Saïd Business School research finds cost-benefit analysis of public investments are highly inaccurate and biased

New research from Bent Flyvbjerg, First BT Professor and Inaugural Chair of Major Programme Management at Saïd Business School, and Dirk W. Bester, professional statistician, finds that cost-benefit analyses of public investments are highly inaccurate and biased.

The bad news is that this means the many cost-benefit analyses underpinning most public investments — from HS2 in the UK to California’s high-speed rail line — cannot be trusted as currently done. The good news is that Flyvbjerg and Bester have developed better methods and data to overcome this.

Professor Flyvbjerg found that amongst forecasters, policymakers and scholars there tends to be a misguided assumption that cost-benefit forecasts are more or less accurate, when in fact they are highly inaccurate and biased. This fallacy, called the “cost-benefit fallacy” results in overestimates of benefit-cost ratios of 50 to 200 per cent, with individual estimates often exceeding these averages. The fallacy makes investments look profitable on paper, when in reality they are often unprofitable. 

The cost-benefit fallacy is shown to be fundamental and pervasive, which has never been documented before.

2,062 public capital investments were considered during the study, these consisted of dams, rail developments, tunnels and bridges amongst other major infrastructure projects. The sample included investments from 104 countries dating between 1927 and 2018, the older investments allowing historical trends to be considered. 

Professor Flyvbjerg recommends that to redress these cost-benefit failings the following method should be followed:

(a) Behavioural science should be used to address biases in cost-benefit analysis

(b) Incentives should be used to encourage this

(c) Independent audits should be used to verify results

(d) Other real-life concerns that need consideration alongside cost-benefit analysis should be integrated through normal democratic processes

Professor Flyvbjerg commented: ‘The findings are crucial for all policymaking based on cost-benefit analysis. Policymakers will make the wrong decisions if they trust cost-benefit analysis as it is practiced today. We don’t stop with pointing out the problem, however, we also present solutions. This research used an incredibly large dataset, the biggest of its kind in the world. But precisely because of the many data our conclusions are firm: the cost-benefit fallacy is a fact that must be overcome if good policymaking is the goal.’

The full research is the ‘Paper of the Month’ in the Journal of Benefit-Cost Analysis is available here

About Saïd Business School, University of Oxford

Saïd Business School at the University of Oxford blends the best of new and old. We are a vibrant and innovative business school, but deeply embedded in an 800-year-old world-class university. We create programmes and ideas that have global impact. We educate people for successful business careers, and as a community seek to tackle world-scale problems. We deliver cutting-edge programmes and ground-breaking research that transform individuals, organisations, business practice, and society. We are a world-class business school community, embedded in a world-class University, tackling world-scale problems. www.sbs.ox.ac.uk