Publication Type: Working Paper
Countries: Bangladesh
Authors: Mushtaq Khan, Mehnaz Rabbani, Faruq Hossain, Rabeena Sultana Ananna
Publication date: March 2025
Keywords: Education
Bangladesh has successfully increased spending on primary education and achieved impressive improvements in enrolment rates; however, the quality of learning outcomes remains a serious concern. The governance of schools is an important determinant of poor learning outcomes as it can result in a loss or diversion of teaching resources. To improve governance, the state has focused on formal arrangements that we describe as ‘vertical enforcement’. Bodies such as School Management Committees, Parent–Teacher Associations and Social Audit Committees operate at the school level, while other bodies operate at higher levels and affect the flow of resources from the treasury to schools.
We argue that the quality and effectiveness of vertical checks at any level depends on complementary ‘horizontal checks’. These horizontal checks occur where other actors have sufficient power and capabilities to check for violations that are in their own interest and then they support or demand effective formal governance. A feasible way of improving poorly performing formal governance is to establish if horizontal checks already operate and if these can be strengthened with policy support.
We investigate three important sources of corrupt leakages associated with poor governance in primary schools in Bangladesh: the inflation of student numbers, the misuse of school development funds, and the failure to check poor teacher allocation and effort. We have used ethnographic observations to see if and how horizontal checks operate at the school level for each of these leakages in a sample of 10 schools. We find that horizontal checks do indeed operate in some schools and that these are effective in reducing the relevant leakages. We have then assessed the learning outcomes of students in our schools in English, Bangla and numeracy and show that the first two, at least, have a significant correlation with our observed measures of horizontal checks.
As public schools are quite similar, the variations in horizontal checks across our schools are not very large. Furthermore, we have not found systematic associations between the strength of horizontal checks and the characteristics of actors or incentives that may be useful for designing interventions to strengthen horizontal checks. If we were to look at a wider range of schools, such as private schools and madrasahs (traditional Islamic schools), we are likely to see greater variations in the strength and types of horizontal checks. We may then discover factors that have a systematic effect on improving horizontal checks in ways that can be influenced by policy. Our findings show the limitations of focusing solely on formal governance arrangements – to improve governance, research effort must shift towards identifying and strengthening effective horizontal checks.
