Latest / The Productivity Podcast with Fexingo: Output, Efficiency, and Long-Term Economic Growth / How One Open-Source Tool Saved Governments Billions
Transcript
- Lucas: You know, when we talk about productivity gains, we usually reach for the dramatic stories—the new factory robot, the supply chain overhaul, the billion-dollar infrastructure project. But some of the biggest ROI stories in the last twenty years have come from something much quieter: governments switching to open-source software. Luna: Quiet but massive. I think of Munich's LiMux project—that was one of the earliest high-profile migrations. How did that actually play out? Lucas: So Munich, back in 2003, decided to move its 15,000 municipal computers from Microsoft Windows and Office to a Linux-based system called LiMux, plus OpenOffice. The stated goals were cost savings and digital sovereignty—they didn't want to be locked into one vendor's licensing fees. By 2013, they had fully migrated. And by 2017, the city calculated they had saved over 10 million euros in licensing costs alone. Luna: Ten million euros—just from licenses. But I remember there were political hiccups, right? The mayor at one point tried to reverse the decision. Lucas: Yeah, in 2014, a new mayor came in who was less enthusiastic. There was pressure from Microsoft, compatibility complaints from some departments… eventually, in 2017, the city council voted to move back to Windows. But the interesting thing is, when they costed out the reversal, they found it would cost over 10 million euros to switch back—so they basically broke even either way. The lesson isn't that LiMux failed; it's that switching costs are real, but so are the savings. Luna: Right. And the French Gendarmerie—that's a much clearer success story. They moved 90,000 workstations to OpenOffice and Firefox starting in 2005, and by 2011 they'd saved about 50 million euros. Lucas: Exactly. The Gendarmerie's migration is one of the best-documented cases. They didn't just swap software—they retrained staff, standardized document formats, and built internal support teams. Their total cost of ownership dropped by about 30 percent compared to staying with Microsoft Office. And they haven't looked back. Today they're on LibreOffice and continuing to save. Luna: So when you scale that—90,000 seats times 30 percent savings—that's not trivial. And it's not just Europe. Brazil's federal government migrated over 400,000 desktops to LibreOffice a few years ago. I've seen estimates of over 120 million dollars saved. Lucas: That Brazil migration is fascinating because it wasn't a top-down mandate initially. It started in 2005 with a few ministries, and then it spread organically as they realized how much they were saving on licensing—especially as the Brazilian real weakened against the dollar. By 2012, it was official policy. And the savings aren't just licensing—they also avoid audit costs, because with open source you're not at risk of being found non-compliant with proprietary licenses. Luna: That's a point people miss. The indirect costs of proprietary software—compliance, audits, vendor negotiations—they add up. And open source removes that friction entirely. Lucas: Which is really the productivity argument in a nutshell: by reducing friction, you free up resources for actual work. And the UK's NHS is a great example. In 2013, they estimated that using open-source software in their GP system—the software doctors use to manage patient records—saved them about 30 million pounds. That's money that went back into patient care, not licensing fees. Luna: And yet, adoption has been slower than you'd expect given the numbers. Why do you think that is? Lucas: I think there are a few reasons. First, switching costs are upfront and visible: you have to train staff, migrate data, deal with compatibility issues. The savings are spread out over years—so a budget-conscious CIO might not want to take the short-term hit. Second, there's a perception that open source is less secure or less reliable, even though in many cases it's actually more transparent and auditable. And third, vendor lock-in is real—once you're embedded in an ecosystem, extricating yourself feels risky. Luna: The security perception is interesting, because the US Department of Defense has been one of the biggest open-source adopters. They created a whole policy in 2018 that said open-source software is part of the DoD's approach to cybersecurity—because they can inspect the code. Lucas: Exactly. And the DoD even publishes its own open-source projects. That's a huge signal. If the Pentagon trusts it, the security argument really weakens. But the perception lingers in the commercial and public sector. Luna: And then there's the economic argument around competition. When governments standardize on open formats, they're basically ensuring that no single vendor can hold them hostage. That's a productivity gain in the long run, even if it's hard to measure. Lucas: Absolutely. And actually, this connects to something I think about a lot with this show. We spend our time walking through the economy and looking at what makes systems more efficient—whether it's a factory in Germany or a bus company in Brazil. And if that kind of thinking has ever made something click for you, or helped you see an old problem in a new way… well, we keep this podcast ad-free because of listener support. So if you find value in what we do, you can join that community at buy me a coffee dot com slash fexingo. It truly makes a difference. Luna: Yeah, it's a small way to keep the conversation going. And speaking of conversations—let's get back to that DoD example, because I think it points to a bigger trend. Lucas: Sure. So the DoD's 2018 open-source policy was driven partly by a report that found that 96 percent of all software applications contain open-source components. The government wasn't adopting open source—it was already using it, just not managing it. So the policy was about formalizing it, making sure they were contributing back and not ignoring security vulnerabilities. Luna: Right. And that's the other productivity angle: when governments actively participate in open-source projects, they get better software over time. The code improves because more eyes are on it. That's a classic network effect. Lucas: It is. And Europe has been leading in policy here. The European Commission has its own open-source strategy—they mandate that all new software developed for the EU institutions be released under an open-source license. That's a huge library of tools that any member state can reuse without paying again. Luna: That's the real productivity win: avoiding reinventing the wheel. If one government builds a tax filing system, another can adapt it for a fraction of the cost. Lucas: And we're seeing that happen. Estonia, which is famous for its digital government, has shared components of its e-governance system with Finland, Ukraine, and others. The initial development cost is high, but the marginal cost of sharing is near zero—that's economics 101. Luna: But then the question becomes: why isn't every government doing this? The evidence is clear. Is it just inertia? Lucas: Partly. But I think there's also a legitimate concern about support. With proprietary software, you have a vendor to call. With open source, you need in-house expertise or a third-party support contract. Some governments don't have that capacity, so they pay for convenience. Luna: That's a fair point. But the savings are so large that it seems like investing in that capacity would pay for itself. Lucas: It would in the long run. But government budgets are often annual, and politicians care about the next election cycle, not the ten-year horizon. That's a structural barrier to productivity that's hard to fix with software. Luna: So it's a collective action problem. Each individual decision-maker sees the upfront cost, but the long-term benefits are dispersed. Lucas: Exactly. And that's where organisations like the Open Source Initiative and the Linux Foundation come in—they try to aggregate demand, share best practices, and make the case to policymakers. There's been real progress, but it's slower than the pure numbers would suggest. Luna: Let's talk about a specific number that might move the needle. I saw a study that estimated if all US state governments migrated their office software to open source, they'd save over one billion dollars per year in licensing fees alone. Lucas: That's a big number. And it's probably conservative because it doesn't count the indirect savings—reduced audit costs, no forced upgrades, longer hardware life because open source software often runs well on older machines. So the true productivity gain is even larger. Luna: And yet we haven't seen a single US state do a full migration. Why? Lucas: I think part of it is lobbying. Microsoft, Oracle, and others have strong relationships with state CIOs. They offer discounts, they bundle products, they make switching seem painful. And there's the Microsoft Office format lock-in—many states have years of documents in proprietary formats. Migration tools exist, but it's friction. Luna: So the open-source movement has had more success in countries where the government has more central authority—like France, Brazil, or the EU institutions. The US federal system makes coordination harder. Lucas: That's a good frame. And even within the US, there are bright spots. The city of Los Angeles moved to OpenOffice in 2009 and saved millions. But it didn't start a wave. The lesson might be that open-source adoption, like any productivity improvement, depends on local champions and specific conditions. Luna: It's also worth noting that the landscape is changing. Cloud-based suites like Google Workspace are now alternatives that compete with Microsoft—that's a different kind of competition, not open source but still reducing lock-in. Lucas: True. And even Microsoft has embraced open source in a big way—they're now one of the largest contributors to GitHub. So the old proprietary vs open binary is blurring. But for governments that want full control and long-term cost certainty, open source remains the clearest path. Luna: So what's the one takeaway for our listener? If you're working in government or advising a public-sector client, what should you do? Lucas: I'd say: start small, measure everything, and focus on total cost of ownership over a five-year horizon. Run a pilot with one department. Track licensing, training, support, downtime, and hardware lifespan. The data will speak for itself. And if it works, the savings can fund the next step. Productivity isn't about big bangs—it's about compounding good decisions. Luna: And maybe share your results. Because the more case studies we have, the easier it is for the next government to take the leap. Lucas: Exactly. That's the open-source ethos in action—transparency and sharing. It's fitting that the method mirrors the message.