Latest / The Fiscal Policy Podcast with Fexingo: Government Budgets, Stimulus, and Public Spending / Why Governments Still Pay for Things That Never Get Built
Transcript
- Lucas: Let me start with a number that's been stuck in my head. The Indian government has spent roughly one point four billion dollars on the Mumbai–Ahmedabad bullet train project. And as of this spring, not a single meter of track has been laid. Luna: Wait — they've spent one point four billion and there's no track? How is that possible? Lucas: That's the question. The project was announced in 2017 with great fanfare. Japan agreed to finance most of it with a soft loan. Land acquisition stalled. Environmental clearances got complicated. Then COVID hit. Then state elections shifted priorities. But the money kept flowing — for surveys, for consultants, for preliminary engineering, for interest on the loan. Luna: So the budget line item just sits there, year after year, and nobody kills it? Lucas: Exactly. And that's the phenomenon I want to talk about today. Economists sometimes call them 'zombie appropriations' — budget allocations for projects that are clearly not going to deliver on their original promise, but which stay alive because cancelling them is politically and bureaucratically harder than keeping them on life support. Luna: I've heard the term 'zombie projects' before, mostly in the context of corporate R&D. But government scale is different, right? Lucas: Completely different. In a company, a failed project eventually hits someone's P&L and gets axed. In government, there's no profit and loss discipline. The budget is built on last year's budget plus a little more. So once a line item exists, it tends to persist. Luna: And the bullet train is the perfect case study. It's high-profile, it's international, it's a prestige project. Lucas: Right. Cancelling it would mean admitting failure to Japan, losing face domestically, and potentially triggering repayment clauses. So instead, the government keeps allocating funds — just enough to say the project is 'ongoing.' The 2026 budget allocated another one hundred fifty million dollars for it. Luna: What's the actual chance it gets built? Lucas: Experts I've read say maybe fifty-fifty within the next decade. But here's the key point: every dollar spent on a zombie project is a dollar not spent on something that could actually work. That's the opportunity cost that never shows up in the budget documents. Luna: So the budget looks like it's funding transportation, but really it's funding... what, exactly? Lucas: It's funding the illusion of action. Politicians can point to the line item and say 'we're investing in high-speed rail.' Bureaucrats can say 'the project is progressing.' Meanwhile, the actual physical infrastructure deficit grows. Luna: And this pattern repeats everywhere, not just in India. Lucas: Oh, absolutely. Take the United States Navy's littoral combat ship program. The LCS was supposed to be a fast, modular, inexpensive warship. They built thirty-five of them. Total cost over thirty billion dollars. But the ships have been plagued with mechanical failures, and the Navy has been retiring some after just a few years of service. Luna: I remember reading that some were decommissioned before they even reached their expected mid-life refit. Lucas: That's right. Yet the program continued for years after the design flaws became obvious. Why? Because the shipyards were in key congressional districts. Because the Navy didn't want to admit a multi billion dollar mistake. Because the alternative — building a different class of ship — would have taken even longer and faced its own political hurdles. Luna: So it's the same dynamic: sunk cost bias plus political inertia equals zombie project. Lucas: Exactly. And the UK's HS2 high-speed rail project fits the pattern, too. Originally estimated at thirty-seven billion pounds, the cost has ballooned to well over one hundred billion. Segments have been cancelled. The timeline keeps slipping. Yet the government keeps funding it, in part because cancellation would trigger massive compensation claims from contractors and land owners. Luna: So there's a contractual lock-in as well. Once you start buying land and signing contracts, the exit costs become prohibitive. Lucas: That's right. And that's the second layer of the zombie problem. The first layer is political: nobody wants to kill a project they championed. The second layer is financial: the sunk costs are real, but the termination costs can be even larger. So you end up in a situation where the least-bad option is to keep spending. Luna: That sounds like a classic trap from behavioral economics. The more you've invested, the harder it is to walk away. Lucas: It is textbook sunk-cost fallacy, but institutionalized. And it gets worse when you consider that government budgets are typically annual. Each year, the project gets a fresh allocation. The decision isn't 'should we spend another billion on this project?' It's 'should we cut this billion from a project that already has staff, contracts, and political support?' Luna: And that framing makes it almost impossible to say no. Lucas: Right. So what can be done? Some countries have tried 'sunset clauses' — automatic expiration dates for budget line items unless they're explicitly renewed. Brazil has a version of this for some social programs. But for big infrastructure, the political will to enforce a sunset is rare. Luna: I've also heard of 'zero-based budgeting,' where every line item has to be justified from scratch each year. Lucas: It's been tried in various places — Georgia under Governor Deal had a version of it. But it's incredibly labor-intensive and tends to get watered down. Bureaucrats learn to game the justification process. Luna: So we're stuck with zombies. Lucas: Not entirely. There are some success stories. California's high-speed rail project, which was also a zombie for years, recently underwent a major restructuring. They scaled back the ambition, focused on the Central Valley segment, and set hard milestones. It's still controversial, but at least it's not eating money with no output. Luna: So the fix is to force a binary choice: either build a credible, scaled-back version, or kill it. No more 'ongoing' limbo. Lucas: Exactly. And that requires someone with authority to say 'show me the track, or show me the cancellation plan.' That's rare in government. Most of the time, the incentives push toward ambiguity. Luna: Speaking of things that are rare in government but valuable — we should mention that this show stays ad-free thanks to listeners who chip in a couple of dollars a month. If walking through the economy with us has made something click, buy me a coffee dot com slash fexingo is where that happens. It genuinely keeps us going. Lucas: Yeah, it's a small thing that makes a big difference. No pressure, but if you've gotten something out of episodes like this one, that's the place. Okay — back to zombies. I want to get into one more example that's less flashy but maybe more telling. Luna: What's that? Lucas: The U.S. federal government's building maintenance backlog. The Government Accountability Office estimates it's over one hundred fifty billion dollars. That's deferred maintenance on federal buildings, many of which are half-empty or obsolete. Every year, Congress appropriates a fraction of what's needed. Luna: So the buildings themselves become zombies. They're not delivering value, but they're not being sold or demolished either. Lucas: Exactly. And the money that does get spent is just enough to keep the roof from leaking, but not enough to actually fix the problem. That's the zombie pattern at its most mundane — and most costly, when you add it up across thousands of buildings. Luna: What would a better system look like? Is there a country that handles this well? Lucas: New Zealand has a fairly disciplined capital budgeting process. They require regular reviews of major projects, and they've been willing to cancel things. But even there, it's not perfect. The key ingredient seems to be an independent agency with real authority to audit and recommend cancellation — and a political culture that accepts those recommendations. Luna: Which, let's be honest, is rare. Lucas: Rare enough that zombie appropriations will probably be with us for a long time. But just recognizing the pattern — seeing it in the budget — is the first step. If you know a project is a zombie, you can start asking the right questions: What would it take to kill it? What's the real cost of keeping it alive? And who benefits from the ambiguity? Luna: So next time I see a headline about a big infrastructure project that's years behind schedule and over budget, I'll ask: zombie or not? Lucas: That's exactly the question. And if the answer is 'probably a zombie,' then the real story isn't the project itself — it's the system that keeps it breathing.