Public spending, early childhood education, and primary school outcomes: insights from quantile regression analysis

Purpose: This study examines the differential effects of public education spending across countries with low, median and high enrolment and dropout rates at pre-primary and primary schools. Design/methodology/approach: We use panel data from 74 low- and middle-income countries spanning 1990 to 2021. Modelling school enrolment and dropout rates as linear functions of public spending and a host of other socioeconomic indicators, we run regressions with pooled OLS, two-stage least squares (2SLS) and the quantile regression approach. We choose quantile regression because it can better capture the varying effects of public spending across countries at different levels of enrolments and dropout rates. Findings: Our results show that public spending is significant in median and high-enrolment countries but is insignificant in low-enrolment countries at the pre-primary level. Public spending influences primary school enrolment across all categories of enrolments substantially. The effect of public spending is insignificant in countries with the lowest dropout rates. Originality/value: Our study makes three important contributions to the literature. Firstly, we revisit and provide new empirical evidence on the question of whether public spending affects “average” school outcomes using a larger dataset over a longer period of time. Secondly, we go beyond these studies by exploring how the effect of public spending varies across countries with low, median and high enrolment and dropout rates, providing insights on not just whether money matters, but for whom it matters most. Thirdly, utilising the quantile regression approach also brings a unique methodological innovation to the literature.