Why Managers Still Matter

We live in a knowledge-based, rapidly changing, networked world. A supposed hallmark of the new economy has been the decline of managerial authority. Modern organizations such as online retailer Zappos have come to favor flat hierarchies with widely distributed authority. We call such businesses “wikified companies,” using the wiki– prefix to denote the loosely structured, bottom-up, egalitarian structure popularized by the Wikipedia encyclopedia project and touted by management thinkers and consultants.1

“Wikifying” the modern business has become a call to arms for some management scholars and pundits. As Tim Kastelle, a leading scholar on innovation management at the University of Queensland Business School in Australia, wrote: “It’s time to start reimagining management. Making everyone a chief is a good place to start.”2

Companies, some of which operate in very traditional market sectors, have been crowing for years about their systems for “managing without managers”3 and how market forces and well-designed incentives can help decentralize management and motivate employees to take the initiative.4 For example, in the 1980s, Johnsonville Sausage, based in Sheboygan Falls, Wisconsin, made a big point of slashing managerial oversight and putting quality control, personnel management, customer relations and even business expansion in the hands of worker-managed teams. As CEO Ralph Stayer explained, “My job … was to put myself out of a job.”5

The Industrial Revolution brought the decline of small-scale, cottage production and the rise of large, integrated businesses; Adam Smith’s invisible hand was replaced with what business historian Alfred D. Chandler Jr. famously described as the “visible hand” of management.6 But now that pendulum appears to be swinging in another direction. What Frederick W. Taylor, a pioneer in the field of scientific management in the late 19th century, saw as rigidly organized factories of docile and obedient workers has been eclipsed by loosely structured teams of highly trained and empowered knowledge workers.7 Indeed, the “visible hand” of management has morphed into a system of loose networks, virtual businesses and peer-to-peer interactions.8

This isn’t to say that strong-minded managers have become extinct. Many highly successful companies in recent years have been run effectively by powerful, opinionated figures. For many years at Apple, for example, CEO Steve Jobs made key decisions in a way that can only be described as dictatorial. Under Jobs, according to author Adam Lashinsky, “only one executive ‘owned’ a [profit and loss statement], and that was the chief financial officer.”9 From one perspective, this points to an extreme concentration of authority. But from another it is entirely consistent with the philosophy behind the wikified company: If functional executives don’t have to worry about a budget, they can focus on applying their own specialist knowledge to meet local demands.

The Lego Group, the Danish toy company, has recently moved to a more centralized business model. As Lego reduced its number of management layers in recent years, it expanded its layer of top management, bringing in functional specialists and moving senior managers much closer to operations. Thus, Lego is developing a new model: becoming both more wikified and more management-driven. Lego may be part of a broader trend. In a study about management hierarchies and compensation at 300 Fortune 500 companies over 14 years, researchers found that while companies were “de-layering,” the size of the executive team (defined as the number of positions reporting directly to the CEO) doubled from an average of five to 10. What’s more, executives were intervening more frequently in operating decisions.10 The result is counterintuitive: Flat management structures can have more micromanagement than vertical hierarchies.

Managerial Authority Is Here to Stay

From our perspective, the view that executive authority is increasingly passé is wrong. Indeed, we have found that it is essential in situations where (1) decisions are time-sensitive; (2) key knowledge is concentrated within the management team; and (3) there is need for internal coordination. (See “About the Research.”) Such conditions are hallmarks of our networked, knowledge-intensive and hypercompetitive economy.

However, for many everyday business activities, employees no longer need a boss to direct them to tasks or to monitor their progress. Such involvement can in fact be demotivating. In a networked economy characterized by dispersed knowledge residing inside the heads of highly qualified specialists, leaders need to let go of the notion that things should be managed from the top. This means that the definition of “authority” needs to change. Managers need to move away from specifying methods and processes, in favor of defining the principles they want people to apply or the goals they want people to meet. In other words, executives can design the rules of the game without specifying the actions of the players. Wikipedia cofounder Jimmy Wales doesn’t control the content of Wikipedia entries, for example, but he and his colleagues did design the structure — such as the format of the entries, the means by which they are revised and the procedures for resolving disputes — within which the enterprise operates.11

Authority doesn’t replace leadership, as some have argued; rather, it is a type of leadership.12 Besides established guidelines for rewards, instruction, rules and communication, something else is needed to help employees react to changes and act when unexpected events occur. For example, in emergency situations, employees don’t want to have to wait for permission from the boss to respond — they need to have a general understanding of “how things are done here.” Effective leaders excel at defining such frameworks. Letting organizational culture emerge and percolate on its own, without deliberate structure and design, can lead to a number of problems — not the least of which is a rough-and-tumble culture that favors certain employees at the expense of others.13

There are other aspects to the frameworks under which employees act and cooperate. Today’s employees are expected to multitask, acquire new skills and collaborate through knowledge sharing. At the same time, companies are made up of employees with increasingly diverse skills and work forces that are increasingly heterogeneous. Many multinational corporations must face an extreme version of this challenge. Some skills are so highly specialized and complex that managers don’t even know exactly what their employees can do.

The new environment suggests the need for a redefinition of the traditional managerial role. Despite all the changes that have occurred, there is a strong need for someone to define the organizational framework within which a business operates. We argue that, in the knowledge economy, the main task for top management is to define and implement these organizational rules of the game.

To be sure, procedures for defining rules and frameworks can themselves be delegated and nested. As management consultant Gary Hamel points out, “more and more of the work of managing and leading — the work of setting priorities, devising strategy, reviewing performance, divvying up work and allocating rewards — is going to be distributed to the edges of the organization.”14 But managers at the center will still maintain a pivotal role in determining both what and how to delegate.

A particularly important aspect of organizational design is what economists call “incentive instruments” — the formal and informal elements that motivate employees. Not only do these instruments have to be designed, they also must work together. A “soft” culture that encourages employees to collaborate may be difficult to align with strong performance incentives. Finding the right mix is crucial.

What’s more, in a fast-moving economy, managers must revisit the design of the company’s incentive instruments frequently because companies change their activities and organizational structures and add new employees with possibly different values. There may be a need to change incentive formulas — how much is tied to individual, team or company performance. Incentives may need to be strengthened or weakened. Lincoln Electric, a global manufacturer of welding equipment headquartered in Cleveland, Ohio, has maintained an elaborate system of performance incentives for more than 80 years. It may seem that the company’s reward system is so well designed that the company basically “runs itself.” In actuality, however, the system has been tweaked continually.15

There are economies of scale and learning economies in the task of adjusting incentives, which suggests it should be centralized. Boards of directors have compensation committees to fulfill this function; companies have managers. Further, when a company’s key assets are knowledge workers whose skills and behaviors are difficult to assess objectively, companies need to increasingly rely on more subjective assessments of performance, which must be carried out and enforced by managers.

When Managerial Authority Is Needed

In knowledge-based, networked economies, market advantage is often based on surprising competitors with new products, processes, organizational forms and business models. Apple’s iPhone, for example, redefined the smartphone category, a shock from which BlackBerry, Nokia and other once-dominant players subsequently struggled to recover. Similarly, taxi companies and hotel chains didn’t anticipate the sudden popularity of ride- and room-sharing services such as Lyft, Uber and Airbnb. Business history is packed with examples of companies that were blindsided: producers of tubes for amplifiers (surprised by the transistor), makers of minicomputers (surprised by the introduction and diffusion of the personal computer), vertically integrated steel producers (surprised by minimills) and so forth.

Many companies that survived major shocks to technology, regulation and global competition have had strong, charismatic leaders with highly authoritative styles. In what was arguably one of the greatest corporate comebacks of all time, Steve Jobs, faced with a major restructuring, rescued Apple by making tough decisions against considerable resistance (for example, cutting its Newton personal digital assistant platform and setting up a partnership with Microsoft). It is no coincidence that many of the best known successful corporate turnarounds are pinned to single individuals, such as Jobs at Apple, Sam Palmisano at IBM and Carlos Ghosn at Nissan.

Clearly, organizations use different approaches to make decisions. Executive teams and self-managing work groups often rely on dialogue and consensus. Dialogue can be followed by voting, as in cooperatives. Alternatively, group decisions can be made by decentralizing decisions as far as possible. Such approaches have their advantages. Dialogue and consensus may give employees a sense of empowerment and psychological ownership, and may produce decisions that commit. Decentralized decision making allows employees to make use of their own specialized knowledge without having to consult with their superiors. But, as we have noted, there are conditions where managerial authority is critical: when there is urgency to the decision making; when decisive knowledge is concentrated in the top management team; and when there needs to be tight interdependence between multiple decisions.

Decision-Making Urgency

Some decisions, such as company-level strategic decisions and those involving large transactions, must be made quickly. Even if the decisions turn out in hindsight to be suboptimal, making them rapidly may be necessary if the costs of delay are significant. Although companies may seek to increase decision-making speed through technology and open, interactive work environments, decisions that need to be made rapidly are still best left in the hands of one senior manager or a small handful of them. These individuals may possess crucial information about what needs to be done; having to take the time to share the information would delay and possibly undermine the decision.

The bottom line is that centralizing decision authority can often reduce the delays resulting from more collaborative and consensus-driven approaches. This seems consistent with the observed tendencies of organizations of all kinds (including armies and hospitals) to rely on centralized authority in cases of unforeseen emergency. More important, centralizing authority can also make better use of what we call “decisive” knowledge.16

In knowledge-based, networked economies, market advantage is often based on surprising competitors with new products, processes, organizational forms and business models.

Decisive Knowledge

Information is needed for making good decisions. The decision maker’s knowledge doesn’t have to be perfect, however, just good enough. Knowledge is good enough if the decision maker would make the same decision even if he or she had access to additional information — information that may be costly to obtain. The economic principle of weighing marginal benefits against marginal costs teaches us that manufacturing quality doesn’t need to be perfect, only good enough: beyond a certain threshold, eliminating manufacturing defects adds more to production costs than to sales revenues. Likewise, after a certain point, the cost of getting additional knowledge may be more than it is worth. For example, if sales in a particular market follow a regular, seasonal pattern, it may not pay to do an expensive demand-forecasting exercise; the additional information would add little to estimates based on historical data. In this case, the existing sales data is good enough to make a rational decision.

Understanding when knowledge is decisive is important for organizational design. Decision makers at the lower end of the organizational hierarchy typically have access to better information about conditions in local markets than do their supervisors. This argues for delegating authority to them. But if the supervisor already has enough information to make an informed decision — in other words, if the supervisor has decisive knowledge — then delegation may be harmful. Delegation allows for the possibility that the local decision maker will use the local knowledge to achieve his or her own goals at the expense of the overall corporate mission. If local knowledge is highly valuable, this risk may be worthwhile. Otherwise, it makes sense to allow those with decisive knowledge to make the call, even if others have marginally better information.

Interdependence Between Decisions

Different factors affecting a decision can be more or less interdependent. For example, accelerating the spread of electric vehicles requires having a number of things in place, including technical interface standards, deals with energy suppliers and well-located recharging stations. Coordinating the involvement of multiple parties takes lots of resources, which is why many companies often prefer to manage complex activities in house. They need things to be tightly coordinated, and working with outside parties can mean letting others have veto power over key decisions.

The economic theory of complementarities describes this kind of interdependence.17 When business activities, relationships, technology and regulation are complementary, it is hard to make incremental changes to the system: changing one element of the system requires changing others. Commercializing genetically modified products, for example, requires companies to invest in the underlying technologies. But it also requires complementary investments — in seed marketing, intellectual property rights protection, processing and delivery mechanisms that separate GMO and non-GMO products, and so on. Not surprisingly, given the need for tight coordination, the biotechnology revolution has required agricultural production and distribution to become more vertically integrated.

Senior managers must be able to adjust organizational structures to adapt to changing environments. By definition, this will involve changing job descriptions and decision-making authority.

Mixing Authority and Delegation

While companies cannot exist and thrive in a dynamic world without authority, they also need to develop mechanisms for decentralizing and delegating. Many companies have board-approved “delegation of authority” policies that specify the division of labor among the company’s leadership team and the board. However, delegation of authority over company assets, including employee time, can be pushed down in the organization. Authority over decisions can also be delegated by making the top leadership group larger. What is the right mix between authority held at the top level and delegated authority over the use of company assets?

Our framework provides part of the answer. (See “How to Mix Authority and Empowerment.”) If fast decision making is essential, the knowledge held by senior managers is highly decisive and activities are highly interdependent, then centralized authority is necessary. Decision-making authority should be in the hands of the CEO and the extended leadership team.

Determining the right size of the leadership team has been the subject of much debate.18 The average large U.S. company has a management team of 10. However, as the number grows, it’s common for factions to form and tension to arise.19 Although Lego thrives with a top management team of more than 20, many of the team members are located in different geographical locations. Fault lines may be less likely to form within a distributed team.

The psychological forces that limit the size of top management teams are less relevant for how decision-making authority should be distributed in the organization. After all, friction between divisions or business units is fairly common and does not necessarily cause problems. However, there are psychological forces that managers should pay attention to.

First, employees develop feelings of entitlement. They come to believe they really do “own” the budgets and accounts they control. Experimental research suggests that individuals attach particularly high valuations to what they own (or believe they own).20 Shifting decision-making authority from middle managers to the top management team may frustrate managers who feel something has been “stolen” from them. However, in dynamic, highly competitive economies, senior managers must be able to adjust organizational structures to adapt to changing environments. By definition, this will involve changing job descriptions and decision-making authority.

Second, no one likes to be overruled. Although everyone knows that senior managers can overrule middle managers, prudent senior managers will intervene only when strictly necessary. Concern about being overruled can cause middle managers to question what authority they really have. Frequent meddling by superiors without an obvious reason can make middle managers anxious and less inclined to act. Research shows that overruling by senior managers harms motivation and productivity.21

To be sure, in a dynamic business environment some amount of overruling is to be expected. It should be minimized, however. For senior managers, this means accepting some decisions even without a detailed justification rather than investing time and effort to gain all the necessary information. If employee decisions are to be overruled, procedural justice concerns are all-important: Employees need to be told carefully and patiently why their decisions are being reversed.

Third, if top management holds too much decision-making authority, it is difficult to harness employee-specific knowledge and motivate employees to make good decisions. But the presence of dispersed, valuable, tacit knowledge is not a trump card that counters all the potential advantages of centralization, even in the wikified economy.

Management practice has always been closely intertwined with broad societal trends. Anti-authoritarianism has been a strong force in many parts of the world since World War II, reinforced by the counterculture of the 1960s. We think the “wikified company” trend reflects a similar distrust of authority and a belief that treating employees as responsible, autonomous and intelligent individuals pays off. This view tracks a more general sense that, thanks to the Internet, mobile telephony and social networks, we are in a new age of collaboration, participation and sharing. Vertical relationships are being replaced by horizontal ones, and hierarchy is giving way to what Harvard Law School professor Yochai Benkler calls “commons-based peer production” not only in business but also in culture, politics and society at large.22

We sympathize with this ethos, which in a number of ways is an improvement over the paternalism of an earlier age. Managers may want to embrace it, not just to be in accordance with the zeitgeist but also because it is likely to pay off, since it leads to a better use of employee knowledge and increases the motivation of those employees who are skeptical of authoritarian management. That said, managers must be careful not to understate the continuing rationale and benefits of managerial authority, even in the wikified economy.

Much innovation activity is shifting toward business model innovation. For example, in the face of declining opportunities for future blockbusters, many big pharmaceuticals companies are trying to remain competitive by developing new business models with a heavy service dimension. IBM’s 2006 Global CEO Study revealed that CEOs saw business model innovation as the most important kind of innovation, and there is no indication that it has become any less important.23 Business model innovation involves strategic and organizational change processes in which decision-making urgency, decisive knowledge and tight interdependence are often key. When such innovation takes a company into uncharted territory, there is massive uncertainty. This only elevates the importance of senior manager intuition and foresight. The top management team not only must define the direction but also needs to get involved in the everyday experimentation and decision making that make the new business model come alive. Many decisions cannot be delegated because so much of the essential knowledge and initiative are concentrated at the top echelons.

For all of these reasons, we think the reports about the death of managerial authority are greatly exaggerated. In today’s world, managerial authority needs to be exercised smartly. Talking about an ethos of participation but acting in ways that are in direct contradiction is a recipe for disaster. Managerial authority still works — but it needs to be accompanied by strong, credible communication and a commitment to fair procedure.