Estimating economies of scale and learning-by-doing effects in a health systems strengthening intervention

It can be challenging to evaluate efficiencies for interventions strengthening health systems, for a variety of reasons. As part of a program review of The Challenge Initiative (TCI) that took place in 2020, we explored whether economies of scale and learning-by-doing effects existed for the health systems strengthening portion of the project, utilizing audited expenditure data and validated reported outcomes. TCI is a project that partnered with local governments in low- and middle-income countries to deliver interventions for family planning; it is relatively unique in terms of both its activities and in how its expenditures and outcomes were tracked. Using TCI data, we estimated a cost function, evaluating economies of scale using as output measures both the number of geographies that joined the project over time and the population of women of reproductive age in those geographies over time, allowing for complex effects by including levels, squared and cubic formulations of the output variables. We also evaluated whether learning-by-doing effects obtained by examining coefficients of time-related variables. Results suggested that while initially there were diseconomies of scale when output was measured using number of geographies, the TCI program began to experience economies of scale as the number expanded, with significant economies of scale experienced beginning at the mean number of geographies. When output was measured using population of women of reproductive age, economies of scale existed throughout. We did not find econometric evidence of learning-by-doing effects.