Regulators Redraw Data Rules Affecting Banks and FinTechs
According to PYMNTS, a new analysis from the European Commission's Joint Research Centre (JRC), reported May 18, 2026, identifies five conditions under which a dominant firm's data holdings become a barrier to competition. Bruno Carballa-Smichowski, a policy researcher at the JRC, lists those conditions as: data centrality to product quality, strong network effects, difficulty of independent reproduction, persistence of value over time, and returns that favour incumbents. PYMNTS also highlights the JRC's distinction between scale (more rows) and scope (more variables), citing the research consensus that adding rows yields diminishing returns, an argument associated with Hal Varian. The article frames these findings as sharpening antitrust and regulatory questions for banks and FinTechs that rely on proprietary transaction, fraud, and risk datasets.
What happened
According to PYMNTS, the European Commission's Joint Research Centre (JRC) published an analysis, reported May 18, 2026, that sets out five conditions determining when a dominant firm's data advantage crosses into an effective barrier to competition. PYMNTS attributes the framework to Bruno Carballa-Smichowski, a policy researcher at the JRC, and lists the conditions as: data must be central to product quality; network effects must compound the lead; the data must be hard to reproduce independently; the data must retain value over time; and additional accumulation must disproportionately benefit the incumbent. PYMNTS reports the analysis also emphasizes the distinction between scale (more observations) and scope (more variables).
Editorial analysis - technical context
Industry-pattern observations: The JRC's five-condition test maps onto common machine-learning realities. In supervised models, incremental observations often yield diminishing marginal gains, while new feature types (scope) can unlock non-linear improvements. For practitioners, that implies marginal utility depends on both sample size and feature diversity. Techniques such as transfer learning, synthetic data generation, and feature engineering are commonly used responses across the industry to mitigate row-based diminishing returns. Privacy-preserving approaches and differential-privacy-aware training also affect the tradeoff between data centrality and reproducibility in model development.
Industry context
PYMNTS frames the JRC analysis as sharpening enforcement questions for financial services because credit scoring, fraud detection, and risk modelling commonly rely on long-running, behaviour-rich datasets. Reporting notes that those use cases align with multiple JRC conditions, for example, multi-year customer histories are both valuable and hard for new entrants to replicate quickly. The piece connects this technical account to wider antitrust and data-governance debates without asserting regulatory outcomes.
What to watch
Observers will watch for:
- •formal regulator guidance or case law citing the JRC framework
- •policy moves on mandated data access or interoperability in payments and credit
- •litigation or merger reviews where scope-versus-scale evidence is submitted
- •vendor and platform responses such as data-sharing consortia, synthetic-data offerings, or technical interoperability standards
Scoring Rationale
The JRC framework directly addresses how data advantages interact with AI-driven financial products, making it relevant to model builders, compliance teams, and platform architects. The story has sector-wide regulatory implications but does not itself announce enforcement actions.
Practice interview problems based on real data
1,500+ SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
