With lower returns, costs matter: Malkiel

(April 23, 2004) Index-investing pioneer Burton Malkiel is projecting considerably lower returns on equities — in fact on all financial assets — something that makes costs matter all the more. But it also means disciplining investor expectations.

Malkiel, the Chemical Bank Professor at Princeton University, was speaking in Toronto yesterday to the annual conference of the U.S.-based National Association of Personal Financial Planners, an organization for fee-only planners. Since he first explored indexing in his book A Random Walk Down Wall Street in 1973, time has shown that indexing “works and it works very well.”

Recalling that book, he suggests he shouldn’t have said that, thanks to the randomness of stock prices, an investor could do as well as anyone else merely by throwing darts at the financial pages. “The right analogy is to throw a towel over the stock pages,” and buy the whole market.

In his original work on indexing, Malkiel argued for buying the whole market because it was broadly efficient. “Information about a stock is immediately reflected in the market,” he says. “Smart people have ensured that stocks reflect all that is known about them.”

As a result, few people consistently beat the market, though Malkiel admits, “If I had known 31 years ago that Warren Buffett was going to be Warren Buffett, then I would have said buy Berkshire Hathaway and forget the index funds.”

Malkiel uses another argument for those not still convinced that active management adds value. “All the stocks in the market have to be held by somebody,” he says. “For investors as a whole, investing has to be a zero-sum game. Investors as a whole are going to get the market return.”

Add in costs, and then most people will underperform the market, he argues. “It would still follow that an index fund with its very low expenses is likely to outperform the average active investor.”

Above-average returns

In his own research, index funds have beat large cap funds 90% of the time. Over one year, it’s 73%. The average outperformance is about 250 basis points (2.5%) which he attributes to fees. Thus, a U.S. index fund has a management fee of 18 basis points, compared to an active management fee of 150 basis points. In addition, actively managed funds typically turn over 100% of their portfolio in a year, which he figures accounts for another 100 basis points.

Given those costs, Malkiel says, “Index investing isn’t, as some critics say, guaranteeing mediocrity. It is likely to give you better than average returns. “

While some funds do manage to beat the market, he adds, there’s no persistence. What works well in one period can flag in the next. “If you were good last year,” he says, “There’s no guarantee you’ll be good this year.” About the only value statistics bring to investing, he suggests, is the predictability of management and turnover costs.

Therefore, instead of buying top-quartile funds to beat the market, he suggests buying those funds that are in the bottom quartile of expenses and turnover. Low-cost funds outperform high-cost funds by about 300 basis points.

Since costs matter, even for active investors, Malkiel maintains that the core of a portfolio should be indexed, with active management reserved for a few active managers or alternative assets.

Index flaws

Active managers have benefited from two flaws in index construction. While capitalization-weighted indexes like the S&P 500 are a more efficient proxy for the market than equal-weighted indexes like the Dow Jones Industrial Average, there is a danger in overweighting a given company. Thus, active managers were able to consistently beat the MSCI Europe, Australasia and Far East index by underweighting Japan. Thanks to a high degree of cross ownership, the Japan component effectively double counted many companies. Another concern was the difference between the market capitalization of a company and the actual number of shares that were freely traded.

All of the index makers have now adjusted market capitalization for the free float of shares.

A more difficult flaw to correct is the division of the market into sectors and styles. There’s a large back and forth between larger-cap and smaller-cap indexes; the Russell 2000, which measures U.S. small-cap companies, turns over 50% of its components a year.

Thus, says Malkiel, the indexing large caps, mid caps and small caps separately “is not the way to do it. You vitiate a lot of the benefits of indexing, and that’s no turnover.” Worse, when it comes to style index, a particular stock, such as JDS, can move back and forth between the value and style indexes. Instead, he says, “Buy everything and then you own both value and growth. I don’t think you should have an index for this and an index for that. It’s best to buy the whole market.”

“Even if you strongly disagree with me,” he says, “what I would strongly urge you to do is use a core-satellite approach.” The core of a portfolio should be indexed, while some active management can be used, “if you think you have a good small cap manager.

He particularly stresses low-cost indexing since equity returns are likely to be lower than their historical average. While the return on U.S. stocks from 1926 to 2003 has averaged 10.4% annually, he expects, based on current dividend yields, that stocks will only average 7.5%. He pegs at a range of 4% to 4.5% for long bonds and, he notes, they reached the outer limit of that range in the past two or three days as 10-year Treasury bills sold off.

Given single-digit returns, 1.5% management fees will reduce returns. The problem, he says, is that like returns, management fees also compound.

The next Buffett

As for other asset classes, Malkiel says hedge fund indexes suffer from a number of flaws that create an upward bias for returns. At the same time, venture capital returns are highly volatile. That leaves real estate, which has lately been overvalued, but could bring in a high single-digit return.

Those return expectations are important he says, because one of the fundamental tasks of a planner “is to set reasonable expectations about what returns are likely in the future. If you’re doing a plan for somebody and you’re saying equities have done 10% in the past, that’s the rate I’ll use, I think that’s a mistake. We are in a single-digit environment.”

In fact, that’s the kind of advice he would expect from his own advisor.

“There will be a Warren Buffett for the next 30 years,” he says. “But I don’t know who it is and you don’t know who it is.” Instead of picking the next Buffett, the planner should “give me really good tax advice, give me a really good retirement plan, and talk to me about rebalancing.”

Filed by Scot Blythe, Advisor.ca, scot.blythe@advisor.rogers.com.

(04/23/04)