German State Prioritizes Politics Over Practical Technology Solutions

Schleswig-Holstein, Germany’s northernmost state, has announced plans to abandon Microsoft’s productivity suite for open-source alternatives by mid-2025, framing the decision as a reduction in both digital dependencies and costs. But the reality is that the government is prioritising political symbolism and cronyism—the opportunity to reject an American company while directing public funds to German businesses—over a sound evaluation of its technology options.

Indeed, Schleswig-Holstein’s decision reflects a growing European trend of prioritizing made-in-Europe digital technology, even when those actions violate international trade agreements, raise costs, and deliver less value. Rather than embracing digital protectionism, EU member states should adopt open, evidence-based, and non-discriminatory procurement of digital tools.

Many European policymakers have been advocating for digital sovereignty—cutting ties with foreign technology firms and replacing them with European alternatives. For example, Schleswig-Holstein’s digitalisation minister has justified its plans as essential to data security, arguing that it had “no influence…on the handling of data, including a possible outflow of data to third countries.” But this claim is simply untrue.

First, any service provider—whether foreign or domestic—must abide by the same data protection laws, including the German Federal Data Protection Act (Bundesdatenschutzgesetz or BDSG), which implements and extends the EU’s General Data Protection Regulation (GDPR). The GDPR includes specific rules that restrict international data transfers to ensure the protection of personal data. Therefore, switching to a domestic provider has no impact on data protection.

Second, Germany, as with the rest of the EU, is part of the World Trade Organization, including its Agreement on Government Procurement (GPA). The GPA requires signatory nations to provide non-discriminatory treatment to suppliers from other GPA parties, meaning government procurement processes must be open, transparent, and based on commercial considerations rather than the nationality of the supplier. While Schleswig-Holstein’s framing of this decision as purely about “open-source” software may technically comply with GPA provisions, it appears to violate the spirit of the agreement by using the open-source argument as a pretext to exclude an American firm in favour of German ones.

Third, Microsoft has an extensive set of protections for European public sector customers. In 2020, the company launched its “Defending Your Data” program, which includes a commitment to legally challenging every government request for public sector of enterprise customer data. In 2023, Microsoft created its EU Data Boundary project to enable European customers to ensure they only store and process their data within the EU.

More recently, Microsoft unveiled additional “European Digital Commitments,” including a 40 percent expansion of European data centre capacity, legally binding promises to fight any government orders to cease European operations, and partnerships with European cloud providers like Bleu in France and Delos in Germany. In other words, the German public sector can run Microsoft technology in German data centres run by German personnel. Yet Schleswig-Holstein’s digitalisation minister’s response to these extraordinary accommodations was to declare, “We’re done with Teams!” This suggests the decision isn’t about sovereignty concerns—it’s about political posturing.

While Schleswig-Holstein officials claim this will save “tens of millions of euros,” this projection is based on incomplete accounting—considering only licensing costs and ignoring the substantial total cost of ownership for open-source alternatives. Open-source software has legitimate benefits and use cases, but government agencies still incur costs when they use it. These costs include dedicated internal or external IT support teams to replace vendor support, custom development for missing features, integration and compatibility maintenance, security patching responsibilities, and ongoing training costs as staff turnover occurs.

Historical evidence suggests caution is warranted. Testing alone can consume 60% of total conversion costs in complex government migrations, whilst training costs average £1,600 per employee with significant hidden operational expenses. Munich’s experience offers a particularly relevant cautionary tale. After pioneering open-source adoption in the 1990s, the city reversed course in 2017, citing substantial ongoing operational costs and integration difficulties.

Schleswig-Holstein’s changes will affect 30,000 public servants and 30,000 teachers. Many of these users may struggle with unfamiliar interfaces, steep learning curves, and collaboration challenges due to interoperability issues, resulting in lower productivity across the government. These operational realities don’t appear in initial budget projections but represent real upfront expenses.

The German state’s decision reflects a broader European trend where policymakers pursue digital protectionism even in the face of substantial costs. For example, the Digital Markets Act and Digital Services Act target U.S. tech companies, resulting in European customers paying between €43 billion to €71 billion more per year due to higher costs. Similarly, even as European regulators have disproportionately pursued fines against American businesses under GDPR, the law has also imposed enormous compliance costs on European businesses—Oxford University research shows an average 8.1 percent decline in company profits.

Schleswig-Holstein’s decision seems to be motivated more by politics than facts. By almost any metric—securing data, keeping data in Germany, or delivering value to taxpayers—remaining with a leading global technology firm with robust domestic operations would be the better option. Embracing digital protectionism is an expensive proposition that will be unlikely to pay off for those who pursue it, especially when ideology trumps evidence-based procurement. Whether choosing open-source or commercial solutions, decisions should be based on rigorous analysis of costs and benefits, not political favouritism.