Uncertainty and the discrepancy between rate of return estimates at project appraisal and project completion

This report analyzes the World Bank's experience with project analysis from a sample of 1,105 projects. The authors compare estimated rates of return at appraisal with re-estimated rates of return at the project completion. Their findings confirm a high degree of uncertainty in project analysis. Only a small part of the discrepancy between estimated rates of return at appraisal and the re-estimated rates of return at project completion can be explained, even with the benefit of hindsight. They conclude that World Bank appraisal estimates of rates of return are too optimistic, and that uncertainties seems to be higher in the directly productive sectors (agriculture and Industry). Estimated rates of return seem more stable for infrastructure projects. One alternative to correcting modal estimates of implementation variables for"bad surprises"might be to set different minimum rate-of return criteria for different types of projects based on observed divergences in rate of return. Traditional methods of project evaluation and selection have been unable to reduce this large measure of uncertainty.