Retail Value of Co-branded Credit Cards for First-time Borrowers

Co-branded credit cards combine financing with relationship marketing, but their reach depends on whom the issuer approves. This paper studies a temporary approval pilot in Peru, linking application records for 91,638 applicants without an observed formal credit history to item-level loyalty transactions recorded independently of card payment. Difference-in-differences estimates comparing approved and rejected applicants show that, over the next twelve months, approval raises average monthly spending at the partner retailer by 61.7\% relative to the approved group's pre-decision average. Among continuing shoppers, a descriptive decomposition attributes 60.6\% of the spending contrast to increased visit frequency. Cardholder-exclusive promotions are closely associated with deeper spending in previously purchased categories, while most of the overall effect occurs in categories applicants had not purchased before, including on trips without promotion redemption. Approval also raises monthly gross margin by about 40\% relative to the approved group's pre-decision average. Promoted items carry negative recorded margins on average, but trips containing them contribute positively because other basket purchases offset those losses---a pattern consistent with basket-level loss-leader economics. These findings highlight why customer selection is a shared strategic concern in co-branded partnerships.