In reality, when making a sequence of decisions, people are often not time consistent. They may plan to make a certain decision at a future date, but do something different when they reach that point in time. Furthermore, different people have different time preferences. In carrying out their decisions, some people may be present-biased and appear as irrationally 'impatient' or 'impulsive'. Some people may be 'future-biased' and only care about what will happen in the end. Such time preferences are often very different from the standard one with geometric/exponential discounting that is still dominant in Economics. Therefore it is very important that, to understand behavior over time, we develop a theoretical framework that can allow for as a large class of time preferences as possible. The main goal of this project is to provide a tool to enable us to incorporate a rich class of non-standard time preferences into the standard economic models in dynamic and strategic settings.
The theoretical framework developed here will be applied to a variety of problems for which, although non-standard time preference can be crucial, researchers have been using either the standard time preference or a restricted class of time preferences due to lack of alternatives. We have already started applying our theory to the standard repeated Cournot competition model and obtained many interesting results about the structure of collusion among firms. The second application is an application to the model of growth with imperfect altruism and bequest. In this model, which is structurally a special case of our model, each generation lives for one period and cares about future generations only imperfectly. As most of the past works use very stylized time preferences, we can contribute to this literature by introducing a richer class of time preferences. The third one is an application to the model of addiction. We plan to apply our method to introduce more general time preferences into the model of addiction to enhance our understanding of this phenomenon.
We believe that a variety of applications of our theory will have broader impacts on our society beyond the field of economic theory. The application on collusion will help us identify some possible patterns of collusive behavior in actual markers, which might have been missed in the view of the standard economic theory. The application on growth and imperfect altruism will provide new insights that are useful for shaping a government's very long-term policy decision (such as environmental policy). Our new insights on addiction could be informative and useful to general public.
