Impacts of interlocked contractual arrangements on dairy farmers’ welfare in Zambia: a robust Bayesian instrumental variable analysis

ABSTRACT While contract farming and interlocked contractual arrangements (ICAs) are generally perceived to resolve persistent market failures and improve smallholder farmers' welfare in developing countries, uncertainties remain as to whether these arrangements enhance welfare because of farmers' low marketed volumes and margins. To account for potential selection bias, non-Gaussian and missing data problems, a robust two-stage Bayesian instrumental variable approach is used to determine the impact of dairy farmers' participation in ICAs on household income and milk revenue. Data are from smallholder dairy farmers in Zambia. We find that male household heads, wealth, experience selling to milk collection centres (MCCs), livestock holding, milking parlour ownership, landholding, and access to marketing information positively affect farmers' probability to participate in ICAs. However, increased off-farm income and distance to MCCs limit their participation. While some socioeconomic variables have significant positive effects of affecting ICA participation on household welfare, we find no sufficient evidence of causal effects of ICAs on household incomes and milk revenue among dairy farmers. Thus, while ICAs enhance smallholder farmers' access to markets, they may not address high rural poverty rates in developing countries. We provide some insights by which performance of ICAs in the dairy sector may be improved.