In Defense of the CEO. By Ray Fisman and Tim Sullivan

In Defense of the CEO. By Ray Fisman and Tim Sullivan
The Wall Street Journal, January 12, 2013, on page C1
Chauffeur-driven limousines, millions in stock options, golden parachutes. It's no wonder bosses' pay and perks can rankle. Here's why the best ones are worth it.
http://online.wsj.com/article/SB10001424127887324081704578233601161769648.html


A $90,000 area rug, a pair of guest chairs that cost almost as much, a $35,000 commode and a $1,400 trash can—these are just a few of the expenses from a remodeling of John Thain's office when he took over as Merrill Lynch's chief executive officer in December 2007. The total bill came to an astonishing $1.2 million—about the price of five average single-family homes.

Those same remodeling expenses contributed to Mr. Thain's resignation just over a year later, after Bank of America BAC bought Merrill, and helped to define the popular image of the CEO as someone who lives a life of extreme privilege: gold-plated faucets, country club memberships and chauffeur-driven limousines, all paid for through corporate largess. Mr. Thain's limo tab included $230,000 for his driver—$85,000 in salary, the rest in overtime and a bonus. This was on top of Mr. Thain's receiving a reported $78 million in compensation for 2007.

It's easy to get upset about perks and pay packages like Mr. Thain's. But even in the face of public and investor outrage, CEO salaries are still on the rise. Progress Energy's CEO Bill Johnson received a $44 million payout when he left the company after its merger with Duke Energy DUK last year, and Abercrombie CEO Michael Jeffries took home over $48 million in 2011—while the company's stock price tanked.

Excessive, decadent? That's a hard call to make without having some idea of what a CEO does. Many CEOs are overpaid or, even worse, paid for incompetence. Still, you can only appreciate the difference between pay-for-performance and pay-for-incompetence by first understanding the CEO's job.

Let's start with the basics: how chief executives spend their time. Among the first researchers to give us a glimpse into the day-to-day life of the CEO was management guru Henry Mintzberg, who followed a handful of business leaders for his Ph.D. thesis at the MIT Sloan School of Management over four decades ago. He discovered that, first and foremost, CEOs go to meetings. Lots of them—it is where his research subjects spent over 80% of their work hours.

The astonishing thing is that the percentage of time CEOs spend in meetings has hardly shifted in four decades, despite innovations like email. A study conducted last year by Oriana Bandiera of the London School of Economics, with Columbia's Andrea Prat and Harvard's Raffaella Sadun, assembled time diaries for hundreds of Indian CEOs. (With other collaborators, they have done similar research on smaller samples of Italian and American executives.)

Unlike Dr. Mintzberg—who did the legwork himself—this group of researchers asked the CEOs' executive assistants to record in 15-minute increments how their bosses allocated their time over the course of a week. Were they working alone or in a group? If in a meeting, how many were in attendance? Was the meeting with employees or with outsiders, via telephone or in person? Despite the vastly different geographies and eras—and differences in customers, products and size of organizations—the CEOs all spent their time in much the same way: in face-to-face interaction.

That time is often marked by interruption. In the five weeks of Dr. Mintzberg's study, he recorded extraordinarily few instances of a CEO alone and without disruption for more than 15 minutes straight. Half their activities lasted fewer than nine minutes—and this was in the pre-BlackBerry age—while only 10% went on for more than an hour. Those hourlong stretches were taken up primarily with hourlong meetings. The more recent studies have found a similar pace of interruption.

Yet saying that the job of someone like Jeff Bezos consists of going to lots of meetings is a bit like saying that Shakespeare wrote words. True, but pretty thin for explaining what made, say, Steve Jobs Steve Jobs.

Meetings remain the focus of the CEO's day because such personal interactions are critical to learning the information necessary to run a company effectively. After all, one of the most important jobs of managers is to decide what information gets passed up through the chain of command. If CEOs were to rely solely on written reports and data sheets from self-serving underlings, they almost would be guaranteed to make the wrong decisions. What manager wants to pass on bad news—so much easier to do in a report than when you're being questioned in detail by your boss? This very problem was at the root of Toyota's response to its problems in 2009 with sudden, unexpected acceleration in its vehicles: Managers were all too willing to paint a rosy picture for the CEO, which hampered his ability to direct the company to respond appropriately.

Harvard Business School professors Michael Porter and Nitin Nohria argue that the skill to extract from underlings the critical details that are needed to inform top-level decisions is part of what makes the best CEOs better than their peers. It works in reverse too. The information the CEO needs to convey is just as prone to being misrepresented and misinterpreted as it works its way through a corporation, across shareholders and among customers. So, in the vast majority of meetings, CEOs are not just uncovering information but also constantly refining their message.

Consider, for instance, founder Tony Hsieh's drumbeat in referring to Zappos as a "service company that just happens to sell shoes." Meetings give him the opportunity to let his stakeholders know exactly what he means. The company hit its billion-dollar sales goal two years before schedule, in 2008, and was acquired by Amazon.com in 2009 for a reported $1.2 billion.

The Porter-Nohria view is backed up by the data. In their time-use study of 354 Indian CEOs—still a work-in-progress—the researchers collected detailed information on the nature of CEOs' meetings, including who attended. Two dominant management styles emerged. "Style 1" leaders, in their taxonomy, spend most of their time meeting with employees; they also tend to hold larger meetings and to include people from a wider set of departments within the organization. "Style 2" CEOs are more apt to spend their time alone, in one-on-one interaction, and outside rather than inside the firm.

Though the researchers are still putting together their findings, they have observed that the first management style, which is inclusive and cross-functional, is typical of CEOs at companies that are more efficiently run and more profitable.

Why don't all CEOs adopt Style 1? It's likely that part of the story is ability: not everyone is up to the task of dealing with the complexities of a bigger conference room filled with disparate participants. It may also reflect a CEO's decision to devote less attention to the company than to cultivating his outside image. In a 2009 study, Ulrike Malmendier of the University of California, Berkeley, and UCLA's Geoff Tate found that companies performed poorly after their leaders were voted "CEO of the Year," because of the distractions that came with the fame, like writing a book and hobnobbing at Davos. A truly great CEO cannot be distracted; she must remain a great intelligence gatherer, a great communicator and ultimately a great decider, and meetings are one of her most important tools.

The existence of great CEOs does not mean, of course, that the average one deserves his millions—although CEOs, never known for their modesty, may think they do. When Dow Jones reporter Kaveri Niththyananthan questioned the CEO of U.K.-based EasyJet, Andy Harrison, about his 2009 compensation of nearly $4.5 million, Mr. Harrison smiled and replied, "I'm worth it." When a congressman suggested to Ford CEO Alan Mulally that he should take a salary of one dollar, given the near-bankrupt state of the U.S. auto industry, Mr. Mulally replied, "I think I am OK where I am"—this in a year when he took home nearly $17 million in compensation. (Mr. Mulally seems to know the value of meetings; he has listed "You learn from everybody" as one of the key attributes of great CEOs.)

What Messrs. Harrison and Mulally no doubt had in mind were their companies' profit numbers. Profits had fallen by 64% the year Mr. Harrison claimed to be worth his millions, but he could point to five straight years of profits as EasyJet CEO—a rare achievement in the airline business. Mr. Mulally's $17 million payday came on the heels of a billion-dollar turnaround that transformed a $970 million loss at Ford into profits of nearly $700 million just a year later.

But are CEOs really so much smarter (and better at running meetings) than the rest of us? Possibly, but that's not the right question to ask. To claim they're worth it, CEOs don't actually have to be all that much better than the runner-up for the job.

In "superstar economies," as in the market for CEOs, even a slight edge in ability can translate into enormous payoffs. That's why Major League Baseball pitchers earn so much more than triple-A players, despite throwing fastballs only a couple of miles an hour faster. When the stakes are in the billions, shareholders should be more than happy to sign off on a multimillion-dollar paycheck, even if the recipient is just slightly better than the next best option.

By the same token, if CEOs' decisions have such a disproportionate impact on corporate profits, you might be willing to pay a lot to motivate them to put in extra hours in the office. And this view helps to explain—if not always to justify—many of the privileges that come with a corner office: the corporate jet that gives CEOs more face time with employees in different locales; the chauffeured limo that frees up time during the morning commute.

As for another controversial perk, what could possibly be the point of paying CEOs for getting fired? The so-called golden parachute goes back to a perfectly reasonable attempt to get CEOs to create even more value for their companies. Introduced by TWA in 1961, the practice took off during the merger wave of the 1980s, when executives started pondering whether it was smarter to seek out merger opportunities to make money for shareholders or to hold on to their jobs. Mostly they opted for keeping their jobs, often to the detriment of the stock price.

As a result, shareholders gave CEOs an escape valve that, the reasoning went, would encourage them to work in the long-term interests of their companies. Even one of the fiercest critics of CEO compensation, Harvard Law School's Lucian Bebchuk, reports in recent research with Alma Cohen and Charles Wang that golden parachutes do motivate CEOs to find merger-and-acquisition opportunities and, as a result, to extract more takeover premiums for shareholders.

So maybe we should be a bit more understanding of Gillette's board, which awarded a severance package worth well over $160 million to CEO James Kilts after the company was acquired by Procter & Gamble in 2005, in what Gillette shareholder Warren Buffett called a "dream deal." (And Merrill's Mr. Thain? He oversaw the company's acquisition by Bank of America at the height of the financial meltdown, a deal that remains shrouded in controversy, in part as a result of $4 billion in 11th-hour bonuses handed out in December 2008. By the time the dust cleared, Mr. Thain walked away with a seemingly modest $1.5 million severance package.)

Yet executives whose ineptitude or laziness makes their companies ripe for takeover also get rewarded sometimes. Indeed, Prof. Bebchuk's study finds that companies where executives are protected by golden parachutes generally trade at lower levels than those where CEOs don't have them. But how should we think about such pay-for-incompetence? Instead of shaking our heads at the injustice, we can consider it an unfortunate side effect of well-motivated incentives. Designing severance packages more carefully is a worthy idea, but simply eradicating them could do real damage.

Before joining the shareholder activists calling for CEOs to be held accountable and stripped of their more obvious excesses, it's worth pausing to think about why those perks exist in the first place. Sometimes it's the result of slick managers who have co-opted their boards, but sometimes it's simply that we can't easily distinguish good CEOs from bad ones before the employment contract is signed. Seeing CEOs make millions for being fired—and even for losing money—may be hard to stomach, but it is collateral damage in the economics of motivating them to run their companies well.

—Messrs. Fisman and Sullivan are the authors of "The Org: The Underlying Logic of the Office," published this month by Twelve.

Corrections & Amplifications

The remodeling of John Thain's office at Merrill Lynch in 2007 included a $35,000 commode (a piece of furniture). An earlier version of this article said the project included a $35,000 toilet.