Statistical methods as a tool to identify bid-rigging: the case of local authorities Cover Image

Statistical methods as a tool to identify bid-rigging: the case of local authorities
Statistical methods as a tool to identify bid-rigging: the case of local authorities

Author(s): Łukasz Ziarko
Subject(s): Public Administration, Methodology and research technology
Published by: Wydawnictwo Naukowe Uniwersytetu Mikołaja Kopernika
Keywords: public procurement; local authorities; bid-rigging; screening methods;

Summary/Abstract: Motivation: Disclosure of bid-rigging is not a trivial process. The main difficulty lies in the secrecy of such an agreement. Also, the signals of collusion can be ambiguous. It, therefore, appears that only a tiny percentage of such collusive agreements are disclosed. Of the disclosed conspiracies, a significant proportion was identified due to information from contracting authorities. We have a catalogue of indicators for collusion, and statistical methods are particularly effective. What is the role of statistical methods in revealing bid-rigging? What determines their use in practice? In order to answer these questions, a questionnaire study was carried out. The survey covered local government units — the dominating group of contracting authorities. Aim: The research aimed to identify the factors characterising local government units that foster statistical methods as a standard bid evaluation tool. Results: The survey results indicate that using a statistical method as standard practice in the process of bids evaluation is related to the level of staff’s professional expertise, the size of the procurement team and the size of the local authority. Concerning the risk of collusive bidding, respondents recognise it but believe that it does not significantly impact the achievement of procurement policy objectives. Despite a sense of responsibility for combating bid-rigging, respondents are reluctant to use statistical methods on a daily basis.

  • Issue Year: 21/2022
  • Issue No: 3
  • Page Range: 655-669
  • Page Count: 16
  • Language: English