The impact of distributional preferences on (experimental) markets for expert services

Credence goods markets suffer from inefficiencies arising from informational asymmetries between expert sellers and customers. While standard theory predicts that inefficiencies disappear if customers can verify the quality received, verifiability fails to yield efficiency in experiments with endogenous prices. We identify heterogeneous distributional preferences as the main cause and design a parsimonious experiment with exogenous prices that allows classifying experts as either selfish, efficiency loving, inequality averse, inequality loving or competitive. Results show that most subjects exhibit non-standard distributional preferences, among which efficiency-loving and inequality aversion are most frequent. We discuss implications for institutional design and agent selection in credence goods markets.