G.3  HM Treasury: Supply Estimates

The Supply Estimates (SE) study dataset includes 283 capital projects with a value of at least £10 m (at 2001 prices) undertaken between 1981 and 1998 and are listed in SE and Departmental Reports. HM Treasury provides the Supply Estimates and the projects cover a range of government departments.

The results of the Supply Estimates study were grouped into project sectors: defence, health, criminal justice, transport, Inland Revenue, Customs and Excise, and Department of Social Security. The Mott MacDonald study was divided into similar sectors to allow comparison of results with the SE study.

Table 13  Comparison of Supply Estimates Publication and MM Studies

The average capital expenditure optimism bias measured in the SE study is about half of that for the Mott MacDonald study. This is due to the presence of one or two projects within each category that experienced very high optimism bias levels as measured in the Mott MacDonald study. These projects tended to be non-standard or innovative and so are expected to have high optimism bias due to their design complexities.

It is unknown if the effect of inflation has been excluded when measuring optimism bias in the SE study. If this is not the case, then the effect of inflation or variation in price indexes may explain the correlation between size of projects and optimism bias. Larger projects tend to have longer works duration as compared to smaller projects and are thus more vulnerable to price fluctuations. There is little difference in the works duration optimism bias.

Analyses carried out on the SE study showed a statistically significant tendency for the cost overrun to increase with the size of the project in all sectors. However, the Mott MacDonald study did not show a relationship between project size and optimism bias. Mott MacDonald's major project experience has shown that as a project increases in size, its complexity also increases and an increased effort is required to control the project in terms of managing project staff, programme, communications, project stakeholders, resources and variations. More project risks within project risk areas are associated with larger projects, which would be expected to contribute to a larger optimism bias. However, the lack of correlation can be explained due to the active mitigation of risks based on previous experience of project managers and/or increased works duration and capital expenditure allowances made during the strategic planning stage as project size increased for the projects studied.

For example, when comparing the construction of a minor bridge to the construction of a major 150-foot bridge over a river, it can be expected that the latter project will be exposed to greater risks e.g. more project stakeholders (local councils, local residents, Environmental Agency), increased ground risks (ground properties may be more variable due to presence of river), increased public relations issues (bridge may affect view of river, local residents may be against possible increase in traffic flow) and changes in construction standards. Without risk mitigation strategies in place, the optimism bias levels of the major bridge construction project are expected to be higher than that of the minor bridge project. However, if the project manager for the major project has had similar construction experience, then he may put in place strategies (e.g. public consultation, more ground investigation) to mitigate expected risks or include risk contingencies, in terms of capital expenditure and time, in the business case estimates. The actions of the experienced project manager could reduce the optimism bias levels for the major bridge construction project.