Detection of collusion in highway construction contract bidding

Lee, T S (1990) Detection of collusion in highway construction contract bidding. Unpublished PhD thesis, University of Wisconsin - Madison, USA.

Abstract

According to the U.S. Department of Justice, many cases of collusion were found in 15 states. It was discovered that 212 individuals and 195 corporations during 1975 to 1980 were engaged in collusive activities. The Department of Justice estimated the cost of collusion for the total highway construction sector in the U.S. was 35-40 billion dollars based on a 10% annual increase in costs over a 40-year time period. Under these conditions, more effective collusion detection methods which are based on a knowledge of the behavior of collusive bidding should be developed. The overall objectives of this research investigation are (1) to develop a basic understanding of the differences between collusive and competitive bidding behavior in the highway construction industry, and (2) to develop statistically based methods for identifying collusive bidding behavior. The achievement of these objectives should permit State Departments of Transportation to develop and implement more effective procedures for detecting and thus reducing collusive bidding activities. In order to build models that identify collusive behavior, historical bidding data on highway construction contracts that includes information on collusive behavior are needed. From the basic bidding data, additional variables for identifying collusion were constructed. The relationships between variables as shown in one way frequency distributions, correlation matrices, cross-tabulations and cross-classifications models, statistical tests for equality of means, and regression models were presented. Goodness-of-fit measures were obtained for the cross-classification and regression models. After identifying potential variables for explaining collusive behavior, discriminant models were developed. In this research, a basic understanding of the differences between collusive and competitive bidding behavior in the highway construction industry was developed. Also, statistically based methods for identifying collusive bidding behavior were developed and tested.

Item Type: Thesis (Doctoral)
Thesis advisor: Smith, R L
Uncontrolled Keywords: collusion; collusive bidding; competitive bidding; construction sector; highway; bidding; correlation matrices; frequency distribution; regression model
Date Deposited: 16 Apr 2025 11:25
Last Modified: 16 Apr 2025 14:15