Cost Assessment of Procurement with Confirmed Violations at the Notice Publication and Bid Submission Stage

Download the report

Abstract

International estimates of losses to corruption in public procurement, the familiar 10 to 25 per cent of contract value, are obtained from the top down and rest on a citation chain that ends in a 2006 practitioners' estimate with no sample and no method. This report measures a quantity of a different nature: the share of the value of procurement procedures carrying a corruption-risk signal that a review authority confirms as a violation before the contract is concluded.

The study uses the CILC open dataset of 427 procurement procedures under Russian laws 44-FZ and 223-FZ, with a combined initial price of RUB 980.5 million (about USD 11.8 million), in 68 regions, with bid deadlines from 30 March to 19 August 2026. The procedures were selected by external participants on signs of restricted competition and carried through to a decision of the Federal Antimonopoly Service; in 369 of them a decision was taken on the merits, in every case before the contract was concluded.

Findings. A violation was confirmed in 92 of the 369 procedures reviewed on the merits, worth RUB 188.0 million (about USD 2.3 million): 24.9 per cent by number, 20.3 per cent by the value of the procedures reviewed and 19.2 per cent of the value of the dataset. The rate is stable across procurement methods (19 to 26 per cent), legal regimes (23.8 per cent under 44-FZ, 27.4 per cent under 223-FZ), types of contracting authority and months, but it ranges from 0 to 55 per cent between territorial review authorities facing the same source of signal. Intervention affected 245 procedures worth RUB 669 million, three quarters of that value being suspensions on complaints ultimately not upheld: this measures the procedural effect of challenge, not violation. At the second instance none of 28 applications was upheld.

The paper separates confirmed violation from procedural effect, explains why the confirmation rate cannot be compared directly with international loss rates or transferred to the national population of procedures without a comparable subsample, and proposes the signal-to-confirmation method as an instrument reproducible on any electronic procurement system.

The report is deposited in English, Russian and Spanish. The underlying dataset is published separately under CC BY 4.0 (DOI 10.5281/zenodo.22236602) together with the anonymisation and packaging scripts, so that every table in the paper can be reproduced without recourse to the primary register.

Recommended citation

Krukovskiy, I. (2026). Cost Assessment of Procurement with Confirmed Violations at the Notice Publication and Bid Submission Stage. Preprint. Consejo Internacional para la Lucha contra la Corrupción. DOI 10.5281/zenodo.22237028. https://cilclegal.org/