More Procrastination

At "Who Has Time For This", David Cowan, IT entrepreneur and venture capitalist, discusses his "ideal configuration" for boards of start-ups in"Control Roulette: a bet on red or black". His view is clear and to the point: small board, with at least three minorities to avoid any a priori majority. After the CEO and a couple of venture capitalists (VC), he recruits outside experts with "skill sets and credentials that would add value to the board".¹ In case of lack of concensus, the CEO wins if he is making money, the backers win if he isn't. This formula may well be valid beyond privately-held start-ups.

Clive Crook writes in the Atlantic Monthly this month, that "The CEOs of too many public companies enjoy the power and rewards of ownership without the risks. Corporate values have deteriorated as a result." No doubt, this is partly structural, brought about as a side effect to changes made to the rules of the game twenty years ago to prevent hostile tackovers. These rules of "governance" are used to construct an index of CEO independence in a study cited by Bradford Plumer in a blog post,

A 2003 Harvard University/Wharton School paper entitled "Corporate Governance and Equity Prices" ranked 1,500 companies in terms of management power, sorting firms into a Democracy Portfolio (firms in which shareholder rights were strongest) and a Dictatorship Portfolio (firms in which managers were subject to less oversight). Shockingly—or not—the democratic firms outperformed the dictatorship firms by 8.5 percentage points per year throughout the 1990s.

This seems to be based on a rather "particular" measure of "governance": an index of protection from hostile takeovers, including measures to protect directors from being fired. Knowing that, it seems less surprising that the companies and directors at higher risk outperformed those who had less to fear. Being told by your fund to "put out or get out" is not synonymous to "strong governance" as I understand it.

So, what to do? If a diva CEO says she needs security to give her best performance, should the owners of the business say "no"? Perhaps not entirely; David Cowan's ink-color could well be a good guideline. But in any case, the diva CEO should be challenged.

Consider the psychological aspect: CEOs are subject to hubris; their positions, when overprotected (under governanced?) may remove opportunities to receive criticism that help most people correct for flawed self-assessment. Good governance is the chance to give CEOs constructive criticism, and improve their steering of the enterprise. This point is made in Dunning, David, Chip Heath, and Jerry M. Suls (2004), "Flawed Self-Assessment", American Psychological Society, Volume 5—Number 3.

Malmendier and Tate (in press) found that CEOs who are
heavily invested in the stock of their companies are the ones most
sensitive to free cash flows—they invest more in new projects
when internal cash flows are high and less when internal cash
flows are low. By using internal cash to finance their projects,
these CEOs avoid having to get outside commitments from financial
markets that may not agree with their optimistic views.
Interestingly, CEOs are also more likely to invest free cash flows
when they hold two other titles, president and chairman of
the board. When CEOs are allowed to accumulate all these titles,
this may be an indicator that they face weak oversight by their
boards of directors. In sum, overconfidence may cause CEOs to
see projects as promising even though other people think those
projects are risky. And they are more likely to pursue those
projects when they can do so without a second opinion from the
external financial markets or an independent, active board of
directors.

Their bibliography refers, on this topic, to

• Hayward, M.L.A., & Hambrick, D.C. (1997). Explaining the premiums paid for large acquisitions: Evidence of CEO hubris. Administrative Science Quarterly, 42, 103–127.
• Malmendier, U., & Tate, G. (in press). CEO overconfidence and corporate investment. Journal of Finance.
• Landier, A.,&Thesmar,D. (2003). Financial contracting with optimistic
  entrepreneurs: Theory and evidence. Unpublished manuscript,
  University of Chicago, Chicago.

Even when the CEO does not have carte blanche, she may not receive constructive criticism if the board members strive primarily for unanimity. Writing on "Why it's so hard to blow the whistle" in the Yale Alumni Magazine, Jeffrey A. Sonnenfeld, associate dean of the Yale School of Management, reminds us of the dangers of "groupthink", the phenomenon described by the late Yale psychologist Irving Janis thirty years ago. He goes on to list ten elements of governance he recommends to help prevent the misconduct created by groupthink:

1. to build a climate of trust and candor;
2. to foster a culture of open dissent;
3. to utilize a fluid portfolio of roles;
4. to ensure individual accountability;
5. to ensure opportunities for the board to assess leadership talent;
6. to evaluate the board's own performance regularly;
7. to seek knowledge rather than marquee names for the board;
8. to avoid joiners who collect boards like trophies;
9. to seek those with a passionate interest in the business; and
10. to ruthlessly purge those with conflicting personal or commercial agendas.

How can one improve CEO performance? Start with David Cowan's prescription, then use Jeffrey Sonnenfeld's checklist for selecting board members and monitoring board effectiveness.

Oh, one last point: if you are starting a business, make sure you sell your plan to backers (like bankers), even if you could finance it entirely yourself. We can all use a bit of governance.
Tags: governance business start-ups

---

1. When I asked him, "Would you ever want a thoughtful but (to the point of significantly underachieving) not ambitious board member?", he replied "...a thoughtful but unambitious underachiever sometimes turns out to contribute more than any other board member. However, you can't know that until it happens, and so I still try to recruit only directors with proven success and skill sets." Hmmm, I wonder, how can a candidate demonstrate that he was a successful eminence grise?