Reversion To The Mean Phenomenon: Part I

Includes: DIA, IEF, QQQ, SPY
by: Larry Swedroe

The Seven Wonders of the Ancient World is a list of remarkable constructions of antiquity. They are the Great Pyramid of Giza, the Hanging Gardens of Babylon, the Temple of Artemis at Ephesus, the Statue of Zeus at Olympia, the Mausoleum at Halicarnassus, the Colossus of Rhodes and the Pharos of Alexandria.

Benjamin Franklin is often credited with adding an eighth wonder, compound interest. The financial world also provides us with a ninth candidate, reversion to the mean (RTM) of abnormal (both high and low) returns. Today, we'll begin a series to help you understand this often misunderstood phenomenon.

When an asset class (or stock) experiences an extended period of abnormally high returns (when prices rise faster than growth in earnings, as they did from 2009-2013) prices are driven to levels that result in future expected returns being lower. Thus, investors should expect returns to exhibit mean reversion. Similarly, when an asset class experiences an extended period of abnormally low returns (when prices fall faster than earnings), prices are driven to levels that result in future expected returns being higher. Most investors recognize this when it comes to bonds, because the future expected returns are easily observed in the form of the current yield. The following example will illustrate this point.

If a 10-year Treasury bond is yielding 10 percent and rates fell quickly to 5 percent, if an investor sold that bond the return would be well above 10 percent. However, future expected returns for investors in that bond are now only 5 percent. In other words, the first year return might be something like 55 percent, and the remaining nine years you earn 5 percent, you get the same 10 percent return over the full period (as if the bond yield never moved). Similarly, if a Treasury bond is yielding 5 percent, and rates rise to 10 percent, if an investor sold that bond the return would be well below 5 percent, but future returns to holders of that bond would be 10 percent. In both cases returns reverted to the mean of 10 percent.

Unfortunately, while the same principle applies to stocks, the expected return isn't as obvious. With that in mind, let's look at an example from the world of equities. For the period 1927-1994, large-cap growth stocks, excluding utilities (Fama-French series) returned 9.4 percent per year. Over the following five years (1995-99) they returned 31.2 percent per year. The effect on valuations (and, therefore, on future expected returns) can be seen by looking at the P/E ratio of the S&P 500 Barra Growth Index. At the end of 1994, the P/E ratio stood at about 19. By the end of 1999, the P/E ratio had risen to 54!1 One dollar invested in this asset class at year-end 1994 would have grown to $3.89 in just five short years (the eighth wonder of the world at work).

It is important to note that these great returns didn't come right after a period of poor returns. In the preceding 10 years large-cap growth stocks returned 15.1 percent. Remember, a period of low returns would lead us to expect higher future returns. However, in the 10 years leading up to 1995, large-cap growth stocks provided returns that were well above the historical average - about 78 percent higher than the return of 8.5 percent they had delivered from 1927-84. This example demonstrates the point that while we you might expect returns to revert to their mean, there's no way to know when they will do so - returns can remain abnormally high or low for a long time. This makes trying to time the market and avoid (or catch) a mean reversion a losing proposition.

While we cannot know when it will happen, we do know that returns will almost certainly exhibit the tendency to revert to their mean. And in the case of large-cap growth stocks, they surely did. In the RTM that occurred in the "lost decade" (2000-09), they lost 3.5 percent per annum, producing a total loss of 30.2 percent.

The sad news is that far too many investors are unaware of this ninth wonder of the world. In fact, as we have discussed, investor behavior tends to exhibit the trait known as recency - the tendency to buy what has done well recently (at relatively high prices) and sell what has done poorly recently (at relatively low prices) - not exactly a recipe for investment success. In fact, that tendency is why so many investors have taken all the risks of stocks and earned bond-like returns.

Those Who Cannot Learn From History

It was Spanish philosopher and novelist George Santayana who said "Those who cannot learn from history are doomed to repeat it." Hopefully, the following example and information will prevent you from repeating the mistake of recency. For the 12-year period 1988-99, the Dow Jones US Select REIT Index returned 7.3 percent, and $1 invested to grow to $2.32. During this 12-year period they underperformed the 20.3 percent per year return of large growth stocks by 13.4 percent a year. As a result, not many people were interested in investing in REITS in 2000.

From 2000-06, the REIT Index returned 23.1 percent per annum. The result was that in just seven years one dollar invested had grown to $4.28, or almost twice the growth in seven years that it had provided in the prior 12 years. And REITS became a favorite investment for many. However, those returns drove the valuations on REITS way up, and higher valuations forecast lower returns. In the seven years from 2007 through 2013 the REIT Index returned just 1.0 percent. That's RTM at work.

Our next post in the series on RTM explains how valuations matter to future returns, explaining the RTM phenomenon.

Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.