Four years ago I was a New York City Teaching Fellow, taking graduate education courses at CUNY’s Lehman College and teaching high school English. I wrote the following essay for a class in American education history. Today, the subject of charter schools dominates the national conversation on public education, with the federal government calling for more of them in the name of “reform.” But if there is one thing we can learn from the past, it is that simply changing the organizational structure of a school system does not guarantee improvement. School officials need to work in tandem with teachers, parents and community leaders to best determine the needs of our students.
Decentralization v. Centralization In New York City Public Schools
by Patrice D. Johnson, August 8 2006
The debate over centralization vs. decentralization of public schools is an old one that the United States and other nations have grappled with for as long as public education policies have been in existence.
The pendulum swings back and forth on this question, and which way the pendulum swings depends upon the prevailing political winds at the time. However, one thing seems certain, in the United States there is no political support for complete centralization of school systems these days. The trend seems be more and more toward decentralization up to the point of funding independent schools or private schools with public money.
According to most studies on the issue, advocates for school decentralization believe that it brings about better relations between school and community, that it provides more efficient maintenance and support for local schools, that it provides more resources for local school needs, and that it responds more effectively than large bureaucracies to the widely varying needs of local schools and communities.
Opponents of decentralization say that schools need a central office to act as an intermediary in their relationship with the state. Decentralization foes point to a lack of accountability on the part of the people who head the decentralized units. They say a decentralized system is prone to fraud. Dealing with corruption and fraud is usually the reason school systems decide to centralize.
A year after New York City Mayor Michael Bloomberg entered office in 2002, he took control of the school bureaucracy and began dismantling the 32 local school boards that had been in place for over 30 years, and he consolidated the system into 10 instructional regions. Observers in the media and academia saw this as a move away from decentralization toward “recentralization” (Keane).
Actually, Bloomberg was just finishing the process begun under his predecessor, Rudolph Giuliani, of removing power from the local school boards and transferring it to the central office (Berger; Kifner). The reaction from some teachers who had worked under the old system was not always enthusiastic. Anecdotes abounded about how Bloomberg and his Schools Chancellor Joel Klein were “micromanaging” the system, even to the point of dictating how teachers should decorate their bulletin boards.
It wasn’t the first time the city moved to greater centralization. A central education system was created for New York’s five boroughs in 1902. But, until the mid-1960s, for example, “public school principals could expel high school students with broad discretion, without due-process hearings to ensure fairness. And public school principals used to hire their own teachers, write paychecks, purchase supplies, and even hire contractors. So more centralization was imposed to reduce fraud and improve accountability” (Cortines).
More recently, Bloomberg, in his second term of office, seems to be rethinking centralization. In January, it was announced that New York City would increase the number of schools in its autonomy zone from 58 to more than 200 (Jones). Already, teachers who work in some of these autonomous schools in the city have expressed concerns about accountability.
To understand Bloomberg’s actions, one has to go back to 1969, when New York passed a decentralization law that led to the creation of 32 semi-autonomous, locally elected community boards. In the 1960s New York City, like Newark and other urban centers, was experiencing “white flight.” In many communities schools were de facto segregated. And even though black and Latino minorities were becoming predominant in many schools, most of the teachers and school administrators remained white. These minority communities felt that the schools weren’t meeting their needs, and they called for decentralization so that they could have a greater say in how their schools were run.
In 1967 New York’s Board of Education began an experiment in community control in the predominantly black district of Ocean Hill-Brownsville in Brooklyn. When the community school board began transferring white teachers out of the district, it sparked a controversy that led to a strike by the teachers union that lasted for over two months in the fall of 1968 (Berger).
A police cordon blocks teachers and students from entering JHS 271 in Ocean Hill– Brownsville on December 3, 1968.
The teachers union, which had been regarded as very liberal and progressive at that point, was now seen as anti-reform, protecting the status quo. The teachers’ main worry was their job security. When it turned out that most of the white teachers who were transferred were Jewish (Berger), there were charges that school board members were anti-Semitic, and all of this contributed to the so-called rift between blacks and Jews in New York City (Kifner).
Eventually, the district fell to state control, and when the New York State legislature, in 1969, passed a law decentralizing all elementary and middle schools in the city, leaving the high schools centralized, it was a compromise law that appeased the union. Decentralization advocates had called for “community control,” but the city Board of Education recognized only “community involvement” (Fiske; Kifner).
By the 1980s reports in the news media were declaring community control a failure. In 1980 The New York Times ran a five-part series, which detailed a pattern of nepotism, patronage, fraud and corruption in the community school boards. It found that after 10 years of decentralization, standardized test scores showed only slight improvement; the city’s 650,000 elementary and junior high school pupils continued to perform at levels far below national norms (Fiske).
However, there were some positive outcomes: With the authority for hiring teachers, principals and district superintendents now in the hands of community boards, the proportion of minority teachers increased to 19 percent from 10 percent, and the proportion of minority principals to 27 percent from 9 percent. In many cases, these professionals, lacking ties to the central bureaucracy, brought a degree of competence and innovation that had not been seen prior to decentralization. Principals reported greater parental involvement, and the community boards were able to implement innovative curricula (Fiske). In districts that experimented with alternative schools and programs, there was some improvement in reading scores (Maeroff). A 1983 study showed an improvement in reading scores for students at every grade between the years 1971 and 1981 (Hess, 218).
But, according to the Times, personnel cutbacks, and questions about the quality of some of the teachers hired by the school boards wiped out some of those gains. Moreover, for several reasons, decentralization hadn’t worked as advocates for community control had hoped:
• Teacher hiring remained largely centralized, and the tenured appointment of principals through merit tests was protected.
• Because the school districts were large and remote, turnout at school board elections was low, only about 7 percent in most instances.
• Ironically, decentralization gained in support from those who first opposed it—predominantly white districts, who learned how to work the system to their advantage, and the teachers’ union, who was able to get its slates elected to the boards (Berger).
Nevertheless, in 1980, 91 percent of the nation’s school districts with 100,000 or more students reported some form of decentralization (Ornstein). In 1988, the Illinois state legislature, with its eye on New York’s experience during the previous decade, passed the Chicago School Reform Act, which reformed the Chicago school system by giving authority to Local School Councils (LSCs). Where in New York, there were 32 communities of 50,000 to 200,000 residents; in Chicago decentralization involved 560 communities (Hess, 218). In Chicago Local School Councils (LSCs) govern all schools. Each LSC comprises two teachers, four parents, two community representatives, and a principal. The LSCs have broad authority over budgeting, principal selection, and curriculum and program selection (Stinnette).
Meanwhile in Detroit, which decentralized its school system in 1970, there was a 41-day school strike in 1971 over the issue of using racial quotas in the appointment of teachers. By 1976, decentralization was declared a failure in improving the quality of education in Detroit, and in 1981, the city voted to recentralize the school system (Hess, 219). In 1988 Detroit began experimenting with site-based management and in the early 1990s reformers of the city’s school system began to push for more of these types of schools, as well as charter schools (McGriff).
By the 1990s, it appeared that the pendulum was once again moving toward centralization. A 1992 paper written by Allan Ornstein, who was researching trends in decentralization and consolidation in urban and rural school districts, claimed that decentralization was on the decline. Ornstein cited data that showed a drop in the percentage of decentralized school districts, from 67 percent (42 out of 66) in 1980 to 31 percent (16 out of 51) in 1988. He said, also, that the majority of districts that were claiming to be decentralized seemed more committed to centralization. He based this on the fact that the central administrative staffs of these school districts were each 10.8 to 21.5 times greater than their respective decentralized staffs. He said the majority of the school districts, 10 out of 16, were moderately more centralized than decentralized. Their centralized administration was 1.5 to 4.0 times greater than their decentralized administration.
Ornstein also claimed that data showed that large urban school districts that claimed to be decentralized, were in fact highly centralized when it came to decisions regarding curriculum, instruction, staffing and teacher evaluation, student testing, graduation requirements, and budgeting. In his conclusion, Ornstein stated, “As of now we have no research evidence that school consolidation or school decentralization improves education” (Ornstein).
Getting back to New York, is the 10-region structure any improvement over the 32-district model? Why are Bloomberg and Klein tinkering with different schemes that would give school administrators more power? The Bloomberg administration has reported an improvement in state reading and math test scores in most grades. Part of this they would attribute to an end to social promotion, the opening of small high schools and an emphasis on performance. Yet, the city’s four-year high school graduation rate of about 53 percent has stayed about the same since Bloomberg took office. And observers, such as education historian Diane Ravitch, doubt that the numbers will increase with “yet another massive upheaval” in the school system’s structure (Herszenhorn).
The trend for large urban school districts has been toward management on the individual school level. This “bottom-up” approach to decision making is variously referred to “school-based management,” “site-based management,” “school-based autonomy,” or “school empowerment.” and this seems to be the direction in which school decentralization has been heading for over a decade now. The premise is that this system works because those who work directly with students know best how to serve the needs of those students (Cotton).
School-based management has been tried in Salt Lake City, Chicago, and throughout California. Salt Lake City’s School Community Council, which included principals, teachers and parents, was given the authority to make decisions on school budget, personnel and programs. However, researchers in one study found that in practice, the school-based councils fell short of their mission, and rather than being policy-making boards, they served as advisory boards to the school districts. Although teachers and parents had equal participation, it was the principals who controlled the agenda. Parents were often left out of the loop, because they didn’t have as much access to information as principals and teachers. The authors of the study concluded that shared governance on a school level did not really alter the relationships between principals, teachers and parents (Hess).
Chicago’s experiment with school-based management was an example of decision-making on the school level and giving parents and community representatives a greater say in how those decisions would be made. Teachers, principals, parents and community representatives shared responsibility at the school level. The result was an improvement in student achievement in some schools, but it wasn’t system-wide, and it was mostly in elementary schools, where there was significant improvement in reading and math by 1999, 11 years after the reform was implemented. Still, about a quarter of Chicago’s elementary schools had fewer students at or above national norms in reading in 1996 than they had in 1988 (Hess).
School-based management strategies often fail to take into account the clashing interests of various players involved in the collaborations. In Detroit, for instance, where school empowerment began in 1990, the teacher’s union was unenthusiastic about the scheme. The union was concerned about contract provisions that might be voted away at the school level. Linking accountability to pay for performance was another major issue of concern to the union (Jelier and Richard Hula, 15-17).
Unable to build coalitions around its reform efforts, the Detroit Public School System—an independent entity over which the mayor has no formal authority—seems to be in perpetual chaos. In 2006 it has lost thousands of students to charter schools or neighboring districts (Associated Press).
When Rudy Crew, New York’s Schools Chancellor under Mayor Giuliani, attempted to wrest control away from the community boards in 1995 and 1996, the State Legislature restructured the city system of school governance, expanding school-based management, while also giving the chancellor increased powers to set performance standards for community school boards, community superintendents and school principals. Thus, in a sense, New York City’s school system was both centralized on one level and decentralized on another.
Similar reforms took place in Chicago in 1995, when the city moved to a mayoral model, yet leaving local school councils intact (Hess, 221). Mayor Richard M. Daley’s decision to stop social promotion and start mandatory summer school to help failing students catch up earned praise from President Bill Clinton, in his 1998 and 1999 State of the Union messages. In 2002 test scores and graduation rates were reportedly increasing, although that had been happening even before Mayor Daley took control. And while most of the improvements in test scores were at the elementary level, only a third of 9th and 10th graders were reading at grade level (Lewin). In 2005 Chicago designated 85 schools as “Autonomous Management and Performance Schools” (Hansen and Roza, 1).
Although attitudes toward school-based management have been positive in some cases, researchers have found no direct link between school-based management and improved student performance. The research suggests that school restructuring should not be an end in itself, but “a means to improving student performance through bringing about improvements in the quality of schooling” (Cotton).
One study suggests that the move away from school district offices to site-based management is premature. “While there are certainly numerous examples of ineffective district offices, those who advocate doing away with them altogether have yet to propose solutions that will raise achievement in more than a small group of schools in any geographic area (and, in particular, in the many urban areas of the country)” the authors write. “While a degree of school-level autonomy is essential in improving instruction for students, and recentralization is certainly not the answer, the role of the district central office in positively influencing those factors that raise the quality of classroom instruction cannot be ignored.” Important to high achievement is the support and services that the central office provides for the schools, according to the report. The ability to provide a higher level of resources to professional development is another key area in which central offices can raise the level of instruction and achievement in schools (Mac Iver and Farley, 29).
“The irony of the new attention being paid to DDS [decentralized decision-making for schools] in the United States is that many public schools claimed to have tried decentralized decision-making—so-called school-based management—in the 1980s and 1990s, yet the performance of U.S. education during that period improved only modestly, at best. SBM did not result in major changes in educational practice” (Hansen and Roza).
Finally, as part of the decentralization vs. centralization debate, the issue of charter schools has emerged as an option of “school choice.” The federal “No Child Left Behind Act” allows states to apply the sanction of charter-school conversion to chronically failing public schools.
Charter schools are publicly funded schools of choice that form a contract, or “charter,” with a public entity (for example a school district, state, or university) in which they are given greater autonomy than other public schools over curriculum, instruction, and operations. In exchange for greater autonomy, they are held accountable for results. (Zimmer and Buddin, 1).
The first charter school law was passed in Minnesota in 1991, and the first school opened there the following year. There are, by a recent count, more than one million students in nearly 3,500 charter schools in 40 states and the District of Columbia.
In 1992, California became the second state to adopt charter schools and now has more charter schools and students than any other state. From an examination of California’s experience with charter schools, the authors of one study concluded that charter schools have performed no better or worse than public schools, have not closed the achievement gap between minority and white students, and have not had the effective competitive effects on traditional public schools. So for all the hope and hype over charter schools, they “are not a ‘silver bullet’ for school improvement” (Zimmer and Buddin, 6).
So far, there has been no research that shows decentralization in the form of school-based management results in markedly higher student achievement than decentralization in the form of community control. Although it seems logical to conclude that decentralization makes sense for large urban school systems like New York’s, it is not a matter of community control versus centralization, which originally was the question this paper set out to answer, but how large is the community that has control. If the community is school based, then there is probably a better opportunity for involvement by parents and community representatives. In Chicago, where this was the case, local school councils seem to have been more successful at keeping centralized power in check.
If there had been more and smaller community districts in New York, perhaps there would have been greater community participation and less fraud and corruption, because there would have been more oversight on the community level and less opportunity for local superintendents to turn those districts into their own personal fiefdoms.
Fraud and corruption are not the necessary result of decentralization, centralization or any other type of structure. U.S. urban history is rife with incidents of patronage, cronyism and nepotism on the part of out-groups who gain some power. To read some of the accounts in the media of what happened in the aftermath of Ocean-Hill Brownsville, one gets the impression the New York City’s educational system had been the victim of community control run amok.
Those who have examined both sides of the decentralization versus centralization issue seem to agree that there should be less focus on changing the organizational structure of the school system and more focus on what goes on in the classroom—decreasing class size, improving the quality of curriculum and the classroom environment; increasing the lines of communication on all levels.
In assessing schools’ success, we need to look at criteria such as teacher retention rates, student dropout rates, and student participation in extracurricular activities and not just standardized test scores.
Changing organizational structures is what corporations do when they want to improve performance, and it does not necessarily work for public schools. Elected officials have repeatedly embarked on programs of school reform to meet political ends rather than educational ones. They should first figure out what it is the various communities need in terms of education and then get the people in those communities to come together to decide what works best for them.
Associated Press, “Detroit schools spending $500,000 to fight pupil loss.” Detroit Free Press 31 July 2006. 6 Aug. 2006 <http://www.freep.com>.
Berger, Joseph, “Board of Education: A Thing of the Past?” The New York Times 18 Feb. 1996: 39. LexisNexis Academic. 25 July 2006.
Cortines, Ramon, “Asking Too Much Of Decentralization.” Education Week 27 Sept. 1995: 34-40. Editorial Projects in Education.
Cotton, Kathleen, “School-Based Management.” Dec. 1992. Northwest Regional Educational Laboratory. 2001. 4 Aug. 2006.
Fiske, Edward B., “Community Run Schools Leave Hopes Unfulfilled.” The New York Times 24 June 1980, Sec A: 1. LexisNexis Academic. 25 July 2006.
Hansen, Janet S and Roza, Marguerite, “Decentralized Decisionmaking for Schools: New Promise for an Old Idea?” Rand Education Occasional Paper 153. 2005. 25 July 2006 <http://www.rand.org/pubs/occasional_papers/2005/RAND_OP153.pdf>.
Herszenhorn, David M., “New York Rethinks Its Remaking of the Schools” The New York Times 9 April 2006. 7 Aug. 2006.
Hess, G. Alfred, Jr., “Community Participation or Control.” Theory Into Practice Autumn 1999: 217-225. Academic Search Premier. 25 July 2006 <http://search.epnet.com/login.aspx?direct=true&db=aph&an=2516171>.
Jelier, Richard W. and Hula, Richard C., “A House Divided: Community Politics and Education Reform in Detroit.” The Urban Review March 1999: 3-29. Academic Search Premier. 5 Aug. 2006.
Jones, Dell, “Schools take a lesson from big business; Decentralizing helps when organizations get unwieldy.” USA Today 9 March 2006: 1B. 29 July 2006.
Keane, Julie Thompson, “The Past and Future of NYC Public Schools: Marilyn Gittell in conversation with Julie Thompson Keane.” The Brooklyn Rail. Winter 2003 <http://www.thebrooklynrail.org>.
Kifner, John, “The Nation: Ocean Hill-Brownsville, ’68; Echoes of a New York Waterloo.” The New York Times 22 Dec 1996, Sec.4: 5. LexisNexis Academic. 25 July 2006.
Lewin, Tamar, “For Mayoral Control of Schools, Chicago Has a Working Blueprint.” The New York Times 12 June 2002: 25. LexisNexis Academic. 6 Aug. 2006.
Mac Iver, Martha Abele and Farley, Elizabeth, “Bringing The District Back In: The Role of the Central Office in Improving Instruction and Student Achievement.” Report 65 Aug. 2003: 1-44. The Center for Research on the Education of Students Placed At Risk. Johns Hopkins University and Howard University. 2003. 7 August 2006
McGriff, Deborah M., “Decentralization: Why, How, and Toward What Ends?” NCREL’s Policy Briefs, 1993. North Central Regional Educational Laboratory. 1995. 26 July 2006 .
Maeroff, Gene I., “Achievement Lacking In Community-Run Schools.” The New York Times 25 June 1980, sec. A: 1. LexisNexis Academic. 25 July 2006.
Ornstein, Allan C., “Trends In Consolidation And Decentralization” Clearing House May/Jun 1992: 332-327. Academic Search Premier. 25 July 2006.
Stinnette, Lynn J., “Decentralization: Why, How, and Toward What Ends?” NCREL’s Policy Briefs. 1993. North Central Regional Educational Laboratory. 1995. 26 July 2006
Zimmer, Ron and Buddin, Richard, “Making Sense of Charter Schools: Evidence from California.” Rand Education Occasional Paper 157. 2006. 6 Aug. 2006.