Customary land certification fees and boundary-dispute escalation among smallholder farmers in rural Rwanda: a difference-in-differences analysis of fee-waiver thresholds

Customary land certification in Rwanda has been promoted as a route to tenure security, yet the fee schedule that governs it may also reshape how neighbours resolve boundary disagreements. No observational data are used or claimed. The analysis treats the before-after-control-impact estimator as the object of study and evaluates its power to detect effects of 3, 5 and 8 percentage points across 8, 16, 24 and 32 clusters, with a between-cluster standard deviation of 6.0, a residual standard deviation of 4.5 and an intraclass correlation of 0.640, over 2000 replications per design cell. Power curves, minimum detectable effect curves, coefficient plots, a scatter fit, a kernel density and a receiver operating characteristic are reported. The design requires 52 clusters to detect a 5 percentage point effect at 80 per cent power. Fee-threshold studies of dispute escalation are feasible only when cluster counts are planned around the intraclass correlation, and it specifies the conditions under which a before-after-control-impact design can separate a fee effect from secular trends in adjudication.