A r c h i v e d  I n f o r m a t i o n

Systemic Reform: Perspecitives on Personalizing Education--September 1994

Why Centralized and Decentralized Strategies Are Both Essential

I have provided some evidence that neither top-down nor bottom-up strategies, by themselves, are effective, but it is necessary to probe deeper by asking why they do not work.

Top-down strategies are problematic because you can't complex change processes from the top. Senge (1990) calls it "the illusion of being in control".

The perception that someone "up there" is in control is based on an illusion - the illusion that anyone could master the dynamics and complexity of an organization from the top (p. 290).

More fundamentally, the forces of educational change are so multifaceted that they are inherently unpredictable. As I stated elsewhere, change is non-linear and complex:

How is change complex? Take any educational policy or problem and start listing all the forces that could figure in the solution and that would need to be influenced to make for productive change. Then, take the idea that unplanned factors are inevitable -government policy changes or get constantly redefined, key leaders leave, important contact people are shifted to another role, new technology is invented, immigration increases, recession reduces available resources, a bitter conflict erupts, and so on. Finally, realize that every new variable, that enters the equation - these unpredictable but inevitable noise factors - produce ten other ramifications, which in turn produces tens of other reactions and on and on (Fullan, in press)

Controlling strategies do not work because there is too much to control. Even strong leadership, and vision-driven change is seriously flawed, because things are constantly changing. We need a new mind-set to manage situations of constant flux (see Fullan, in press, Beer, Eistenstat & Spector, 1990, and Stacey, 1992).

Given these difficulties of attempting to control change from afar, it is understandable that local participation and site-based management appears to many to be the solution. Yet this alternative is also fraught with fundamental deficiencies. First, there is ample evidence that organizations in general, are not likely to initiate change in the absence of external stimuli. Schools in particular are not known for their innovativeness.

Second, we saw in the previous section, when schools do have the opportunity to control the change process (through site-based management, for example), they do not necessarily take productive action. They are more likely to get bogged down and/or make superficial structural changes.

Third, in decentralized systems it is difficult to discern let alone maintain quality control (on the other hand, accountability fares no better in centralized systems).

Fourth, one could speculate, that it is possible for a given school to become highly innovative, despite the district it is in. I would venture to add, however, that is not possible for such a school to stay innovative despite the district. District action or inaction - personnel transfers, hiring decisions, budget decisions, and the like - inevitably take their toll.

Stacey (1992) summarizes the problem of decentralized decision-making:

The whole point of flexible structures and dispersed power is to enable those below the top level in the management hierarchy to detect and take action to deal with a large number of changes affecting an organization that operates in a turbulent environment. This is supposed to enable the organization to learn about its environment and so adapt to that environment faster than its rivals do. However, studies have shown that widening participation and empowering people by no means guarantees that organizational learning will improve (p. 175).

When two alternative positions - opposite solutions really - are both found to be basically flawed, it normally means that a paradox lies behind the problem. What is required is a shift of mind-set from either/or to both/and thinking. Beer et al (1990) summarize the situation.

The top-down approach possesses some allure. It holds the promise of producing rapid change toward an elegantly conceived end state that is symmetrical and complete. Thus, managers can lead their employees in the desired direction. But the unilaterally directive approach also has traps into which renewal can fall. Employee commitment to the newly aligned organization may be low, and employee knowledge of how things get done in the organization may not be considered in the solution.

A bottom-up approach that allows, even demands, participation by employees seems to address many of the failings of unilateral top management direction. But it can suffer from a different set of problems. A participative approach to change may be too slow and ill defined to respond effectively to short-term business demands. It presents top managers with the problem of how to incorporate their perspective and knowledge into new solutions. It raises questions about the motivation and skill of employees to develop an ambitious solution that will "force" them, the employees to change their ways. Even worse, participative approaches to change can be derailed by resistant managers, unions, and workers.

Our examination of revitalization efforts in 26 plants and business units across the six companies reveals that effective renewal occurs not when managers choose one alternative or the other. Instead, effective revitalization occurs when mangers follow a critical path that obtains the benefits of top-down as well as bottom-up change efforts while minimizing their disadvantages (p. 68-69).

Pascale (1990) draws a similar conclusion in examining the turnaround at the Ford Motor Company in the 1980s:

Change flourishes in a 'sandwich'. When there is consensus above, and pressure below, things happen. While there was no operational consensus at the top as to precisely what should be done at Ford, the trips to Japan caused many senior managers to agree that the problems lay in the way the organization worked. This might not have led anywhere, however, were it not for pressures for change coming from the rank and file (p. 126, 128 emphasis in original).

Finally, research on effective and collaborative schools shows that such schools do not go it alone, but are actively part of a wider network in which external and internal to the school influences are equally important. Collaborative schools, for example, are more likely to seek outside ideas, more likely to be linked to their districts, and more likely to engage state-level policies proactively (Louis & Miles, 1990, Rosenholtz, 1989, Nias, Southworth & Campbell, 1992). Baker et al's (1991) study of 48 school districts in Illinois confirms that internal development and external involvement must go together. Thirteen of the 48 districts were classified as engaged in "systematic improvement" on a sustained basis. It is no accident that all 13 successful districts were found to be users of external support from regional educational service centers and several other sources. By contrast in all 8 cases that had no external support there was no evidence of school improvement. Time and again we find that seeking external support and training is a sign of vitality.

External linkages do not necessarily involve connections to a hierarchical center, but in effective systems they include two way interaction with those in authority. Thus, the boundaries between external and internal systems, between top-down and bottom-up levels become effectively permeable and mutually influential in successful organizations.


[Coordinating Top-Down and Bottom-Up Strategies for Educational Reform] [Table of Contents] [Coordinating Top-Down and Bottom-Up Strategies]