Clustering Bank Financial Performance: Persistence and Migration among Indonesian Listed Banks

Abstract

Indonesian listed banks operate within the same financial system, yet their financial profiles can differ substantially and may evolve differently over time. This study identifies financial-performance clusters among Indonesian listed banks and examines whether banks remain within or migrate between these profiles. The analysis covers 47 banks listed on the Indonesia Stock Exchange during 2022–2025, resulting in 188 bank-year observations. Net profit margin (NPM), return on assets (ROA), and debt-to-equity ratio (DER) were standardized and analyzed using Ward’s hierarchical clustering followed by K-Means clustering. Four distinct financial profiles were identified: a Severe-Loss Profile, a High-Profitability and Low-Leverage Profile, a High-Leverage Profile, and a Moderate-Profitability and Lower-Leverage Profile. Despite substantial cross-sectional heterogeneity, cluster membership was remarkably persistent over time. Of 141 possible annual bank transitions, 129 (91.5%) remained within the same cluster, while only 12 (8.5%) represented migration. At the bank level, 38 of 47 banks (80.9%) maintained the same financial profile throughout all three annual transitions, with only nine banks migrating at least once. These findings show that financial heterogeneity among Indonesian listed banks coexists with substantial temporal persistence. By extending financial-ratio clustering from static classification to longitudinal tracking, this study provides a simple framework for identifying both persistent financial profiles and changes in banks’ relative financial positions.