12 Oct 2007Working Paper Summaries

Mental Accounting and Small Windfalls: Evidence from an Online Grocer

by Katherine L. Milkman, John Beshears, Todd Rogers and Max H. Bazerman

In the course of daily life, people occasionally receive small windfalls. Every so often we are handed a gift certificate for $5 off a meal, find a $10 bill on the street, or win $20 in an impromptu game of poker. According to standard economic theory, these types of small windfalls should have no noticeable effect on spending decisions because such windfalls constitute meaningless changes to lifetime wealth. However, if you have ever been the recipient of a small windfall, you may remember thinking about ways to spend this unexpected cash, buying items you might not have otherwise purchased. This kind of behavior can be interpreted as an example of "mental accounting" as theorized by economists Richard H. Thaler and Hersh M. Shefrin. This paper presents evidence supporting some of the implications of a theory of mental accounting in the domain of online grocery shopping. Key concepts include:

In the domain of online groceries, the redemption of a $10-off coupon increases an individual's spending, as predicted by Thaler and Shefrin.
The increase in spending stimulated by the redemption of a $10-off coupon is focused on groceries that customers would not purchase in the absence of such a coupon.

Author Abstract

We study the effect of small windfalls on consumer spending decisions by examining the purchasing behavior of a sample of online grocery shoppers over the course of a year. We compare the purchases customers make when redeeming a $10-off coupon they received from their online grocer with the purchases the same customers make when shopping without a coupon. The standard permanent income or lifecycle theory of consumption predicts that grocery spending will be unaffected by the use of a $10-off coupon, while a simple mental accounting framework predicts that such a coupon will increase spending on groceries. Controlling for customer fixed effects and other relevant variables, we find that grocery spending increases by $1.59 with the use of a $10-off coupon. In addition, even though the receipt of a $10-off coupon does not correspond to a meaningful increase in wealth, the extra spending associated with the redemption of such a coupon is focused on "marginal" grocery items, or grocery items that a customer does not typically buy.