What Drives Public Contract Prices in a Developing Economy? Institutional, Procedural, and Market Evidence from Kenya

Public procurement absorbs a large share of government spending in developing economies, yet contract-level evidence on price determinants in Africa is scarce. Using 14,181 Kenyan contract awards from the World Bank's ProACT database, we estimate log price models with product market and year fixed effects and buyer-clustered inference. We show that three common uses of these data mislead: indicator scales are reversed, a concentration indicator mechanically contains the contract's own value, and a timing variable is mislabeled as a market code. Corrected estimates indicate open tenders and direct awards are priced roughly eight and five times higher than requests for quotations, statistically indistinguishable, revealing threshold-based sorting and noncompetitive awards at large scale. Supplier experience robustly lowers prices, including within suppliers, while market share modestly raises them. Short decision periods and missing tender calls carry no robust price premium. Red flags therefore need unit price benchmarks before guiding audits.