The numbers behind the headline
On 15 September 2026, Estonian public broadcaster ERR reported that e-residents had registered more than 4,200 Estonian companies so far this year, about 36% more than the same period of 2025 and 47% more than 2024. Around 9,000 people joined the programme in 2026. In the first seven months, the state collected €57.6 million in direct revenue from e-resident businesses: €35.3 million in labour taxes, €19.5 million in dividend income tax and €2.8 million in state fees, as summarised here. These are programme-reporting figures, not audited accounts, and they should be read that way.
Even with that caveat, the trend is consistent with the full-year record. The e-Residency programme's own 2025 report puts 2025 economic impact at €124.9 million, from 5,556 new companies and 13,828 new e-residents. It says e-residents now found roughly one in five new Estonian companies, and that the programme has 135,000+ e-residents from 185 countries. Minister Erkki Keldo's framing is that every euro spent returned more than twelve. If 2026's seven-month pace holds, the programme is on track for another strong year.
Why this matters beyond Estonia
E-Residency is the clearest working test of a proposition many governments only discuss: that a state can sell access to its legal and administrative system as a digital service, without requiring physical presence. The results so far argue for the open model. Most e-resident companies are small, many are founded by people from Germany, France and Ukraine (the top three application countries in 2025), and the state earns tax revenue on activity that would otherwise have gone to no European jurisdiction at all.
The composition of the revenue is also informative. Labour taxes are the largest component, which suggests that many e-resident companies employ people or pay their owners salaries in Estonia, not merely park profits. That is a different and more durable story than a pure tax-shelter model.
The strongest case for tighter controls
The sceptics' case deserves a fair hearing. Remote onboarding means the state cannot always confirm who it is dealing with. When Estonia's interior ministry drafted rules to restrict e-residency for nationals of high-risk countries, it cited concerns from MONEYVAL, the Council of Europe's anti-money-laundering committee, and an official's worry that "we may not receive answers, meaning we don't really know who this person is". The restricted list, based on Financial Intelligence Unit assessments and EU and FATF risk designations, covers 32 entities including Iran, North Korea and Syria. Russia and Belarus already face stricter renewal rules since 2022. A system that grows 36% in a year can also grow its abuse surface by 36% if screening does not scale.
That is a legitimate argument, and the response to it should be better evidence, not retreat.
What proportionate looks like
The policy answer is risk-based screening at the point of identity issuance and at the point of company formation, rather than caps on the programme or blanket country bans that punish legitimate founders. Three things follow.
- Publish audited outcomes. The programme reports revenue it can attribute to e-residents, but methodology matters: attributing a company's entire tax take to e-residency overstates the counterfactual if some founders would have used Estonia anyway. Independent audit by the National Audit Office or an academic team would turn a promotional figure into a policy-grade one. The same openness should cover enforcement data, such as how many e-resident companies are flagged or liquidated.
- Treat conversion as a signal. ERR noted that two thirds of new e-residents do not immediately register a company. The programme's own materials say 34% of those who joined in early 2026 had already formed one, a record. Low conversion is not a failure, but it means headline sign-up counts are a weak proxy for economic value, and screening should focus on the companies that actually form and transact.
- Keep the burden on activity, not identity. Banks, company service providers and the FIU already carry anti-money-laundering duties. Strengthening them is more proportionate than rationing digital identity itself.
The 2027 test: mobile e-Residency
The programme is moving to a cardless model. From 1 January 2027 the state fee becomes a flat €165, and the programme's roadmap says applicants will submit biometrics through a smartphone app that complies with EU eIDAS requirements, removing the embassy visit. Physical cards continue during the transition, and over 50 pickup locations remain.
This is the right direction for access, and officials have estimated it could raise company formation by around 20% and add €3 million to €9 million in annual tax revenue, according to ERR's coverage of the 2025 results. But the in-person pickup was also, in practice, a vetting checkpoint. Removing it means remote biometric capture and liveness detection must carry the full weight of identity assurance. That is where Estonia's credibility is on the line, and it is why the supplier's technical standards and the programme's fraud rates should be public.
Bottom line
The 2026 figures strengthen the case that open digital identity can be a growth policy. They do not by themselves prove the programme is risk-free, and the data are unaudited. The proportionate course is to keep the doors open, scale risk-based screening alongside volume, and let independent auditors confirm what the numbers claim before other governments copy the model.