http://knowledge.wharton.upenn.edu/article/1041.cfm: "Another paper that shows that governance is not eaily quantified."
Short Version: Larcker, Tuna, and Richardson provide more more evidence that corporate governance is not easily quantifiable and what works for one firm may not for other firms.
Consistent with Coles, Daniel, and Naveen, but opposed to what many so-called corporate governance experts are trying to sell, is a recent paper by Larcker, Tuna, and Richardson. They find that corporate governance is endogenous and that there does not appear to be an easily quantifiable means of differentiating good and bad governance.
The authors did try to quantify goverance. They examined data on "more than 2,100 public firms" to find what factors led to good governance and what variables wee tied to poor governance. Somewhat surprisingly, they could not find any relationships: ""Our overall conclusion is that the typical structural indicators of corporate governance used in academic research and institutional rating services have a very limited ability to explain managerial decisions and firm valuation."
Why? The most likely explanation is by trying to find single factors that fit all firms, they are losing the relevant importance. For example (in the spirit of Coles, Daniels, and Naveen) what works well at a diverse firm (a large board) may not work well as a single-line firm. Or while insiders are good for R&D intensive firms, insiders may magnify agency costs at slow growth firms.
As Richardson summed up "The recipe book is big, and there's a different recipe for each company." And to those that are using governance report cards to evaluate or instruct firms, Richardson adds "As far as we can tell, there's no evidence that those scorecards map into better corporate performance or better behavior by managers."
What does this mean to financial research? My best guess is that using the "structural indicators" (example number of outsiders on board, board size, etc) will only have meaning in a controlled context. For instance, it will no longer be enough to measure governance by saying there are X% of outsiders on the board, but rather one will have to say "within the same industry, one firm has more insiders than the other firm." Which will be much more time consuming and difficult, but maybe if we have the proper matching, then we can begin to see what does and does not work.