The Expert and The Charlatan: an Experimental Study in Economic Advice

How do people choose what economic advice to heed? We develop a set of validated multiple-choice questions on economic policy problems, to examine empirically the persuasiveness of expert versus populist advice. We define populism as advice that conforms to commonly held beliefs, even when wrong. Two (computerized) advisers suggest answers to each question, and experimental participants are incentivised to choose the most accurate adviser. Do participants choose the high-accuracy adviser (`the Expert'), or the low-accuracy one (`the Charlatan'), whose answers are designed to be similar to the modal participant's priors? Our participants overwhelmingly choose the Charlatan, and this is only slowly and partially reversed with sequential feedback on the correct answer. We develop Bayesian models to determine optimal choice benchmarks, but find that behavior is best explained by a naive choice model akin to reinforcement learning with high inertia.