When people talk about decentralization and empowerment, they usually mean delegating decisions to a lower level in a hierarchy. But what if power isn't delegated to lower levels but instead originates there? How much energy and creativity might be unlocked if all people in an organization feel in control? This more radical kind of decentralization may become more important in the future.
In a free market, for instance, no one at the top delegates decisions about what to buy or sell to the different players. Instead, a buyer and a seller can exchange almost anything on which they mutually agree (subject to their financial constraints and their abilities). The marvel of how overall coherence emerges from these countless decisions between two parties is what Adam Smith called the "invisible hand" of the market.
A similar kind of radical decentralization comes from the notion of subsidiarity. This principle of social philosophy, derived from Roman Catholic teaching, holds that any task should be performed in the smallest possible unit: for example, at the local level before the regional level, and at the regional level before the national level. The principle of subsidiarity is increasingly viewed as desirable for political organizations (for example, in the European Community) and in business organizations.
In essence, this principle turns the whole notion of delegation upside down. Instead of all legitimate power being derived from the top of an organization and delegated down, all legitimate power originates at the bottom and is delegated up only when there are benefits in doing so.
To understand these ideas more precisely, consider the following two dimensions of centralization: (1) Who makes the most important decisions?, and (2) Who can overrule the decisions made by others? We can use these two dimensions to develop a decentralization continuum ranging from highly centralized systems at one end to highly decentralized systems at the other (see figure 3.3). (Of course, it is possible to have other combinations of these two factors, but most real-world examples fall somewhere along this continuum.)
Figure 3.3: The Decentralization Continuum
In highly centralized systems, for example, central decision makers make most of the important decisions and can overrule most other decisions that they delegate to local decision makers. Traditional military organizations, for example, embody an extreme form of centralized control; high-ranking officers make all important decisions for their troops and can overrule their subordinates' most trivial decisions. On the other hand, in highly decentralized systems, local decision makers make most of the important decisions and can overrule most decisions that they delegate to the central decision makers. For instance, Internet users can communicate however they want with other users, as long as their computers follow the standard protocols (or interconnection procedures) that the Internet governing boards have approved. A subgroup of users can even create a new protocol.
In between these two extremes are mixed systems, in which both central and local decision makers make some important decisions, and in which each makes certain decisions that cannot be overruled by the other. The U.S. Constitution, for instance, spells out a mixed system of relationships between the federal government and the state governments in which each has certain important powers that the other cannot overrule.
Most discussions about empowerment stop halfway, at the middle of the decentralization continuum. By definition, you cannot empower someone unless you have the right to make or overrule the decisions you are delegating. But radical decentralization is not something that people at the top do for people at the bottom; it is something that starts at the bottom.
Visa International An example of a radically decentralized organization in which the ultimate power rises from the bottom is Visa, whose users are its owners. Dee Hock, Visa founder, calls this company an "inverted holding company". Rather than one company that owns numerous others, Visa is a company owned by the banks and other institutions that issue Visa cards. They are simultaneously its owners and its customers. In many cases, they are also its suppliers.
The Visa organization was consciously designed as a "federal" system and includes a series of regional, national, and international organizations, each with its own members and board of directors. Each organizational level receives its power from the levels below rather than from above. Decisions are made by votes at the various board levels, typically with a sixty to ninety-day cycle for an issue to pass through all levels. For instance, Visa members have voted on a service charge to themselves for all Visa transactions and certain other transaction fees for processing services, if they choose to use them. However, the member organizations are free to use any Visa product, to leave the whole Visa organization if they so choose, and to offer competing products. (In fact, most banks offer the primary competing product, Mastercard.)
While such a highly decentralized structure would be inefficient for some purposes, it has been extremely successful for Visa. In the approximately twenty years since its founding, Visa has become a global organization with more than 23,000 member institutions, 300 million customers, and a $650 billion annual sales volume. More important, this decentralized structure has been able to provide essential centralized services (such as a global transaction clearinghouse and global brand management) for members that are, in many cases, direct competitors. From this fundamentally decentralized structure, therefore, a very successful global organization has emerged.
The Internet An even more extreme example of a decentralized structure is the Internet, which has no ownership relationships at all. The Internet has been doubling in size every year since 1988 and now has more than 30 million users. Clearly the Internet (or communication networks like it) will have a profound effect on how electronic commerce—and business, in general—is conducted in the next century. But in addition to being a technological enabler of other organizations, the Internet's technical architecture and governance structure themselves provide models for structuring highly decentralized organizations.
For example, no one is in charge of the Internet, and everyone is. No one, for instance, can unilaterally decide to shut the Internet down or deny access to any particular person or organization. Instead, anyone who follows the agreed-on rules for communicating on the Internet can connect to any other node and thus be connected to the entire worldwide network. And anyone who is connected to the network can be a service provider or a service user connected to anyone else. In the rare cases when Internet protocols need to be changed, a combination of elected and volunteer boards approves them. In addition, any group of users that wants to experiment with new protocols is free to do so. In fact, new protocols that the Internet boards adopt are generally accepted only after they have been widely— and successfully—used in experiments. The role of the "center" (that is, the standards boards) is simply to establish the framework through which its "members" interact—not to tell them what to do.
It is easy to imagine Internet-like principles being used in other kinds of organizations. For instance, even though global consulting firms such as McKinsey & Company are mixed organizations overall, they have structures that use some Internet-like principles. McKinsey has established a strong organizational culture that includes norms about selecting and promoting people and expectations for working with others. But its management does not tell individual partners what kind of work to do, which clients to work for, or which people to select for their teams. Instead, the partners make largely autonomous decisions about what they will do and how they will do it. When it works well, this highly decentralized effort—within the overall interaction framework that the firm provides—results in an extremely flexible, global organization.
See, for example, Handy 1992.
Interestingly, even military organizations are now moving away from this extreme form of centralization. See, for example, Smith 1994.
See, for example, Breuner 1995; Nocera 1994.