Corporate savior

The Corporate savior ideal casts the chief executive as the decisive source of corporate success: appoint the right person at the top, and the company will prosper. The book argues that this faith grew alongside calls for decentralization, even though the strongest case for organizational intelligence points in the opposite direction: decisions improve when knowledge and judgment are distributed, communicated, and combined. The executive celebrated as a miracle worker is also made the convenient scapegoat when uncertainty defeats prediction.

The rise of the corporate savior

The contrast sharpened in the 1990s. Companies paid more attention to decentralization and bottom-up methods, yet investors, journalists, and the public increasingly treated CEOs as superheroes. In the 1940s, the average American would not have known Alfred P. Sloan; by the 1990s, Jack Welch was a household name. The transformation had begun in the 1980s, when Lee Iacocca became a symbol of resurgent American capitalism, and accelerated as executives with only a few successful years were hailed as visionaries. Rakesh Khurana described the expectation succinctly: companies wanted their CEOs to be “corporate saviors.”

The language of individual creation made the belief concrete. Cisco’s John Chambers was said to have created “$300 billion in shareholder value,” as though one executive had alone built the company’s position and generated the rise in its stock price. The latter attribution had some basis in investor behavior: a Burson Marsteller poll found that 95 percent of investors said they would buy a stock based on what they thought of its CEO. Executive pay rose with the aura of exceptional personal agency, but the book distinguishes the real influence a chief executive may have from the much larger claim that one person can reliably direct a company’s fate.

The miracle worker’s double bind

The 1990s exalted CEOs while shortening their tenure. Chief executives who failed to improve the bottom line or deliver on promises could be removed quickly, even as golden parachutes softened their departures. The paired judgments—superhero and failure—were not contradictory accidents so much as two sides of the same expectation: leaders were treated as miracle workers, then dismissed when they did not perform miracles.

The book questions the premise that any individual can consistently forecast or choose well in conditions of genuine uncertainty. Researchers disagree about how much CEOs affect company performance at all; even those who believe their effect is substantial allow that it can be positive or negative. Jeff Skilling had a major impact on Enron, but that fact does not make his appointment a good decision. The difficulty is not simply identifying whether a CEO mattered. It is judging an executive apart from the company and circumstances in which decisions were made, especially when outcomes are hard to measure cleanly.

Available measures offer little support for a general presumption of exceptional executive judgment. About 80 percent of new products introduced in a given year did not survive their first twelve months. Corporate profit margins did not rise over the 1990s despite soaring executive compensation. Roughly two-thirds of mergers destroyed shareholder value: the acquiring company would have been better off not making the deal. Mergers are especially revealing because they are yes-or-no decisions, generally initiated by the CEO and approved by the board, with comparatively clear outcomes. Their frequent failure suggests that CEOs, in general, are not extraordinary decision makers.

The reversal from genius to failure

Gary Wendt’s changing reputation illustrates how readily a record is mistaken for a permanent personal quality. While running GE Capital under Jack Welch, he was regarded as the country’s smartest non-CEO. A journal described his mind as “as focused as a laser beam,” and the cash generated by his division made him seem like GE’s secret weapon. When he was hired to lead the troubled Conseco in 2000, he received $45 million on signing and the prospect of a $50 million bonus. Conseco’s stock tripled in his first year as investors waited for him to work his magic. Two years later, they were still waiting: Wendt resigned abruptly, Conseco went bankrupt, and its stock traded for pennies.

The same pattern appeared in the collapse of other once-celebrated executives. Kodak, Xerox, AT&T, and Lucent supplied further examples; Bernard Ebbers turned WorldCom from a small phone company into a global telecommunications giant, then saw it become bankrupt and notorious for the indictments of senior executives over cooked books. These reversals do not show that the executives became stupid. The book’s point is that they remained as smart and skilled at the end as at the beginning, but that skill was never enough to produce the right answer most of the time. Almost no one is that capable under uncertainty.

Sydney Finkelstein’s warning captures the problem of treating past performance as a dependable forecast: “CEOs should come with the same disclaimer as mutual funds: Past success is no guarantee of future success.” The book connects the warning to a broader habit of attributing success to innate ability rather than circumstance or chance. That attribution can sometimes be reasonable; in assessing corporate performance, it is dangerous because it encourages observers to read a favorable result as proof that the leader possessed a repeatable gift.

Uncertainty, luck, and the limits of attribution

Armen Alchian’s 1950 argument explains why a successful outcome cannot by itself establish superior foresight. In an economy with enormous numbers of people and companies striving to get ahead, success may arise from fortuitous circumstances. His illustration imagines thousands of travelers leaving Chicago, choosing destinations and routes at random, while only one road has a gas station. One traveler makes it out; that outcome does not mean the traveler knew more than everyone else, only that he took the road that happened to work.

Alchian was not claiming that most successful businesspeople are merely lucky or that skill is irrelevant. His point was narrower and more consequential: it is hard to know why a company has done well. The CEO-savior story turns that uncertainty into a simpler tale of personal mastery. It assigns a complex result to a visible individual, then treats success as evidence of genius and failure as evidence of incompetence. The book instead asks readers to distinguish capability from outcome and to leave room for circumstances that cannot be controlled or reliably anticipated.

What corporate intelligence requires

The critique of executive hero worship belongs to a larger account of how organizations gather and use knowledge. Decentralization can encourage specialization and independence, and people near a problem often possess tacit knowledge grounded in a particular job or experience. But dispersing decision-making also carries a risk: valuable information uncovered in one part of a system may never reach another. An organization needs both local knowledge and a way to make it collectively useful. The issue is not simply whether authority sits at the top or elsewhere, but whether the structure allows different contributions to be brought together without erasing their diversity.

The book makes this distinction through the U.S. intelligence community before September 11. Agencies had reasons to specialize: different perspectives could sharpen analysis, and information dismissed as irrelevant by one group might reveal an important connection to another. The failure was not decentralization itself, but decentralization without a mechanism to aggregate information and judgments. Richard Shelby favored a truly central intelligence agency, which might have made one agency responsible for the information, but would have risked sacrificing diversity, local knowledge, and independence. The proposed correction is not a single mind at the center; it is aggregation that lets a larger community’s knowledge inform decisions.

NASA’s response to the foam strike on Columbia shows another way hierarchy can silence the knowledge an organization already possesses. The Debris Assessment Team engineers had serious concerns and insisted that images of the Orbiter’s wing were needed for an informed analysis. Their concerns never received a serious hearing from the Mission Management Team. The meetings put a question from Linda Ham to one respondent at a time, but did not systematically invite other members to analyze particular issues. The resulting discussions were at once over-structured and under-structured: participation followed a narrow pattern, while the team failed to ensure that a broad range of members spoke. It also began with a conclusion and reinterpreted incoming information to fit it. The scene makes the corporate argument tangible: organizational intelligence depends on channels for independent analysis, not merely on placing a powerful decision maker above a group.

The hierarchy of the corporation can obstruct those channels as surely as the worship of a CEO can distort judgments about results. At one GM plant in Van Nuys, a supervisor showed two assembly-line workers the consequences of failing to install a sunshade bracket: repair workers had to tear out the carpet and weld the bracket into place. One worker, who had been doing the job for two years, had not known what the part she was welding was for. In another measure of the distance between authority and work, more than fifteen layers of management separated Ford’s chairman from a factory-floor supervisor, compared with five at Toyota. A chart reproduced in In Search of Excellence showed that a new product idea at an unnamed company had to pass through 223 possible paths in the bureaucracy before acceptance. Layers can therefore hide both the work itself and the knowledge that might improve it.

The Corporate savior ideal is consequential not because leaders are irrelevant, but because their prominence can obscure the conditions that make sound decisions possible. If executives are assumed to possess uniquely reliable judgment, organizations have less reason to cultivate dissent, share local knowledge, or aggregate the analysis of many people. The book’s alternative is not the fantasy of a leaderless company; it is an organization that does not confuse the person at its apex with the intelligence of the whole. In that view, exceptional leadership is neither guaranteed by past success nor a substitute for structures that let people see, question, and combine what they know.