Bank Business Model Similarities Based on Financial Ratio Networks: Evidence from Deposit Banks in Türkiye
Journal of Risk and Financial Management, cilt.19, sa.8, 2026 (Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 19 Sayı: 8
- Basım Tarihi: 2026
- Doi Numarası: 10.3390/jrfm19080642
- Dergi Adı: Journal of Risk and Financial Management
- Derginin Tarandığı İndeksler: Scopus, ABI/INFORM, EconLit
- Anahtar Kelimeler: bank business models, cosine similarity, deposit banks, financial profile similarity, financial ratios, network analysis, similarity networks, Turkish banking sector
- Akdeniz Üniversitesi Adresli: Evet
Özet
Bank business models are commonly classified into fixed groups, although banks may simultaneously resemble multiple peers and these relationships may change over time. This study examines the evolution of ratio-based business model similarity among 25 deposit banks in Türkiye from 2014 to 2024. Forty-three financial ratios were standardized within a year and used to construct weighted cosine-similarity networks; complementary analyses separated strong- and weak-profile links, tested constrained null models, controlled for bank scale and balance-sheet structure, and assessed threshold, outlier, and equal-dimension sensitivity. Under the baseline 0.40 threshold, network density was lowest in 2022 and 2023, with greater component fragmentation in 2022; winsorized specifications instead identified 2023 as the lowest-similarity year, showing that the exact extreme-year ranking is outlier-sensitive. Weak-profile links exceeded strong-profile links in every year and clustered beyond the baseline constrained-null expectation (p < 0.001). State-owned banks showed higher weighted strength after observable scale controls were applied, whereas evidence of degree centrality was weaker. Profitability and liquidity remained the densest ratio-family networks, while asset quality was the most consistently differentiating under equal-dimension comparisons. This framework provides a reproducible screening approach for identifying changing bank-business-model proximity and shared relative weaknesses, without measuring causal risk transmission, absolute financial strength, or performance superiority.