New Research Sheds Light on the Implementation and Impact of Decentralization in Houston School District


Adopted decades ago, the model was called into question recently. New research begins to answer those questions.

HISD Central office


Adopted decades ago, the model was called into question recently. New research begins to answer those questions.

Around the time the Houston Independent School District was implementing the final phase of a roughly decade-long reform aimed at decentralizing many decisions down to the school level, the superintendent wrote in the Houston Business Journal in 2000 that he had high hopes for the effort. Given more autonomy, including over their budgets, principals, wrote the superintendent, could start with a “blank page” to build the program most responsive to their students’ needs. Part of the switch, and one of the last pieces to be implemented, were funding changes meant to put more budget discretion in the hands of schools instead of a central administration. In his article, Rod Paige wrote that he hoped principals would have control over roughly 80 percent of their schools’ budgets.

Paige was a business-minded man and the idea of decentralization, one that gained traction across the country, was borrowed from business management theory.

But now, several decades on from the change, district officials and board members are again asking questions about how the district’s schools operate and are funded.

“We were asked to look into four research questions around decentralization,” Ruth López Turley, director of the Kinder Institute’s Houston Education Research Consortium, told the HISD school board at a workshop meeting Tuesday. The questions surrounding decentralization got to the heart of thorny issues like funding and equity. “Sit up and listen, folks,” read the start of one recent headline from Houston Chronicle’s then columnist Lisa Falkenberg when the district’s former superintendent floated what she described as “a drastic change in philosophy, a centralized model of governance.”

Now, with an interim superintendent in place, though the district is not considering a move back to a fully centralized system, those questions about decentralization’s implementation and impact remain for the district and board, who heard the results of a series of studies from HERC at its Tuesday workshop.

To address those questions, researchers at HERC undertook four studies. First, they looked at whether the district’s implementation of decentralization matched with best practices laid out in the broader research literature. Then they examined whether current principals felt empowered and informed enough to make the sorts of decisions decentralization argues principals are best suited to make. To determine its impacts, the researchers then looked at whether student test scores changed after decentralization and how funding functioned as a result of decentralization.

“We found that overall decentralization was implemented with fidelity,” said Turley, noting that it was in line with the approach adopted by other districts around the country, “but the following areas deviated somewhat from accepted theory and those are small schools, magnets and teacher salaries.”

And even areas where postdoctoral fellow Jodi Moon found partial implementation of what would be expected given the theory, there were still noteworthy gaps. So, for example, while the district’s funding formula relies on a base allotment plus weighted funds dependent on various student characteristics, those specific weights have gone almost unchanged after being implemented nearly 20 years ago. In addition, Jay Aiyer, another researcher involved in the studies and an assistant professor at Texas Southern University, said, those weights “are essentially reflections” of the state’s own weights.

Because decentralization, in theory, relies so much on principals to make critical decisions for their campuses, another study looked at principal feedback on issues relating to things like budgeting and staffing. Here, Moon said she found that most felt they understood how to use data and communicate with teachers to assess student needs, as well as how to make budget decisions based on that.

Unlike Paige’s goal for 80 percent of a campus’ budget falling to a principal’s discretion, though, the researchers said that a little less than half currently does. And the biggest area for improvement, according to the principals, was in budget analysis support.

Using student test scores during a three-year window after the district adopted the new weighted-student funding formula and completed its phase-in of decentralization, researchers sought to test whether the change affected student scores.

As best they could tell, decentralization did not make a measurable impact on test scores. Comparing the district’s performance on standardized state tests to that of a set of roughly comparable districts across the state, researchers found that though scores generally improved over the study period, it was not likely related to decentralization since scores also improved in those matched districts not undergoing decentralization.

Still, as the researchers and one board member noted, test scores are just one possible effect of such a change. The researchers also said that it’s possible more noticeable changes would’ve been detected looking at other outcome measures or over a longer period of time but that measuring that consistently would’ve been difficult as the state-administered tests changed.

“Focusing on testing is not equity,” said Trustee Elizabeth Santos, reflecting on a larger conversation the board finds itself in the middle of as it considers the direction of the district.

In the final study, researchers sought to see the effectiveness of the decentralized funding model, looking at spending and key personnel staffing across school levels, school sizes and various student body demographics. Drawing on general fund budget data from roughly 1999 to 2016, the researchers found more of a hybrid funding model than a purely decentralized one.

Still, researchers found that the funding formula as it was conceived was largely working as intended. But that raised another important question for the board and district; is the way it was intended to work, the way it should work?

Middle schools and high schools, for example, tended to spend more per student than elementary schools and enrollment size was the strongest predictor of general fund budgets. At the same time, added subsidies for designated small schools—also given to magnet schools—seemed to support those schools so that they had more or less the same sorts of key personnel, like nurses, librarians and other positions, on campus as non-small schools.

The researchers did note one gap when analyzing personnel data from roughly 2013 to 2016: schools with higher shares of economically disadvantaged students were less likely to have librarians. But board members suggested that there were inequities not being captured in the data. “Why when I walk onto some campuses is it clearly not equitable,” said Santos.

Presented as averages and not campus-level comparisons, the budget analysis could not speak to the sorts of concerns that have arisen as the district continues to face a possible state takeover because of several long-struggling schools.

“Our results speak to the average,” said Daniel Potter, associate director at HERC, noting that identifying trends in spending and staffing across higher levels might not capture the kinds of differences Santos described and saying that board input was welcomed for future research.

With two of the studies not yet published, the findings come at an important time for the district. “We did start some initial discussions for the weights and the way they are applied,” explained Trustee Diana Dávila, who said that the interim superintendent, Grenita Lathan, had formed a committee to look at budget decisions. Several board members floated the idea of a tiered autonomy system that would allow some principals to have more control than others depending on specific factors. Dávila also encouraged the input of parents, in addition to principals. “They would tell you that this system is not working for their children,” in terms of access to different resources, she said.

Though the studies found that decentralization was more or less implemented as intended when it was adopted several decades ago and that the funding strategy was working as initially conceptualized, Potter said, “It pushes the question as well as the challenge: is the current strategy doing what we want it to do?”

Mailing Address

6100 Main St. MS-208
Houston, TX 77005-1892


Physical Address

Rice University
Kraft Hall
6100 Main Street, Suite 305
Houston, TX 77005-1892

Featured Sponsor

Support the Kinder Institute