Prudential saving: evidence from a laboratory experiment

Abstract : Prudence is a behavioral attitude that is broadly applicable to settings involving risk. It has particular importance in intertemporal choice theory, where it can be interpreted as the intensity of intertemporal substitution. Prior laboratory experiments to elicit prudence have addressed it in a pure-risk sense, by examining behavior in static lotteries and other gambles. It is tempting to impute these results into an intertemporal context, leveraging the fact that “risk aversion” and “elasticity of intertemporal substitution” are directly mappable under expected utility. However, many empirical studies of intertemporal behavior suggest that the two ideas may be distinct. To address prudence in its intertemporal sense, we instead design a small-scale laboratory experiment around a two-period consumption/savings model in order to detect prudence via savings choices. The utility concept in this model disentangles risk preferences from intertemporal preferences, and suggests the type of exogenous variation to present to subjects in the experiment. The experimental design involves a “multiple price list” with scenarios involving income risk and interest-rate risk. In each scenario, subjects must choose how much of their first-period income to save for the second period. The design also implements field-like wealth levels and real time lags to ameliorate the possibility of the decisions being a laboratory artifact. We estimate risk and intertemporal preferences at the individual level using a subject’s savings data and the model’s structural Euler equation. Ex- cluding outliers, the average coefficient of relative risk aversion is 2.06, the average elasticity of intertemporal substitution is 0.75, and the average coefficient of relative prudence is 3.90. These averages mask a good deal of subject-level heterogeneity, as the respective coefficients of variation are, at a minimum, 70%.
Document type :
Poster communications
Complete list of metadatas
Contributor : Archive Ouverte Prodinra <>
Submitted on : Friday, October 2, 2015 - 8:41:23 PM
Last modification on : Wednesday, March 21, 2018 - 4:08:11 PM


  • HAL Id : hal-01208938, version 1
  • PRODINRA : 255269



Aj Allen Bostian, Christoph Heinzel. Prudential saving: evidence from a laboratory experiment. SFB 649 Workshop "Risk Preferences and Decisions under Uncertainty", Jan 2013, Berlin, Germany. 44 p., 2013. ⟨hal-01208938⟩



Record views