Using Shiller’s CAPE Ratio as a System Filter
It isn’t common for ideas from value investors to make appearances on system trading sites. Many traders believe that the indicators and strategies that belong to these schools of thought are mutually exclusive. That doesn’t have to be the case, though.
Before my evolution to system trading, I spent a number of years studying William O’Neil’s CANSLIM approach to discretionary trading. In a nutshell, the CANSLIM strategy uses earnings based fundamentals to screen for top stocks and then uses technical signals for entry and exit points. I believe that this can be a profitable approach for the right trader, but I struggled with the discretionary decision making part of the strategy.
Despite the fact that CANSLIM didn’t work well for me, I have always been attracted to the idea of combining fundamental and technical data in a manner similar to what O’Neil suggests. Fast forward to this week, when I listened to Michael Covel’s second podcast interview with Mebane Faber (who is surprisingly not the old British guy that I assumed he was).
Faber’s CAPE Asset Allocation
Faber spent a good portion of the interview discussing his use of Shiller’s CAPE Ratio to give a PE Ratio type of value to more than 30 different countries and then invest in the least expensive countries and avoid the most expensive countries. While Faber was discussing this from a long term value perspective, there is certainly a quantitative aspect to what he is doing, along with lots of numbers.
According to Faber’s blog, had you invested exclusively in the three most expensive countries according to their CAPE ratios from 1999 through 2009, you would have lost 40%. However, if you had invested exclusively in the three least expensive countries, you would have made 24%. This strategy would have had you buying countries like Thailand, New Zealand, and Chile in 1999. Faber also points out that 1999 was a terrible time to start a ten year investment strategy because of the bubbles that occurred in 2001 and 2008.
According to Faber’s blog, these are the ten cheapest CAPE countries as of March 31, 2013:
Faber’s most expensive CAPE countries as of March 31, 2013 are:
A More Systematic Alternative
I like the general concept behind Faber’s approach. Investing in countries like Greece, Argentina, Italy, and Spain is clearly a good value play at this time. The hangup for me is investing in these countries as part of a long term buy and hold portfolio. This would give us no option or plan to get out of a position if the bottom were to fall out of one of the countries.
Drawing on my CANSLIM experience, it occurred to me that I could use this CAPE approach as a filter for a trading system. This would combine the CAPE fundamental data with technical entry and exit points. Rather than use the CAPE rankings for asset allocation, I could use them to make sure that the odds were stacked in my favor.
My system would have to be fairly simple and have a long term outlook, so The 50 Unit EMA System or the 10/100 Moving Average Crossover System would work. The idea would be to trade one of those systems using the approximately 30 countries that Faber is tracking, then only taking the entry signals that are generated by countries in the bottom half of the CAPE rankings. (We could adjust this filter to any percentage based on backtesting results.)
Combining this value based filter with a trend following system could possibly give us the best of both worlds. We would be exclusively trading countries that are both undervalued and trending up, or overvalued and trending down. While I believe the upside of this type of system could be significant, I am still concerned about the level of risk it would be taking on by trading poor quality countries. It would take significant backtesting information to convince me that this risk was acceptable.
Sector Based CAPE Strategy
Faber has also discussed using his CAPE approach to place values on sectors. This could be done in exactly the same manner as the country-based approach. At the end of March, Faber posted the sector rankings based on CAPE values. The top five sectors were Financials, Energy, Utilities, Telecom, and Materials. The bottom five sectors were Industrials, Healthcare, Consumer Staples, Technology, and Consumer Discretionary.
We could use a simple system with a CAPE style filter for trading these sectors just as we discussed earlier with different countries. This would provide us with a system that was focused on trading undervalued sectors in uptrends or overvalued sectors in downtrends.
About Shiller’s CAPE Ratio
CAPE stands for Cyclically Adjusted Price-to-Earnings Ratio. The concept was developed by Yale professor Robert Shiller. It is based on the argument made by Benjamin Graham and David Dodd that earnings should be smoothed over a ten year period because year-to-year volatility was too high.
CAPE is calculated by dividing the current price by the average earnings over the past ten years, adjusted for inflation.