Is there a net economic loss from employing reference class forecasting in the appraisal of hydropower projects?

Abstract

This paper investigates the potential effects of the use of reference class forecasting on the World Bank's financing decisions, and quantifies the net economic impact of such decisions in the long run. A set of 57 World Bank-financed hydropower projects constructed between 1975 and 2015 was selected based on data availability. The findings show that reference class forecasting can help reduce net losses by preventing some hydropower projects with negative economic net present values from being executed. However, it also leads to the forfeiture of even larger amounts of net economic benefits by causing the rejection of some projects that are found, from ex-post analysis, to be economically worthwhile. Furthermore, because of the increased ex-ante rejection of projects, the loss of potentially economically positive projects from the portfolio of hydro dam projects is greatly increased. The errors in the estimation of economic net present values of these hydropower projects are highly positively correlated to the errors in the estimation of the benefits and only weakly negatively correlated to the errors in the estimation of costs.