Scarred Consumption

Working Paper: NBER ID: w24696

Authors: Ulrike Malmendier; Leslie Sheng Shen

Abstract: We show that personal experiences of economic shocks can “scar'” consumer behavior in the long run. We first illustrate the effects of experience-based learning in a simple stochastic life-cycle consumption model with time-varying financial constraints. We then use data from the Panel Study of Income Dynamics (PSID), the Nielsen Homescan Panel, and the Consumer Expenditure Survey (CEX) to estimate the long-term effects of lifetime experiences on consumption. We show that households who have lived through times of high local and national unemployment, or who have experienced more personal unemployment, spend significantly less on food and total consumption, after controlling for income, wealth, employment, demographics, and macro-economic factors, such as the current unemployment rate. The reverse holds for past experiences of low unemployment. We also estimate significant experience-based variation in consumption within household, i. e., after including household fixed effects. At the same time, lifetime experiences do not predict individuals' future income. The Nielsen data reveals that households who have lived through times of high unemployment are particularly likely to use coupons and to purchase sale items or lower-end products. As predicted by the experience-based learning model, the effects of a given macro shock are stronger for younger than for older cohorts. Finally, past experiences predict beliefs about future economic conditions in the Michigan Survey of Consumers (MSC), implying a beliefs-based channel. Our results suggest a novel micro-foundation of fluctuations of aggregate demand, and explain long-run effects of macroeconomic shocks.

Keywords: scarred consumption; economic shocks; consumer behavior

JEL Codes: D12; D83; D91


Causal Claims Network Graph

Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.


Causal Claims

CauseEffect
high local and national unemployment (F66)decrease in food consumption (D12)
high local and national unemployment (F66)decrease in total consumption (E21)
low unemployment experiences (J64)increase in consumption (E21)
personal unemployment experiences (J64)decrease in food consumption (D12)
personal unemployment experiences (J64)decrease in total consumption (E21)
past economic conditions (N13)consumption behavior (D10)
past unemployment experiences (J65)current consumption (E20)

Back to index