The Style Roulette and RAFI™ Strategy

By John West

JULY 2010 Read Time: 10 min

The first half of 2010 has been a roller coaster ride in global equity markets. The S&P 500 Index and MSCI All Country World Index posted gains of 5.4% and 3.2%, respectively, in the first quarter. But, as we’ve been suggesting for some months, the consequences of a global addiction to debt-financed consumption—sovereign, corporate, and household—started to take their toll in the second quarter, with the S&P 500 declining 11.4% and the MSCI All Country World falling 12.0% in U.S. dollar terms. Adding to this ride, value and growth styles have been flip-flopping in past years. If the current narrow value outperformance experienced in the first half of the year holds, 2010 will mark the fifth straight calendar year in which style leadership has shifted between growth and value.

This growth–value whipsaw is nothing new and begs several questions. Is there a better way to play the style game than splitting our equities evenly between growth and value? How reliable is the value premium? If we were able to magically win repeatedly in “style roulette,” what would be the rewards over a buy-and-hold index? Where does the Fundamental Index™ approach, with its inherent contra-trading against the market’s most extreme bets, fit within all of this? In this issue we explore some of these facets of equity style investing.

The Value Premium
The size and value premiums have been well documented in the literature, culminating in Fama and French’s highly regarded work in the early 1990s.1 Table 1 provides empirical evidence on the size and reliability of the value premium relative to the broad market through May 2010. As Table 1 shows, excess returns for value investors ranged between 0.6% and 1.8% per annum, depending on the market. However, the variability in these results—that is, the tracking errors for value investing relative to the broad market—are substantial. Clearly, every so often a value approach will substantially underperform the capitalization-weighted broad market.

Of course, most U.S. investors experienced this first-hand during the tech bubble and collapse in the late 1990s and first part of the “Naughties.” It was a boom or bust period depending on which side of the style fence our portfolio sat (and significantly contributed to the large tracking error figures above). To prevent being blindsided again by another volatile value–growth cycle, U.S. investors attempted to diversify their largest bet—domestic equities—employing both growth and value approaches. This strategy, of course, reduces the equity portfolio’s variability to the broad cap-weighted market, but also gives up the value premium. So, what other choices are there for investors to win the style roulette game?

Winning the Style Roulette Game
Suppose we could perfectly time exposure to the winning equity style. If value wins in the year ahead, we’ll be in value and if growth wins, we’ll be in growth. To quantify a perfect run at the “style roulette wheel,” we built portfolios that were always on the winning side of the style bet each year.

Not surprisingly, the perfect foresight value–growth strategy leads to wonderful excess returns for all four strategies tested. Table 2 displays the premium such perfect timing portfolios would have produced. In U.S. large company equities, placing 100% of our funds in the winning style would have produced an annualized return of 16.5% for the 1979–2009 period—500 bps ahead of the cap-weighted broad market! Substantial excess returns also were available in global and non-U.S. strategies, measured in somewhat shorter time frames. These are impressive figures but, like a run of 20 straight “reds” at Monte Carlo, unattainable to us mere mortals.

So, what is the winning style roulette strategy? Use the RAFI methodology to contra-trade against whichever style tilt is most currently in vogue.

How can something that’s always got a value tilt beat an approach that always has the winning growth or value tilt? The answer: By having a deep value tilt when value is really cheap, and sharply trimming the value tilt when it’s more fully priced. Consider Figure 1, which graphically explores this contra-trading by style. Here, we compare the two-year excess return of value or growth with a Value Tilt index. For the Value Tilt index, a value of “1” indicates a value tilt approximately equal to the All Country World Value Index, while a “0” indicates a neutral style bias approximating a broad cap-weighted All Country World Index.

What’s this worth? In the United States, the Fundamental Index approach produces an excess annualized return of 220 bps, equivalent to 43% of the perfect style roulette strategy, for the 1979–2009 period studied (see Table 2). This result is nearly half of a clairvoyant, and unattainable, style timer! Not bad, but look at what happens to the non-U.S., global, and emerging markets strategies. Remarkably, the RAFI methodology captures more than 100% of a perfect style timing strategy! Of course, there is more to the Fundamental Index methodology than systematic contra-trading between styles; dynamic exposures to size (small versus large), economic sectors, countries and regions also are additive to portfolio returns. 

The intuition behind why the Fundamental Index approach is so successful is simple. Remember what happens in the cap-weighted index when one style outperforms meaningfully—it comprises more of the index! Thus, when growth was taking off globally in the late 1990s, the weights for the tech and telecom winners increased in the cap-weighted index. The comparable weights in a Fundamental Index portfolio, paying no attention to rising valuations, were pared back at the annual rebalance point to be in line with each company’s fundamental size. The resulting Fundamental Index portfolio looked more and more like a deep value index (note the yellow line briefly goes above “1” during this time period).

Meanwhile, when value goes on an extended winning streak (as seen in the early 1990s and mid-2000s), the Fundamental Index portfolio lightens its value orientation. How? The same way it extended it in the late 1990s— it trims its recent winners back to their fundamental size! (The yellow line troughs in the 0.2 range). The changing nature of the RAFI value exposure—resembling a deep value portfolio after strong growth runs and a mild value orientation after value outperformance—certainly appears to add value to portfolio returns when mean reversion occurs.

Legendary gambler Nick “the Greek” Dandalos once said: “Remember this: The house doesn’t beat the player. It just gives him the opportunity to beat himself.” The style merry-go-round similarly provides ample opportunity for investors to be their own worst enemy, even within the supposed balanced broad market indexes. Make no mistake—cap-weighted index funds are stealthy returns chasers loading up on past winners.

The Fundamental Index concept contra-trades against the market’s recent popularity contest winners. This approach clearly means that we sell whatever is newly most beloved and buy what is newly most loathed, which can hardly be easy or comfortable. This simple periodic realignment back to financial size remarkably captures over 100% of a perfect style timing strategy in three major non-U.S. equity asset classes. So, rather than play the roulette wheel at our own risk, we’ll take our chances at the rebalancing table.

Featured Tags

Learn More About the Author

Partner, Head of Client Strategies


1. Eugene F. Fama and Kenneth R. French. 1992. “The Cross-Section of Expected Stock Returns.”Journal of Finance, vol. 47, no. 2 (June):427-465.