Posted by: Aaron Pressman on February 13, 2006
I can’t count the number of times that I’ve heard some talking head say that because stock market returns will be under 10% a year for the foreseeable future it’s a “stock picker’s market.” It always makes me laugh, because what is the job of a portfolio manager if not to pick stocks? It’s really more of a dig against investing in index funds. Lehman analyst Chip Dickson has a different take on that bit of conventional wisdom today. He argues that it’s actually become less of a stock picker’s market lately, at least for large cap stocks, because of the shrinking difference in returns between individual issues.
With the dispersion of returns shrinking, there’s less difference to portfolio returns from individual stock picks. Instead, macro and sector selection have become more important, he says in a report out today. In other words, if hypothetically all of the stocks in the telecom industry each fall about 10%, it’s more important to avoid all of them than to have picked AT&T over Verizon.
The dispersion of stock returns is simply the standard deviation, or average difference in returns, over any period. Dickson looked at the historical stocks returns of the 1,000 biggest stocks by market capitalization since 1969 and of the stocks in the S&P 500 since 1990. He calculated the standard deviation both using an equal weight for each stock and weighted by market capitalization. In all cases, the current dispersion is well below the level of the late 1990s and at the low end of the historical range. At the height of the Internet bubble, the average difference in 12-month returns for the 1,000 biggest stocks was over 100%. Today, it’s close to 40%.
Dickson also looked within sector groups, reasoning that areas experiencing higher dispersion of returns present more opportunities for smart stock selection while areas showing clumped performance offer fewer rewards. The highest levels of dispersion currently are in the materials, diversified financials, pharma and biotech and energy areas. However, all four areas are at or above twice their historical average levels. If the market reverts to the mean, as it usually does, stock picking will become less rewarding in those industries.
As a final aside, Dickson found three groups moving in different directions of some interest. Real estate investment trusts have shown low and steady levels of return dispersions since the 1980s. Not a great area for stock picking by this measure so you’ve got to get the macro call right. Hardware and software have typically shown very high differences in returns since 1969 but over the past 12 months have not. That may mean they’re due for an upswing which would reward those picking individual names over a simple weighting to the whole group. And though utilities used to show little dispersion, that’s been increasing as deregulation has increased.