The question of discounting in assessing values of future utility streams has played an important role in many fields of economics. Discounting is one of the key determinants affecting inter-temporal trade-offs and in making an explicit account of the distant future, which therefore encompasses motives related to sustainability, transmission to offspring and altruism. Many models of preferences over time assume the objective is obtained by a separable, exponentially discounted utility. Such (rational) models are normative.
They prescribe how decisions should be made in order to avoid inconsistencies or non-optimal solutions, e.g. normative representations of preferences over time are also proposed by social sciences and include concepts such as just savings or sustainable development. On the contrary, behavioural models are descriptive. They try to accommodate behaviour that is consistent with empirical or experimental observations. With the notable exception of exponential discounting, both models of preference over time (normative and behavioural) typically lead to time-consistency problems, i.e. optimal plans under such preferences are time-inconsistent, and a decision maker has no incentive to follow the optimal plan in the future.
The question of design and computation of optimal among time consistent plans (and thus the one that will be actually followed) has received much attention in economics. Taking this literature into account, the aim of the project is: to identify the impact of selected forms of behavioural discounting on the existence and characterisation of optimal decisions in dynamic stochastic economies and to develop tools for its constructive (and numerical) analysis, including time-consistent decision rules applying to sustainable growth.