A compact group of profitable lenders controls most financial assets. SEPA, open banking and consolidation will sharpen competition, but they will not quickly dislodge the bank-centric model.
Nineteen banks, one dominant channel
Serbia’s financial market is less a system of competing capital pools than a banking system with satellites. At the end of 2025, 19 banks controlled 90.4 per cent of RSD7.75tn in financial-sector assets. Insurers accounted for 5.8 per cent; leasing companies, voluntary pension funds, payment and e-money institutions, and virtual-asset providers divided most of the remainder. For households and companies, the bank branch, app and credit committee still set the price of money.
The concentration is visible at the top. A TheBanks.eu compilation of public 2025 figures indicates total-asset shares of 15.32 per cent for Banca Intesa, 14.22 per cent for OTP banka Srbija, 11.05 per cent for Raiffeisen, 10.85 per cent for UniCredit, 10.53 per cent for AIK Banka, 10.16 per cent for NLB Komercijalna and 8.60 per cent for state-owned Banka Poštanska štedionica. On that basis the top six held about 72.1 per cent and the top seven 80.7 per cent. These are rounded third-party shares rather than an official NBS league table, but they describe the competitive reality.
Ownership matters as much as size. Intesa, OTP, Raiffeisen, UniCredit and NLB connect Serbia to larger European groups; AIK is a Serbian private consolidator; Poštanska štedionica gives the state a broad retail and public-sector channel. The National Bank lists eight systemically important banks: those seven plus Erste. A corporate borrower can run a competitive syndication, but it is usually negotiating within the same small circle.
Competition in Serbian finance is intense inside banking and shallow outside it.
The balance sheets are unusually comfortable
The sector enters its next phase from a position of strength. Non-performing loans were around 2 per cent by the end of 2025 and regulatory capital was near 21 per cent, according to NBS indicators. In 2024 return on assets reached 2.8 per cent and return on equity 20.3 per cent. High interest income, deposit funding and good asset quality provided a cushion that many western European peers would envy.
Credit continues to grow. Corporate loans rose by RSD27bn in the first quarter of 2026, with working-capital lending the main use. Yet foreign-currency exposure has not disappeared. The dinar share of bank receivables was 39.7 per cent in the first quarter, while 52.6 per cent of newly approved corporate and household loans were in dinars. The direction is positive, but balance sheets remain sensitive to euro funding and euro-linked cash flows.
Profitability is likely to normalise as rates, regulation and competition change. The state has shown willingness to cap or influence selected retail lending costs. Deposit competition is stronger than it was when liquidity was abundant, and customers expect instant onboarding and cheaper payments. Banks can defend margins through scale, data and cross-selling, but the easy part of the earnings cycle is probably over.
Payments reform will move the battleground
Serbia became operationally connected to the Single Euro Payments Area in May 2026 through the NBS and 18 banks. That should reduce the friction of euro transfers for businesses and households and make fee comparisons with EU providers more direct. Amendments aligned with the EU’s second Payment Services Directive have also opened the architecture to account-information and payment-initiation models, subject to licensing and technical execution.
The immediate winners will not necessarily be stand-alone fintechs. Large banks can use open interfaces and SEPA rails to improve treasury products, merchant services and cross-border payments. Telecoms, e-commerce platforms and specialist payment companies can attack the customer interface. The strategic risk for incumbents is becoming a regulated balance sheet behind somebody else’s app; the risk for challengers is discovering that compliance, fraud controls and customer acquisition cost more than software.
Consolidation has not finished. Raiffeisen Bank International’s 2026 voluntary offer for Addiko attracted acceptances for 56.16 per cent of the shares, but remained subject to closing conditions. Any resulting change in Addiko’s Serbian business would be another step in a long reduction from 33 banks to 19. The next deals are more likely to rearrange mid-sized franchises than break the grip of the top tier.
Insurance is concentrated; long-term capital is scarce
Insurance offers a second concentrated market. In the first quarter of 2026, total premium rose 11.6 per cent to RSD50.2bn, but 82.9 per cent remained non-life. Dunav held 25.6 per cent of premium, Generali 19.8 per cent, DDOR 10.1 per cent, Wiener 9.5 per cent and Triglav 8.4 per cent. The top five therefore controlled 73.4 per cent. State-controlled Dunav gives government a material position alongside international groups.
Voluntary pension funds are smaller and more concentrated still: the three largest held 80.8 per cent of net assets at end-2025, and 68.5 per cent of assets were invested in government bonds. This is useful demand for state paper but a narrow source of growth capital for companies. Leasing helps finance equipment and vehicles, yet it does not change the system’s centre of gravity.
The medium-term trend is clear. Serbia will have faster payments, more digital distribution and fewer banking entities, while asset quality starts from a strong base. What it will not automatically gain is a deep market for equity, venture, pensions or corporate bonds. Until that changes, the leading banks will remain not just financial intermediaries but the main allocators of corporate opportunity.
