Hospital efficiency and productivity: bias-corrected evidence from control function estimation

Abstract

Purpose This study investigates the efficiency of hospital services production in Afghanistan, a conflict-affected and resource-constrained setting, to identify key drivers of hospital productivity and provide actionable insights for healthcare policy and management. Design/methodology/approach We use 2010–2020 panel data from 29 national public hospitals in Afghanistan. To account for endogeneity and unobserved heterogeneity, we estimate hospital production functions using Olley–Pakes (OP) and OP-generalized method of moments (OP-GMM) control function estimators. We also implement an instrumental variables strategy (geographic distance) to address the endogeneity of bed occupancy. Findings This study demonstrates increasing returns to labor and material inputs, suggesting gains through improved staffing, while capital inputs show diminishing marginal productivity. Productivity-enhancing bed occupancy reflects the importance of capacity utilization. Conversely, higher mortality rates highlight quality inefficiencies in care delivery. This study advances the performance management literature by applying control function estimators in a fragile context. It integrates methodological rigor with practical relevance, providing new insights into how hospitals convert limited resources into effective service and identifying resource combinations that maximize efficiency and accountability. Originality/value To our knowledge, this is the first study to apply OP and OP-GMM methods to healthcare in Afghanistan. By addressing endogeneity in input selection, we provide more accurate productivity estimates, offering critical insights for hospital administrators and policymakers. Findings guide investments in labor and materials, improving service utilization and linking productivity to quality metrics. These results have wider implications for fragile health systems in other low-income countries.