Liquidity constraints, income variance, and buffer stock savings: Experimental evidence

Abstract

We provide a direct, experimental test of the buffer stock model of savings behavior using a three-period intertemporal model of consumption/savings decisions. In one treatment, liquidity in the second period is constrained (thus, borrowing is not possible), while in the unconstrained treatment, there is no such constraint. Our main contribution is to test the buffer stock model's prediction that the liquidity constraint in the second period increases savings in the first period relative to the absence of such a constraint. A second treatment variable considers differences in the variance of the stochastic income process that agents face (high or low), resulting in a 2x2 experimental design. While we do not find evidence for the predicted effect of the liquidity constraint on first-period savings, we do find strong support for almost all the other predictions of the model, e.g., the impact of a higher variance of income on savings behavior and differences between period 1 and period 2 savings amounts. In further analyses, we find that we can rationalize the observed departures from model predictions by some combination of debt aversion, heterogeneity in cognitive abilities, and/or learning.

Keywords: Experimental Economics, Inter-Temporal Optimization, Liquidity Constraints, Consumption, Saving, Debt Aversion

JEL Classification: C91, D92, E52