Published on 15 October 2026, ECB Working Paper 3300 reports an incentivised experiment conducted in November 2025 with 761 participants in France, Germany, Italy and Spain. A high holding limit increased average experimental allocations. Automatic funding at payment reduced average holdings while increasing participation. The paper is research, not an ECB decision on digital euro design.
Event / source publication date: 15 October 2026The following is Human Fintech’s analysis
The balance sheet is only one view of the product.
A wallet with little money sitting in it can still be valuable at checkout. Consider a hypothetical customer who funds purchases from a linked bank account only when needed. That customer may make frequent payments without maintaining a substantial wallet balance. Measuring success through stored funds alone would miss the service being provided.
This distinction also changes how a reader should interpret claims about deposit displacement. The ability to pay through a new instrument does not, by itself, reveal how much money users will move out of banks for extended periods. Timing, funding arrangements and the reason for holding a balance all matter. Transaction activity and persistent holdings deserve separate measures.
A ceiling can communicate more than a constraint.
A product limit is also information presented to a person making a choice. A generous allowance might look like a normal amount to keep, even to someone whose intended balance was below the old ceiling. Designers and observers should therefore avoid assuming that a higher cap affects only customers previously constrained by it.
The practical reading question is whether an interface clearly distinguishes a legal maximum, a provider restriction and a suggested amount. A user should not have to infer that distinction from a large number on a screen. Clear explanations cannot remove every behavioural response, but they make the economic choice easier to understand.
Convenience has dependencies, and evidence has boundaries.
Automatic funding deserves scrutiny beyond its promise of a smoother checkout. A linked account can be unavailable, a permission can expire, or a transfer can fail. For an everyday reader, the useful questions are simple: which account supplies the shortfall, what consent is required, and what happens if funding fails? These are analytical considerations, not findings that the experiment tested operational resilience.
The study uses stylised choices and an online participant sample rather than observing a live currency launch. Conditions were presented in sequence, and experimental participation is not the same as opening and regularly using a real account. Its findings are a reason to question a simple balance-based adoption story, not a forecast of euro demand or a recommendation for a specific holding cap.
White Fintech.
Part VII of White Fintech covers payment businesses, including company structure, risks, compliance and economics. Explore the contents and the free sample.
Complete English edition, PDF + EPUB. Read the first 30 pages free, with no sign-up.