Verified by 
“Most of our managers are independently wealthy, and it's therefore up to us to create a climate that encourages them to choose working with Berkshire over golfing or fishing.”
– Warren Buffett, An Owner’s Manual, 1996
In its lead story of December 6, 2014 THE WALL STREET JOURNAL wrote about the “abrupt” retirement of United Technologies Corporation CEO Louis Chenevert. According to the story, senior executives and Board members found Mr. Chenevert too “disengaged from the details” of running a global business. For example, executives reported difficulty scheduling meetings with him. On the other hand, the CEO found time to travel from the company’s global headquarters in Hartford to Taiwan for the purposes of checking on the construction of his personal 110 foot yacht.
An “unnamed person familiar with Mr. Chenevert’s thinking” said the former CEO had been “passionate” about his job. Perceptions of disengagement were inappropriate conclusions based on job-related travel requirements.
CEOs and Golf:
Is CEO disengagement a problem? One way to measure disengagement is to calculate the number of hours CEOs of public companies play golf per year and then to compare golf time with other measures. Professor Lee Biggerstaff of the Department of Finance at the Miami University of Ohio and his team did this study in 2014.
They found that CEOs play more rounds of golf when (1) they have lower equity-based incentives (2) and weak governance systems on Boards of Directors. As CEOs begin to play 37 or more rounds of golf per year, their firms have lower operating performance. Assuming a round of golf is a 4-6 hour commitment, 37 rounds of golf represents 19 days of disengagement.
Data was obtained from the golfing records of 363 S&P 1500 CEOs from a database maintained by the United States Golf Association. The researchers examined golf records from 2008 to 2012. Thirty-seven rounds of golf per year would place a CEO in the top quartile of the survey. One CEO in the database played 146 rounds in a single year. Using our calculations, that is 73 days of disengagement.
“When I Play Golf I am Conducting Business:”
We have heard CEOs frequently make this case. One CEO we knew bragged that he would give up his office before he would give up his Golf Club membership: so many of his company sales were accomplished on the golf course.
Is this assertion valid or a rationalization?
There is a clear relationship between CEO golf behavior and CEO compensation. And this is something that members of Boards of Director Compensation Committee meetings should take seriously: CEO golfers in the bottom quartile of golf participation tend to own a significantly bigger equity stakes. More “skin in the game” is associated with more time at work. And less “skin in the game” is associated with more golf time.
The study measured operating performance in terms of Return on Assets (ROA). Top quartile CEO golfers have 100 basis points lower performance than the companies with the lowest quartile CEO golfers. Professor Biggerstaff’s team found negative correlation within firm changes of ROA and changes in CEO golf activity. In other words, over time within the same firm, if the CEO started playing less golf, ROA would increase. Over time within the same firm, if the CEO started playing more golf, ROA would decrease.
Governance Matters:
We predict that the publication of Biggerstaff’s research will achieve the following result: the United States Golf Association will be pressured to keep its database confidential.
That is what will happen.
What should happen is this: members of Boards of Directors should look at this issue as a “canary in the coal mine” study. Just as it is dangerous to have CEOs micromanaging corporate operations, it also is dangerous to have CEOs not be sufficiently involved.
What is the right balance between CEO uninvolvement and micromanagement? This is a topic that Board members of Compensation Committees seldom discuss. Perhaps they should.
Is it appropriate to ask CEOs to submit to the Compensation Committee an accounting of personal time? An implication of this study is that Board members have a fiduciary responsibility to shareholders to request such information. Board members have a fiduciary responsibility to ask the CEO’s direct reports if they believe that they have enough access to the CEO.
Before the publication of this study, perhaps such questions might be considered “unreasonable” by CEOs and by Board members. With the publication of this study, is it not prudent to request this information?
CEOs Are Not Enlightened Philosophers:
CEOs of both for profit and nonprofit companies are not enlightened philosophers. They are leaders. And our experience is that these leaders can often described as intelligent, assertive, competitive, self-absorbed, and eager to improve their personal net worth.
CEOs have no trouble rationalizing their behavior.
It is the role of external members of the Board to patiently listen to such justifications from a skeptical perspective.
It may be the role of the Compensation Committee to put limits on outside behavior and to ask questions about whether the proper financial incentives are in place for CEOs.
Warren Buffet is Right:
The reader should not come away from this piece thinking that this is an article about golf. It is a piece about CEO commitment to excellence in the job. Warren Buffet’s quote at the beginning of this article is perfect. CEOs can easily find ways to rationalize the appropriateness of time spent golfing, fishing, sailing, skiing, or working for worthwhile charities.
If the Board finds that the CEO is focusing too much time on causes that do not directly provide value for investors or directly advance the mission, then the Board should ask itself if it has created the wrong incentive program or has the wrong CEO.
Trusting CEO justifications makes for collegial Board meetings.
“Trust but verify” is a Russian proverb that makes for better governance.
References:
Biggerstaff, L., Cicero, D. C., & Puckett, A. (2014). FORE! Analysis of CEO Shirking SOCIAL SCIENCE RESEARCH NETWORK, August 2014.
###
