One of the hottest ideas in school reform right now is decentralization. Its supporters blame school district bureaucracies and the rules they spawn for much of what is wrong with our schools. They are convinced that if we decentralize school districts, we will get responsive, high-achieving schools.
According to the critics, bureaucrats and bureaucratic rules create a one-size- fits-all approach to teaching and learning, even though different neighborhoods and communities want different things for their students and students vary in the kind of education they need. As for the people who work in these big, centralized systems, they are rewarded for following rules and regulations rather than getting results with students.
Decentralization is touted as the panacea for these problems. Its supporters tell us that it loosens the grip of bureaucratic rules and empowers people in a community to decide what kind of schools they want and to make sure they get them. And in decentralized schools, principals and teachers don't have to spend a lot of time and energy jumping through bureaucratic hoops. So they are better able to fulfill the real mission of the schools--helping students learn.
All of this sounds pretty convincing, but we don't have to take it on faith. New York City has just reached the end of a 25-year experiment that allows us to test many of these assertions about decentralization. In the late 1960s, New York City's system of 1,000 schools was broken up into 32 community school districts. The complaints that led to this breakup sound very familiar. People said that central office administrators didn't know or care what was going on in local schools or communities. They said that these bureaucrats were in no way representative of the diverse groups in the city, and they were not doing a good job of educating the kids. Decentralization was the solution of choice. Its supporters figured that when the system was broken up and each community could elect its own board, hire its own administrators, and set up its own curriculum, people would at last have schools tailored to their own needs and desires. As a result, student achievement would soar. Besides, they said, nothing could be worse than the current system.
What happened bore no resemblance to what supporters had hoped for or expected. Over the years, some New York City community school boards became legendary for their corruption and indifference to academics. Of course there were districts where superintendents and community school boards worked together and achieved good things for their students. But in many others, the board functioned chiefly as a source of patronage and jobs. Board members routinely handed out jobs to relatives and friends or sold them for money or drugs. Even board members who may have been honest were often ignorant of the basic facts about their schools. When the New York Daily News interviewed 236 of the 288 community board members, it found that a majority did not know how many students they had in their district, what percentage of students were reading and doing math at grade level or the amount of their district budgets--even though they had just voted on the budgets. Why didn't communities get rid of these crooks and incompetents? Because it is easy to get reelected when the turnout for an election is under 10 percent. As for student achievement, it did not improve, and it may have gotten worse.
A couple of months ago, the New York State legislature decided it was time to return a good deal of the power that had been in the hands of community school boards to the chancellor of the New York City schools.
Does the New York experience refute the current wisdom that decentralization is good and centralization is bad? Should we be opposing, on principle, the push for charter schools and other similar schools run by businesses or community groups, which are largely independent of school districts? Not necessarily. Decentralization may work when certain things are highly centralized--when there are rigorous, grade-by-grade academic standards set at a state or national level, assessments based on those standards, and accountability for the way public money is spent, to name a few. The Japanese education system demonstrates this very well. But decentralization can't work without at least these basics in place.
Supporters of decentralization assume that we will improve student achievement by changing the way schools are governed, but where is the evidence? Decentralizing the New York City system did nothing to improve student achievement. Kentucky's decentralized schools yielded such poor student achievement that, when the system was overhauled several years ago, state-defined standards and assessments were introduced. The American education system is already the most decentralized in the industrialized world, and the result is a system in which some kids get a pretty good education and others get garbage--often in the same school district. Some panacea.