Production of educational output: time series evidence from socio-economically heterogeneous populations – the case of South Africa 1910-93

This paper uses the natural experiment South Africa provides in order to examine the impact of educational inputs on educational output in different learning environments. Co-integration analysis is employed in order to establish that inputs into the education process do make a difference to educational output. Such an impact is not only strong, but is present even where variation in the input measures is small. By separately estimating production functions for relatively homogeneous population groups in the socio-economically heterogenous South African population, we control not only for background variables, but establish that the capacity to exercise political control over the decision-making executive implies that institutions matter. This paper uses the natural experiment South Africa provides in order to examine the impact of educational inputs on educational output in different learning environments. Co-integration analysis is employed in order to establish that inputs into the education process do make a difference to educational output. Such an impact is not only strong, but is present even where variation in the input measures is small. By separately estimating production functions for relatively homogeneous population groups in the socio-economically heterogenous South African population, we control not only for background variables, but establish that the capacity to exercise political control over the decision-making executive implies that institutions matter.