No abstract

We propose an experimental design to investigate the role of information disclosure in the market for an experience good. The market is served by a duopoly of firms that choose both the quality and the price of their product. Consumers differ in their taste for quality and choose from which firm to buy. We compare four different treatments in which we vary the degree to which consumers are informed about quality. Contrary to theoretical predictions, firms do not differentiate quality under full information. Rather, both tend to offer products of similar, high quality, entailing more intense price competition than predicted by theory. Under no information, we observe a "lemons" outcome where quality is low. At the same time, firms manage to maintain prices substantially above marginal cost. In two intermediate treatments, quality is significantly higher than the no-information level, and there is evidence that prices become better predictors of quality. Taken together these findings suggest that information disclosure is a more effective tool to raise welfare and consumer surplus than theory would lead one to expect. (JEL L15, C91, D82)

No abstract

No abstract

The present research analyses the ‘social visualization’ tool Sense.us, a commercial interactive Web application in which U.S. Census data are being visualized. Sense.us was developed as a tool for social data exploration and interaction, in which it would be worthwhile to pay attention to the socio-cultural values that have driven the collection and categorization of the underlying U.S. Census datasets. It is argued that closer attention to value-driven U.S. Census statistics would greatly enhance the social appeal of Sense.us, and would be a logical next step in the development of online social visualization tools. In order to allow for explicit socio-cultural values of statistics in online visualizations, three strategies are offered: pro-active annotation; more attention to visual aesthetics; and, a tighter integration of user profiles and represented data.