Credence goods markets - Theory and experiments

Credence goods (like a medical treatment or a car repair) have the following characteristics. Though consumers can observe the utility they derive from the good ex post, they cannot judge whether the quality they received is the ex ante needed one. Moreover, consumers may also be unable to observe which quality they actually received. An expert seller is able to identify the quality that fits customers` needs by performing a diagnosis. He can then provide the right quality and charge for it, or he can exploit the information asymmetry by defrauding customers through providing an insufficient or unnecessary product (undertreatment, respectively overtreatment) or by charging for more than the expert has actually provided (overcharging). In sum, credence goods markets may suffer from a variety of inefficiencies. We study credence goods markets both in theory and by means of a large-scale experimental study. In the theoretical part we extend existing models of credence goods markets - which almost always assume rational and profit-maximizing subjects - by allowing for non-standard preferences, covering, e.g., intrinsic motivation, a preference for honesty, guilt aversion, or inequality aversion. We furthermore consider reputation building of expert sellers and competition between sellers. In such a theoretical framework we can examine key questions such as a) whether the existence of non-standard preferences can be expected to alleviate the problems associated with the provision of credence goods, b) whether experts with non-standard preferences will be driven out of business by those experts with standard preferences, c) whether in a repeated interaction subjects with one type of preferences will have an incentive to mimic subjects with the other type of preferences, or d) whether reputation building or competition increases the efficiency in credence goods markets. We will then expose the theoretical predictions to an extensive experimental test. Such a behavioral test is important, because it will clarify whether the factors identified in theory as capable of curing the inefficiencies in credence goods markets are behaviorally relevant and how different factors (such as liability, verifiability of actions, reputation building or competition) compare to each other in their effects on the efficient provision of credence goods. The insights from both the theoretical and the experimental part of the project will provide guidance for economic policy makers (i.e. market designers) on how an optimal design of the institutional and legal framework of credence goods markets should look like.

This project has investigated theoretically and experimentally the economics of credence goods (like medical treatments, car repairs and taxi rides in an unknown city). A core characteristic of such goods is that experts have better information about the appropriate quality than their customers. Moreover, consumers are often unable to observe the quality they actually received. As experts typically provide both diagnosis and treatment, this leaves scope for fraud. Two types of fraud are particularly important in this kind of market, provision of an inappropriate quality (under-or overtreatment), and charging for a more expensive quality than provided (overcharging). In a large-scale laboratory experiment, we have investigated which institutional framework protects customers particularly well from being defrauded. While theory predicts that liability or verifiability of the provided quality yields efficiency, we find that liability has a crucial, but verifiability at best a minor effect. Allowing sellers to build up reputation has little influence, while competition yields maximal trade, but does not lead to higher efficiency as long as liability is violated. The crucial impact of liability rules is remarkable because such rules only eliminate the problem of undertreatment without directly affecting the problems of overtreatment and overcharging. Evidence indicating that overtreatment may lead to substantial efficiency losses in real markets for credence goods is provided in a follow-up study. It shows that the overtreatment rate increases sharply when professional car mechanics act as sellers in our lab experiments. The failure of verifiability to yield efficiency is investigated further in a theoretical follow-up study. It argues that in economic decisions humans are often motivated by items beyond material self-interest and shows theoretically and verifies in lab experiments that non-trivial "distributional preferences" (as altruism, inequality aversion, spite, or envy) of sellers can explain the failure of verifiability to increase efficiency. An important insight of that study is that instead of choosing doctors, mechanics or computer specialists exclusively according to their training, customers on markets for credence goods should worry more about the attitudes of these experts towards their customers. In a further study we combine theory and lab experiments in an investigation of the impact of 'soft' factors on market performance. An important insight from that study is that non-binding promises reduces the amount of fraud significantly although standard theory predicts that it does not. In several field studies we have investigated determinants of fraud by real experts in real markets for credence goods. In one of these studies we show that taxi drivers exploit their information advantage over passengers in a systematic way: While passengers with inferior information about optimal routes are taken on longer detours, lack of information on the local tariff system increases the likelihood of manipulated bills. An important insight of that study is that conveying to an expert seller the impression of possessing relevant information (be it true or not), or at least refraining from revealing one's lack of information, can alleviate the problems associated with the provision of credence goods. In a second field study in a market for taxi rides we show that taxi drivers exploit their information advantage more if they expect that the expenses will be reimbursed by their employer.