CILA: Enhancing accuracy in insurance forecasting with behavioural and decision science
Part of the Current Issues in Life Assurance (CILA) 2023 Webinar Series For those forecasts characterised by a lack of data, extreme events or high uncertainty, expert judgement is a powerful, under-utilised ally for both costing and reserving decisions. The quality of expert judgement can vary widely, though. Behavioural economics studies indicate that cognitive biases and inconsistencies (noise) undermine the quality of expert judgement. We aimed to identify the degree of noise and bias in various re/insurance decisions and test the effectiveness of a custom-built toolkit for boosting the accuracy of expert judgement. We achieved substantial improvements in forecasting accuracy in a pre-post analysis (5-40%) when applying our toolkit to a variety of forecasts.