Optimism Bias and World Bank Project Performance

Abstract

This paper examines the correlates of optimism bias and its impact on World Bank project performance. We measure optimism bias in different ways using estimated Economic Rates of Return (ERR) of projects at approval and closure. We examine over 2,800 World Bank projects that were appraised between 1960 and 2019. We find that approximately 60% of projects in the sample were prone to optimism bias. Correlates of optimism bias include both project and country characteristics. Findings also indicate that the incidence of optimism bias reduces the chance of a satisfactory project performance rate at the time of evaluation by 17–20%. Recommendations include embracing complexity and uncertainty in considering projects for approval, providing organizational incentives for ensuring projects are successful rather than ERRs being accurate, shifting some resources from appraisal to implementation, and changing the nature of project supervision.

Keywords:

• Optimism bias
• economic rate of return
• project performance
• project appraisal
• project evaluation

JEL classification codes:

• D61
• D83
• F35
• O22

Acknowledgments

Thanks to Professors Peter E.D. Love, Alberto Posso, Ashton de Silva, Osvaldo Feinstein, Christophe Bredillet, Michele Alacevich, Anna Zhu, editor Oliver Morrissey, and two anonymous reviewers for comments and suggestions. All errors are our own. The dataset is available upon request by email to authors.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 While there are micro-economic papers that look into project characteristics such as project preparation and supervision (e.g., Kilby, 2000), they only measure how much effort or time is devoted to those activities (e.g., World Bank staff weeks). In so doing, these papers fail to unravel the ‘black box’ of management processes and thus do not explain how outputs are really derived from inputs in projects (Ika, 2015). As such, as Coase (2012) argues, they leave World Bank staff with their own management acumen, gut-feelings, and rules of thumb for making projects successful.

2 Thanks to two anonymous referees for identifying this purely mechanical effect.

3 As noted earlier, the ‘optimism bias debate’, notably in infrastructure project settings, has focused a good deal on its incidence.

4 There were 138 projects in the database that had an ERR at approval (133 also had an ERR at closure) of less than 10%; this is not surprising; for example, in their sample, Pohl and Mihaljek (1992) found 46 such projects out of 1,015 projects).

5 There were 154 projects assigned with this value. Results remain consistent to their inclusion.

6 Financial Intermediary Loans provide long-term resources to local financial institutions to finance real sector investment needs. Specific Investment Loans support the creation, rehabilitation, and maintenance of economic, social, and institutional infrastructure. Sector Investment and Maintenance Loans focus on public expenditure programs in particular sectors. They aim to bring sector expenditures, policies, and performance in line with a country’s development priorities (World Bank, 2015). The dataset does not include any measure of project complexity. While project size (e.g., financial cost) could have been an ‘imperfect proxy’ for complexity (Denizer et al., 2013), this information was not available for a large number of projects. The same goes for the only proxy for uncertainty in the dataset: risk to development outcomes.

7 We also included an implementation duration squared variable to account for a non-linear relationship, but it remained statistically insignificant.

8 While inflation can impact significantly on project costs and benefits, it is unanticipated inflation, a sort of unforeseen uncertainty (Pohl & Mihaljek, 1992), which is important to the analysis of optimism bias.

Additional information

Funding

The research has been funded by the Major Projects Observatory of the Telfer School of Management (uOttawa).