RAFI Strategy and Quality

By Rob Arnott

SEPTEMBER 2009 Read Time: 10 min

Despite moderately high valuations, the equity markets roll on in the mother of all recoveries. The six months ending August 31, 2009, were the best six months of performance for the S&P 500 Index since 1975, a mirror image to the previous six months, the worst since 1932. Within equities, the Fundamental Index™ methodology continues to add substantial value with almost all FTSE RAFI® applications worldwide; nearly 98% of the apples-to-apples comparisons have exceeded the returns for the corresponding capitalization-weighted indexes on a year-to-date basis through August 2009. The breadth of Fundamental Index outperformance was exceeded only by the magnitude. For example, year-to-date the FTSE RAFI All World 3000 Index achieved an excess return of 15.3% over the MSCI All Country World Index.

Are the FTSE RAFI portfolios adding value or is capitalization weighting underperforming? The answer depends on the frame of reference. From the market vantage point, RAFI™ portfolios are active, value-tilted portfolios, so they are delivering an impressive positive alpha this year, outpacing all previous years except 2000. But, there’s another frame of reference: the economy. From the vantage point of the broad sweep of companies that comprise the broad publicly-traded economy, the market is making constantly-changing active bets on which companies have strong growth prospects deserving a premium multiple and that face challenges worthy of a deep valuation discount. From this economic frame of reference, the cap-weighted market is an active growth-tilted portfolio with the RAFI approach contratrading against these constantly changing market bets.

The excess returns for 2009 have been largely driven by the annual rebalance of the FTSE RAFI series in March when the Fundamental Index portfolios added to stocks whose share prices had fallen substantially more than their underlying economic fundamentals. Invariably, these net purchases occurred in the most distressed areas of the equity markets—financials and consumer discretionary stocks. True to mean reversion, these left-for-dead companies have led the market comeback. A recent Barron’s article indicated the “…junk rally, in which financially dodgy companies that were granted a reprieve by healthier credit markets have seen their stocks double and triple and price.”1

On the heels of RAFI outperformance in a low quality rally, some question whether the Fundamental Index approach is simply a permanent bet on “junky,” lower quality companies. Our research indicates this is not the case. Rather, the RAFI methodology offers a mirror image of the cap-weighted market’s bets relative to the economy. Whatever the market is bidding up—faster than its economic footprint is growing—we sell. Whatever the market is punishing with lower valuation multiples, we buy. At any snapshot in time, the RAFI portfolio looks like a value-tilted active portfolio relative to the market; but it’s incremental return comes largely from this contratrading against the markets most extreme bets and constantly shifting expectations, rather than from the value tilt per se.

Recent Quality Performance
Although there is no definitive industry-wide classification system, quality stocks are generally associated with larger companies that have stable earnings, stable dividends, and low debt. On the other hand, low quality stocks are associated with companies that have more unpredictable earnings and higher debt. With this unreliability, lower quality stocks tend to trade at lower price-per-share levels. Thus, share price is a simple proxy for quality. By this admittedly crude measure, the past five months was unquestionably a low-quality rally as evidenced in Figure 1. We’ve whimsically referred to it as “garbage floating to the surface.”

Stocks in the Russell 1000 Index priced below $5 per share (as of March 31, 2009) surged over 116% in the subsequent five months. Meanwhile, stocks priced above $50 per share—a loose proxy for higher quality—managed to post a more pedestrian gain of 22% over this period. Because these stocks were priced so low, the cap-weighted Russell 1000 only had an allocation of 1.5% as of March 31 while the FTSE RAFI US Large Company Index, fresh off of its March rebalance, held 11.2% in stocks priced under $5—many of them familiar blue chips, such as Citigroup and Ford Motor, that have since fallen on hard times.

Quality over the Long Term
The RAFI approach has unquestionably benefited from the massive 2009 rebound in low-priced and distressed stocks. But, is this a byproduct of recent markets or does the RAFI methodology have a structural bias to low quality stocks? To better ascertain this issue, we turn to Standard and Poor’s Quality Rankings.2 The rankings “…attempt to capture the growth and stability of earnings and dividends record in a single symbol.” The single symbol was a letter grade from A+ (highest quality) to C and D (lowest qualities). In the analysis, Standard and Poor’s produced calendar year returns for all the rankings from 1986 to 2004. What do we find over these 19 years?

In Table 1, “Quality” years are those where all A-rated stocks outperformed all stocks ranked B, C, and D on a market-value-weighted basis. On the flipside, “Junk” years are lower quality issues outperforming A-rated stocks. We find the RAFI methodology’s win rate and annualized excess returns are not much different for “Quality” versus “Junk” years. In the 11 years when quality outperformed junk, the RAFI methodology produced an average annualized excess return of 2.5%. The RAFI methodology also outperformed in junk years, albeit by a somewhat smaller (!) margin of 1.8% per annum. In both environments, the RAFI methodology won between 55–63% of the time (in an admittedly small sample).

Over the entire period 1986–2004, the FTSE RAFI US Large Company finished ahead of the S&P 500 by 220 bps per annum, which is almost identical to the simulated results for the longer period January 1962–June 2009 of 210 bps per year. Further, high quality stocks actually outperformed low quality stocks by 100 bps per annum as all A-rated stocks achieved an annual return of 13% versus 12% for stocks rated B through D. Based on this data, it is hard to say that the RAFI methodology has a systematic tilt to low quality companies. Indeed, the correlation of excess returns for high quality over low quality and excess returns for the RAFI approach over the S&P 500 was positive at 0.16.

The Fundamental Index approach has recently benefited from larger exposures to the distressed companies that have led this market higher. These were still large companies despite having next-to-nothing in market capitalization! But does that mean the Fundamental Index approach will always have a structural bias to low quality? We think not. The key to RAFI methodology outperformance is contratrading out of what has risen most in price relative to fundamentals and into the recently downtrodden—out of the beloved and into the loathed. Capitalization weighting naturally does the opposite—riding recent winners to higher and higher allocations and losers to lower and lower weights. This can happen between growth and value, amongst sectors, between countries and regions, and, based upon this brief overview, between quality subsets of the equity markets.

Does the RAFI strategy have a value tilt? Of course. Does it have a small-cap bias? A little bit most of the time, but it’s a reciprocal of the cap-weighted market’s bias toward small growth companies. Does it have a low-quality bias? Perhaps somewhat, but it becomes pronounced only when the market is punishing lower-quality companies with Armageddon pricing. Isn’t that the right time to load up on them?!

Featured Tags

How to Invest

Select filters Strategies, Asset Classes, Regions


Asset Classes


Learn More About the Author

Chairman & Chief Executive Officer

1. Santoli, Michael. (2009). “Big Stocks May Be Ready for Their Close-Up,” Barron’s (September 21).
2. Standard and Poor’s Quality Rankings–Portfolio Performance, Risk, and Fundamental Analysis, October 2005.