Marketing Lessons from a Hurricane: Research Finds the Effects of Brand-switching Only Temporary

New study says brand loyalties have staying power

Key Takeaways:

• Brand loyalties endure even after crisis pressures to switch.
• “State dependence” is not a driving force of persistence in brand choice.

BALTIMORE, MD, September 13, 2023 – Using data from pre- and post-hurricane purchasing of consumer-packaged goods (CPG), new research has found that while consumers may switch brands in the lead-up to a natural disaster, they will likely switch back to their previously purchased brands when the crisis has passed. This has implications for the extent to which companies can expect to capture consumers using temporary price promotions.

The study, which is published in the INFORMS journal Marketing Science, is called “Identifying State Dependence in Brand Choice: Evidence from Hurricanes.” The authors of the study are Julia Levine of Johns Hopkins Carey Business School, and Stephan Seiler of Imperial College.

“Choice persistence” describes the phenomenon whereby consumers are more likely to repurchase brands and products that they have purchased in the past. Researchers have speculated that consumers do this not only because they prefer those brands and products, but also because of “state dependence.” State dependence refers to the causal effect of past choices or circumstances on current outcomes. This can be thought of as a form of inertia – you might repurchase the same brand not because you like it more than other brands, but simply because you purchased it last time.

If state dependence is the driving factor behind choice persistence, companies might be able to capture consumers with temporary price discounts and retain them even after reverting in price. If preferences are the driving factor, companies should only expect to retain those consumers if they are selling a better product. It is therefore important to ask, “what is driving choice persistence: preferences or state dependence?” This study looks at purchases made before, during, and after a hurricane to find out.

Choice persistence is observed in many contexts, but in some situations, such as when you are stocking up on essentials for a possible hurricane, you might have no choice but to switch to any available brand. Whether you stick with that brand after the hurricane tells us something about the drivers of choice persistence: After the hurricane has passed, if consumers continue to purchase the brands that they switched to during their hurricane preparations, then persistence in brand choice can be explained, in part, by state dependence (i.e., there might be a causal effect of the last brand choice on the current brand choice). However, if consumers revert to their pre-hurricane choices, then persistence in brand choice can be explained by preferences. 

The researchers utilized data that included information on the locations and timing of hurricanes and combined it with consumer-level purchase data. They observed 14 hurricanes over the course of 12 years that affected thousands of households participating in the panel.

“We applied our analysis to data from a common hurricane staple: bottled water,” said Levine. “We observed large demand spikes in the period leading up to a hurricane, presumably resulting in stock-outs. These stock-outs resulted in an increase in brand-switching behavior – consumers were much more likely to try brands that they hadn’t purchased before. However, after the hurricane, consumers immediately revert to their pre-hurricane choices. Therefore, it takes more than just inertia to create persistence in brand choice.”

Link to Study

About INFORMS and Marketing Science

Marketing Science is a premier peer-reviewed scholarly marketing journal focused on research using quantitative approaches to study all aspects of the interface between consumers and firms. It is published by INFORMS, the leading international association for operations research and analytics professionals. More information is available at www.informs.org or @informs.

# # #