In Houston, Compaq Computer Corp., struggling through rough weather, has ousted its chief executive. In New York, Salomon Inc. is regrouping under new management after revealing its employees' illegal bidding in Treasury security auctions. These two seemingly unrelated events are travails that only a corporate governance expert could love. Why? They may give life to an idea that has long been a goal of boardroom reformers: splitting the job of chairman and chief executive.
At Compaq, the board was able to act speedily -- firing founder Rod Canion for failing to pull the personal-computer maker out of a sales-and-profit tailspin -- largely because it has a separate chairman, venture capitalist Benjamin M. Rosen. At Salomon, the board was kept in the dark long after former Chairman and CEO John H. Gutfreund learned of Salomon's transgressions. Directors didn't hear of the problem until shortly before the public did. Now, Interim Chairman Warren E. Buffett has said he would like to remain chairman after he relinquishes the CEO post.
'MY BOARD.' Most U. S. corporations function more like Salomon than Compaq. At some 80% of the companies on BUSINESS WEEK's list of America's 1,000 most highly valued corporations, one person rules the roost as chairman and chief executive. He controls the meetings of directors, who are theoretically monitoring his performance on behalf of shareholders. It's no wonder CEOs often confuse who works for whom. As John Nash, president of the National Association of Corporate Directors, put it recently: "Their attitude is, 'It's my company and it's my board.' They don't get it that it's not."
To remedy that attitude, many governance experts prescribe a job split. "The CEO should not be chairman of the board," says Harold M. Williams, former chairman of the Securities & Exchange Commission. "Control of the agenda and pace of the meeting is a powerful control." Harvard business school Professor Jay W. Lorsch calls a split "the single most significant thing to do." In Pawns or Potentates, he argues that a change would reinforce directors' independence, help prevent corporate crises, and underscore the notion that managers serve at the pleasure of the board, not vice versa.
Trouble is, getting a CEO to give up his almighty perch is hard, especially when no one has really studied whether companies work better with nonexecutive chairmen. Yet common sense says the idea has merit. As Compaq showed, troubled times can demand tough action that a nonexecutive chairman can best direct: You can hardly blame a CEO for refusing to recognize the need for new talent at the top.
Even in normal times, boards could function better if someone other than the CEO supervised board business. Outside directors admit privately that they often don't have the necessary information to ask the right questions, raise the right issues, or make the right judgments. Buffett, often called "the sage of Omaha" (where he is chairman and CEO of Berkshire Hathaway Inc.), has described the information gap with an analogy: "As Henry Kissinger once said about the State Dept. memos, they give you three options: The first leads to nuclear war, the second to unconditional surrender, and the third is what they want you to choose." In general, Buffett has said, "things will come to light with a nonexecutive chairman." (Buffett is unlikely to give up the chair at Berkshire, since with a 45% stake, he would be in control no matter what his title.)
WORTH TRYING. Nonexecutive chairmen wouldn't run companies but would merely manage board business. They would consult with the CEO to set a meeting's agenda, see that directors get requisite information in time, then run the meetings. They would supervise the selection of new directors and make committee assignments. And they would ensure that outside directors regularly evaluate the CEO's performance. As Rosen says: "The last thing I want to do is be a manager."
Splitting the jobs of chairman and CEO is no panacea. It hasn't always worked well. A few years back, RJR Nabisco had a nonexecutive chairman: Charles E. Hugel, then CEO of Combustion Engineering Inc. But that didn't stop CEO F. Ross Johnson from wasteful ways or from trying to buy RJR with a lowball bid.
Such instances don't mean the idea isn't worth trying, though. Much will always depend on the personalities involved, the CEO's managerial style, and the board's culture. But there seems to be no cogent argument that it would hurt corporate performance, and it might increase the odds of getting the best from management. Compaq and Salomon have put the idea in the air. Since making the switch will take time, it's up to those who want to see better corporate governance to keep it there.