G.2  HM Treasury: Central Unit of Procurement (CUP)

Public departments provided an annual return to HM Treasury's Central Unit of Procurement (CUP) recording progress on works projects. This was used to monitor performance, establish trends and plan CUP's programme of future guidance. The CUP study investigated construction project cost and time overruns and was based on information provided in 1994-95. Projects are reported first in the year that construction starts, then annually until completion and, finally, after full commercial settlement. For most departments, only projects with a forecast outturn cost greater than £1m were reported. A total of 807 projects were included in the study with an average value of £10.9 m. All cost estimates have been brought to a common cash price basis by removing, where necessary, the calculated effect of inflation (using tender and cost price indices where appropriate).

Table 11  Comparison of CUP and MM Studies

There is a significant difference in optimism bias levels recorded by the two studies. The Mott MacDonald study is based on projects implemented in the last 20 years whereas the CUP study is based on projects implemented more recently (in the last 5 to 10 years). The CUP study was carried out every year from 1990 to 1994 and the results show that both the capital expenditure and works duration optimism bias levels for the approval stage decreased from past to present (See Table 12). The average results for the Mott MacDonald study are similar to the results of the CUP study for 1990 to 1991. However the average results of the Mott MacDonald study are higher than the CUP results recorded for 1994 to 1995. This discrepancy is attributed to the fact that the Mott MacDonald study results are averaged over 20 years.

Table 12  Results of CUP study from 1990 to 1995