Nonparametric Estimation of Habitual Brand Loyalty
96 Pages Posted: 8 Apr 2022 Last revised: 14 Nov 2023
Date Written: November 13, 2023
Abstract
Testing for and measuring habitual brand loyalty (HBL) is one of the earliest areas for empirical research on switching costs, and a classic questions in quantitative marketing. Using a partial identification approach, we derive the nonparametric identified set for the extent of HBL in a consumer population making multinomial brand choices. Several behavioral restrictions implied by popular random utility models can be applied to tighten the identified set without assuming a parametric model or utility maximization behavior. We also test whether consumers rationally plan these brand habits by testing for forward-looking choice behavior. We prove that the canonical dynamic discrete-choice model has ``built-in'' exclusion restrictions that semiparametrically identify the discount factor through moment conditions that do not require parameterizing consumers' flow utilities. Case studies of several large consumer goods categories reveal a non-trivial extent of HBL in consumers’ brand choices. Alternative explanations like learning and search costs are ruled out. Consumers are found to be forward-looking, but typically more impatient than would be implied by the real rate of interest. The long-run price elasticities from a dynamic discrete-choice model are found to be larger in magnitude than those from a model with myopic choices.
Keywords: habitual brand loyalty, dynamic potential outcomes, dynamic discrete choice, discount factor, partial identification
JEL Classification: D11, D12, L66, M3
Suggested Citation: Suggested Citation