Ben Graham Trading Rules Backtest

Our recent Ben Graham post generated a lot of interest and a LOT of emails.

One of the more interesting questions asked was, “How does the Ben Graham criteria work as a trading rule?”

In order to explore this question we went ahead and examined the relationship between the Shiller 10-yr CAPE, the # of stocks fitting the Ben Graham Criteria of <10 P/E and >.5 (TA-TL)/TA, and 10-year real returns.

Next, we analyzed what would happen if you attempted to build simple exposure rules using the Shiller CAPE, or the # of Ben Graham stocks in the market. What follows is a quick and dirty analysis of our findings.

Here are the results (all based on a monthly-rolling basis):
Click to get the “Trading Rule Analysis” pdf file.

Key Takeaways:

1. The Shiller CAPE seems to do a good job predicting real returns for the next 10-years (when ratio is high, returns are low, and vice versa).
2. The # of Ben Graham stocks in the market also seems to do a good job predicting 10-year returns (when there are 30+ names fitting the criteria, returns are generally good, when there are very few bargain stocks, returns are poor)

–There is no doubt that using the Shiller CAPE or looking at the number of “cheap stocks” in the universe do a decent job predicting long-term market returns (10-year, at least). Lesson learned: valuations matter for expected returns–duh.

…however, we next examined some naive trading rules to try and see if there are any interesting trading rules to use from this data.

1.) A trading rule that shifts in 50% cash when the CAPE is hitting 75 percentile or when there are less than 15 Ben Graham stocks in the universe, doesn’t exactly help your ultra-long term buy-and-hold returns over the 1965-2010 period.

2.) We also tested more extreme trading rules–shift 100% into cash if CAPE is in 95 + percentile or there are less than 5 Ben Graham stocks. This strategy keeps you protected during the 1987 debacle, and the 1995-2002 bull/bear debacle, but it doesn’t get you out of the 2008 crash–d’oh!

The results are hypothetical results and are NOT an indicator of future results and do NOT represent returns that any investor actually attained. Indexes are unmanaged, do not reflect management or trading fees, and one cannot invest directly in an index. Additional information regarding the construction of these results is available upon request.

Conclusions?

We have examined very simple trading rules, that are admittedly too simplistic. If someone were to put their thinking cap on, I am certain there are some interesting trading/exposure rules one could devise using various market valuation methods such as CAPE or # of Ben Graham stocks.

Good luck hunting for new ideas.