Latest / The Fiscal Policy Podcast with Fexingo: Government Budgets, Stimulus, and Public Spending / Why Public Rail Projects Always End Up Late and Over Budget
Transcript
- Lucas: So, the California high-speed rail project was originally supposed to cost thirty-three billion dollars and open in 2020. As of August 2026, the latest estimate is over one hundred twenty-eight billion, and there's still no opening date. Luna: That's nearly four times the original budget. And the state's been working on it since 2008. Lucas: Right. And California is not alone. This is a global pattern. You see it in Germany with Stuttgart 21, in the UK with HS2, in France with the Grand Paris Express. Public rail projects have a remarkable track record of blowing their timelines and budgets. Luna: So what's going on? Why can't governments seem to get these projects right? Lucas: Let's start with the obvious one: cost estimates are often done before the design is complete. In California, the initial estimate was essentially for a concept, not a detailed engineering plan. Once you start drilling tunnels and buying land, the real numbers come in. Luna: But don't they do feasibility studies? I mean, they must have some idea of the costs. Lucas: They do, but those studies are often based on optimistic assumptions about geology, land acquisition, and labor costs. And there's a well-documented phenomenon called 'strategic misrepresentation'—that's when planners deliberately underestimate costs to get the project approved. Politicians want to announce a ribbon-cutting, not a cost overrun. Luna: So it's almost like a political incentive to lowball the numbers. Lucas: Exactly. And once the project is underway, it's very hard to stop it. The money's already been spent, the political capital's invested, and the contractors are on site. So you just keep pouring in more. Luna: But there's also the sheer complexity, right? Building a rail line through dense urban areas can't be easy. Lucas: That's a big factor. The more the line interacts with existing infrastructure—utilities, subways, roads—the more surprises you get. In Germany, Stuttgart 21 involves digging under the main train station, and they've hit unexpected groundwater issues and had to deal with ancient Roman ruins. That's not something you can plan for in a cost estimate. Luna: Roman ruins? That's a wildcard. Lucas: And it's the kind of thing that eats into both time and money. So you've got optimistic estimates, political pressure, and technical unknowns. But there's another layer: public projects don't have the same incentives to control costs as private ones. Luna: What do you mean by that? Lucas: If a private company builds a rail line and it goes over budget, shareholders punish them, and managers might lose their jobs. With a public project, the cost is spread across all taxpayers. There's no single owner who's losing money. And the people making the decisions are often not the ones paying for it. Luna: That's a classic agency problem, right? The people who decide don't bear the full cost. Lucas: Precisely. And that's why you see such a difference in performance between public and private projects. There are studies that show private infrastructure projects tend to be more predictable in their budgets. Not because they're inherently easier, but because there's a hard budget constraint. Luna: But wait, we've had some public projects that succeeded, like the Channel Tunnel. That came in close to budget, didn't it? Lucas: Actually, the Channel Tunnel went over budget too. It was originally estimated at about five point five billion pounds in 1987, and ended up costing around nine point five billion. That's a seventy percent overrun. But it did get built and it's been operating for decades. Luna: So even the 'successes' have overruns. What's the worst case you've seen? Lucas: There's a famous example from Denmark: the Great Belt Fixed Link, which connects the islands of Zealand and Funen. That went over budget by more than fifty percent and took longer than planned. But it's hard to say which is the absolute worst because you have to adjust for inflation and currency. Luna: But in terms of sheer dollar amount, California has to be up there. Lucas: It's the most expensive public infrastructure project in U.S. history. And the interesting thing is that the original plan was for a line connecting Los Angeles and San Francisco in two hours and forty minutes. Now, they're building a segment in the Central Valley, which is the flattest part of the route. And even that segment has been scaled back. Luna: So you're building the easiest part first, but the whole line is still years away. Lucas: Right. And that's another pattern: projects often get split into phases, and the first phase gets built, but the rest never materializes because costs have escalated so much. Or the phase that gets built isn't the one that delivers the original benefits. Luna: It's like the sunk cost fallacy applied to public works. You've already spent so much that you have to keep going, even if the benefit no longer justifies the cost. Lucas: And that's a tough position for politicians. If you kill the project, you're admitting that billions were wasted. If you continue, you can say you're 'committed to finishing what was started.' Luna: So what could be done differently? Any ideas? Lucas: One thing that's been suggested is to require independent cost-benefit analysis before a project is approved, and to have that analysis be binding. Also, benchmarking against similar projects—like the cost per kilometer for rail in different countries—can help set realistic ranges. Luna: But wouldn't politicians just ignore that if it doesn't fit their narrative? Lucas: They might. But if the public is aware of the benchmarks, it becomes harder to sell a cheap dream. Transparency is a powerful tool. Luna: It's like that saying, 'trust but verify.' For public works, maybe we need 'build but verify.' Lucas: That's a good way to put it. And it's interesting that some jurisdictions are starting to adopt these practices. For example, the UK's Infrastructure and Projects Authority has been publishing more detailed cost data, and they've seen some improvement in the accuracy of estimates. Luna: But we're still talking about billions of dollars at stake. It's not like we've solved the problem. Lucas: No, it's a structural issue. And it's not just rail—it applies to bridges, highways, public buildings. But rail is the poster child because the costs are so visible and the delays so long. Luna: And that's exactly why we're covering it. If you're a taxpayer, these overruns affect you directly. Speaking of which, if you find this kind of analysis useful, and you want to support our work, a simple way is to buy me a coffee dot com slash fexingo. It's just a small gesture that helps keep this show independent and ad-free. Lucas: And we appreciate that more than you know. So, back to the numbers: I was reading a study that looked at over three hundred public rail projects worldwide, and the average cost overrun was about forty-five percent. That's not a one-off; it's the norm. Luna: Forty-five percent on average. So when a project comes in at twice the estimate, that's actually on the higher end, but not unheard of. Lucas: Exactly. And the interesting thing is that the overruns are even higher for projects that are particularly long or complex. The study found that overruns tend to increase with the length of the project, both in terms of distance and time. Luna: So the bigger you dream, the more you pay. Lucas: And that's why some experts advocate for a different approach: incremental delivery. Instead of building one massive line, you build in segments, each with its own cost-benefit analysis, and you only continue if the benefits still justify the costs. Luna: But wouldn't that be less efficient? You'd have to start and stop construction, and you might lose economies of scale. Lucas: It's a trade-off, for sure. But the alternative—committing to a massive project with uncertain costs—has proven to be a recipe for fiscal disaster. The incremental approach gives you options and limits your downside. Luna: So what does this all mean for the future of public rail? Are we going to see fewer mega projects? Lucas: I think we'll see a shift toward more modest, better-planned investments. And I think the public is becoming more skeptical of big promises. When you see a politician tout a thirty-three billion dollar project, you might want to ask, 'What's the real number going to be?' Luna: And that's the takeaway: be skeptical, demand transparency, and maybe don't hold your breath for that high-speed rail from LA to San Francisco. Lucas: It may come, but it will be later and costlier than advertised. That's the almost inevitable reality.