Here’s a simple question: which asset class has better historical returns, stocks or one-month U.S. treasury bills?  That’s pretty much a no-brainer.  Most anyone would tell you the answer is stocks.  And they’d be right.  Since 1926 treasury bills have returned a paltry 3.5% annually on average, vs. over 10% for the S&P 500. That’s a huge difference.  However, we’re talking about aggregate stock market returns. What about individual stocks?  Hendrik Bessembinder of Arizona State University, in a working paper entitled “Do Stocks Outperform Treasury Bills,” found that 58% of common stocks have holding period returns less than those of one-month treasuries over their full lifetimes.  Even more astounding: more than half of all stocks have negative lifetime returns.

How can this be possible?  It’s because the cross-sectional distribution of long-term stock returns demonstrates positive skewness.

Wait!  Don’t leave yet!  I promise to explain this in very simple terms.  Let’s start by reviewing some basic statistics.  Remember from high school what a normal distribution is?  It looks like a bell curve when displayed graphically.  And it’s symmetrical, which means that the average of all the data points is identical to the median.  (The median, if you recall, is simply the point at which exactly half the other data points are above and half are below).

Skewness refers to the degree of asymmetry of a normal distribution.  To illustrate this using a practical example, open up the home sales section of your local newspaper.  You will usually find both the median sales price and the average sales price reported.  More often than not, the median price is higher than the average price.  Why?  Because typically among all the homes that were sold during the reported period, there were some that were priced especially low relative to the average.  This had the effect of pulling the average price down below the median price.  Graphically it’s the same as grabbing the lower left point on the bell curve and pulling it further to the left.  Such a curve is said to exhibit negative skewness.  Another way to think about this is that there are outliers in the curve (in this case the exceptionally cheap houses) that are skewing the average price lower.

Bessembinder found that there is significant positive skewness in the contribution of individual stock returns to aggregate U.S. stock market performance.  He studied the performance of the 26,000 stocks that have come and gone since 1926 and which collectively have been responsible for lifetime shareholder wealth creation of nearly $32 trillion.  Amazingly, he found that just 86 stocks accounted for over half of that total market return over that period, and only 1,000 (less than 4% of the total) accounted for all of it.  In other words, the entire gain in the U.S. stock market since 1926 is attributable to the best-performing four percent of listed stocks.

This flies in the face of modern portfolio theory’s capital asset pricing model (CAPM), which expects each individual stock’s return premium (i.e. excess over U.S. treasury returns) to be the stock’s beta times the expected market-wide premium.  Bessembinder explains that CAPM relies on the assumption that stock returns are normally distributed (the average and median is the same).  But they’re actually positively skewed, and that skewness increases for longer holding periods due to the effect of compounding.  Therefore the median excess return for the majority of stocks is actually negative.

Now back to stock picking.  Bessembinder’s study indicates that if you pick the right stocks, you can make a killing in the stock market.  But if you pick the wrong ones, you’ll do worse than if you had just bought treasuries or bank CDs.  And there are a lot more wrong ones than right ones.  This study also explains why active mutual funds attempting to outperform the market by selectively picking stocks have underperformed the market average over time.

Since Bessembinder’s work has poked some holes in the key theory that has been the mainstay of the financial planning profession for decades, should you now avoid the stock market entirely?  Not if you want to participate in that $32 trillion of wealth generation!  But your best strategy is probably to buy some kind of market index fund so that you are participating in the entire market.  You may end up holding a lot of losers, but those few winners will help put you over the top!  (Different index funds may weight each stock differently, but choosing the right index fund is beyond the scope of this posting).

Here’s a link to the study:  https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2900447.