The Non-Linear Effect of Fintech Adoption on Bank Performance: Evidence from Jordan (2013-2023)

Abstract

Financial technology has altered the competitive conditions facing commercial banks, yet the evidence on whether adoption raises or lowers bank performance remains contradictory. This study asks whether the relationship is non-linear rather than uniformly positive or negative. It uses an unbalanced panel of thirteen conventional commercial banks listed on the Amman Stock Exchange over 2013 to 2023, giving 140 bank-year observations. Adoption is measured by digital transaction intensity, the share of digital transactions in total customer transactions, drawn from central bank payment system data. Return on assets, return on equity, net interest margin and the cost-to-income ratio are estimated using two-way fixed effects with Driscoll-Kraay standard errors and a quadratic adoption term. The results show a convex, U-shaped association between adoption and profitability, with a turning point near one-third of transactions. Adoption is associated with a lower cost-to-income ratio and with compressed net interest margins, and the initial profitability decline is milder for larger banks. The pattern holds across seven robustness and endogeneity checks. Because the curve is conditional on bank and year effects, and unconditional profitability had not regained its early-period level by 2023, the estimate locates a turning point rather than a completed recovery