Essays on Imperfect Information Processing in Economics

Economic agents generally operate in uncertain environments and, prior to \nmaking decisions, invest time and resources to collect useful information. \nConsumers compare the prices charged by di..erent firms before purchasing \na product. Politicians gather information from di..erent sources before \nimplementing a policy. Doctors conduct several tests on patients before subscribing \na treatment. Since having access to valuable information requires \ntime, e..ort, and often money, a problem typically addressed by economists \nis how much information a decision-maker should collect before being su..- \nciently convinced about one state of the world. Implicit in economic models \nthat aim to identify the optimal amount of information to be gathered is the \nidea that, once information is available, there are no costs involved in interpreting \nand transmitting it. However there are plausible reasons to believe \nthat the activity of information processing is far from being perfect. \nLooking at the quantitative dimension of information, a larger amount \nof information usually implies that more resources are needed in order to \ninterpret it. In this respect, with the rise of the Internet, the cost of generating \nand transmitting information has fallen so dramatically that the \nscarce resource is not anymore information itself, but the attention needed \nto understand it. For example, the cost of sending an electronic manuscript \nto a journal is much lower than the opportunity cost of the time it takes \nto a referee to read and understand the paper. Modelling explicitly such \nopportunity costs may shed light on phenomena such as market congestion \nand information overload.