#908: The Underground Paradox: Why Subways Cost Billions Today

Why does a mile of subway cost billions today? Herman and Corn explore the hidden complexities and rising costs of modern urban transit.

0:000:00
Episode Details
Published
Duration
00:21:58
Audio
Direct link
Pipeline
V4
TTS Engine
LLM

AI-Generated Content: This podcast is created using AI personas. Please verify any important information independently.

In the latest episode of My Weird Prompts, hosts Herman Poppleberry and Corn dive deep—literally—into the mechanical world breathing beneath our pavement. Inspired by a listener’s prompt comparing modern coal mining to urban transit, the discussion centers on a perplexing economic mystery: why was it seemingly easier and cheaper to build massive underground rail networks in the 1900s than it is today? As cities grow denser and the need for efficient transit becomes more desperate, the cost of digging has reached a staggering tipping point that threatens the future of urban mobility.

The "Open-Heart Surgery" of Modern Construction

Herman begins the discussion by contrasting the "cut and cover" methods of the 19th century with the surgical precision required today. In the 1860s, builders of the London Underground simply ripped up streets, laid tracks, and roofed them over. It was disruptive and dangerous, but it was fast. Today, Herman notes, digging under a city like New York or London is akin to performing "open-heart surgery on a patient running a marathon."

The modern underground is no longer an empty void; it is a dense thicket of fiber optic cables, high-voltage power lines, and ancient sewage systems. To add a new line, engineers must navigate this invisible maze without stopping the city above. This complexity is a primary driver of the "cost premium" that plagues modern projects.

The Billion-Dollar Mile

The hosts highlight the jarring disparity in construction costs across the globe. Corn brings up the Second Avenue Subway in New York, where the first phase cost a jaw-dropping $2.5 billion per mile. In contrast, Herman points to Madrid, Spain, where the Metro Sur—a 25-mile circular line—was completed for roughly $58 million per mile.

The difference, Herman explains, isn't necessarily in safety standards, but in "soft costs" and management. While the US and UK often rely on expensive outside consultants and "bespoke" architectural designs for every station, countries like Spain use standardized designs and keep expertise within the government. By owning their tunnel-boring machines and maintaining a dedicated internal team, Spanish authorities avoid the massive profit margins and legal entanglements that come with heavy outsourcing.

The Weight of Safety and History

A significant portion of the discussion focuses on the "legacy of fear"—the accumulation of safety regulations that follow major disasters. Following events like the 1987 King’s Cross fire, regulations regarding ventilation, fire suppression, and emergency exits have become non-negotiable. While these rules save lives, they also require massive vertical shafts and high-tech hardware that drive costs into the stratosphere. Herman argues that some modern codes may be over-engineered, based on worst-case scenarios that don't account for modern, less-flammable materials, yet these rules are rarely rolled back.

Furthermore, in historic cities, the earth itself holds secrets that can halt a multi-billion dollar project in its tracks. The duo discusses the Thessaloniki Metro in Greece, which faced decades of delays due to archaeological finds. Every shovelful of dirt in such cities risks hitting a Roman road or a Byzantine church, requiring expensive, time-consuming documentation and preservation that above-ground projects simply don't encounter.

The Environmental and Political Horizon

The conversation takes a turn toward the physical limitations of old infrastructure. In London, the very ground is changing; a century of train activity has saturated the surrounding clay with heat. Cooling these ancient tunnels now requires massive new ventilation systems, often requiring the purchase of some of the world's most expensive real estate just to "dig a hole for a fan."

Politically, subways present a unique challenge. Herman describes the "political horizon problem," where the costs of a project are concentrated in a single current budget, but the benefits—increased GDP and reduced traffic—are spread across decades. For a politician, justifying a twenty-year construction project that won't be finished until they are long out of office is a difficult sell, leading many to favor cheaper, surface-level options like Bus Rapid Transit (BRT).

Is Technology the Answer?

As the episode concludes, Herman and Corn look toward the future, specifically at innovations like Elon Musk’s The Boring Company. While Musk promises to reduce costs through smaller tunnels and automated digging, Herman remains skeptical about throughput. A subway's power lies in its ability to move 90,000 people per hour; moving individual cars in small pods doesn't solve the mass transit crisis. However, the engineering goal of making tunnel-boring machines faster and more continuous is a legitimate path toward changing the economics of the underground.

Ultimately, the hosts conclude that while the "tipping point" of cost is real, the verticality of the underground remains essential for the world's densest megacities. The challenge for the next century won't just be how we dig, but how we manage the bureaucracy, regulations, and historical layers that make our "hidden machines" so expensive to build.

Downloads

Episode Audio

Download the full episode as an MP3 file

Download MP3
Transcript (TXT)

Plain text transcript file

Transcript (PDF)

Formatted PDF with styling

Read Full Transcript

Episode #908: The Underground Paradox: Why Subways Cost Billions Today

Daniel Daniel's Prompt
Daniel
After listening to the episode about coal miners, I started thinking about underground transit systems like the London Underground and the New York City subway. Given that this old infrastructure is constantly being retrofitted with modern safety hardware, what are the economics of these systems today? Is it still feasible to build new, large-scale underground transit systems in major world capitals, or has the cost of meeting modern safety standards become prohibitive? Have we reached a tipping point where it’s only viable to develop above-ground infrastructure while maintaining existing underground systems?
Corn
You know, Herman, I was walking through the Old City yesterday, looking at all those layers of history literally piled on top of each other, and it made me think about how much we take for granted what is happening beneath our feet. We walk on these ancient stones, but in cities like London or New York, there is this whole other mechanical world breathing underneath the pavement.
Herman
It is the ultimate hidden machine, Corn. And it is a machine that never sleeps, even when it is breaking down. Herman Poppleberry here, by the way, for those of you who might be joining us for the first time on My Weird Prompts.
Corn
Right, and we have a really interesting jumping-off point today. Our housemate Daniel sent us a prompt after listening to our last episode, episode eight hundred ninety-two, where we talked about the brutal reality of modern coal mining. He was thinking about the parallels between those dark, deep mines and our urban transit systems. Specifically, he is curious about the economics. Is it even feasible to build these massive underground networks anymore, or have the safety requirements and the sheer cost of retrofitting old tech made them a relic of the past?
Herman
It is a brilliant question because it touches on something called state capacity. Why could we build hundreds of miles of track in nineteen hundred, but we struggle to build two miles today? Daniel mentioned the London Underground and the New York City subway, which are the grandfathers of this tech. But he also brought up newer projects like the one in Thessaloniki, Greece, which finally opened after decades of delays.
Corn
It really feels like we are at a crossroads. We want the density and the speed of underground rail, but the price tag is becoming something that even the wealthiest nations on earth are starting to choke on. So, Herman, let us start with the basics. Why was it so much easier, or at least seemingly more viable, back in the late nineteenth and early twentieth centuries?
Herman
Well, the short and somewhat grim answer is that we cared a lot less about people back then. When the first sections of the London Underground—the Metropolitan Railway—were being built in the eighteen sixties, they used a method called cut and cover. They literally just dug up the street, laid the tracks in a trench, and roofed it over. It was incredibly disruptive, it destroyed entire neighborhoods, and the safety standards for the workers were... let us just say they were non-existent compared to today.
Corn
Right, no OSHA back then. No environmental impact reports.
Herman
Exactly. And the trains themselves were steam-powered at first! Can you imagine the air quality in those tunnels? They had to have these huge vents to let the sulfurous smoke out. But because there was no existing infrastructure to worry about—no fiber optic cables, no complex sewage systems, no high-voltage power lines—they could just rip through the earth. Today, when you dig under a city like New York or London, you are basically performing open-heart surgery on a patient who is currently running a marathon. You cannot just stop the city.
Corn
That is a great analogy. The "patient" is the city, and every time you want to add a new line, you are trying to navigate a maze of existing pipes and foundations. But Daniel’s point about the economics is what really strikes me. If you look at the Second Avenue Subway in New York, the first phase, which was only about two miles long and added three new stations, cost roughly four point five billion dollars. That is over two billion dollars per mile. How does that happen?
Herman
It is a combination of factors that people in the industry call the "cost premium." Part of it is the sheer complexity of the modern safety hardware Daniel mentioned. We are not just talking about tracks and a tunnel anymore. We are talking about massive ventilation systems that can clear smoke in seconds in case of a fire. We are talking about full accessibility—elevators and escalators at every single entrance to meet disability acts. We are talking about fire-suppression systems, communication networks that work deep underground, and positive train control systems that prevent collisions automatically.
Corn
And those things are not optional. You cannot build a "budget" subway today because the liability would be astronomical. If anything went wrong, the legal and human costs would dwarf whatever you saved on construction. But I wonder, is it just the hardware? Or is it the way we manage these projects?
Herman
Oh, it is definitely both. There was a fascinating study by the Transit Costs Project at New York University. They looked at why some countries build much cheaper than others. For example, Italy and Spain actually build subways for a fraction of the cost of the United States or the United Kingdom. In Madrid, they built the Metro Sur, which is a twenty-five mile circular line with twenty-eight stations, for about fifty-eight million dollars per mile. Compare that to the two billion per mile in New York.
Corn
Wait, fifty-eight million versus two billion? That is an insane disparity. What is Spain doing differently? Are they just ignoring safety?
Herman
Not at all. Their safety standards are just as high. The difference is often in the "soft costs." In the United States and the United Kingdom, we spend an enormous amount of money on consultants, legal battles, and what they call "bespoke" designs. Every station has to be a unique architectural marvel. In Madrid, they used standardized station designs. They bought their own tunnel boring machines instead of leasing them. They had a dedicated, experienced team within the government rather than outsourcing everything to five different private firms that all want a profit margin.
Corn
So it is a management and procurement issue as much as a technical one. But let us look at the "tipping point" Daniel asked about. If the cost is this high, are we going to see a shift toward above-ground infrastructure? I am thinking about things like light rail or even what they call Bus Rapid Transit, where you give buses their own dedicated lanes and stations.
Herman
We are already seeing it. If you look at mid-sized cities that are growing rapidly, they are almost never looking at heavy rail subways. They are looking at BRT because you can build fifty miles of BRT for the price of one mile of subway. But there is a catch. Density. If you are in a city like Hong Kong, Tokyo, or Manhattan, you simply do not have the surface space. You cannot move ninety thousand people per hour on a bus lane. You need the verticality of the underground.
Corn
That is the paradox, right? The cities that need subways the most are the ones where it is most expensive and difficult to build them. And then you have the maintenance of the old systems. Daniel mentioned retrofitting. Think about the London Underground. Some of those tunnels are over a hundred and fifty years old. They were built for small, four-car steam trains, and now we are trying to cram high-frequency electric trains with air conditioning into them.
Herman
And air conditioning is a huge problem! People do not realize that the clay around the London Underground has been absorbing heat from the trains for a century. It is actually getting harder to cool the tunnels because the ground itself is saturated with heat. To fix that, you need massive new ventilation shafts, but where do you put them? You have to buy some of the most expensive real estate in the world just to dig a hole for a fan.
Corn
It makes me think about episode seven hundred fourteen, when we talked about the challenges of keeping technology waterproof and resilient in harsh environments. A subway is basically a giant, inhabited humid pipe. The salt from the streets in winter leaks down, corroding the steel. The water table is constantly pushing against the walls. It is a miracle they work at all.
Herman
It really is. And to Daniel’s point about the economics, we have to look at the "second-order effects." Even if a subway line costs five billion dollars to build and loses money on every ticket sold, the economic value it creates by allowing millions of people to get to work without sitting in traffic is worth tens of billions to the city’s Gross Domestic Product. But the problem is that the "cost" is concentrated in a single government budget, while the "benefit" is spread out across the entire economy. It is a hard sell for a politician who has to justify a twenty-year construction project that will not be finished until they are out of office.
Corn
That is the "political horizon" problem. But what about the technology of digging itself? We have seen companies like The Boring Company, started by Elon Musk, claiming they can reduce the cost of tunneling by a factor of ten. They want to use smaller tunnels and more automated machines. Is that a real solution, or is it just high-tech hype?
Herman
Well, the jury is still out. The main issue with the Boring Company’s approach so far is throughput. They are mostly moving individual cars or small pods. That is not a subway. A subway's power comes from mass. If you are only moving two thousand people an hour, you haven't solved the transit problem for a major city. However, their focus on making tunnel boring machines faster and more continuous—meaning they dig and install the tunnel segments at the same time—is a legitimate engineering goal. If we could make digging sixty percent cheaper, the economics of the underground would change overnight.
Corn
I also wonder about the "safety hardware" Daniel mentioned. Is there a point where we have over-engineered these systems? I am playing devil’s advocate here, but are we demanding a level of perfection that makes the "good" impossible?
Herman
It is a sensitive topic. No one wants to be the person who says "we should have fewer fire exits." But there is a legitimate argument that some of our modern codes are based on worst-case scenarios that are extremely unlikely, or that they do not take into account how modern materials are less flammable than they were fifty years ago. For example, in the United States, the distance required between emergency exits in a tunnel is much shorter than in many European countries. That means you have to dig more vertical shafts to the surface, which is one of the most expensive parts of the project. If you could double the distance between those shafts without significantly increasing the risk to passengers, you could save hundreds of millions of dollars.
Corn
It feels like a legacy of fear. We have these disasters in our history—like the King's Cross fire in London in nineteen eighty-seven—that fundamentally change the regulations. And those regulations never go away; they only accumulate. It is like the geological layers of the Old City we were talking about. Each new safety rule is a new layer that the next project has to dig through.
Herman
Exactly. And let us not forget the archaeological layer! Daniel mentioned the Thessaloniki Metro. One of the main reasons that project took decades was because every time they moved a shovelful of dirt, they found a Byzantine church or a Roman road. You cannot just steamroll over history in a city like that. You have to stop, bring in the archaeologists, and document everything. That is a massive cost that above-ground infrastructure mostly avoids.
Corn
So, looking at the "tipping point" idea... are we seeing a world where only the super-wealthy cities or the highly authoritarian ones can build subways? I mean, look at China. They have built thousands of miles of subway in the last twenty years. Is that just because they have cheaper labor and fewer legal hurdles?
Herman
Labor is part of it, but it is also "state capacity" and "scale." Because they are building so much, they have an entire industry that is incredibly efficient at it. They have standardized everything. They have a massive fleet of tunnel boring machines and a workforce that knows exactly what to do. In the West, we often treat every new subway line like a unique, artisanal project that we have forgotten how to do since the last time we tried it twenty years ago.
Corn
It is like we are relearning how to build a pyramid every generation. It is incredibly inefficient. But I want to pivot back to the "above-ground" alternative. If you are a city planner in twenty-twenty-six, and you are looking at the costs we are discussing, how do you justify going underground? Is there any scenario where it is not just a "luxury" for the elite?
Herman
It is justified when you look at the cost of the alternative. If you do not build the subway, what happens? You build more roads. You deal with more pollution, more accidents, and more lost time. If you value a human hour at, say, thirty dollars, and your subway saves a million people thirty minutes a day, that is fifteen million dollars a day in economic value. Over thirty years, that pays for a lot of tunnels. The problem is that we are very good at measuring the cost of a project and very bad at measuring the cost of doing nothing.
Corn
That is a profound point. The "cost of inaction" is invisible on a balance sheet. But let us talk about the human experience of these retrofitted systems. Daniel mentioned the "haunted" aspect of the London Underground. Beyond the paranormal, there is a psychological weight to being in a space that is clearly a patchwork of nineteen-hundreds brickwork and twenty-twenties digital screens.
Herman
It is a "palimpsest," Corn. A surface that has been written on, erased, and rewritten. You see these old tiles from the nineteen-twenties next to a high-definition advertising screen. It creates this sense of "technological vertigo." And for the workers, it is even more intense. Imagine being a technician trying to fix a modern fiber-optic signaling system that is laid over a copper wire network from the nineteen-fifties, all inside a tunnel that was hand-dug in eighteen-eighty.
Corn
That sounds like a nightmare. It is like trying to install a modern operating system on a computer made of clockwork. And yet, we have to do it because we cannot afford to let these systems fail. If the New York City subway stopped for a week, the city’s economy would literally collapse. It is a "too big to fail" piece of infrastructure.
Herman
Which brings us to the "tipping point" Daniel asked about. I think the tipping point is not that we will stop building underground, but that we will change how we build. We are seeing a move toward "automated" metros—trains without drivers—which allows for much higher frequency and smaller stations. If the trains can run every ninety seconds with perfect precision, the platforms do not need to be as long. Smaller platforms mean smaller stations, which means lower costs.
Corn
So the solution to the high cost of modern safety and hardware is... more technology? It is like we are doubling down on the complexity to solve the problems created by the complexity.
Herman
In a way, yes. But it is also about simplicity in design. The most successful modern systems, like the ones in Copenhagen or Singapore, are highly automated and very standardized. They do not try to make every station a cathedral. They treat it like a utility. I think that is the shift we need to see in the West. We need to stop treating transit as a "monumental" architectural project and start treating it as a "logistical" one.
Corn
I think that is a really key takeaway. If we want to keep our cities livable, we cannot give up on the underground. But we might have to give up on our "artisanal" approach to building it. We need to find that middle ground between the "no-safety" era of the eighteen-sixties and the "over-regulated" era of today.
Herman
And maybe we need to look at what Daniel’s prompt was really hinting at—the human cost. Whether it is a coal mine or a subway tunnel, being underground is inherently taxing on the human body and spirit. If we are going to ask people to spend an hour of their day in a tube under the earth, we owe it to them to make it safe, efficient, and, dare I say, a little bit pleasant.
Corn
Even if it is "haunted" by the ghosts of nineteenth-century steam engines.
Herman
Especially then! It adds character. But seriously, the economics of these systems are the economics of the future of the city. If we can't figure out how to move people efficiently beneath the surface, our surface cities are going to become increasingly clogged and unworkable.
Corn
It is a fascinating challenge. I am really glad Daniel sent this in. It is one of those things you use every day—or at least you see in the news every day—but you rarely stop to think about the incredible layers of math, law, and engineering that are keeping those tunnels from collapsing.
Herman
And the incredible amount of money. Never forget the money. It is the fuel that moves the dirt.
Corn
Well, on that note, I think we have dug deep enough into this one for today. If you are listening to this and you have your own "weird prompt" about infrastructure, or history, or anything else that keeps you up at night, we would love to hear from you. You can find us at myweirdprompts.com. There is a contact form right there on the site.
Herman
And if you have been enjoying our deep dives—whether it is coal mines, missile defense systems like we discussed in episode eight hundred eighty-nine, or today's look at the underground—we would really appreciate it if you could leave us a review on your podcast app or on Spotify. It genuinely helps other curious people find the show.
Corn
Yeah, it makes a huge difference. We are available on Spotify and pretty much everywhere else you get your podcasts. Just search for My Weird Prompts.
Herman
Thanks to Daniel for the prompt, and thanks to all of you for listening. It is a wild world out there, both above and below the ground.
Corn
Stay curious, everyone. This has been My Weird Prompts. We will see you next week.
Herman
Until then, watch your step on the platform.
Corn
Nice one, Herman. Bye, everyone.
Herman
Goodbye!
Corn
You know, I was just thinking about that Thessaloniki project again. Did you see that they found a headless statue of Aphrodite while they were digging?
Herman
I did! Can you imagine being the project manager and having to call the mayor to say, "Hey, we found another goddess, add six months to the timeline"?
Corn
It is a tough job, but someone has to do it.
Herman
Better them than me. I prefer reading the papers about it from the comfort of our living room.
Corn
Exactly. Alright, let's go get some coffee. I think I've had enough of the "underground" for one morning.
Herman
Lead the way, Corn.
Corn
Thanks for listening to My Weird Prompts. Check out the website for the RSS feed and more. We'll be back soon.
Herman
See ya!

This episode was generated with AI assistance. Hosts Herman and Corn are AI personalities.