On 13 August 2026 the European Central Bank (ECB) published new figures on the use of cash. 92 per cent of companies with physical outlets accept cash, against 90 per cent in 2024. Card acceptance stands at 88 per cent against 87, and acceptance of mobile payments has jumped from 36 to 68 per cent. The survey covers 8,205 companies across all 21 euro area countries, with fieldwork between February and April 2026.
The figures have widely been read as evidence that cash is staging a comeback. They do not support that reading, for three independent reasons: one concerns the source, one the unit of measurement, and one the period covered.
What: survey on the use of cash by companies in the euro area, published on 13 August 2026
Base: 8,205 companies in 21 countries, fieldwork February to April 2026, conducted by Ipsos European Public Affairs
Scope: four consumer-facing sectors with at least one employee: retail trade, restaurants and cafés, hotels, and arts, entertainment and recreation
What is measured: whether a company accepts a means of payment, not how often it is used
Not included: any reference to the digital euro. The study makes none
First: this is not the study it is taken for
The ECB runs several survey series on payment behaviour, and they measure different things. The best known is the Study on the Payment Attitudes of Consumers in the Euro Area (SPACE), last published in December 2024. It surveys consumers and measures what people actually pay with.
The figures published now come from a separate series on the use of cash by companies, with waves in 2021, 2024 and 2026. It surveys companies and measures the acceptance side.
The distinction is not academic. Attributing the figures to the consumer study assigns them to a survey that does not exist for 2026, and implies that they describe measured payment behaviour. What they describe is measured willingness to accept.
Second: acceptance and usage move apart
The report states the unit plainly: “of all companies that receive customer payments at physical locations, 92% accept cash, 88% accept physical cards and 68% accept mobile payments.” What is counted is companies, not payments.
The difference is substantial. The 2024 consumer study supplies the usage figures at the point of sale: cash accounted for 52 per cent of transactions, against 59 per cent in 2022, and 39 per cent of value. Cards lead by value at 45 per cent. In Germany the cash share came to 51 per cent of transactions and 26 per cent of value.
Placed side by side the picture is clear: almost every shop takes cash, a little over half of transactions use it, and the share is falling. Anyone inferring from the 68 per cent mobile acceptance that two thirds of payments run through a phone has swapped one unit for another.
Third: the series has three points
Comparison with 2024 suggests an upward trend. It disappears once the first wave is included. In 2021, 96 per cent of companies accepted cash. The 2024 survey recorded the change: “the acceptance of cash has decreased by 8 percentage points (from 96% to 88%)”.
The series therefore reads 96 per cent in 2021 (all companies surveyed), 88 to 90 per cent in 2024 (depending on the base) and 92 per cent in 2026 (companies with a physical outlet). The current figure remains well below the starting point. What looks like a reversal across two points is, across three, a partial recovery after a fall.
A word of caution on the bases: the 2021 and 2024 values in the 88 per cent reading cover all companies surveyed, while the 90 and 92 per cent figures cover companies with physical outlets. The series should not be mixed without saying so. The direction is unaffected.
What the figures do show
Beyond the misreadings, the survey holds findings of more practical interest than the headline suggests.
The spread between countries is wide and moving. In Greece and Italy 99 per cent of companies take cash, in Belgium 81 and in Cyprus 76 per cent. Cyprus and Slovakia each gained nine percentage points against 2024, while Belgium lost ten and Ireland nine. For the Belgian decline, which runs against the general direction, the survey offers no explanation.
The self-checkout is the quiet driver. 13 per cent of companies operate them, and only 52 per cent of those accept cash there. The ECB warns explicitly about the consequence: “even though most companies accept cash and say that they do so, among consumers the perceived acceptance of cash may go down because companies have reduced the ease of paying with cash.” Measured acceptance stays high while perceived acceptance falls.
Crypto-assets play no role at the till. Among companies that accept online payments at all, 0.2 per cent take crypto-assets. For a debate that turns heavily on tokenised money, that is a sobering number.
The jump in mobile acceptance deserves a caveat. From 36 to 68 per cent in two years is remarkable, particularly since the same measure rose by only seven points between 2021 and 2024, from 30 to 37 per cent. Part of the difference follows from the change in reference group; the jump itself does not. Anyone using the figure should flag the comparability caveat alongside it.
The bridge to the digital euro, and what it does not carry
The project has progressed further than the public debate often suggests. The European Parliament settled its position on 9 July 2026 and the Council its mandate back in December 2025. Negotiations between the institutions opened on 13 July 2026. The aim is political agreement by the end of the year, a pilot with 36 payment service providers from September 2027 and a launch in 2029. Holding limits, bank compensation and whether merchants must accept remain open.
For the case behind the project, the new figures are awkward. The argument that digital central bank money is needed because cash is disappearing from daily life does not gain from rising willingness to accept it.
The ECB had already moved its reasoning, well before these figures appeared. Piero Cipollone, member of the Executive Board, spoke in Riga on 1 April 2026 about resilience and autonomy in payments, putting it this way: “As payments are vital to daily life and the economy, the mere threat of disconnecting payment systems could give others leverage over Europe.” The argument is dependence on a handful of non-European providers, not the disappearance of cash. A gap of a good four months rules out any reading in which the shift responds to this survey. The figures supply, after the fact, the reason why the older argument could no longer be sustained.
Worth noting: the survey itself draws no such bridge. It makes no reference to the digital euro anywhere. The connection is made in this assessment, not in the survey.
The other side
Criticism from the European banking industry aims less at the objective than at the route. Daniel Baal, President of the Fédération Bancaire Française and of the Confédération Nationale du Crédit Mutuel, told an audience in Brussels on 15 April 2026 that the digital euro “dans sa conception actuelle, ne répond pas à l'enjeu de souveraineté des paiements et est paradoxal à plus d'un titre”. He puts the additional investment for European banks alone at 18 billion euros and points to funds that would leave bank balance sheets.
Martina Weimert, CEO of the European payments initiative Wero, sees an acceptance duty for merchants as a “distortion of competition” between public and private means of payment.
Both objections come from parties with an interest in the answer, which places them without devaluing them.
Recommendations
Now: The 92 and the 68 per cent are company acceptance rates; the 52 per cent cash share comes from a consumer study two years earlier. Anyone placing these figures side by side in a deck or client material should name the unit each time. The error is common and specialist readers spot it at once.
This month: Only around half of self-checkouts accept cash. For anyone serving retail clients with till systems this is a concrete point of leverage: the merchant's measured acceptance and the customer's experienced acceptance part company precisely here.
By year end: What becomes expensive for banks turns on holding limits, the compensation model and any acceptance duty. All three are under negotiation. A preparation programme should hang on those three variables rather than on how quickly cash recedes.
In planning: The gap between 76 per cent in Cyprus and 99 per cent in Greece is 23 percentage points, and movement between waves reaches double digits. Anyone running payments across several euro area countries should plan country by country. A euro area average describes no single market accurately.
Christian Schablitzki
Strategy & Management Consultant · Agentic AI expert for financial institutions
More than 20 years in investment banking and derivatives trading, followed by over 10 years advising financial institutions. Currently Partner at Infosys Consulting in Germany. Certified in Google AI, Generative AI Leader (Google Cloud) and IBM RAG and Agentic AI.
LinkedIn profile →Keep reading – every 14 days in your inbox.
Capital markets insights, regulatory updates and AI trends. Concise, well-founded, free.
GDPR-compliant. Unsubscribe at any time.