Procurement and Governance Choices for Collaborative Infrastructure Projects

Abstract

Collaborative approaches to infrastructure procurement are increasingly popular around the world due to their potential to provide improved project performance compared with more traditional approaches. The problem is that project outcomes continue to be unpredictable. Previous research has shown that this is the case regardless of whether the chosen procurement approach is based on price or nonprice selection of the project team. This is a major choice that clients make, but the presented research shows that governance choices for project execution are more important. This is significant because clients tend to focus more on procurement choices and typically do not differentiate governance based on those choices. This needs to change, and the authors show that optimal governance configurations vary on the basis of the chosen type of team configuration. For example, three specific governance arrangements for workshops are highlighted for single teams, and two specific governance arrangements for risk/reward sharing are highlighted for multiple teams. This study identifies governance actions that are associated with superior time and cost outcomes on collaborative infrastructure projects in Australia run by experienced public-sector clients under the two procurement scenarios. Based on a survey of 32 senior managers, independent sample t-tests were conducted to compare the application of governance actions among three distinct groups of projects, based on type of team selection and type of project outcome. The study provides evidence of the most effective approaches to project governance in a country that is a world leader. The results provide much needed recommendations for improved project performance based on large-scale quantitative analysis, which before now has not existed. Overall, the study recommends more attention be paid to noncontractual governance under both approaches to team selection, although the specific actions recommended vary.