Trying Decentralization for Real This Time
There seems to be a flurry of new interest, especially among state and local officials, in school-based budgeting and its natural corollary, decentralizing decisionmaking authority to schools. ("Weighty Decisions," Jan. 06, 2005.) In June, for example, the Chicago public school system announced that, beginning with the new school year, the district will give 85 of its schools more autonomy over some instructional and financial decisions.
We could imagine the reaction of many educators and district officials to this news: “Been there, done that in the ’80s and ’90s with school-based management. Didn’t work then. Won’t work now.”
They are wrong, however, at least about whether the United States has really tried decentralizing decisionmaking authority to the school level (we call it decentralized decisionmaking for schools, to distinguish it from school-based management). And until we really try it, we can’t possibly know whether it will accomplish what its new advocates hope: improving student achievement by empowering school-level administrators to make key decisions about budgets and resource use.
It is true that many public schools and districts initiated reforms under the school-based-management banner during the latter part of the last century, yet the performance of American education during that period improved at best only modestly. School-based management did not result in major changes in educational practice. Bruce Bimber of the RAND Corp. pointed out a decade ago that the reason decentralization had failed to accomplish much could be either its assumption that institutional structure (such as where decisionmaking authority is located) can affect educational outcomes is wrong, or the attempted decentralization efforts never produced real changes in institutional structure in the first place. His research, and findings from numerous other studies, support the second explanation.
There are four ways in which school-based management failed to change institutional structure.
First, it created only minimal changes in authority. Under school-based management, constraints on schools were not effectively relaxed, and the rule-based operating environment in most schools changed little. Fragmentary approaches to decentralization meant that the effect of eliminating constraints in one area was nullified by continuing constraints on other decisions. Decentralization had limited effects because it treated functions (budget, personnel, instruction, and operations) as separable and kept many of them under central management. Since many decisions are in fact highly interdependent, authority ostensibly granted in one area was inherently limited by policies or practices in others. This was so much the case that some schools in Montgomery County, Md., actually withdrew from the district’s school-based-management plan, claiming they had lost rather than gained freedom of action.
A second shortcoming of school-based management was its lack of focus. Widely different plans created fuzziness about what the concept of decentralization meant and who (principals, unions, parents, teachers, school boards, or superintendents) was in charge of what. The goal of improving student learning—the centerpiece of decentralized decisionmaking for schools—was confounded under school-based management with other objectives, such as the democratization of decisionmaking. For example, highly visible decentralization efforts in the United States (New York City in the late 1960s, Chicago in the late 1980s) were motivated as much, or more, by a political goal of stimulating new neighborhood-based political activity as by an educational goal.
By contrast, Paul T. Hill has pointed out that a strategy of decentralized decisionmaking for schools aimed squarely at raising student achievement finds the justification for school autonomy in a theory of comparative advantage: Teaching and learning occur at the school level, and school professionals are in the best position to organize teaching, adapt to the present needs of children, and set priorities on the use of child and adult time.
A third shortcoming of school-based management was the absence of the supportive infrastructure that Australian educator Bruce J. Caldwell found was essential to successful decentralization.
Based on 25 years of experience with effective and ineffective decentralization efforts, Caldwell argues that real decentralized decisionmaking requires “consistent decentralization to the school level of authority and responsibility to make decisions on significant matters related to school operations within a centrally determined framework of goals, policies, curriculum, standards, and accountabilities.”
Finally, school-based management wasn’t integral to district strategy. Experience shows that decentralization is likely to be effective only if it is the centerpiece of a district’s reform effort and if it embraces the full range of decisions affecting resource use. While many U.S. school districts claim to have pursued decentralization over the past 20 years, for all but a handful, it was one among many reforms implemented simultaneously—some of which actually required greater centralized decisionmaking.
School-based management thus failed to make significant structural changes because its implementation was generally fragmentary and incomplete. But despite its notable failure in the last round, the theoretical reasons for pursuing decentralization remain as powerful today as when they were first developed. Public education systems are bureaucracies—and bureaucracies, with their codified rules and standardized procedures specifying how work is to be done, are increasingly seen as problematic in environments, such as schools, with shifting circumstances, unpredictable client needs, and a lack of clarity on the links between tasks and desired outcomes. Bureaucracies discourage creativity and innovation and encourage members to focus on compliance with rules.
Much of the interest in decentralization among educators was inspired by the experience of business firms that, when faced with unprecedented levels of global competition in the late 20th century, dramatically reformed their traditionally top-down structures in imitation of so-called “quality” approaches to management. Organizational and management scholars cited both theoretical and empirical evidence to support claims that decentralized organizations perform at higher levels than centralized ones.
In addition to the traditional reasons for thinking decentralization would be good for schools, recent analysis of the nature of work in the 21st century suggests that if education remains top-down and rule-bound, it will have an increasingly difficult time competing for talented leaders, who are likely to prefer (and to be able to find) jobs with more flexibility and autonomy. A new RAND study predicts that the forces of technology and globalization will persist in changing the nature of business organizations, which continue to evolve away from command-and-control leadership styles to decentralized management.
Will today’s decentralization efforts face the same defeating forces that trivialized school-based management? On some fronts, the answer is certainly yes. But there have also been some changes in the context of education that set the stage for more realistic attempts at effectively decentralizing decisionmaking authority to schools.
The standards-based-reform movement and the creation of new federal and state accountability systems have resulted in a crucial context for decentralized decisionmaking to schools that was previously lacking. The advantages of moving decisions downward in an organization are unlikely to be realized unless the front-line decisionmakers have agreed-upon goals, clear performance incentives toward these goals, and good systems of accountability. Earlier school-based-management efforts lacked these conditions. While strong performance incentives are arguably still lacking, there has been clear movement in the direction of creating the other enabling conditions related to goal-setting (via standards) and accountability, including vastly expanded data systems providing information on student achievement at the school and classroom levels.
Moreover, the growth in school choice options for parents creates for at least some public schools a new competitive environment that serves implicitly as a performance incentive. The sanctions specified under the federal No Child Left Behind Act for persistently underperforming schools, for example, require districts to consider turning these schools into charter schools, which are by definition more autonomous than traditional public schools. These developments may make districts more willing and/or comfortable than they have historically been with providing more autonomy to school-level personnel.
Finally, education finance efforts at the federal, state, and local levels are laying the groundwork for decentralized decisionmaking by exploring school-based, rather than district-based, allocation formulas. The federal government’s Title I program allows for school-based decisions about how compensatory education funds are used. Some states and districts are considering or using so-called weighted-budget allocation policies that focus on student and school needs, rather than on specific educational inputs that are distributed according to district-determined formulas.
That said, it is important to emphasize that most districts and schools are currently not ready for decentralized decisionmaking, an important caveat for policymakers who might in their newfound enthusiasm be tempted to insist on its quick implementation. New budgeting and tracking systems would need to be developed. District personnel would have to learn how to use these systems and be helped to shift from a command to a service mentality. Principals would have to develop new skills to exercise their new roles as both resource allocators and instructional leaders of their schools. Every district moving toward decentralized decisionmaking for schools would have a host of operational decisions to make.
Nevertheless, the current interest in this concept offers a new window of opportunity for testing seriously this time the proposition that real improvement in student learning can be achieved within the public school system by radically altering the locus of decisionmaking and shifting authority over key decisions to the school level. Given the lack of success to date in moving public education to significantly higher levels of performance, it might be time to give decentralized decisionmaking for schools a fair trial.
Vol. 24, Issue 44, Pages 40, 52Published in Print: August 10, 2005, as Trying Decentralization For Real This Time