Most founders begin the jurisdiction question the wrong way. They ask, “Where is it fastest?” or “Where is it cheapest?” Regulators do not think like that, and over time, the market usually punishes founders who do.
Choosing a jurisdiction for EMI licensing is not mainly a search for the lightest-touch country. It is a search for the place where your business model, governance, AML profile, local substance, and growth plan can survive both authorisation and supervision. The EBA’s 2025 follow-up peer review is especially telling here: it says divergences across Member States still remain in governance, internal controls, AML/CFT, and local substance, creating risks of regulatory arbitrage and forum shopping.
That is the real point. A founder does not choose a jurisdiction only for the application. They choose it for the life of the institution that follows.
Mistake 1: Choosing the jurisdiction by headline timeline
This is probably the most common mistake.
Some founders compare jurisdictions as if they were buying delivery speeds. But the EBA’s follow-up review shows authorisation timelines vary widely across Member States, from around 4–6 months in some jurisdictions to as much as 27 months in one case, with an EU median of about 9.5 months. More importantly, the report says delays are most often linked to incomplete or low-quality applications, applicants’ underestimation of regulatory requirements, later changes during assessment, and business-model complexity.
So “fast jurisdiction” is often a misleading idea. A strong file in a demanding jurisdiction can move better than a weak file in a supposedly fast one. Founders who choose based only on nominal speed often discover that the real bottleneck was never the country. It was the quality and coherence of their own project.
Mistake 2: Assuming EU rules mean identical supervisory reality
Another classic founder error is assuming that because EMI licensing sits inside an EU framework, every Member State will read the same business in the same way.
That is not how the supervisory landscape currently looks. The EBA says progress has been made, but expectations still diverge across Member States in key areas such as governance, internal controls, AML/CFT, and local substance. It also notes that those divergences can create an unlevel playing field and encourage regulatory arbitrage.
This does not mean the rules are random. It means founders should stop looking for a mythical “best jurisdiction” in abstract terms. A jurisdiction may be workable for one EMI model and a poor fit for another. A founder should be asking whether the local supervisory culture matches the actual complexity of the product, the target geographies, the management setup, and the expected pace of growth. That conclusion is an inference, but it is directly supported by the EBA’s finding that divergence persists precisely in the areas that define whether an EMI can be run soundly.
Mistake 3: Ignoring local substance and effective management
A lot of founders still think they can choose a jurisdiction first and figure out local presence later.
That approach has become harder to defend. The EBA’s 2025 follow-up report says all supervisors now report verifying, at authorisation stage, that applicants are effectively managed and controlled from the home Member State. At the same time, the report says expectations about how applicants should prove that still diverge, which is exactly why local substance remains a major jurisdiction-choice issue.
This matters because jurisdiction is not just a flag on the licence. It is supposed to be the place from which the EMI is genuinely managed and controlled. Founders who pick a country because it looks commercially convenient, while planning to run the real institution somewhere else, often create a mismatch between strategy and supervisory reality. The EBA’s own language links this directly to risks of forum shopping.
Mistake 4: Overestimating how simple passporting will be
Passporting is one of the most misunderstood parts of the EMI discussion.
Yes, the EU single-authorisation principle means a financial institution authorised in its home Member State can in principle provide authorised services across the Single Market through a branch or the freedom to provide services. But the EBA also makes clear that passporting still involves home-host notifications, specified information, standard forms, templates, procedures, and cooperation between competent authorities.
In other words, passporting is not the same thing as “one licence and frictionless expansion everywhere.”
That becomes even clearer when cross-border distribution gets larger. The EBA notes that, in some cases, host Member States may require an EMI or PSP to appoint a central contact point if thresholds are met, including when the number of establishments reaches 10 or more or when annual e-money or transaction volumes exceed EUR 3 million in the host state.
So founders who choose a jurisdiction purely because “we can passport later” often fail to model what cross-border supervision, structure, and notifications will actually look like once the business starts scaling.
Mistake 5: Treating AML as secondary in the jurisdiction decision
Another serious mistake is treating AML/CFT as something to solve after choosing the jurisdiction.
That is backwards. The EBA’s 2025 AML/TF risk opinion says an increasing number of competent authorities assess the inherent risk in the e-money institution sector as significant to very significant. The same opinion also highlights deficiencies in ongoing monitoring, transaction-monitoring calibration, customer identification, and suspicious reporting, including a case where an EMI faced withdrawal of authorisation in 2024 after serious breaches.
This means the AML profile of your business should influence the jurisdiction choice from the beginning. If the model involves higher-risk channels, agent or distributor complexity, fast-moving digital flows, or difficult geographies, founders should assume that AML scrutiny will be central, not peripheral. Choosing a jurisdiction because it seems commercially easy while ignoring how your AML risk will be perceived is one of the fastest ways to build a weak licensing strategy.
Mistake 6: Looking at legal cost, not total cost of supervision
Some founders optimise for the cheapest legal setup and forget that authorisation cost is only the front door.
The EBA authorisation guidelines show how broad the application burden really is: the applicant must provide detailed information on the business model, governance, internal controls, safeguarding, shareholders, managers, and more. The guidelines also stress that the information must be true, complete, accurate, and up to date, and that applicants should provide extracted relevant sections rather than vague references to internal materials.
That tells you something important about jurisdiction choice. The cheaper jurisdiction on paper may not be cheaper in practice if it turns out to require repeated revisions, more remediation, more governance rebuilding, or more difficult post-licensing supervision for your specific model. This is partly a strategic inference, but it follows directly from the fact that the application and supervisory burden is much wider than a filing fee or a lawyer’s quote.
Mistake 7: Following old market folklore instead of current supervisory signals
A lot of founders still rely on outdated market narratives: which jurisdiction was popular three years ago, where other fintechs used to go, or which country had a reputation for being easier.
That is risky because supervisory conditions change, application mixes change, and regulators react to market behaviour. The EBA’s follow-up review says application volumes and timelines shifted materially between recent periods, and that 2022–2024 saw more new entrants with lower-quality applications than earlier waves dominated by Brexit-related relocations.
That means yesterday’s “easier jurisdiction” can become today’s crowded, slower, or more sceptical one. Founders who choose based on stale industry gossip often end up solving the wrong problem.
Mistake 8: Choosing the jurisdiction before defining the actual EMI model
This is the strategic mistake underneath all the others.
The EBA’s authorisation guidelines for EMIs make clear that the level of detail in the file should be proportionate to the applicant’s size, internal organisation, and the nature, scope, complexity, and riskiness of the services it intends to provide. They also require the application to be complete and tailored to the specific applicant.
That means jurisdiction should come after the business model is properly defined, not before. A wallet-led consumer EMI, a B2B embedded-finance infrastructure play, and a cross-border merchant-flow model may all sit inside the EMI perimeter, but they do not create the same governance, AML, operational, or passporting pressures. Founders who choose the country before they lock the real operating model usually end up redesigning both at once.
What a better jurisdiction choice looks like
A better approach is less glamorous and much more effective.
First define what the EMI will actually do. Then test where it can be genuinely managed and controlled. Then assess how the local supervisory approach aligns with your governance maturity, AML profile, and cross-border strategy. Then model passporting realistically, including home-host procedures and possible additional host-side expectations. Finally, price the full life of supervision, not just the filing stage. That sequencing is an inference, but it is the one most consistent with the EBA’s current findings on local substance, application quality, supervisory divergence, and cross-border processes.
AMS Europe helps founders assess jurisdiction fit through the real lenses that matter: business model, local substance, governance, AML/CFT, and passporting strategy.
Final thought
The biggest mistake founders make when choosing a jurisdiction for EMI licensing is assuming they are choosing a place to submit documents.
They are not. They are choosing the supervisory home of a regulated institution. That home has to fit the real business, not the imagined shortcut. And the more the EMI depends on cross-border growth, technology, partners, and high-risk flows, the more expensive a bad jurisdiction choice becomes later.
FAQ
Is there one best jurisdiction for EMI licensing in Europe?
Not in any universal sense. The EBA’s 2025 follow-up review says supervisory expectations still diverge across Member States in governance, internal controls, AML/CFT, and local substance, so the right fit depends on the actual business model.
Should founders choose based mainly on speed?
Usually no. The EBA says timelines vary widely, but delays are often caused by incomplete or low-quality applications, changes during assessment, and business-model complexity rather than jurisdiction alone.
Does passporting remove the importance of jurisdiction choice?
No. The EBA explains that passporting still depends on home-host notifications, standard forms, procedures, and supervisory cooperation. In some cases, host states may also require a central contact point when activity reaches certain thresholds.
Why does local substance matter so much?
Because supervisors now report verifying that applicants are effectively managed and controlled from the home Member State, and the EBA warns that divergent expectations here can fuel forum shopping and regulatory arbitrage.
Should AML influence the jurisdiction decision?
Yes. The EBA says in its ML/TF risk opinion more competent authorities now assess inherent ML/TF risk in the EMI sector as significant to very significant, and it reports continuing weaknesses in monitoring, customer identification, and suspicious reporting in the sector.
