There are significant differences between the findings of the Mott MacDonald study and those of the three similar studies. Some explanations for these have been identified, although a full reconciliation cannot be provided. Although the evidence is somewhat mixed, the following conclusions are relevant:
• All the major studies that have researched this area in detail have found optimism bias, although of varying magnitudes.
• The detailed review of the projects in the Mott MacDonald study has shown that, if key project risks are not managed, then high levels of cost and works duration overruns are very likely to occur. The other studies also found many instances of very high optimism bias, although with lower mean and median values overall.
• The aim of the guidance provided in this paper is, ultimately, to prevent high levels of cost and works duration overruns. The prescribed adjustments, therefore, tend to be based prudently on the higher levels of optimism bias that the Mott MacDonald study has found, rather than the lower levels found in the other studies. The emphasis is on setting high initial optimism bias levels, which can be reduced if good practice in project management can be demonstrated.
• The upper bound optimism bias guidance, nevertheless, are lower than the Mott MacDonald study findings, given some recent improvements in procurement, and the omission of the most significant outliers in the Mott MacDonald study.
• The other studies did not investigate in such depth the causes of optimism bias, which is a key part of the Mott MacDonald analysis, and of the prescribed guidance.