Policy discourses posit an accountability deficit as an underlying cause of a “learning crisis” in many low-income countries. Many studies understand this perceived deficit from a principal-agent perspective, arguing that incentives facing teachers and schools often do not align to the interests of parents and students. Such perspectives underlie many randomized controlled trials, which associate interventions with outcomes, but which also produce varying or inconsistent results across contexts. This paper seeks to study the accountability of schools and teachers more directly, looking at how it varies across public and private schools and how it relates to students’ literacy and numeracy abilities. We report results from a mixed methods study conducted in Mumbai and Kathmandu. Our results show that there are some relationships between accountability and learning outcomes, but these appear to be specific to the context. Quantitative data also show that differences between public and private models of schooling are negligible when students’ social backgrounds and school composition are considered. Qualitative data show that accountability processes create a significant burden on staff time and embed complex power dynamics that are not always productive. Taken together, these results problematize policies that seek to improve learning through “demand-side” approaches such as privatization. They show that the dynamics of accountability are a complex system, like the motion of a “double pendulum,” and therefore simple conceptual approoaches such as the principal-agent model are of limited academic and practical utility.
- Mixed methods
- South Asia