Rob Axtell
Krasnow Institute
George Mason University

Other things being equal, people will accept less reward in the near term instead of more in the long run. This behavior is called discounting the future (or simply discounting) and is a cornerstone of both practical finance (interest rates!) and theoretical models in economics. It is known that the only consistent way for an individual to discount the future is exponentially at constant rate. To do anything else induces changes in behavior part way through the time period over which discounting is occurring, leading to inconsistent behavior. However, a wide variety of behavioral experiments clearly demonstrate that people are not exponential discounters, but rather discount the near term more than the long run, which may lead to inconsistent behavior over time. In this talk I will present a novel theory for such ‘hyperbolic’ discounting and demonstrate that, far from being exceptional, ‘anomalous’ discounting is a generic feature of human behavior. I will show how experimental data on discounting behavior can be represented mathematically using the theory. The formalism can be given ‘multiple selves,’ ‘competing substructures,’ and risk-based interpretations, and is capable of rationalizing old adages like ‘A bird in the hand is worth two in the bush.’ The mathematical formalism leads naturally to social questions, concerning how groups or whole societies discount collectively. I will show that fMRI, laboratory, survey, and natural experiment data concerning how people value the future can be rationalized in terms of the new theory, and will conclude by applying it to a broad range of real-world problems including climate change and the term structure of interest rates.