Demand for public goods.

摘要

The first chapter of this dissertation explores stated choice data from an experiment which specifies explicit risks of failure and redundancy for a publicly-sponsored research and development program. These data are used to calibrate a discounted expected utility model. The choice model is generalized to accommodate the possibility of subjective scenario adjustment or selective inattention by respondents which is then used to counterfactually simulate the demand for the public good in the absence of these distortions.;The second chapter uses fMRI data to examine the motives that people have for contributing to public goods. One possible reason people make contributions is called "pure altruism" and it is satisfied by increases in the level of the public good. Another possible explanation is called "warm glow" and it depends on the level of the individual's own voluntary contribution. Consistent with a pure altruistic motive, mandatory (tax-like) transfers of the subjects' own funds to a charity elicit neural activity in areas linked to the processing of rewards. Moreover, these responses can predict subsequent decisions in choice situations where subjects can accept or reject a proposed transfer. However, consistent with a warm-glow motive, neural activity in these same regions is higher in voluntary treatments. This suggests that both pure and warm-glow altruism play a role in the demand for public goods.;The third chapter explores how the exertion of mental effort changes between choice situations in which people make tradeoffs between their own funds and contributions to public goods. Specifically, it develops a model which predicts that more mental effort will be exerted when the difference in utility between the available alternatives is small. Empirically, individual-specific preferences are calibrated and activations in prefrontal cortical areas are tested against the prediction of the model. Consistent with the prediction of the model, activity in the rostral medial frontal cortex is higher in choice situations where the utility difference between the available alternatives is small.