Buy and Hope: More Failure

February 18, 2011

From academia comes a study that suggests that simply buying a market index does not live up to reasonable return expectations, even over a long period of time. (The study is summarized here by CXO Advisory. You can access the full paper here.) S&P 500 constituents were measured from 1991-2010 by sampling the members in 2004 and 2010. The reasonable return expectation was set up by the authors using the 10-year Treasury bond yield, plus a risk premium of 4.0% to 4.7%. This is a standard way that long-term returns are estimated in academic finance, by the way. (Back in the olden days, when people thought stocks were actually risky, a risk premium of 6% was the rule of thumb.) So how did stocks actually measure up during that 20-year period? According to CXO Advisory:

Even though the aggregate market value of S&P 500 components grows from $2.8 trillion in 1991 to $11.4 trillion in 2010, these firms underperform reasonable investor rate of return requirements (T-note yield plus a modest equity risk premium) by $4.5 trillion.

Even using a pedestrian risk premium, blue chip stocks-in aggregate-did not compensate you very well for the risk. The authors also point out in the abstract that:

41% of the companies included in the S&P 500 in 2004 or 2010 created value in 1993-2010 for their shareholders, while 59% destroyed value.

Even over a 17-year holding period, the majority of companies destroyed shareholder value. Keep in mind that these were not fly-by-night outfits, but the bulk of the largest public firms in America. Even buying blue chip companies over the long run is likely to cause you to lose your shirt. Buy and hope, unless you get lucky, is not going to work.

Should you give up on stocks entirely and hide your money under the mattress? We think not. After all, 41% of the stocks created shareholder value—and some of them did exceptionally well. The stocks that created the most shareholder value were the ones that went up the most in price. If we focus our attention on those stocks with the most powerfully rising prices, we can identify the companies that created the most shareholder value. Hmmm…come to think of it, that’s the basic definition of relative strength.

In contrast to buy-and-hope (internally we sometimes refer to this investment solution as sit-and-take-it), systematic application of the relative strength return factor has been shown to add a lot of value to the investment process.

Posted by:


Peter Lynch’s Cocktail Party Indicator

February 18, 2011

An absolute classic, by way of Barry Ritholtz at The Big Picture. I know, not quite as exciting as the Sports Illustrated Swimsuit Indicator. Hard to say what stage we are at, but it’s clearly not stage four!

Posted by:


Putting The Last Two Years In Context

February 18, 2011

Now that the market has doubled since its Great Recession low, it is getting more popular to argue that this could be it for this bull market. That argument is a little hard to believe when put in the context of the chart below which shows the 10-year S&P 500 annual compound returns from 1926-2010. The data is displayed in a distribution with the lowest 10-year returns on the left hand side of the chart and the highest 10-year returns on the right. As shown below, the 10-year annualized return of the S&P 500 at the end of 2010 was a scanty +1.4%. This reading of +1.4% is the first to register in positive territory since 2008. However, it remains the twenty-fourth lowest reading over this 1926 to date time frame (out of 300 quarterly observations) and in the first decile of the distribution since 1926!

TenYear Putting The Last Two Years In Context

(Click to Enlarge)

Printed with permission from The Leuthold Group.

Posted by:


Sector and Capitalization Performance

February 18, 2011

The chart below shows performance of US sectors and capitalizations over the trailing 12, 6, and 1 month(s). Performance updated through 2/17/2011.

 

Posted by: