Answers are grounded in this paper and cite their sources.
Open contracting data are only as useful as they are complete. Using 14,181 Kenyan contract award records from the World Bank's ProACT database, we study which information public buyers fail to publish. Composite transparency scores conceal the answer: 88 percent of records are complete, and in conventional complete case samples the score varies only through the awarded price. Analyzing fields separately, we find that gaps in award dates and supply types, common before 2019, disappeared once reporting changed, while undisclosed prices rose from 4.9 percent of contracts in 2018 to 8.8 percent in early 2021. Logit models with product market and year fixed effects show that framework agreements are 21 percentage points more likely to lack a price and regional agencies 7 points less likely. Buyer identity explains more than twice as much variation in price nondisclosure as contract characteristics, and ten entities account for two thirds of unpriced awards.
No takes yet. Share an insight, caveat, or question.
Ndambiri et al. (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: