Do Co-branded Credit Cards Increase Customer Loyalty?

Abstract

Firms have a growing interest in leveraging cobranded credit cards to boost spend and
engagement among their loyalty program members. Our goal in this study is to estimate
the causal impact of cobranded credit card adoption by consumers. Using a unique
and comprehensive longitudinal data set from a North American airline containing
detailed records of both card adopters and non-adopters, we employ a multi-pronged
empirical strategy to address selection bias, both from observables and unobservables.
We use propensity score matching to develop a sample with covariate balance, and
employ difference-in-differences estimation with a two-way fixed effects specification.
To deal with time-varying unobservables that may result from consumers’ future travel
expectations at the time of the adoption decision, we leverage a finite planning horizon
feature in our institutional setting to estimate treatment effects over several phases of
time after adoption. For robustness, we also compute Rosenbaum sensitivity bounds,
and conduct analysis that excludes customers who are more likely to be farsighted in
their travel planning.

We find statistically significant and economically meaningful effects of card adoption
on a multitude of behaviors. Specifically, revenue lift was 42% compared to non-
adopters 12 months after adoption, which translates to a spend increase of nearly $200
per year. The revenue lift was driven by a larger number of transactions rather than
higher price per transaction. Adopters also engage more strongly with the loyalty
program, and focus on redemption with the brand rather than its partners. Finally,
card adopters who experienced the highest increase in flight spend, tended to live near
hub airports of the airline firm or were already existing members of the loyalty program.
Put together, these findings add a novel perspective on the impact of cobranded credit
cards on customer loyalty.

Keywords: Cobranding, Loyalty Programs, Selection Bias, Airlines, Quasi- Experimental Methods