The subsidy-technical efficiency nexus: A meta-analysis of empirical findings

The subsidy-technical efficiency nexus in the agricultural sector is a topic of interest among researchers, but its exploration in Africa is limited. The impact of these subsidies on technical efficiency is mixed and poorly understood. This meta-analysis uses ordered probit regression to identify potential drivers of the differences from 56 global studies and 203 observations to explore the global relationship between subsidies and technical efficiency in the agricultural sector. Inclusion criteria encompassed peer-reviewed original studies (2000-2023) published in English that examined subsidy impacts on technical efficiency across diverse agricultural enterprises. We employed the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) through Scopus, Web of Science, and Google Scholar databases. Despite the administration of input-related subsidies in most African countries, only two studies (11 observations) were found from this region, highlighting the need for further research. Given the varying contextual factors specific to Sub-Saharan Africa (SSA), such as agricultural systems, institutional environments, and policy frameworks, this is crucial for understanding this nexus from a developing world perspective. Most studies (43.8%) show a negative relationship between subsidies and technical efficiency, suggesting that subsidies may not effectively improve technical efficiency. Hence, policymakers should critically examine the design and implementation of subsidy programs to identify potential barriers that may negatively affect technical efficiency. The relationship is sensitive to subsidy type, region, enterprise type, method of analysis, sample size, GDP per capita and how the subsidy proxy is modelled in the analysis. The meta-analysis highlights the need for methodologically rigorous approaches and sufficiently representative samples, which are critical for generating policy-relevant insights. Future research should compare analysis methods and different subsidy proxy modelling with the same data sets. Although contextualising these findings to SSA would be highly difficult, valuable lessons can be drawn for future research, especially in African countries.