Towards a New Theory of Consumer Choice: Cognitive Limitations, Categories and Loss Aversion

The consumer choice is one of the most important problem in the field of economics. The consumer seeks for a basket of goods that maximize the utility under the budget constrain. The classical models in economics assume that individuals have unlimited ability to analyze information, but recent literature demonstrates direct violations of transitivity and independence, which suggest that people often do not maximize utility functions (see, Thaler 1999, Kahneman & Tversky 1974). Thaler (1999) argues that the individual divides the products into consumption categories and that money is not fungible between categories.
My research so far presents a novel explanation to loss aversion. I present a model according to which the consumer’s goal is to find a feasible basket that induces a payoff above a certain threshold (”satisficing”). This problem is NP-Complete, and therefore solving the problem entails a significant computation cost. I show that this cost can induce loss aversion, and, moreover, that the induced loss aversion is non-monotonic with respect to the size of the loss. I further show under the division of goods to categories with separate sub-budgets, smaller losses can create loss aversion.
In my future research, I plan to model the consumer choice under categorization, and to analyze behavior that combines maximizing, satisficing and the use of heuristics. In addition, I will define and study a novel notion, category elasticity, to define flexibility in re-assigning the budget between categories.
Next, I plan to develop formal results that show that categorization is a good heuristic. Finally, I plan to do experiments to check my assumptions and demonstrate the empirical validity of my predictions.