Existing measures of corruption often suffer from bias and are too broad to guide policy or test theories. This paper proposes three new indirect indicators of high-level corruption in public procurement, using contract and organisation-level administrat The first is a composite score expressing the probability of corruption occurring in public procurement tenders based on the incidence of ‘red flags’ associated with barriers to competition and unusually high winner market share. The second is a binary variable marking companies whose public procurement market success depends on the government in power, generated by tracking companies’ market volumes from before to after the government changes. The third is a binary variable marking those public procurement winners whose owners or managers are or were public officials, building on the literature on conflict of interest and the revolving door. These new ‘objective’ indicators are used to demonstrate the utility of the approach taken for the study of other questions, such as the role of development aid (including EU Funds) in institution-building, the effect of elite composition on patterns of state capture, and the impact of civil service pay on high-level corruption. Creative Commons Attribution-Non-Commercial-Share Alike 2.0 UK: England & Wales; http://creativecommons.org/licenses/by-nc-sa/2.0/uk/