Dynamic Consistency, Sample Selection and Attrition: A Panel Experiment on Individual Discount Rates
Dr Hong Il Yoo, Associate Professor in Economics, Durham University Business School, Durham University
Research Seminars
Tuesday, 04 June 2019, 2.30pm to 3.30pm
University of Oxford Old Road Campus, Richard Doll Building
Hosted by HERC
Abstract: Temporal stability in time preferences is a fundamental hypothesis that underlies the predictions of many economic models. We evaluate this hypothesis using data from a longitudinal field experiment in Denmark, with a representative sample of the adult population. Our experimental design allows us to identify a range of exponential and non-exponential discounting models, whilst controlling for the non-linear curvature of the utility function and the effects of sample selection and attrition.
Detecting dynamic consistency or inconsistency in economic decisions requires the use of panel data, making the issue of endogenous selection and attrition inherently pertinent to the present analysis. We propose a structural econometric model that fully harnesses the flexibility that our experimental design offers, in a framework that accounts for endogenous sampling and unobserved heterogeneity in preferences. The model captures within-individual correlation in preferences over study periods, and between risk and time domains.
Our econometric strategy allows us to distinguish temporal stability in the population distribution of time preferences from within-individual correlation in time preferences. It also allows us to address a broad range of questions in a coherent manner, including whether risk averse individuals tend to me more patient; whether non-expected utility decision makers tend to be non-exponential discounters; and whether more patient individuals are more likely to participate in a field experiment.
Biography: Hong Il joined the Business School in September 2013. He works in the field of Applied Microeconometrics. His research focuses on the analysis of individual preferences, specifying and estimating econometric models which allow people's preferences to be inferred from data on the economic decisions they have made (e.g. job choices). Hong Il has used data from stated preference surveys to address non-market valuation questions in health economics and environmental economics. More recently, he has become increasingly involved in the structural modelling of data from economic experiments. Prior to his current appointment, he studied economics at the University of New South Wales, Australia.