Sarbanes Oxley, Changes in Board Structure, and D&O Insurance

A consistently anti-Sarbanes-Oxley site is The FRAUDFiles Blog.  Last week, for example, the blog noted a study discussing the changes in boards as a result of SOX and the increase in D&O insurance costs.  As the Blog concluded:  "Clearly there is a one significant cost of SOX noted here - an increase in the D&O premiums.".  The post is here. 

The post brought to our attention an article discussing these costs by James Linck and Jeffrey Netter, both professors of Banking and Finance at the University of Georgia, and Tina Yang, from Clemson, titled:  "The Effects and Unintended Consequences of the Sarbanes-Oxley Act, and its Era, on the Supply and Demand for Directors" and dated Feb. 14, 2007.  The piece contains a number of interesting facts about changes in the board of directors in a post-SOX era.  For example: 

• Salaries of independent directors have risen, with the median increasing $33,000 from 2001 to 2004.  We have, of course, discussed this trend on a number of occasions;
• Boards have become larger and more "independent." Thus, for the period from 1998 to 2004, larger firms (over $700 million in market value) saw an increase in the number of independent directors on the board, from 71.9% to 75.6%, and in the size of the board, from 9.37 directors to 9.98 directors;  
• The number of executives as a percentage of the total pool of directors has declined, falling from 54.76% to 41.61%.  Much of that decline has, however, been picked up with the increase in the number of retired executives (16.36% to 22.76%, the category includes both retired executives and directors over the age of 70);   
• The number of companies adding at least one director to the audit committee have increased from 18.09% in 1999 to 36.16% in 2004.  One can surmise that a good part of this increase arose from the need for a financial expert on the audit committee.  
• While there was no material change n the number of board meetings as a whole in the post-SOX world, audit committee meetings doubled.
• The cost of D&O insurance has increased.

The data contains no real surprises.  The most interesting thing is the attempt in the paper to assign responsibility for these changes and the costs of implementation.  At first, it seemed as if the authors would attribute the changes to a wide range of events that have taken place since 2001.

• "While our study examines the specific impacts of SOX on directors and boards, we recognize that these changes are more broadly the result of a changed landscape for corporate governance.  During the time when SOX was deliberated and passed, other significant events were occurring, such as the large drop in stock prices, the start of an economic recession, and a series of corporate scandals.  These events influenced the corporate governance landscape, and let to SOX and the related rule changes on the NYSE and NASDAQ."

But no.  The paper went on to say this:  "However, we also believe SOX signifies a turning point, and our results are consistent with the conjecture that firms are making substantial adjustments directly in response to SOX and the contemporary rule changes made by the NYSE and NASDAQ."  Yet the conclusion to the paper states that "[w]e find evidence consistent with the theory that director workload and risk increased post SOX."  On the one hand, the costs are associated with SOX compliance; on the other hand, they are associated with increased risk.

These conclusions are in some ways contradictory as is the evidence in the paper.  First, many of the changes discussed in the paper cannot be attributed to SOX.  The paper discusses changes in the number of independent directors and in the makeup of the pool of directors.  With the exception of the audit committee, SOX said nothing about independent directors or board composition.  This fell to the exchanges.  Moreover, large public companies have been increasing the number of independent directors beyond the number required by the exchanges.  In other words, some of the increase in the number of independent directors (and, as a result, the costs associated with these directors), has nothing to do with meeting legal requirements but represents a volitional decision by the company.

As for the increased risk on the part of directors, SOX can't really be given responsibility for this, even if the rates have risen.  SOX made almost no changes at the board level.  In fact, it is a weakness in the Act that there was no increase in board responsibilities or strengthening of board duties.  The study confirms this by noting that there has been no increase in board meetings since the adoption of SOX.  In other words, the board as a whole is working at about the same level as before SOX.  While the Act did step up the role of the audit committee by providing some specific responsibilities, including the power to hire and fire the independent accountants, there is no proof that this has increased exposure, although it probably has increased workload.  

As for the increase in risk, the rise in D&O insurance at least suggests that there is the perception that this is the case, although premiums can rise for reasons unrelated to shifts in risk.  So, was FRAUDFiles right in suggesting that this is a cost attributable to SOX?

Hardly.  One suspects, however, that a good portion of the increase was an overreaction to the huge fraud suits brought in the pre-SOX days and, perhaps, to some fear of increased exposure under Delaware law.  The same thing occurred during the tender offer era in the early to mid-1980s.  When things calmed down, insurance costs receded.  As the evidence has shown, the number of fraud suits in the post SOX era are falling, not increasing, and any hint that Delaware would impose more rigorous standards has evaporated.  To the extent that D&O policies are tied to risk, premiums should eventually begin to fall. 

It is difficult to isolate the costs directly related to SOX.  In the aftermath of Enron and Worldcom, many of them would likely have occurred, even in the absence of any specific legislation.