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

Hydropower is currently the largest renewable energy source worldwide, providing about 16.4% of global electricity. With the continued calls to expedite the transition from fossil fuels to renewable energy sources, investments in hydropower have increased significantly. Investment in the hydropower is marked by high upfront costs, complex financing models and long-time horizons. Governments have had to come up with innovative ways to address project financing and reduce risks. Moreover, the construction of hydropower plants is generally complex due to a lack of reliable and adequate information during project appraisal stages. This increases sources of uncertainty, such as cost overruns, which could result in huge losses. Thus, improving the cost estimation accuracy for high-power power projects is very significant. Traditional methods for forecasting and evaluating hydropower project focus on the project’s unique characteristics. While this approach is widely recognized, it is less effective as it ignores the relevant information and lessons learned from past similar projects. Therefore, combing the characteristics of the project and the relevant formation from previous projects could produce realistic estimates – an approach known as reference forecasting.

There are three main underlying root causes of benefit shortfalls and cost overruns, namely, psychological factors, politico-economic factors and technical factors due to the use of less effective forecasting techniques. Since technical factors are less important in forecasting estimation errors, employing de-biasing instead of error correction during cost estimation could prevent cost overrun. In particular, correcting potential biases using reference class forecasting has been identified as a promising approach and is applicable to hydropower projects.

On this account, Professor Glenn Jenkins from Queen’s University and President of Cambridge Resources International Inc. together with Professor Godwin Olasehinde-Williams from University of Istanbul and Dr. Saule Baurzhan from Eastern Mediterranean University in Turkey investigated how economic net present value (ENPV) of a set of World Bank-financed hydropower projects could be affected in the long run by using reference class forecasting of investment costs. A set of 57 World Bank-financed hydropower projects constructed between 1975 and 2015 were selected in their study. The main objective was to determine the long-term economic effect of raising the estimate of the ex-ante real cost. Their work is currently published in the journal, Renewable and Sustainable Energy Reviews.

The authors demonstrated the effectiveness of reference class forecasting in reducing the net losses by preventing the construction of some hydropower projects that are highly likely to have negative economic net present values. However, it also resulted in the rejection of some projects at the ex-ante appraisal stage thus forfeiting substantial amounts of net economic benefits that could have been yielded is such projects were to be implemented. The analysis revealed that about 70% of projects encountered cost overruns, with an average cost overran of 24% for all the projects.

The increased ex-ante project rejection further increased the possibility of losing potentially economically positive hydropower projects. For these hydropower projects, the errors in estimating the economic net present values positively correlated to those in estimating the benefits and negatively correlated to those in estimating the costs. Furthermore, it was worth noting that hydropower projects could still yield greater net benefits despite suffering huger cost overruns. Despite the benefit of reference forecasting in reducing the risk of cost overrun, it could also lead to the rejection ex-ante of viable projects. Thus, maintaining high standards of project appraisal quality rather than heavily relying on class forecasting could help address the inherent politico-economic and psychological challenges in infrastructural development.

In summary, Professor Glenn Jenkins and colleagues estimated the cost overrun in 57 World Bank-financed hydropower projects, and a reference class forecast was employed to determine the potential outcomes of these projects. The estimates revealed the frequent occurrence of cost estimate errors in hydropower projects. In a statement to Advances in Engineering, the corresponding author, Professor Glenn Jenkins who is an expert in public finance and policy, investment appraisal and economic development explained that their new findings would improve the appraisal and planning of infrastructural projects such as hydropower projects.