THERE'S A BATTLE

brewing in the indexing industry, and it's being fought on the fields of ETFs.

Historically, index funds beat most actively-managed funds, because, as passive investment vehicles, there's little portfolio turnover. That translates into lower transaction costs and eliminates the need to pay a high-priced active fund manager, keeping expense ratios substantially lower than active funds. ETFs, meanwhile, post even smaller expense ratios than comparable mutual funds and with greater tax efficiency.

However, because indexes track the market or slices of the market, there can only be so many. That's causing problems. In order to capitalize on the growing demand for ETFs, fund issuers and index creators end up slicing the market into narrower and narrower segments. For instance, iShares offers eight different ETFs for the financial-services sector alone.

How to improve on indexing is at the heart of the battle. On the one hand is the status quo, with adherents saying that indexes should be weighted by market capitalization. Market-cap weighting multiplies a stock's price by the total number of shares outstanding, or its float, the actual number of shares that trade. This total market value then gives the stock its proportional weight in the index. "Indexes should be cap-weighted because the market itself is cap-weighted," says David Blitzer, managing director and chairman for the Standard & Poor's index committee.

But others espouse a new theory basing indexes on fundamental metrics, like dividends or earnings. Recently, in an opinion piece in The Wall Street Journal, Jeremy Siegel argued that a security's price is not always the best estimate of a company's true underlying value. Stock prices often move on factors unrelated to fundamentals, or market "noise," such as speculators, momentum traders, insider buying or just enthusiastic investors chasing returns by piling into stocks that have already rallied strongly. Such factors could make a stock overpriced to its fair value, which would effectively overweight it in the index. Meanwhile stocks underpriced to their fair value would be underweighted in the index.

Because of the passive nature of index investing, as new money enters the fund more would be allocated to overvalued stocks and less to undervalued stocks, which runs counterintuitive to common-sense investing, he argued. Instead, indexing based on fundamentals, he said, was a better approach. (For Siegel, it wasn't an academic exercise: He wrote the piece to coincide with the launch of WisdomTree Investments' 20 ETFs based strictly on the fundamental metric of dividends. The professor of finance at the Wharton School of the University of Pennsylvania is a strategy advisor to the firm.)

Fundamental indexing says weighting a stock by a fundamental metric, such as sales or dividends, delivers significantly better returns than cap-weighted indexes by removing market noise. Leading the charge is Rob Arnott, chairman of Research Affiliates, a research-intensive asset-management firm based in Pasadena, Calif. As editor of the Financial Analysts Journal, Arnott published research showing how market-cap weighted indexes are inefficient compared with what he calls fundamental indexes. Arnott's Research Affiliates Fundamental Index, or RAFI, measures companies by four metrics: book value, sales, gross dividends and cash flow.

"I've never seen an idea gain so much traction as early as this one has," says Arnott. "But it's hitting a nerve because it addresses a concern people have had for not just years, but decades. Ever since the S&P 500 launched in 1957, people have expressed concerns that it's inherently designed to overweight the overvalued and underweight the undervalued, even though we can't know which stocks are overvalued."

owned by Dow Jones, which, along with Hearst Corp., is a co-owner of SmartMoney.com) published a sharp rebuttal from two of the biggest names in indexing, Princeton University professor Burton Malkiel and John Bogle, founder of the Vanguard Group and the first index fund. The pair wrote that, even in inefficient markets, cap-weighted indexing remains the optimal investment strategy primarily because ETFs based on fundamental indexing charge management fees well above the typical index fund. "While index funds also incur expenses, they are available at costs below 10 basis points," they wrote. "The expense ratios of publicly available fundamental index funds range from an average of 0.49% (plus brokerage commissions) to 1.14% (plus a 3.75% sales load), plus an undisclosed amount of portfolio turnover costs."

Bogle and Malkiel argued that fundamental indexes experience more portfolio turnover than typical index funds, incurring higher transaction costs and capital gains. And finally, they dismiss the outperformance of fundamental indexing over the past five years compared with market-cap weighted indexes to fundamental's propensity to give greater weighting to value stocks and companies with smaller capitalizations. It just so happens that those two sectors have been in favor over the past five years.

"The people behind fundamental indexes claim that they can beat the market, and essentially they are saying they are smarter than the market," says S&P's Blitzer. "But it's not a free ride. The further you move away from cap-weighting, the more risk you take on."

Arnott agrees that ETFs based on fundamental indexing have higher expense ratios than many market-cap weighted ETFs. But the performance makes that moot. "Would you pay 0.15% for zero added value or 0.6% for 2% to 3% of additional returns?" he says. "Most investors would cheerfully choose the latter."

Who's right? Well, the jury is still out. Independent research, though, appears to back Arnott and the fundamental indexers. Robert Schwob, chief executive of Style Research, a research and consulting firm in London, says he initially found Arnott's claims "unbelievable." So, his firm constructed portfolios based on earnings, dividends, sales or book value going back five, 10 and 20 years across markets in the U.S., the United Kingdom, Europe, Southeast Asia and Japan. The result? "We found that over the length of the horizon, fundamental indexing did outperform on an average of 2% to 2.5% per annum," says Schwob.

Tracking the RAFI is still in its infancy. Currently only one ETF follows the RAFI: The PowerShares FTSE RAFI U.S. 1000 Portfolio, which tracks 1,000 companies, was launched in December. So far, fundamental indexing is winning. Year-to-date through Tuesday, the RAFI-based index rose 6.3%, compared with the 3.3% gain on the SPDR Trust, which tracks the S&P 500. Meanwhile the iShares Russell 1000 Index fund is up just 3% year-to-date.

Expect more RAFI-based funds. PowerShares has registered with the Security and Exchange Commission 10 new ETFs based on the FTSE RAFI indexes. They're expected to launch before the end of this year.