Mental Accounting and Consumer Spending

Citation:

Chip Heath and Sue O'Curry (1994) ,"Mental Accounting and Consumer Spending", in NA - Advances in Consumer Research Volume 21, eds. Chris T. Allen and Deborah Roedder John, Provo, UT : Association for Consumer Research, Pages: 119.

Advances in Consumer Research Volume 21, 1994      Page 119

MENTAL ACCOUNTING AND CONSUMER SPENDING

Chip Heath, University of Chicago

Sue O'Curry, DePaul University

Consumers obtain income from a variety of sources and incur many types of expenses. The tendency to categorize both income and expenses is pervasive, and can have a profound impact on patterns of spending, as well as other uses of money. Subjective feelings of wealth, response to sales promotions, self-control problems, and post-purchase evaluations are just a few of the consumer issues addressed by mental accounting.

Heath and Soll showed that consumers tend to set budgets for classes of expenses and monitor expenses against the appropriate budget. They stop purchasing items in the category when total expenses during a particular time period exceed the category budget. Within a class of expenses, purchases differ in their prototypicality. For example, a movie is a prototypical entertainment expense, while a hardback novel is less protoypical. The purchase of a prototypical item decreases future expenses in a category more than the purchase of a less prototypical item with an equivalent monetary price.

Experiments showed that a current expense lowers future expenses within the same category but has little effect on other categories of expenses. Within a category, the effect of one expenditure on another is a function of their similarity. The experiments control for other factors such as satiation and income effects.

O'Curry demonstrated a tendency to assign income to different accounts. Most consumers are familiar with the injunction to "buy something special for yourself" that accompanies many gifts of money. From a normative standpoint, gifts of money, earnings from work, tax refunds, and income from all other sources should be spent the same way.

The first experiment examined differences in spending when income is derived from price decreases compared to cash income. If the gain of real income is the same in both cases and price ratios are unaffected, the pattern of spending should also be the same, except for an increase in quantity purchased when prices fall. However, price decreases may be perceived as a gain to the budget for the product class, rather than as a gain to overall wealth. This experiment demonstrated that consumers may be more likely to buy higher quality or complementary goods when an increase in income is derived from a price decrease rather than an increase in money income. Consumers may also code income and expenses by how "serious" or "frivolous" the source of income or category of expense is. The second experiment demonstrated a tendency for consumers to spend income from frivolous sources on frivolous uses.

Nye tested the hypotheses that (1) the category to which an expense is assigned will affect the consumer's effort to "recover" that expense and (2) sunk costs may be most salient soon after they are incurred. These hypotheses were tested in a natural field setting, in which subjects incurred genuine out-of-pocket costs. Subjects purchased coupon booklets offering a variety of discounted products and services from local merchants. The price of the coupon booklet (sunk cost) was manipulated. A second manipulation encouraged subjects to recategorize the booklet price as a charitable contribution rather than as an investment. Greater coupon usage by subjects who paid the higher price indicated a sunk cost effect. When subjects were able to categorize the booklet price as a charitable contribution rather than an investment, the sunk cost effect was eliminated. The sunk cost effect was strongest immediately after the coupon booklet was purchased. Subjects paying the higher price tended to recover their investment quickly.

Eric Johnson concluded the session with a discussion of the papers and suggested directions for further work.

----------------------------------------