A launch date that is a starting line, not a finish
On 6 October 2026 the Bundestag's Committee for Digital Affairs and State Modernisation heard experts on the Digitale Identitätengesetz (DIdG, Bundestag document 21/7408), the national law supporting Germany's EU Digital Identity Wallet. The government wants the wallet available on 2 January 2027. According to the Bundestag's own summary of the hearing, the invited experts welcomed the wallet but asked for changes to the bill.
The strongest case for the government's approach is simple. The EU's amended eIDAS Regulation (Regulation (EU) 2024/1183) obliges member states to offer citizens a wallet, and Germany has no good domestic alternative: the existing eID card function has seen limited everyday use. A voluntary wallet that lets people prove identity or age from a phone, and that works across borders, could remove a great deal of paperwork. The bill's own estimate, as reported at its 23 September first reading, is that annual compliance costs in administration would fall by about €121.7 million, against roughly €13.5 million in one-time costs for business. Those are government projections, not outcomes.
We support that ambition. But the hearing showed that a fast launch and a trustworthy launch are not the same thing, and the gap is mostly fixable by drafting rather than delay.
The case for staging
Torsten Lodderstedt of SPRIND, the federal innovation agency involved in building the wallet, argued for a phased rollout. He said the first version would already be better than previous options and described 2 January 2027 as the start of scaling, not the end of development. The Federal Data Protection Commissioner's representative, Andreas Hartl, accepted the legal basis for a first version that will not yet meet every eIDAS requirement, as long as the government communicates clearly what the first version can do and when the remaining requirements will be met.
This is a reasonable, pro-innovation position. Identity infrastructure that waits for perfection never ships, and an ecosystem needs real issuers and real relying parties to mature. Germany also runs a sandbox with SPRIND where companies and public bodies test wallet functions, which began its first cohorts in December 2025 according to Bitkom. Staging is how you learn from real use.
The condition is that staging must not mean deferring the protections that are hard to retrofit.
Where the draft falls short
Pseudonymous use. Hartl called the missing option for pseudonymous use unfortunate from a privacy standpoint. This is not a niche demand. The EU regulation's recitals say wallets should include a functionality to generate user-chosen and managed pseudonyms for authentication, and that accessing services under a pseudonym should not be prohibited (Regulation (EU) 2024/1183, recitals 22 and 60). A wallet that forces full identification where a pseudonym would do pushes users toward over-disclosure, which is the opposite of what a free and open internet needs.
Opaque security architecture. Prof. Jiska Classen of the University of Potsdam told the committee that much of the security architecture is not public, has not been security-assessed, and is not set out in the law. She also noted that smartphones carry a high risk of data extraction, partly because many Android devices no longer receive updates. For a credential meant to reach the 'high' assurance level on mobile phones, that is a serious flag. The remedy is cheap relative to the risk: publish the architecture, commission independent assessment, and make the results available before, not after, mass adoption.
Weak guardrails on who may ask for data. Lina Ehrig of the Verbraucherzentrale Bundesverband said the draft does not use the flexibility EU law allows to secure a privacy-friendly, consumer-oriented design, and argued that data queries should be limited to what is strictly necessary. Bianca Kastl warned that without oversight of provider registration the result could be a 'Cookie banner 2.0', where every service asks for more than it needs and users click through. Thomas Lohninger of epicenter.works rated the draft as leaning negative and doubted the wallet would earn enough trust to be recommended.
There is a fair counterpoint. Arno Fiedler of the Verband Sichere Digitale Identität praised the draft's data protection level and its experimentation clause, and Bitkom's Susanne Dehmel asked for more precision on attestations and the wallet provider's role, a call for clarity rather than restriction. Municipal bodies, meanwhile, said the wallet must work in local administrations by 2 January but that the draft under-addresses municipal integration.
Proportionate fixes, not a brake
None of the concerns requires abandoning the timetable. Proportionate regulation here means a short list of targeted amendments:
- Write pseudonymous use into the first version or into a dated roadmap in the law, so staging does not become indefinite deferral.
- Publish the security architecture and an independent assessment ahead of the mass launch, with the Federal Office for Information Security (BSI) and the Bundesnetzagentur, which would be central supervisor, given clear duties to report on residual risks.
- Hard-wire data minimisation in relying-party registration: a service should have to justify each attribute it requests, and the Bundesnetzagentur should be able to revoke registrations for over-asking.
- Keep the wallet strictly voluntary. The bill states that use is voluntary and does not replace existing identification methods. That commitment is the single most important safeguard against identity requirements creeping into everyday life, and it should survive amendment untouched.
The risk of getting this wrong is asymmetric. A wallet launched on time but distrusted will see low uptake, leaving the promised administrative savings on paper. A wallet that is a week later but ships with pseudonymity, published security design and tight provider limits could become the trusted default. The committee has the time between now and January to choose the second path.
What to watch
The committee must still report the bill back to the Bundestag, and the practical test is whether amendments address pseudonyms, security transparency and registration safeguards. Also watch the municipal readiness question and the timeline for the central issuing and verification service, which the Deutscher Landkreistag said would not be operational until later in 2027. If the launch is only the start of scaling, the law should say what scales, when, and with which protections.