Direct Elicitation of Risk-Controlled Time Preferences

Abstract

Empirical studies of intertemporal choice often nd individuals are surprisingly impatient. However, almost all previous research has confounded risk and time preferences by assuming risk neutrality - often implicitly. This is in stark contrast with signicant evidence showing that individuals are in fact risk averse even at low stakes. The assumption of risk neutrality in the presence of risk aversion will yield upwardly biased discount rates, and this in turn may be the primary reason for high discount rate estimates in past studies. One solution to the problem is to jointly estimate risk and time preference. We propose a new method for eliciting discount rates in which we do not need to elicit individual risk attitudes and make fewer assumptions about the form of the utility function. We test this new method by conducting a controlled experiment with real payments, and nd signicantly lower discount rate estimates in comparison to standard techniques. We report the distribution of risk and time preferences for our sample and perform statistical analysis of the individual-level data. Also, we compare an estimate from the risk-controlled discount rate elicitation task to one obtained using the joint estimation technique of Andersen et al. (2008). We nd consistency between our risk-controlled discount rate estimate and that obtained from the joint-estimation procedure.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 2 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.