Sector ETF Rankings: Weekly Update for May 8, 2011

by: Diffusion

Sector ETF Rankings

This section is included for completeness. Readers may want to skip to the next section if they are familiar with our methodology article “ETF Ranking: A New Fundamental Approach That Drives Short-Term Return."

The ETF ranking is an extension of our newly designed stock ranking system that ranks every stock based on its valuation, financial condition and return on capital. Although the ranking system is fundamental based, it actually drives short term return. We observed that stocks with higher ranks had a strong tendency to outperform those with lower ranks over a period of one week. The data show that moving up 10 rank points translates to an extra annualized return of 1.7% in the past 10 years, if ranks range from 0 to 100. As a mater of fact, the S&P 500 Index returned an annualized 2.5% in the past 10 years.

Extending our fundamental ranking system to ETFs is straightforward. The rank of an ETF is the weighted average over the ranks of stocks in its portfolio. Although we don’t have historical data for ETF ranking, we believe it is effective because the ranking on individual stocks is effective.

ETFs are often better choices than individual stocks for investors and traders. An ETF is naturally diversified, thus it reduces volatility. Popular ETFs such as SPDR sector ETFs enjoy abundant liquidity, which also minimizes bid / ask spread. Sector ETFs are ideal trading vehicles.

Of the nine SPDR sector ETFs, three are defensive (XLP, XLU and XLV). The rest are offensive. A key difference traders want to be aware of is the beta. Offensive stocks or ETFs typically have higher beta and move faster than the market. Defensive ones are on the opposite side. In a bull market, traders like to chase offensive ETFs because they rise faster than the market. In a bear market high beta ETFs are supposed to fall faster. Traders can still be offensive by shorting them.

Action Plan for the Week

We list this week’s ranks of offensive sector ETFs in the table below, together with last week’s ranks.


Rank as of 5/7/11

Rank as of 4/30/11



















The rank of XLE fell and that of XLB rose in the past week. As a result, XLB now has the highest rank.

Price change affects valuation, thus rank changes when price changes. Besides, rank changes when new fundamental numbers enter the ranking system, which is normal in an earnings season. Last week the price of XLE saw s larger drop than that of XLB. All else being equal, XLE’s valuation should be better so its rank should rise, contrary to what we have seen last week. This is alarming because it indicates that we were seeing bad fundamental numbers in the energy sector during earnings season. In our opinion, XLE’s big drop is due to the fundamental deterioration, not to the plunge of crude oil dubbed by the media.

If a trader holds a long position of XLE and short position of XLF as we recommended last week, the expected return is proportional to the spread between the ranks of XLE and XLF. The spread is at 67 - 21 = 46, one point better than that of the week before, which was at 69 - 24 = 45. Spread wise, the long XLE / short XLF thesis is still valid. Nonetheless we would like to caution that the spread may shrink if the fundamentals further deteriorate. Therefore, we will switch from XLE to XLB in the coming week.

Depending on a trader’s market view, we recommend three trading plans:

  • Bullish: Long XLB, the one with the highest rank. Although the ranking system is designed for one week return, traders may hold it for four to six weeks because the ranks of ETFs change slowly.
  • Bearish: Short XLF, the one with the lowest rank.
  • Neutral: Long XLB and short XLF for the same dollar amount.

We pick the last one mainly due to our neutral market view. As mentioned before, spread is an important factor for this plan. At 68 - 21 = 47, the spread is fairly large. Because 10 rank points translated to 1.7% annualized return, our expected annualized return is 1.7% x 47 / 10 = 8.0%. It is noteworthy that this number is derived from the past 10 years. The S&P 500 returned merely an annualized 2.5% in that period. An 8.0% return is really more than three times the market return. That said, historical returns do not guarantee future performance.

It is healthy that the spread between XLB and XLF expanded compared to that of the week before, which was at 65 - 24 = 41. Because the spread is proportional to the expected return, any shrink would be a warning sign.

APPENDIX: Between Fundamentals and Short Term Returns

Graham’s famous saying on voting and weighing machines is well perceived and helps divide market players into two camps: Traders who pursue short term returns and value investors who patiently wait for fundamentals to play out in the long run. Putting fundamentals together with short term returns may sound questionable considering the enormous contrast between the two camps.

We think the market is governed by “fact” in the long term and by “behavior” in the short term. Popularity is an important character of social behavior. We believe that combining popularity with fundamentals may help us reach a delicate compromise between traders and value investors. In our ranking system we choose trailing-12-month P/E to represent valuation because it is entrenched in the mind of almost every market player and it is quoted in almost every popular financial website. Value investors may disagree and use instead the Shiller P/E, the average P/E over past 10 years. It has merits, but it is not popular, thus is less likely to drive short term return.

After careful evaluation, we selected three components in our fundamental ranking system: valuation, financial condition (to manage risk), and return on capital (for better returns on investment).

Please read our methodology article for more details.

Disclosure: I am long XLB. I'm also short XLF.