Does "Repayment-by-Purchase" Help Consumers Pay Down Debt?

Journal of Marketing Research Scholarly Insights are produced in partnership with the AMA Doctoral Students SIG – a shared interest network for Marketing PhD students across the world.

Consumers worldwide grapple with the significant challenge of repaying loans, credit card bills, and other expenses. While the act of spending often brings consumers joy or a sense of accomplishment (such as gaining admission to a top-tier university or purchasing a first car), the enjoyment fades over time, giving way to the burden of repayment. Although making incremental payments can somewhat ease the burden, consumers face a dilemma: how to determine the optimal repayment amount without risking accruing unnecessary interest charges and potentially damaging their credit score. Can consumers pay off credit card debt by addressing specific purchases? For example, are consumers more likely to pay off debt if they can choose to repay an airfare? Or a hotel stay? Or even the entire category of travel?

Firms such as Chase, American Express, and Citibank have introduced such “repayment-by-purchase” options that allow consumers to make payments toward specific purchases rather than their aggregate debt. Under repayment-by-purchase, consumers select specific items (e.g., a coffee at Starbucks) or categories of purchases (e.g., restaurants) to repay and make payments specifically directed toward “eliminating” these purchases from their bill.

In a recent Journal of Marketing Research study, researchers collected data including interviews, a field experiment, and five laboratory experiments, revealing that the “repayment-by-purchase” format can increase repayment by more than 20%. These findings have multiple positive implications for consumer welfare and satisfaction with banks.

The “repayment-by-purchase” format can increase repayment by more than 20%.

In contrast, interviews with frontline bank officials reveal skepticism regarding the potential adverse effects of such features on their interest earnings and credit fees.

To learn more about the challenges of balancing consumer welfare and a bank’s profitability, as well as the impact of personalized payment categories on purchase decisions, we got in touch with one of the study’s authors, Grant E. Donnelly. Grant emphasized the potential of repayment-by-purchase to empower consumers while acknowledging the complexity and user experience challenges of implementing this intervention:

Q: Currently, most banks offer repayment-by-purchase over time instead of repayment-by-purchase at once. Would you consider this as a compromise banks have found to incorporate your findings (i.e., purchase-specific repayment opportunities) while protecting their revenue-generating mechanisms (i.e., time = interest)? Also, can you think of a possible way that makes the situation win-win for both?

A: Banks need to balance profit maximization with customer well-being. Repayment-by-purchase is a tool that empowers consumers to manage their debt. Our research with the Commonwealth Bank of Australia showed that this feature can increase both repayments and financial well-being. Customers repay what they bought within the statement period, leading to more repayments for the bank. Consumers feel more in control of their finances, which outweighs the opportunity costs of interest for the bank.

Consumer empowerment is the biggest advantage of this approach. This can attract new customers who value financial control and provide  a unique selling proposition for banks.

Q: In today’s attention economy, practitioners seek a sleek user experience. However, balancing repayment-by-purchase or category increases users’ efforts to pay their statements. What do you think convinces banks to implement your intervention?

A: The intervention increases user keystrokes, but users who opted to log in more often are boosting overall app engagement. We used up to 14 categories developed with the Commonwealth Bank of Australia to make the design sleeker and more digestible. Categories were presented by debt size, from smallest to largest.

The main effort was on the bank’s side, redesigning the app interface for categories or purchases and not for the consumers. This increased engagement and provided a meaningful user experience.

Q: For which credit card customer segment do you think your intervention is most effective? Transactors, Revolvers, or Dormants? And why is that? Also, can you see the intervention bearing any cross- and up-selling potential for other credit products of the bank?

A: Our intervention is most effective for Revolvers and Dormants. Transactors, who pay their bills in full each month, are not influenced by this intervention. Revolvers, who make partial or minimum payments, benefit from the increased engagement and the sense of accomplishment provided by the intervention. Dormants, or those at risk of default, are also positively impacted as the intervention encourages them to engage and make small payments, improving their financial well-being. Younger customers showed more engagement than the older ones.

Regarding cross- and up-selling potential, consumer banks in the U.S. offer payment plans for smaller purchases, leveraging the sense of accomplishment and empowerment (i.e., choosing tickets ranging from $50–$1000s to pay off over time). This can retain consumers as Revolvers without impacting their credit scores negatively.

Q: Although you have used both “category”-based approaches and “item”-based approaches for your studies, from a consumer perspective, what approach do you think makes decoupling easiest? How many partitioned entities (i.e., labels, items) have you seen as the most effective? What role do you think the diagnosticity of category labels plays?

A: Imagine working an hourly paid job. If you wanted a certain t-shirt, you could frame the goal as “working an extra shift” and therefore couple the purchase with the expense on your statement item.

If the label of the product or category makes the purchase salient to some extent, the decoupling will work equally well. For example, in our paper, we show that the effect does not work well if we show one purchase by a transaction ID. Effective decoupling requires making the purchase salient through meaningful labels. Too many categories can lead to decision paralysis. We’ve found that around 14 categories work well, providing sufficient granularity without overwhelming the consumer.

Q: Repayment-by-purchase demonstrates that an original decoupled procedure yields positive outcomes through recoupling. Are there any other contexts outside financial services in which recoupling affects consumer behavior?

A: Recoupling has also been shown to be a powerful intervention in other contexts. In physical exercise, for example, coupling exercises with weight loss equivalents has a positive impact on exercise behavior. Also, charitable donation behavior is affected by demonstrating the impact of the donations. In the sustainability realm, researchers have shown that coupling recycling outcomes (i.e., future products) with the touchpoint of either recycling or trashing an item positively impacts recycling intentions.

Q: The Bank of Australia implemented your repayment-by-category intervention for six months. What were the successes of this project at the Bank of Australia? What were the challenges faced by the practitioners during implementation? What do practitioners need to be cautious of?

A: In general, banks underlie rigorous regulations. For this intervention, we were lucky to work with a bank that focuses on innovation and sustainability with the consequence of operating a behavioral unit in-house. This unit is responsible for experimenting to enhance consumer welfare and generate knowledge for the bank. Taken together, this commitment made the implementation of the repayment-by-purchase and expensive adjustments of the app interface smoother.

Source: Grant E. Donnelly, Cait Lamberton, Stephen Bush, Zoë Chance, and Michael I. Norton (2023), “‘Repayment-by-Purchase’ Increases Consumer Debt Repayment,” Journal of Marketing Research, 61 (3), 411–29. doi:10.1177/00222437231182372

Go to the Journal of Marketing Research