Spending Choices and School Autonomy: Lessons From Ohio Elementary Schools

---

July 2007
Marguerite Roza, Tricia Davis, Kacey Guin

Download Full Report (PDF: 881 K)

SFRP Working Paper 21

Education policy is experiencing a revival of interest in school-based decision-making. Giving schools more autonomy reflects the notion that school personnel are better equipped than district administrators to make efficient and effective use of their resources to meet student needs. But for policymakers, several questions remain: If school leaders had more autonomy over resource decisions, would that result in any real difference in how schools use resources? When provided with greater spending autonomy, what kinds of choices do school leaders make? Are those choices meaningfully different from current practice? What kinds of constraints continue to serve as barriers to change? Answers to such questions could help policymakers design more effective school autonomy policies.

This study examines resource allocation patterns across elementary schools and how these patterns differ depending, in part, on various levels of autonomy over resources at the school level. It provides insight into how schools spend their funds, make trade-offs among competing demands for resources, and decide what proportion of their resources should be devoted to staffing. Using data on school allocations, staffing patterns, and salary information from the 2004-2005 school year, we analyzed over 500 schools across 8 large districts in Ohio, including charter and private schools within district boundaries, allowing for comparisons among different school types in the same labor markets.

First, we identified four key areas of school-level resource autonomy. These areas included:

• allocation of personnel to responsibilities;
• allocation of resources across functional categories (e.g., instruction, pupil support, administration);
• control over the number of teachers; and
• ability to determine teacher compensation.

Different school types were placed on an autonomy continuum based on these four areas. At the least autonomous end of the spectrum, centralized schools are restricted in how they allocate personnel and resources, the number of teachers they can employ, and the amount of teacher compensation they can provide. By contrast, schools in decentralized districts have the autonomy to allocate personnel and resources and to determine the number of teachers employed. Charters and private schools can take advantage of all four categories of resource autonomy, including determining teacher compensation.

Second, we utilized fiscal and staffing data to examine patterns of resource use across different school types. Several key findings emerged from this analysis. Schools with more autonomy over their staffing, including charter and private schools, have more teachers per pupil and use fewer non-teaching professionals such as librarians, counselors, and nurses. Private schools, however, have more administrative staff per student than any of the public schools, including charters. They also use a greater number of partial FTEs to meet the needs of their school.

We also found that schools of different types make different trade-offs between teacher salaries and FTEs. Charter and private school leaders seem to opt for lower salaries and more staff, while the opposite pattern is evident in centralized and decentralized schools. These choices are likely influenced by autonomy over teacher salaries. In public schools, whether centralized or decentralized, teacher salaries are on fixed schedules that cannot be varied at the school level. For charter and private schools, salaries are included among resource decisions that school administrators can make. While public centralized and decentralized schools have fewer teachers per pupil, they also have higher salaries, which drives up their cost. Since classroom teacher salaries comprise the bulk of instructional expenditures, centralized schools spend more per pupil on instruction even though they have fewer teachers per pupil. Restrictions on teacher salaries, then, serve as a formidable barrier to some of the resource allocation changes schools might pursue.

It is important to emphasize that most current decentralization efforts have not yet created all the conditions necessary for fully decentralizing control of resources. Findings from this study indicate that constraints on school autonomy matter. Identifying innovations or efficiencies from resource data alone is difficult, yet the patterns prevalent in schools with fewer resource constraints may point to more efficient resource uses. Additional research is necessary to better understand all of the constraints different schools face, as well as how effective and efficient they are at using their resources.