Final Report Summary - INVEXPECTATIONS (Investors' expectations: Measuring their nature and effect) Economic predictions often rely on the expectations that the economic agents are assumed to have. The project contributes to the measurement and analysis of “irrational” expectations, where agents deviate from having an accurate knowledge of the relevant economic processes. We formulate and test multiple systematic deviations from accurate expectations. For example, it is found that cumulative price growth is especially hard to understand if the growth rate is random. Not only do investors tend to underestimate the average growth, but they also neglect the possibility that the randomness in growth leads to “skewed” results, where small increases may be much more likely than large increases and the average is far away from the typical outcome. As another example, we consider group contexts where agents learn from other agents and find there is a strong neglect of redundancy: human decision makers are often unable to recognize that the same information keeps being repeated. Instead, the typical pattern is that early statements are repeated often and gain more and more weight as the group size increases. We also measure expectations about relevant economic variables, and investigate how they can be influenced. For example, we find that trust in others, measured by a person’s investment in a joint project with another person, can be installed by influencing the beliefs in the trustworthiness of others. Similarly, information about past stock market returns can result in an increased investment level.
