Search This Blog

Wednesday, April 29, 2009

Fundamental Indexing: A Better Mousetrap ?

Forbes recently reviewed the performance of Robert Arnott's fundamentally weighted index as well as those of some others (hArnott's firm's website is here. As the article notes the premise behind the fundamental index is that the more traditional cap weighted indexing places too great a weight on stocks with high valuations, leaving it exposed to what many have dubbed "bubble" risk. There is some merit to that argument. However, as many have noted, these "new" indexing techniques tend to look fairly similar to value weighted and cap weighted indexes designed to increase the weighting of more value and smaller capitalization stocks. This methodology is nothing particularly new, Professors Eugene Fama and Kenneth French have written about this methodology for decades and have pursued an indexing investment strategy based on this with the fund management firm Dimensional Funds Advisors.

And as we have noted in earlier posts, value weighted indices, particularly those in the large cap area have suffered greatly as of late. In the current environment the methodologies used by all of these various indices have generated large weightings for financials.

When the various quantitative screens for these various indexers use data reported in financial statements as their inputs they generate similar results and cannot take into account any analysis of those numbers. Thus outlooks for writedowns on book value, restatements of earnings, or dividend cuts cannot be incorporated into the index. While this approach may have the virtue of avoiding any subjectivity, under the extreme distress of the financial services industry in the current crisis, it has certainly hurt performance. Another variable that seems not to have been taken into account is any kind of limit on industry concentration for the funds. Those that formed these indices might argue that industry is not a relevant factor that determines returns and therefore one need not worry about over concentration in a particular industry. I find much fault with that assertion, certainly on a risk management basis.

Arnott had the unfortunate luck to unveil his fundamentally weighted indices in precisely the market environment least congenial to his methodology. You can follow the links to the forbes article with the performance numbers.

From the Forbes article (my bolds, my comments in italics)

Money & Investing
Can You Out-Index The S&P
Evan Hessel 04.27.09, 12:00 AM ET

As the former global equity strategist for Salomon Brothers, Robert Arnott built a reputation for knowing his way around the stock market. So it didn't go unnoticed when five years ago he boldly proclaimed he'd come up with a formula to beat index funds over the long haul.

The crux of Arnott's argument is that traditional index funds are flawed because the basis for determining how much of each individual stock is held in them is its total stock market value. When technology was on a tear, indexes such as the S&P 500 and Nasdaq were left loaded with the likes of Cisco Systems and, which promptly crashed. When financials got hot a few years later, indexes emphasizing large-capitalization companies had heavy weightings of Lehman Brothers and Citigroup.

Certainly, Arnott figured, there must be a better way to populate indexes. He crunched the numbers at Research Affiliates, his Newport Beach, Calif. advisory firm, and decided the best basis for filling indexes is not a stock's total market value but what he calls its ''economic weight.'' That he defined as a secret-sauce algorithm comprising revenue, cash flow from operations, dividends and book value.

The method fuses passive indexing with a value tilt. High-flying stocks with tiny earnings, but huge earnings multiples, get light weightings.

Arnott back-tested his index with 42 years of data and proclaimed he'd devised a way to beat the S&P 500 by 2 percentage points a year over the long haul. His pitch enticed Charles Schwab, Pimco, PowerShares and Nomura to license his indexes for a bevy of products that go by the appealing sounding moniker ''fundamental index funds.'' Investors have entrusted $18 billion to them.

The results to date: Yet more evidence that your best bet is still a plain vanilla index fund, like the benchmark S&P 500 one launched by Vanguard founder John Bogle in 1976 and essentially unchanged since. Such funds won't shield you from bear market losses. But with minimal fees, tax-efficient structures and the collective wisdom of the market baked in, marketcap- weighted indexes are still the performance champs over the long haul.

In the three years since its launch Arnott's Research Affiliates Fundamental Index 1000 has lagged the Vanguard 500 Index Fund and the Russell 1000 funds by 1.9 points each. His Rafi 1000 for Europe and Asia has bested the benchmark MSCI Europe Australia & Far East Index by 0.6 percentage points. Over the past year Schwab's Fundamental U.S. Large Company Index Fund is down 42.6% after fees of 0.5%. Vanguard's 500 Index Fund costs a quarter as much and is down 38.1%.

Arnott has some prestigious company in his quest to best traditional indexes. Wharton finance professor Jeremy Siegel has built (and marketed) similar products at WisdomTree Financial with similarly mixed results. WisdomTree's LargeCap Dividend Fund ETF has lagged the S&P 500 by 2 percentage points a year since its June 2006 inception. Its Earnings 500 Fund has trailed the S&P 500 by 0.2 percentage points annually since its February 2007 inception.

In tacit recognition that the dividend methodology can generate a very high weighting in financials, Wisdomtree has announced plans for a dividend weighted ex financials index etf.

Not surprisingly, given that they are attacking a bedrock claim of investment theory (that no mechanical formula can beat the market), the vendors of fundamental indexes have got a skeptical reception in academia. Harvard Business School's Andre Perold joked that Arnott's strategy is ''fundamentally flawed.'' Burton Malkiel, a Princeton professor known for his support of the efficient markets theory, suggested Arnott was just another stock picker in drag.


Malkiel and Bogle attacked Siegel's approach on the Wall Street Journal's editorial page in 2006. They argued that Siegel's claim of superior performance for a small-cap value stock index over five years would be disproved over the longer term. Siegel countered with a column claiming that value-oriented funds have beaten cap-weighted funds since 1925, the earliest date for which comprehensive stock data are available. So far the skeptics are winning the battle over fundamental indexing, but three years is, to be sure, too short a time to judge a system that is supposed to prove itself over a period of decades.

Arnott is sticking to his guns and attributes his underperformance to lousy timing. Research Affiliates rebalances its portfolios each March, selling stocks its models indicate are overweighted relative to their economic value and replacing them with cheaper ones. Stocks whose weightings were increased in March 2008: American International Group, Wachovia, Washington Mutual, Citigroup and Bank of America.

It may be "lousy timing" that hurt Arnott's indices but in my view it is the lousey timing resulting from the fact that his methodology is heavily depended on those financial ratios that have been least reliable when reported by financial institutions. The weighting on the companies listed above went up because the data used was so unreliable yet the (lower) market price reflected skepticism with regard to earnings, book value and future dividends. As efficient market and market cap advocates like Malkiel would point out, the market price reflected the prospective outlook for the companies. On the other hand the fundamental indexing formula saw the stocks as "underpriced" because it was using historical financial data and the prevailing market price.

''Investing against the market hurts on the way down,'' Arnott says. ''But there is always potential for huge gains when it reverts to the mean.''

Talking his book, Arnott argues that financials are undervalued now. If you don't count (presumably nonrecurring) writedowns of assets, then banks and insurers represent one-quarter of the stock market's earnings but only 10.3% of its value.

The above statement illustrates my observation: "If you don't count writedowns of assets" and assume no more will occur than financials are undervalued. Of course if you anticipate more writedowns or that financials will have to raise more equity in moves which could dilute current shareholder value,then the stocks are either fairly or overvalued. In other words Arnott thinks that "Mr.Market is "wrong" in its current valuation of financials.....

Arnott set up Research Affiliates in 2003. His strategy is influenced by two acclaimed finance gurus, Dartmouth's Kenneth French and the University of Chicago's Eugene Fama. Both make an exception to the efficient market theory for value stocks. These, they think, offer the best long-term returns.

Offering new fodder to critics who believe the only real value of trying to out-guess the market is its marketing appeal, Arnott has also created a fundamental index on steroids. Dubbed Enhanced Rafi, it aims to give investors a first peek at which companies' earnings may rise and fall by measuring ''net operating assets''--an amalgam of easily inflated accruals such as intangibles, goodwill, accounts receivable and capitalized investment as a percentage of total assets.

The theory is that companies with outsize net operating assets are more likely to have booked dubious accounting gains that could eventually be written down. Arnott's team admits the analysis penalizes firms in industries with high capital expenditure requirements and says it only underweights firms with the stock market's highest ratios. By this measure, General Electric deserves underweighting because its net operating assets represent 60% of its total assets.

(another indexing methodology that works except when it doesn't...we'll see)

Arnott is nothing if not ambitious and prolific. He is working on an alternative to prevailing indexing methodology in the fixed income arena as well, a variant on "fundamental indexing". More on that in the future.

Pimco has used Arnott's recent creation to roll out the Pimco Fundamental IndexPlus Total Return fund. It combines Enhanced Rafi with an actively managed bond portfolio in a bid to beat the S&P 500. The fund has failed miserably over the past 12 months, falling 48% after its hefty 1.7% in fees. Arnott blames a poorly performing bond portfolio. Vanguard's low-cost S&P 500 fund fell 38%.

No comments: