Good to Great to Maybe Not

The article by Niendorf and Beck criticizing Collins’ land mark 2001 book Good To Great, raised a few interesting points.  In their response to Collins’s book they question his research methods.  They claim that he is guilty of data mining and that he searched for trends to justify things he wanted to say about the companies he believed had made the transition from good to great.  Some of these points are well made, especially those about data mining.  However, I am not sure that all of their criticisms are valid.

One of the criticisms the authors spend a lot of time looking at is the statistic Collins used about the “great” returns being made to stockholders by the GTG companies.  Niendorf and Beck point out that the returns of the 11 GTG companies are only one third as large as those of the companies with the biggest share holder returns on the Fortune 500.  However, in making this comparison the authors fail to take into account a number of factors that could contribute to this disparity.  Firstly, there are other ways that companies can add value to the portfolios of their stockholders.  One of these is by paying out dividends.  Dividends are quarterly (or annual) payments made to shareholders.  Usually these are paid out as a percentage of every share that the stockholder has in the company; ie. 59% = $0.59 for every share owned.  Couldn’t it be possible that Collins’ 11 pay dividends?  This is certainly the case with Kimberly-Clark.  In 2009, they raised their dividends 3.4 percent to pay out 60 cents per share (NYSE).  This is the 37th year in a row that they have raised their percent payment on dividends.  While it may be true that Kimberly-Clark only had a stock price growth of 6.2% during the study period, they were paying dividends.  It is likely that these high payouts kept stockholders happy.  Also, it is an understood negative of dividends that company paying them will not be able to invest as much in the company, and therefore will not grow as quickly.  Obviously, the stockholders who hold shares of Kimberly Clark are okay with this as they continue to invest.

Another aspect of shareholder returns that the authors failed to mentioned, were the ages of the companies.  Younger companies will almost always have more shareholder growth than older ones.  There are two reasons for this.  Firstly, younger companies have not been around as long and they may be in an emerging marketplace. Because of this, they have more room to grow than older, more established companies.  Secondly, younger companies are riskier to invest in than older ones.  Because of that, they have to offer greater incentives to get people to invest in them; incentives like high shareholder returns and robust growth.  When one remembers these factors, the disparity between the GTG 11 and the companies listed by Niendorf and Beck is obvious.  Some of the companies selected by Collins are ancient. Fore example, Wells Fargo was started in 1852!  NVR on the other hand, was only started in 1980.  One would expect a company started during the Reagan years to post higher returns than one started during the Franklin Pierce administration.

Niendorf and Beck make some good points.  Collins and his team were clearly guilty of data mining and used statistical patterns to verify foregone conclusions.  However, Niendorf and Beck also used statistics that may have been less than comprehensive to make their points.  This article does serve to show that statistics can be manipulated to make any point that the authors want. Unfortunately for them, they don’t prove this in quite they way they were hoping.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: