Thursday, March 14, 2013

Say on Pay in the US: The Early Evidence

Joe and I have been debating the merits of the new Swiss Say on Pay regulation.   (See Mad as Hell and I'm Not Going to Take This and Swiss Say on Pay - The Dangers

The US now has it's own, non-binding Say on Pay legislation, but it is a relatively new development.  A few years ago, my colleague Jay Cai and I performed the first empirical analysis of Say on Pay in the United States in the academic literature.

The starting point for our analysis was April 20, 2007, when underdog Presidential Candidate Barack Obama introduced the legislation into the Senate.  It had just passed the house on the same day.  Thinking about the impact of this legislation on shareholder wealth you can immediately envision three possible impacts.

First, it could be beneficial, prodding boards to rein in excessive pay and to be more careful in the compensation packages they offer.  In this case, the share price of firms would increase with the bill.

Second, such a law could be detrimental to shareholders.  Boards are responsible for executive compensation and can always hire additional expertise to advise them as needed.  Outside pressures from shareholders, some of whom may have narrow agendas, could produce unnecessary complications within the boardroom which could reduce shareholder wealth.

Finally, we might find that the mere introduction of the bill in the Senate (and it's passage in the house) have no impact on shareholder wealth.  After all, we are talking about the introduction of a bill, not its passage.  Moreover, even if it passed the Senate, then-President Bush said he would veto it.  Besides, it is an advisory vote - boards don't have to do anything even if a majority of shareholders disagree with the level of executive compensation.  (In this regard, I note that previous evidence on advisory votes finds them to have little impact on shareholder wealth.)  So in my mind, the probabilities were stacked towards no impact.  

What did we find?  Some interesting results.

First, we find that one size doesn't fit all with this legislation.  Our results indicate that situations where the law appears beneficial and situations where shareholder votes are likely to be seriously considered are associated with increases in shareholder wealth.  In other situations, the bill would appear to reduce shareholder wealth.  

So where was it beneficial?  In firms with high, abnormal CEO compensation, those with low pay for performance sensitivity and for those firms that had been responsive to shareholder votes in the past.  The latter result is in line with the fact that firms don't have to implement changes, even if a majority of shareholders disagree with pay packages.  

But Say on Pay could have been voluntarily implemented by boards if they deemed it beneficial.  They didn't need legislation to do this.  Moreover, shareholders themselves could have proposed Say on Pay to their boards.  In fact, this did happen.  We found numerous cases where activists (primarily unions) had sponsored Say on Pay proposals.  Unfortunately, they appeared to target the wrong firms.  The firms they targeted were large firms that, on average, appeared to have reasonable performance, governance and pay characteristics. As a result, the share price dropped for firms targeted in this way.  Our research suggests that say-on-pay creates value for companies with inefficient compensation, but can destroy value for others. 

The complete paper can be downloaded here: Shareholders’ Say on Pay: Does It Create Value? 

No comments:

Post a Comment