I recently read an article on Seeking Alpha entitled 10 Low Debt Tech Stocks With Strong Corporate Governance. The article screened stocks with low debt to equity ratios that were also ranked as low risk according to Institutional Shareholder Services (NYSE:ISS). I have read many research articles showing a positive correlation between strong corporate governance and strong stock returns, but I have never looked into the services of a company like ISS to do the work for me. Their scoring process made me curious. I wondered if using such a service would be an effective way of finding great market performers, as the Seeking Alpha article implied. So, I decided to examine this further.
I began by doing a Google (NASDAQ:GOOG) search for the ISS website. In the search results, I found a research report paid for by the ISS to show the correlations between their scoring process and stock returns. One thing in the report immediately bothered me. It was a phrase in the fine print that was being used over and over again. Next to many of their tested factors the fine print said, "Becomes positively (or negatively) significant... when Spearman Correlations are used." I am not a statistician, but when I'm reading about a statistical process and it is described as, "Becomes significant," then I skeptically wonder if the data is being manipulated in order to produce the results that are necessary to sell their service. Naturally, my next question was, "What are Spearman Correlations?" Another quick Google search resulted in an example explaining Spearman Correlation:
As an example, let us consider a musical (solo vocal) talent contest where 10 competitors are evaluated by two judges, A and B. Usually judges award numerical scores for each contestant after his/her performance. A product moment correlation coefficient of scores by the two judges hardly makes sense here as we are not interested in examining the existence or otherwise of a linear relationship between the scores. What makes more sense is correlation between ranks of contestants as judged by the two judges. Spearman Rank Correlation Coefficient can indicate if judges agree to each other’s views as far as talent of the contestants are concerned (though they might award different numerical scores) – in other words if the judges are unanimous.In other words, Spearman Correlation ranks perceived governance among companies of the same industry and can be called, "a correlation coefficient between the ranks." Does it make sense to use this type of ranking system?
For example, if I would like to know the strength of Dell's (DELL) corporate governance policies then what I am obviously seeking is to learn how their policies compare to a set of defined best practices. I am not interested in knowing if Dell's governance system is better in relation to an industry competitor, like Hewlett-Packard (NYSE:HPQ). What if both companies had weak governance policies, but Dell's, while still bad, were better than a more atrocious situation at Hewlett-Packard? At best, Dell's higher governance score would be utterly worthless, because the system is simply giving the better score to the lesser of two evils. At worst, the score would be misleading because it instills a sense of false confidence in the company. (This is a hypothetical example and not a valid opinion of either Dell or Hewlett-Packard's actual corporate governance systems.)
To make matters worse an article out of Stanford revealed that these governance watchdog companies, like ISS, do not even have direct contact with the companies that they are assessing. The ratings are constructed quantitatively. This begs the question, is it possible to rate a governance system without actually speaking to the company? David Larcker, codirector of the Rock Center for Corporate Governance at Stanford and the Stanford Business School’s James Irvin Miller, Professor of Accounting, said that, “There’s an industry out there that claims you can. But for the most part, we found only a tenuous link between the ratings and future performance of the companies.”  In their independent study, the Stanford researchers examined 15,000 ratings of nearly 7,000 firms from 2005 to 2007 and found that in all cases that the ratings provided by the corporate governance rating agencies showed very weak correlations that, "did not appear very useful."
To conclude, always be skeptical of research until you have personally verified it to be sound and objective. Moreover, do not rely on corporate governance ratings unless there is evidence that the rating agency has actually had contact with the target company's managers and directors and have reviewed the company's policies themselves. It is my opinion that such a rating cannot be measured on a purely quantitative basis. It appears that investors will still have to evaluate governance systems for themselves by examining publicly available documents and by contacting shareholder relations in an effort to identify the implementation of best practices.
Disclosure: I am long GOOG.