How many times do you turn on your TV or radio these days, only to hear people yapping back and forth about current stock market valuations? One side says that stocks are cheap and that we should be mortgaging our house to buy them - the other says valuations are stretched to ridiculous levels and that the bubble is about to burst. The purpose of this three part series is to try and make sense out of all the noise - examining how "valuation levels" might be utilized (or ignored) in developing a long term investment strategy. Better put, let's try to determine whether or not we can use valuation metrics as a basis for adjusting our asset allocation models in hopes of achieving better long term, risk-adjusted returns....
Unfortunately, I have to share that while we spent several months doing research for this discussion, our compliance team ultimately determined that we could not share any of the specifics as it might violate SEC regulations against "back-testing" and "cherry-picking" in marketing materials. While I adamantly disagreed with their premise, it was a losing battle. So for better or worse, this series will contain much less numerical evidence than I would have preferred, and I will have to leave it to each of you to do your own research.....just don't publish it anywhere as you apparently will be in big trouble! While on the subject, I will throw out my first disclosure that NONE of what I am discussing here should be taken as ANY representation of expected returns or performance. Got it???
Onward and upward - this segment of our series will focus on a specific valuation metric developed by Dr. Robert Shiller (who ultimately won a Nobel Prize for his work on the subject) called the "Shiller PE/10 Ratio". For the sake of space and time, I will give a very brief description of the metric while encouraging you to read more about it on your own. Simply put, it is a Price to Earnings ratio that incorporates the past 10 years of earnings and adjusts them for inflation. The "simple PE Ratio" of a stock or index only uses the most current earnings data. For both, the higher the ratio the more "overvalued" a security or index is perceived to be and vice versa. There are many that believe the Shiller PE to be infinitely more valuable than the traditional measure - and some that doubt it's relevance. I will leave you to do your homework and form your own opinions....but we are believers.
So we decided to examine the usefulness of this metric when applied to a tactical allocation strategy. Our allocations used only stocks and bonds - stocks represented by the S&P 500 Index and bonds by using constant maturity, 10 year Treasury returns. We studied what would happen if we reduced equity exposure by a set percentage from our target allocation when Shiller PE valuations were extremely elevated (top quartile historically), then added equity exposure by a set percentage when valuations were abnormally depressed (bottom quartile) - keeping our target allocation in tact for any year in which valuations were not at "extreme" levels on either side. Our rebalancing occurred only once per year, based on the Shiller PE level of the S&P 500 on January 1st of each year.
We then compared the results against leaving the target allocation static regardless of valuation levels. The first analysis went back 50 years (1963) for no other reason than it was a nice round number. We then examined what would happen if we began the analysis immediately preceding a major market correction (1973). Finally, we tested it with a start date immediately preceding a multi-year run of above average stock returns (1993). The results were the same for each of the periods tested - both the annual returns and annual risk adjusted returns for the tactical models were meaningfully higher than those of the static allocation. At least in the studies we ran - using the Shiller PE as our metric of choice - valuation certainly seemed to matter!
To add some present-day relevance, the historical mean for the Shiller PE is approximately 16.5, with the median slightly below 16. The low point for the metric (1920) was a shade under 5, and the high point (1999) a touch above 44. We began the year 2000 ("bubble burst #1") close to 44, and began 2007 ("bursting bubble #2") just over 27. So, where do we sit today? As of August 21, 2014 the Shiller PE on the S&P 500 is 26.37. We will discuss this information and what it might tell us in part 3 of this series....
I would like to take a moment to again apologize for not being able to furnish the specifics of our research. We considered scrapping the whole rant, but ultimately I believe the relevance and importance of the topic enough to warrant at least a general discussion. Better yet, I would encourage anyone reading this to conduct their own research, selecting parameters that are most relevant to your personal situation. If you do, PLEASE reach out to us with the results - we would love to hear it!!
As you know by now, the opinions expressed in this rant are solely my own, are not intended to provide specific advice or recommendations for any individual, and do not necessarily reflect those of LPL Financial. Furthermore, investing involves risk - including potential loss of principal. I remind you that past returns (even those we are unable to discuss) are never a guarantee of future results. And last but not least, all indices discussed are unmanaged and may not be invested into directly.
Most important, we want to thank all of our clients for their continued trust and confidence in our team at Round Hill Wealth Management.
Thanks for listening,
Doug