Rural school feeding procurement decentralisation and smallholder market participation in Kenya: a difference-in-differences analysis of county-level devolution thresholds

Kenya's devolution of school feeding procurement to county governments was expected to draw smallholders into institutional food markets, but the causal claim rests on a difference-in-differences design whose operating characteristics have not been examined for the clustered, threshold-defined county samples actually available. Power was estimated across 2000 design cells defined by cluster counts of 8, 16, 24 and 32 and effects of 3, 5 and 8 percentage points. The analysis found that 56 clusters are required to detect a 5 percentage point effect at 80 per cent power, that the minimum detectable effect at 80 per cent power falls steeply as clusters increase from 8 to 32, and that the estimator's confidence intervals widen enough at low cluster counts to make the 3 percentage point effect indistinguishable from zero. The contribution is a design-characterisation result: county-level devolution thresholds in Kenya cannot support credible difference-in-differences inference on smallholder participation at the effect sizes the policy literature reports, and the paper specifies the cluster counts and effect sizes under which such inference becomes defensible.