing small cap into value, growth, and blend categories. The oldest small-cap-value index fund is the DFA U.S. Small Cap Value, which ranks 7th out of 58 funds at five years. It incepted in 1993, but itís not much of a stretch to extend its record backwards by tacking on the performance of the Fama-French small value index, which it tracks closely. (Actually, almost all DFA funds beat their underlying indexes, so this is a more-than-fair approximation.) At 10 years, this fund would rank first of 14 funds, and at 15 years, first of nine funds.
For small blend, it all depends on which index you use. The Vanguard Small-Cap Index fund ranks only 26th of 36 funds at 10 years, and the underlying Russell 2000 Index would have ranked 12th of 16 funds at 15 years. But DFA U.S. Micro Cap ranks 6th of 36 at 10 years and 7th of 17 at 15 years, and DFA U.S. Small Cap, exactly in the middle of the pack at 10 and 15 years. It turns out that thereís a problem with the Russell 2000 Indexóit is rebalanced every June 30. Since it is defined as the 1001st through 3000th stocks ranked by market cap, and since it is the most widely used small-cap index, savvy traders can easily predict which stocks will be added and dropped from the index, bidding these stocks up or down before June 30, adversely impacting the indexers who must buy or sell these stocks after June 30, lest they incur increased tracking error.
The DFA funds do not suffer from these disadvantages. Neither does the S&P 600, as it is committee-chosen and thus impossible to predict which stocks will be added and dropped; were it a fund, it would rank 60th of 160 funds at five years, and 13th of 36 at 10 years. In any case, the mediocre performance of small-blend indexing is largely an illusionóadjust for survivorship bias, and even the Vanguard Small-Cap Index fund beats the average actively managed fund by a significant margin.
Finally, small growth, as weíve noted before, is a real problem for indexers. The Fama-French small growth index would have ranked 34th out of 53 funds at 10 years and 22nd out of 25 funds at 15 years. Weíve been down this road beforeómomentum effects are strongest in the small growth arena. An index fund that kicks out its strongest performers, which a small growth and to a lesser extent a small blend fund does, suffers accordingly. So, yes, donít buy a small-cap-growth index fund. But the larger point is simply not to buy small growth funds at allóthis is a miserable asset class, with long-term historical returns lower than all other market segments.
Finally, REITs. Surely, this a specialized, inefficient area where savvy analysts should be able to pick out underpriced securities. Well, no. DFA Real Estate Securities ranks 13th of 60 at five years; the Vanguard REIT Index offering, 28th. Again, add in survivorship bias, and the Vanguard fund too outperforms by a handy margin.
At first blush, indexing foreign stocks also appears to be a loser: the MSCI-EAFE, were it a fund, would have ranked 45th of 77 foreign entries at 10 years, and nearly dead last at 15 years.
The problem, of course, can be explained in one wordJapan. Indexing the foreign market in 1989-1990 would have resulted in a portfolio consisting of nearly two-thirds Japanese equity, something that even the most ardent indexer would not likely do, and which killed the EAFE index for over a decade and a half.
So, letís break things down by region. At 10 years, the Vanguard European Index fund ranked 7th out of 19; at 15 years, the index itís based on would have ranked first of seven. Pacific Rim? Thereís no fund with a 15-year track record, but the MSCI Pacific-ex-Japan Index would rank second of four at 10 years and 20th of 46 at five years.
Emerging Markets? Now, if you buy the argument that active management adds the most value in inefficient markets, it should do so in this arena. The best index data are for the DFA Emerging Markets and Emerging Markets Value funds. At five years, they rank 30th and 5th of 106 funds, and at 10 years, adding on their strategies before their 1994 incepts, they would rank second and first of seven funds. (The Vanguard Emerging Markets fund ranks 46th of 106 at five years; since they donít precisely track the MSCI Emerging Markets Index, itís difficult to judge just how well they would have done at ten years.)
Lastly, I canít help myself from adding in a value twist: At five years, DFA International Value ranked 122nd of 423 diversified foreign funds, even with its heavy Japanese market weighting. Adding on its strategy before the 1994 incept, it would rank 15th of 79 at ten years, and 4th of 29 at 15 years.
No one really expects the truth from anyone at Fidelity. But, Mr. Pozen, youíre in the big leagues now. People expect that when a Harvard professor opens his mouth heís at least had an ever-so-brief look at the data. Better luck next time.
Copyright © 2002, William J. Bernstein. All rights reserved.
The right to download, store and/or output any material
on this Web site is granted for viewing use only. Material may not
be reproduced in any form without the express written permission of William
J. Bernstein. Reproduction or editing by any means, mechanical or electronic,
in whole or in part, without the express written permission of William J.
Bernstein is strictly prohibited. Please read the disclaimer.
The right to download, store and/or output any material on this Web site is granted for viewing use only. Material may not be reproduced in any form without the express written permission of William J. Bernstein. Reproduction or editing by any means, mechanical or electronic, in whole or in part, without the express written permission of William J. Bernstein is strictly prohibited. Please read the disclaimer.