Responsive Reforms Can Lead to Learning Gains: How Brazil’s Municipality of Sobral Turned Around Its Education System

This blog series reviews major policy reforms and interventions to showcase real-life examples of the ‘five actions to accelerate progress in learning’ identified by Pritchett, Newman & Silberstein (2022). In this fifth blog, I discuss how Sobral turned around its education system by setting a clear goal and adapting its approach until progress was made, aligning to the fifth action of ‘adapt what you adopt as you implement’ under the five actions.


Image of Rastee Chaudhry

Rastee Chaudhry

RISE Directorate

Blavatnik School of Government, University of Oxford


Despite facing high poverty levels, a five-year drought, and a national recession, the Brazilian municipality of Sobral showed remarkable improvements in its education system by becoming the highest-ranked municipality in Brazil in terms of basic education (Crouch, 2020).

The improvements in Sobral’s basic education were the result of an iterative and adaptive approach toward improving education outcomes. This involved designing and adapting reform efforts that centred on a single goal: to ensure all students were literate by the end of Grade 2. The reforms were implemented in three phases, each tailored and adapted based on feedback, and all working towards the same goal. The story of Sobral shows that significant results can be achieved by maintaining a shared commitment to a goal and being responsive and adaptive in response to feedback.

What happened?

Sobral is a municipality located in Ceará, the fifth-poorest state in Brazil (Loureiro et al., 2020). The municipality is home to 200,000 people and faces high poverty levels and severe resource constraints (Crouch, 2020). Further, the municipality recently experienced a five-year drought, and Brazil overall faced a national recession (Crouch, 2020). Despite these substantial contextual challenges, Sobral has shown remarkable improvements in its education system by becoming the highest-ranked municipality in Brazil with regards to basic education, undeterred by unfavourable economic conditions (Crouch, 2020).

The improvements in Sobral’s basic education services were the result of a commitment to improving educational outcomes, which was pursued through an iterative and adaptive approach to reforms centred on ensuring that all students were literate by the end of Grade 2. Education reforms in Sobral can broadly be grouped into three phases, with each phase showing noteworthy education gains in a short period. These phases are described below.

Phase I (1997-2000): Improving enrolment and infrastructure

In 1997, Sobral elected a new mayor, Cid Gomes, who was deeply committed to improving education. Gomes recognised that educational outcomes in the municipality were not ideal, which he attributed to children not being in school (McNaught, 2022). Further, Gomes hypothesised that there were two drivers of poor educational outcomes: (1) teacher hiring was not meritocratic and (2) schools did not have basic inputs (McNaught, 2022).

Gomes’ administration addressed these concerns with his first set of education reforms.

  • Firstly, technical criteria for teacher recruitment were developed, and all teachers who did not meet these basic technical requirements were laid off (McNaught, 2022). As a result, 1,000 teachers were laid off and replaced with meritocratic hires. Surprisingly, this move faced little opposition from the teachers’ union. This was because transparent technical criteria were used both to remove teachers who did not fulfil the criteria (generally due to being politically appointed) and to appoint new recruits.
  • Secondly, the municipal government embarked upon a series of activities to improve school inputs to expand access and encourage enrolment (McNaught, 2022). These activities included building new schools, expanding and renovating existing schools, and increasing the availability of books and teaching and learning material in schools.

This two-pronged approach resulted in improved enrolment and better infrastructure in schools.

While this first set of reforms was successful in improving enrolment and infrastructure, a reading assessment of almost 12,000 students conducted in 2000 showed that substantial proportions of students lacked foundational literacy skills (McNaught, 2022). Specifically, 60 percent of students in Grade 2, 40 percent of students in Grade 3, and 20 percent of students in Grade 4 could not read.

These results were shocking and demonstrated to the municipal government that while more children were in school, they were still not learning much. In other words, the government’s hypothesis—that getting more children into school would lead to gains in education—was missing a big piece of the puzzle and had to be revised.

Phase II (2001-2004): Emphasising learning

The results from the reading assessment in 2000 were a wake-up call for the municipal government. The Gomes administration, which was re-elected in 2000, shifted its focus to improving literacy and centred its reforms around one goal: to ensure that all children in Sobral could read by the end of Grade 2 (McNaught, 2022).

With a clearer understanding of the problem and the context, the municipal government adapted their initial reform efforts, designing a series of comprehensive and interconnected reforms that would collectively work towards their goal. The reforms were designed to improve literacy and spanned four areas: (1) strengthening institutional management, (2) strengthening school management, (3) improving pedagogy, and (4) providing professional incentives (McNaught, 2022).

  • Firstly, institutional management was strengthened by creating a culture of monitoring, including biannual external assessments of student learning in Grades 1 to 4. These assessments informed the targeted support provided to students, teachers, and schools.
  • Secondly, school management was strengthened by ensuring meritocratic hiring of principals and replacing principals that did not meet basic technical selection criteria. In addition to this, ‘pedagogical coordinators’ were hired to monitor student learning, and more autonomy was provided to schools.
  • Thirdly, pedagogy was improved by designing new teaching and learning materials, setting periodic learning goals, and providing monthly professional development opportunities to teachers.
  • Fourth, incentives were provided in the form of prizes and salary bonuses for teachers, principals, and pedagogical coordinators for meeting literacy goals.

Following the second set of reforms, substantial improvements in learning were seen after only a few years of the reform. In 2001 (prior to the second phase of reforms), only 33.7 percent of 6-year-old students and 49.1 percent of 7-year-old students could read (McNaught, 2022). In 2004 (after only 3 years of implementing reforms to improve literacy), 88.9 percent of 6-year-old students and 91.7 percent of 7-year-old students could now read. These massive gains in literacy were truly remarkable, and were correlated with the government listening to feedback, understanding the problem, and adapting their approach.

Phase III (2005 onwards): Nationally competitive gains in learning

Despite these significant gains in literacy, Sobral’s municipal government remained committed to improving educational outcomes and set their aspirations higher. In 2005, the federal government of Brazil began measuring the quality of education in its schools through a basic education development index (called IDEB after its Portuguese acronym). Out of the 5,570 municipalities in Brazil, Sobral ranked 1,366th for the quality of education in its schools (McNaught, 2022). These results signalled to the Sobral government that there was significant room for improvement and that aspirations should be set higher.

To improve learning further and compete at the national level, the leadership of Sobral reviewed the state of education in the municipality, the existing reforms, contextual factors, and options for optimising the system to further improve the quality of basic education. Lessons from previous experiences led the team to identify key points that would generate improvements in the education system in Sobral.

Overall, the government decided during this phase to take a two-tiered approach to improving learning: (1) constantly assessing learning and being data-driven and (2) providing more targeted and focused support to teachers (Crouch, 2020).

  • Firstly, the learning levels of all children were now assessed twice a year. Assessment results were used in a formative manner to inform education reforms.
  • Secondly, teachers were supported with professional development opportunities as well as the setting of clearer and more precise teaching objectives, the provision of more detailed lesson plans with better sequencing, and the unification and alignment of teaching materials.

This approach was constantly modified and iterated upon based on results from the learning assessments.

The 2017 IDEB rankings placed Sobral first out of all 5,570 municipalities of Brazil for both primary and lower secondary education (Loureiro et al., 2020). Sobral had risen 1,365 places in IDEB scores, largely a result of an unwavering commitment to learning and a willingness to adapt education reforms in response to feedback.

What can we learn?

The case of Sobral shows that if leadership truly commits to improving educational outcomes and is willing to adapt its approach based on feedback, then meaningful and tangible improvements are within reach. Each phase of reforms implemented by Sobral’s municipal government showed noteworthy results in just a few years and within the tenure of the elected municipal governments. Further, the story of Sobral demonstrates that such improvements could be achieved even in the face of adverse economic conditions: the improvement in education scores seen in Sobral was 80 percent higher than would be expected given the expenditure on education by the municipality (Crouch, 2020).

It is important to recognise that education reforms in Sobral also benefited from numerous factors that created an enabling environment for improving literacy. Firstly, although Sobral was resource-constrained and had smaller financial outlays than other school systems, its educational improvements were also facilitated by the introduction of state funds for education financing and by decreasing fertility rates which allowed per-student spending to increase substantially (because funds were now spread over fewer students). Secondly, Phase I of the reforms put in place inputs that became enabling factors for more learning and better teaching in schools during Phases II and III. Lastly, the devolution of primary and lower secondary education from the state to municipalities was another enabling factor that enabled the municipal government to make decision policy changes (Cruz & Loureiro, 2020).

Still, it is clear that the implementation of an adaptive approach aligned around a single goal of increasing learning is a promising model for other governments to follow. The case of Sobral shows that while there is no ‘silver bullet’ for improving educational outcomes, a willingness to listen to feedback and iterate on reforms creates space for substantial results to be achieved.

Policy Implications

The Sobral example shows that if you remain committed to your goal and are responsive to feedback, significant results are possible even in resource-constrained contexts. This case yields the following policy implications:

  • Set clear and relevant goals and remain committed to them. Ensure that you set goals for your educational system that are clear, easy to understand, and—most importantly—contextually relevant and appropriate. Stay committed to your goals and design your reforms around them.
  • Continuously monitor progress towards your goals. Introduce a regular method of data collection that measures progress towards your goals. Share these results with all persons in your education system in a manner that is transparent and fosters accountability while being formative rather than punitive. Use the results to further tailor your reforms and approach.
  • Do not be afraid to adapt your approach. Reforming your education system is a dynamic process, and your approach should change as new information arises or as feedback is collected. Pay attention to feedback from the system and tailor your approach accordingly.


Crouch, L. 2020. Systems Implications for Core Instructional Support Lessons from Sobral (Brazil), Puebla (Mexico), and Kenya. RISE Insight Series. 2020/020.

Cruz, L., Loureiro, A. 2020. Achieving World-Class Education in Adverse Socioeconomic Conditions: The Case of Sobral in Brazil. World Bank, Washington DC.

Loureiro, A., Cruz, L., Lautharte, I., Evans, D.K. 2020. The State of Ceará in Brazil is a Role Model for Reducing Learning Poverty. World Bank, Washington DC.

McNaught, T.  2022. A Problem-Driven Approach to Education Reform: The Story of Sobral in Brazil. RISE Insight 2022/039.

RISE blog posts and podcasts reflect the views of the authors and do not necessarily represent the views of the organisation or our funders.