Humans often lie strategically. We study this problem in an ultimatum game involving informed proposers and uninformed responders, where the former can send an unverifiable statement about their endowment. If there are some intrinsically honest proposers, a simple message game shows that the rest of them are likely to declare a lower-than-actual endowment to the responders. In the second part of the paper, we report on an experiment testing this game. On average, 88.5% of the proposers understate the actual endowment by 20.5%. Regression analysis shows that a one-dollar gap between the actual and declared amounts prompts proposers to reduce their offer by 19 cents. However, responders appear not to take such claims seriously, and thus the frequency of rejections should increase. The consequence is a net welfare loss, that is specific to such a "free-to-lie" environment.

This paper analyses the consequences of young researchers' scientific choice on the dynamics of sciences. We develop a simple two state mean field game model to analyse the competition between two paradigms based on Kuhn's theory of scientific revolution. The dynamics of the model are driven by the scientific choice of young researchers at the beginning of their career. Despite the possibility of multiple equilibria, the model exhibits at least one stable solution in which both para digms coexist. The occurrence of shocks on the parameters may induce the shift from one paradigm to the other. During this shift, researchers' choice is proved to be having a great impact on the evolution of sciences.

This paper estimates the relation between the size and quality of scientists' co-author networks and individual characteristics (notably productivity) in the context of institutional changes in French academia in the mid-1980s. The analysis employs the Two-Stage Residual Inclusion (2SRI) framework to handle endogeneity in individual productivity relative to the quality of co-authors. Data is taken from a novel database of French academic economists. The main finding is that the size and quality of authors' networks are positively related to their productivity; this is understood as evidence of assortative matching. Other effects on coauthor networks (life-cycles, specialities) are also identified.

Criticism is mounting on business schools for their excessive focus on research and for neglecting teaching. We show that if students have imperfect information about a school's overall capabilities and if business schools differ in their research productivity, the least productive schools may do as much research as the top-tier ones only to manipulate students' expectations. In turn, the most productive schools might resort to excess research in order to signal their type in the eyes of future students. This signaling equilibrium is characterized by a relative neglect of teaching by the top-tier schools. Such a situation is socially inefficient as compared to the perfect information case.