coordination. IBM lost an early lead in the personal computer business in part because new initiatives required so many approvals—from levels and divisions alike—that new products were overdesigned and late to market. The same problem hindered Hewlett‐Packard's ability to innovate in the late 1990s.
Too Loose Versus Too Tight
Another critical structural dilemma is how to hold an organization together without holding it back. If structure is too loose, people go astray, with little sense of what others are doing. But rigid structures stifle flexibility and encourage people to waste time trying to bolster or beat the system.
We can see some of the perils of a loose structure in the former accounting firm Andersen Worldwide, indicted in 2002 for its role in the Enron scandal. Efforts to shred documents and alter memos at Andersen's Houston office went well beyond questionable accounting procedures. At its Chicago headquarters, Andersen had an internal audit team, the Professional Standards Group, charged with reviewing the work of regional offices. Unlike other large accounting firms, Andersen let frontline partners closest to the clients overrule the internal audit team. This fostered local discretion that was a selling point to customers, including Enron, but came back to haunt the firm. As a result of the lax controls, “the rainmakers were given the power to overrule the accounting nerds” (McNamee and Borrus, 2002, p. 33). The august firm collapsed as a result.
The opposite problem is common in managed health care. Insurance companies give clerks far from the patient's bedside the authority to approve or deny treatment or to review medical decisions, often frustrating physicians and patients. Doctors lament spending time talking to insurance representatives that would be better spent seeing patients. Insurance providers sometimes deny treatments that physicians see as urgent. In one case, a hospital‐based psychologist diagnosed an adolescent as likely to commit sexual assault. The insurer questioned the diagnosis and denied hospitalization. The next day, the teenager raped a 5‐year‐old girl.
Goal‐less Versus Goal‐bound
In some situations, few people know what the goals are. In others, people cling intently to targets long after they have become irrelevant or outmoded, like the Japanese soldiers who hid on islands in the Pacific for decades after World War II, unaware that the war had been over since 1945. Conversely, the Salk vaccine virtually eradicated polio in the 1960s. This lauded medical breakthrough also brought to an end the existing strategy of the March of Dimes organization, which for years had championed finding a cure for the crippling disease. The organization rebounded by shifting its strategy to focus on preventing birth defects.
Irresponsible Versus Unresponsive
If people abdicate their responsibilities, performance suffers. However, adhering too rigidly to policies or procedures can be equally harmful. In public agencies, “street‐level bureaucrats” (Lipsky, 1980), who deal with the public, are often asked, “Could you do me this small favor?” or “Couldn't you bend the rules a little bit in this case?” Turning down every request, no matter how reasonable, alienates the public and perpetuates images of bureaucratic rigidity and red tape. But agency workers who are too accommodating create nagging problems of inconsistency and favoritism.
Greatest Hits from Organization Studies
Hit Number 5: Michael C. Jensen and William H. Meckling, “Theory of the Firm: Managerial Behavior, Agency Costs, and Ownership Structure,” Journal of Financial Economics 3 (1976), 305–360.
This classic article, fifth on our list of works most often cited by scholars, focuses on two central questions:
What are the implications of the “agency problem”—that is, the conflicts of interest between principals and their agents?
Given those conflicts, why do corporations even exist?
An agency relationship is a structural arrangement created whenever one party engages another to perform a task. Jensen and Meckling's particular focus is the relationship between a corporation's owners (shareholders) and their agents, the managers. Principals and agents both seek to maximize their own interests (utility in economics‐speak), but their respective interests often diverge. If you are a sole proprietor, a dollar of the firm's money is a dollar of yours as well. But if you are an employee with no ownership interest, you're spending someone else's money when you pad your expense account or schedule a business meeting at an expensive resort.
One rationale for linking executive compensation to the price of the company's stock is that it may reduce the agency problem, but the impact is often marginal at best. A notorious example is Tyco's chief executive, Dennis Kozlowski, who reportedly spent more than $30 million of company money to buy, furnish, and decorate his palatial apartment in New York City (Sorkin, 2002). Nonexecutive shareholders hate this kind of thing, but it is difficult for them to stay abreast of everything management does, and they can't do it without incurring “monitoring costs”—time and money spent on endeavors like supervision and auditing.
One implication the authors draw is that the primary value of stock analysts is their sentinel function. Their ability to pick stocks is notoriously poor, but their oversight puts heat on managers to serve shareholder interests. The article also concludes that, despite agency conflicts, the corporate form still makes economic sense—managers cost more than owners wish, but they still earn their keep.
The agency problem is a persistent feature of cooperative activity. Relationships between a team and individual members, or between a boss and a subordinate, is like that between principal and agent. If members of a team share rewards equally, for example, there is an incentive for “free riders” to let someone else do most of the work. Principals face a perennial problem of keeping agents in line and on task.
STRUCTURAL CONFIGURATIONS
Structural design rarely starts from scratch. Managers search for options and images among the vast array of possibilities drawn from their accumulated wisdom and experiences of others. Templates and frameworks can offer options to stimulate thinking. Henry Mintzberg and Sally Helgesen offer two abstract images of structural possibilities.
Mintzberg's Fives
As the two‐dimensional lines and boxes of a traditional organization chart have become increasingly archaic, students of organizational design have developed a variety of new structural images. One influential example is Mintzberg's five‐sector “logo,” depicted in Exhibit 4.1. Mintzberg's model clusters various functions into groupings and displays their relative size and influence in response to different strategies and external circumstances. His schema provides a rough atlas of the fundamental terrain that can help managers get their bearings. It assists in sizing up the lay of the land before assembling a structure that conforms to prevailing circumstances. One of the distinctive features of Mintzberg's image is expanding the typical two‐dimensional view of structure into a more expansive rendering that provides a sharper image of the intricacy and issues of organization design.
Exhibit 4.1. Mintzberg's Model.
Source: Mintzberg (1979, p. 20). Copyright ©1979. Reprinted by permission
