Systematic Relative Strength

From Eddie Elfenbein comes what I consider to be among the biggest reasons for employing a tactical approach to asset allocation—an approach that seeks to overweight or underweight U.S. equities depending on their relative strength.

I want to revisit the Ibbotson data again to look at the often-cited claim that the stock market has historically returned (capital gains and dividends) 7% per year greater than inflation. This claim has been repeated so much by Jeremy Siegel that some have called it Siegel’s Constant.

Here’s the long-term real return in blue, with a 6.9% trendline in black.

I think this is a dangerous idea for several reason. For one, the long-term real return of stocks is no longer 7%. Thanks to two giant market crashes in less than a decade, the long-term number has fallen to 6.67% (that’s since 1925). Furthermore, much of that gained comes from the post-war boom. Over that last 50 years, that real return has been a much more reasonable 5.45%.

Another problem is that this data series is highly volatile. Too volatile to call the results a constant. There have been long periods when investors haven’t made a dime from the stock market in real terms.

Looking at the data, there seems to be periods of 15 to 20 years when stocks boom, or when they bust. Notice how the blue line runs above or below the black line for long periods at a time. Still, I would say that we don’t have enough data to make the claim that the stock market does this regularly.

During those 15-20 year periods when stocks boom it makes sense to seek to overweight them and during the extended periods when stocks are out of favor it is quite likely that an investors will benefit by seeking to overweight other asset classes. This theory of tactically allocating among different asset classes was put to the test here in a white paper published by John Lewis. It doesn’t work every year, but over time the results speak for themselves.