Implicitly Assuming that the CEO is Not a Crook Misses the Problem

The experiment did not allow for fraud, which makes it a poor pairing with White's speech. The experiment is lab study that has the usual strengths and weaknesses. It does show that simply telling the test subjects to play a role in which the CEO is their friend is associated with the subjects being willing to take actions that harm the corporation's long term interests in order to increase the CEO's bonus. We should consider, of course, how vastly more powerful real friendship reinforced by mutual self-interest would likely be in the real world. (The authors made the first point in their paper.) But the experiment's design removes the key elements of reality that explain why boards typically support the most fraudulent of CEOs -- the ability of the CEO to select directors who are known to defer to the CEO and support extreme executive compensation, the gains to directors from modern director compensation and other mutual back-scratching arrangements, and the perverse incentives of a Greshams's dynamic produced by control fraud.

It would help the discussion greatly if we discussed the key real world examples of the types of fraudulent practices thousands of board members signed off on routinely that drove the crisis. Consider the extraordinary growth in the use of loan brokers, particularly those specializing in liar's loans. Combining these two fraud-friendly practices was sure to produce massive losses -- but also record (albeit fictional) income in the near term that would maximize corporate bonuses. Here is George Akerlof and Paul Romer's 1993 warning about loan brokers' reputation.

"Loan brokers, who match borrowers and lenders in exchange for a commission, have a deservedly bad reputation. The incentive to match bad credits with gullible lenders and to walk away with the initial fees is very high. It can also take several years for this kind of scheme to be detected because even a bad creditor can set aside some of the initial proceeds from a loan to make several coupon or interest payments" (George Akerlof and Paul Romer 1993: 46).

Akerlof and Romer are talking about an odious reputation that was well-established by the 1980s, the time period they are discussing -- three decades ago. The typical fraudulent mortgage lender in the U.S. adopted a series of business practices that were nonsensical for honest lenders but optimized accounting control fraud.

• Extort, by creating a Gresham's dynamic, appraisers to inflate market values
• Make liar's loans known to be overwhelmingly fraudulent
• Originate the loans through loan brokers with perverse incentive systems
• Sell the fraudulently originated mortgages through fraudulent "reps and warranties" to the secondary market

Yes, CEOs shape their executive compensation to be so perverse that it "successfully" reduces the firm's value through means such as cutting R&D investments. That is very bad. Note that the research demonstrating the harm of the perverse incentives has not led to ending those perverse incentives even though we know how to design vastly better incentives. But the four bullet points above cause financial catastrophe and they too are driven by the criminogenic environment that the perverse executive and professional compensation systems that the CEOs design create. The good news is that if we concentrate on the indefensible creation of criminogenic incentives through modern executive compensation the policy steps we would take to reduce fraud would also fix the CEO's perverse incentive to under invest in R&D.

By assuming fraud out of existence, however, we end up with proposed remedies that are certain to fail.

"There are two messages in this study. One is for regulators: Simply disclosing a conflict or friendship does not eliminate its potential to create problems.

The other is for investors.

'Shareholders should take a more active role in finding out what kinds of relationships their boards and C.E.O.s have,' Mr. Rose said, 'and recognize the potential traps created by them.'"

No, there are additional and far more important "messages" that come from the combination of the study and White's speech. Once we realize that CEOs that "embrace" fraud will select directors to aid their frauds we realize that directors that are primarily chosen by the CEO are inherently susceptible to being suborned by CEOs and are not reliable barriers to fraud. We can then begin to think of far more effective reforms that Rose suggests.

• Get rid of the perverse incentives of modern executive, director, and professional compensation
• Restore the fiduciary duty of loyalty that has been eviscerated by hostile courts
• Forbid the elimination of the fiduciary duty of care by states such as Delaware that are engaged in a race to the bottom on corporate governance and strengthen that duty
• More generally, end that race to the bottom
• Do not rely on any reform that allows shareholders to remove minimum corporate governance standards
• Spur a competition in integrity and effectiveness among directors (fyi, more female directors is associated with increased effectiveness)
• Require publicly held firms to separate the CEO and Board Chair positions
• One thing we know does not make sense based on behavioral finance research is to try to put the onus on tens of millions of shareholders to become more sophisticated investors in the manner that Ross proposes