Municipal service charge arrears disclosure rules in Nairobi's middle-income condominium estates create a sharp informational threshold: once arrears exceed a statutory reporting band, managing agents must publish the delinquency schedule to all owners. A regression discontinuity design around that threshold would appear to identify the effect of disclosure on owner withholding, but the clustered structure of estate-level assignment makes the design's power fragile in ways that standard power formulas obscure. the present analysis characterises the operating characteristics of a sharp regression discontinuity estimator applied to a before-after-control-impact framework under a declared variance structure: between-cluster standard deviation of 6.0 percentage points, residual standard deviation of 4.5, and an intraclass correlation of 0.64. Power was estimated across 2000 replications per design cell for cluster counts of 8, 16, 24, 32 and effect sizes of 3, 5 and 8 percentage points.56 clusters are required to detect a 5-percentage-point effect at 80% power, a requirement that exceeds the number of estates in most Nairobi middle-income jurisdictions. The intraclass correlation of 0.64 dominates the design's operating characteristics, and the minimum detectable effect at conventional cluster counts is large relative to plausible disclosure effects. The regression discontinuity design is defensible for this substantive problem only when the cluster count is expanded through cross-jurisdictional pooling or when the estimand is redefined around within-estate variation.
No takes yet. Share an insight, caveat, or question.
Otieno et al. (2026) studied this question.