We use a panel dataset to test the stability of measured discount rates over time in response to changes in both macroeconomic events and household-level labor market outcomes. While discount rate measures are constructed to capture a rate of time preference, our evidence is inconsistent with such an interpretation. Our results more closely align with the interpretation that standard methods to elicit discount rates reveal the market interest rate faced by an individual rather than their pure rate of time preference. It follows directly from such an interpretation that factors which influence the interest rate at which a household can borrow and lend, such as the inflation rate and household income, ought to be correlated with the elicited discount rate – a prediction supported by our data. (c) 2012 Elsevier B.V. All rights reserved.