The symbiotic relationship and collaboration between commercial banks and fintechs in Turkey

The study explores how the rapid growth of digital technologies and fintech after the 2008 financial crisis has transformed banking, creating both threats and opportunities for traditional banks. Fintechs, with agile structures and fewer legacy constraints, can disrupt bank business models through customer-centric, data-driven, and cost-efficient services. Banks respond through strategies of competition, collaboration, acquisition, or exit, depending on the perceived technology- and market-centered threats posed by entrants. In Turkey, major banks recognize fintechs as significant competitive threats and have emphasized strengthening cooperation with fintechs. The research question focuses on which forms of collaboration (investment-related versus product-related) Turkish banks prefer with fintech startups and in which fintech segments these collaborations are concentrated. The study posits that large banks may prefer investments to gain control over innovation (Hypothesis 1) and may partner more with payment-service fintechs given their prominence (Hypothesis 2). It aims to provide empirical evidence from Turkey's innovation-friendly regulatory context.

The paper reviews innovation strategy literature (make, buy, ally) and bank–fintech collaboration studies. Borah and Tellis (2014) show firms’ strategic choices for innovation and associated payoffs; announcements of make or ally often yield more positive outcomes than buy. International evidence suggests banks’ collaboration intensity increases with digital leadership and that large banks tend toward product-related collaborations, while investments are often made in smaller fintechs (e.g., Lars et al., 2021). Network analyses (Brandl & Hornuf, 2020) find product-related ties prevalent in Germany. Case studies identify motives: for banks—outsourcing, innovation, competitiveness, cost efficiency, service quality; for fintechs—market access, trust, scale, licenses. Collaboration has been linked to efficiency gains in banking (Ntwiga, 2020). In Turkey, regulation since 2013 has fostered payments/e-money, open banking, digital onboarding, and crowdfunding, with payments comprising the largest fintech vertical. From this literature and context, two hypotheses are formulated: Hypothesis 1: While large banks mainly invest in fintech startups, small banks are expected to establish product-related alliances (grounded in incomplete contracting and control considerations). Hypothesis 2: Large banks establish more alliances with fintechs in the payment service segment, while small banks are likely to collaborate with fintechs in other segments.

Design: Cross-sectional quantitative study using hand-collected data on bank–fintech alliances in Turkey in 2021. Sample: 24 active commercial banks; 430 observed bank–fintech alliances. Data sources: Banks’ official websites and annual reports; press announcements; FinTech Istanbul network; Startups.watch to identify fintechs; fintech firm data from Google, LinkedIn, Crunchbase. Classification: An alliance is labeled investment-based if the bank acquired a minority/majority stake or established full ownership; product-related if governed by contract-based cooperation without equity. Variables: Dependent variables—Investment (dummy=1 if investment-based; 0 if product-related); Payment service (dummy=1 if fintech operates in payments; 0 otherwise). Bank characteristics—ln(Bank total assets), ln(Bank age), Bank type (domestic=1, foreign=0). Fintech characteristics—Fintech age (years), Fintech employees (rank: 1: 1–10, 2: 11–50, 3: 51–100, 4: 101–1000, 5: >1000), Fintech headquarters (Istanbul=1, otherwise=0). Models: Cross-sectional probit regressions. Model 1 (Type of Alliance): Pr(Investment=1)=F[ln(Bank total assets), Fintech employees, Bank type, ln(Bank age), Fintech age, Fintech HQ]. Expected signs: ln(Bank total assets) positive, Fintech employees negative (per prior literature). Model 2 (Fintech Segment): Pr(Payment service=1)=F[ln(Bank total assets), Fintech employees, Bank type, ln(Bank age), Fintech age, Fintech HQ]. Expected sign: ln(Bank total assets) positive (anticipating greater collaboration with payments fintechs). Descriptive statistics: Of 430 alliances, 7% are investments; Payment service mean=0.45; Fintech age mean=13.28 years; 82% HQ in Istanbul; Fintech employees mean rank=2.94; ln(Bank total assets) mean=26.12; ln(Bank age) mean=3.69; 47% domestic banks.

- Of 430 bank–fintech alliances in 2021, only 7% are investment-based; the vast majority are product-related collaborations. - Model 1 (Investment vs. product-related): ln(Bank total assets) has a negative and statistically significant coefficient (−0.104, p<0.01), indicating that larger banks in Turkey are less likely to invest and more likely to engage in product-related collaborations. Fintech employees has a positive but statistically insignificant effect; other controls (Fintech age, Fintech HQ, ln(Bank age), Bank type) are not significant. - Model 2 (Payment service vs. other segments): ln(Bank total assets) is negative and significant (−0.033, p<0.05), implying larger banks are more likely to ally with non-payment fintechs. Fintech age is negative and significant (−0.032, p<0.01), suggesting younger fintechs are more prevalent in the payment segment. Fintech HQ (Istanbul) is positive and significant (0.510, p=0.003), indicating Istanbul-based fintechs are more likely to collaborate with banks in payments. Fintech employee size, ln(Bank age), and Bank type are not significant. - Descriptive context: 45% of alliances involve payments fintechs; fintechs average 13.3 years of age, with most headquartered in Istanbul and employee size typically in the 11–100 range.

The findings address the research questions by showing that, contrary to expectations derived from incomplete contracting arguments and some international evidence, large Turkish banks predominantly favor product-related collaborations over equity investments. This suggests banks prioritize expanding service portfolios and accessing new distribution channels without assuming ownership risks, possibly reflecting a Turkish fintech ecosystem that is still maturing for investment plays. Additionally, large banks’ preference for non-payment fintech partnerships may reflect that many banks have already internalized robust payment infrastructures, reducing the need for partnerships in that domain. Younger fintechs’ concentration in payments and the higher likelihood of Istanbul-based fintechs partnering in payments align with the regulatory and ecosystem developments that have fostered payments and open banking in Turkey. Overall, collaborations appear to be a strategic response to technological threats, enabling banks to innovate and serve new customer segments while managing risk and regulatory constraints.

The study contributes empirical evidence on bank–fintech collaborations in Turkey, showing that large banks chiefly opt for product-related alliances rather than investments and tend to partner with non-payment fintechs, while younger, Istanbul-based fintechs are more involved in payments collaborations. Practical implications include: (i) fintech entrepreneurs seeking capital may find better alignment with smaller banks or investors, while those seeking customer access and independence may benefit from product partnerships with large banks; (ii) policymakers’ restrictive bank licensing can channel innovation into alliances, underscoring the need for vigilant oversight as collaborations proliferate. Future research should evaluate the success determinants and outcomes of bank–fintech alliances, and examine fintechs’ perspectives on motivations and benefits of partnering with incumbents.

The primary limitation is the omission of alliance duration and dynamics; some collaborations may be temporary or end due to banks’ in-house technological progress, acquisitions by competitors, or expiration of agreements. This cross-sectional snapshot (2021) may not capture temporal evolution, causality, or long-term performance impacts.