Are expert judgments a reliable tool for predicting farmer and food consumer decisions? : experimental evidence from a forecasting survey

Abstract: Often trusted to provide sound recommendations and advice, experts from academia and industry are often relied upon throughout industries around the globe, and the food and agriculture industry is no different. We therefore ask, how accurate are these experts, and are they able to accurately forecast behavior from varying food chain actors such as farmers and consumers? Do these experts have a preconceived bias to one side or the other? These questions become increasingly important when considering policy developments such as the EU Farm to Fork strategy, which seek to integrate the consumer-facing food industry and the producer-forward agriculture industry, two policy realms that have historically remained relatively independent of one another. Utilizing a novel hands-on prediction-based approach to collect global results from food, agriculture, and economics experts and non-experts, we analyzed over 2,300 predictions from 87 respondents to determine the accuracy of expert predictions against actual values and behavior, individually and on average. Results are compared against similar predictions from bachelor’s and master’s students currently enrolled in agriculture-based programs at a Swedish university. We document three primary results; firstly, industry experts do not appear to be more knowledgeable of either given side of the agricultural system, secondly, there is an inverse relationship between higher education and forecast accuracy, and lastly, those with higher self-reported confidence levels showcased incrementally lower predictive accuracy across treatments. We further explore what these findings imply for the future role of experts in the food system.