Charter Cities Podcast Episode 28: Cost Differences in Railway Infrastructure Projects Globally with Alon Levy
A conversation is about the influence of politics and culture on the cost of building rail-based infrastructure projects across the globe.
Key Points From This Episode:
• How Alon got interested in infrastructure while riding NYC trains.
• Definitions of different rail-based transport types to be found in cities.
• Cost differences of constructing different rail-based transport across the globe.
• Cultural elitism and why there is a cost premium on American rapid transit.
• The high cost of rail transport construction in countries whose planning logistics happen in English and are inherited from America.
• A deterioration in building practices leading to higher construction costs in America.
• The spread of a design-build plague in America and from America outward.
• The role of perceived externality and NIMBYism in producing cost overrun and surplus extraction.
• Hollowing out of the public sector and the overabundance of informal pressures in the private sector.
• The issue of environmental protection laws being enforced by lawsuits rather than internal bureaucracy.
• The evolution of high-cost building techniques in New York.
• Perspectives on effective bureaucracy and the politics of railway building practices in Italy, France, Germany, and Spain.
• How the cost of mega infrastructure projects will evolve considering China and America’s influences.
• The role of isomorphic mimicry and cultural abnegation in inheriting poor building practices.
• Different cultural practices around how close to the city to put that stop’s station.
• Why optimizing for security instead of transportation effectiveness is paranoid.
• Problems with Biden’s infrastructure plan including the budget for State of Good Repair.
• The hallucination that the Anglosphere is the best; American tendencies to point out imagined problems in other cultures as an argument against adopting cheaper methods.
• How Lagos or low-income countries should approach building a subway.
• Perspectives on the internet and outside voices influencing on-the-ground challenges.
Mark: Hello and welcome to the Charter Cities Podcast. I’m your host, Mark Lutter, the Founder and Executive Director of the Charter Cities Institute. On the Charter Cities Podcast, we illuminate the various aspects of building a charter city, from governance to urban planning, politics to finance. We hope listeners to The Charter Cities Podcast will come away with a deep understanding of charter cities, as well as the steps necessary to build them.
You can subscribe and learn more about charter cities at chartercitiesinstitute.org, follow us on social media, @cci.city on Twitter and Charter Cities Institute on Facebook. Thank you for listening.
Our guest today is Alon Levy, a fellow with the NYU Marron Institute. Their research focuses on public transportation and how to apply best practices from cities around the world.
Mark: Welcome to the show, Alon. So, to start, talk about how you got interested in infrastructure?
Alon: I lived in New York. I was in grad school between ’06 and ’11. When I moved to New York, I mean, it was interested in politics in general. I mean, growing up in Israel can't be not interested in politics. But I didn't necessarily care about infrastructure more than like literally any other issue. In New York, the subway is omnipresent in culture, let's say so for example, or in socialization. I mean, if I'm in Morningside Heights, and everyone I socialized with is in Brooklyn, and it's a four-seat ride, it’s an interesting comment about the subway, right?
Mark: What is what is a four-seat ride mean? I don't even know what that means.
Alon: Oh, so a one seat ride is when you can get from –
Mark: Just one train. So, seat is basically a number of transfers. So, four seats, you have to do three intermediate transfers to get there.
Alon: That is correct. So, New York is not very good at two-seat rides. There are a lot of cities that make sure that you can get from most stations, I mean, all stations to all station with just one transfer. New York is very bad. There are a lot of three-seat rides and sometimes you can even snag a four-seat ride. Thinking about it forced me to think a lot about the subway. It seems like a cool example of public works, cool example of how to just expand non-car access, and then I stumbled across the cost differences.
Mark: So, I think that's, I guess, probably where I know you maybe many of our listeners know you from is this looking at what the cost differences are between mostly subways, but guided major infrastructure in OECD countries. What are those cost differences? How did you find them? Why had nobody really been talking about them before you came on?
Alon: Okay, so one of the cost differences is there's a systematic difference in construction costs for subways. There's also a difference for light rail, which is smaller but still serious.
Mark: How do you define light rail versus a subway? What's the technical difference in definition?
Alon: So, the American definition is just whether you're on vehicles that could run in streetcar mode or not. So, there are American lines that are called light rails that are functionally Rapid Transit like the Green Line in LA. They're not subways, basically, never underground. So, Rapid Transit is designed around longer trains that are supposed to be always absolutely separated from traffic. So, this means that you have zero lane sharing. You could have grade crossings, it's uncommon, but this does happen. The Chicago L has some. The New York City Subway had some into the ‘60s, but whenever you have grade crossings, they have to be separated by railroad grade so the train has absolute priority.
Light rail, before defining light rail, I'm going to find a streetcar or a tramway. A tramway is a train certainly shorter, it runs in the street, it could usually get dedicated lanes, but it could sometimes run in mixed traffic and for example, European tramways. They try to give the new ones dedicated lanes when possible, but sometimes they have to compromise and not do that on some segments. They cross streets like normal traffic. So, this means that they don't generally have railroad crossings, they never have absolute priority crossings. This leads to lower speed. However, because you're running on the surface, the construction costs are lower. So that is the tradeoff.
American light rail is this kind of hybrid where usually you're on as a tramway in the center of the city and then you usually upgrade grade but with railroad crossings, further out. Usually you're going to use a disused rail right of way like in for example, Portland, and then there are the weird ones like you have lines that run light rail vehicles but are functionally rapid transit. For example, the Green Line in LA, you have lines that are kind of fast in the center and slow outside city center. They're called subway surface line, like Boston, as is the Green Line, Philadelphia, where it's literally called Subway Surface Lines. San Francisco where it's called Muni Metro. It's very common in Germany.
So, this is the difference, the subway part is what I studied the most, and the reason is that these are usually mega projects. It costs a lot to build a subway. The result is that these are going to be publicly debated. So, if I want to know how much it costs to build, for example, Metro Line 14 in Paris, I can put, for example, Paris Metro Line 14 cost. I mean, I think if you put in Google now, you will get to me, but 10 years ago, you didn't. Or you can put this in France, and you will get two media articles that will tell you how much it cost because it's widely reported, because the line which cost at the time may be a billion euros, and I mean, there's an inflation since. I mean, a billion euros is a lot of money that will be posted in mass media.
So, this is why I did subway. I mean, you can also do this for light rail, it's just much harder searched or more light rail lines being built. Sometimes they're not going to be reported except in trade media. Sometimes they're going to be bundled with extras that are needed to winner out the extras to think that you can do and we're probably going to eventually extend to that. But right now, we're doing rapid transit, not just subway, they could be elevated. But that's our database, and you see a large American cost premium, especially in New York cost premium.
Mark: Why does that cost premium exist?
Alon: So, I have a bunch of explanations, none of which is fully satisfying it. So, if you look at how subways are built in the United – so, in the United States, the only line that we did a really deep dive on is the Green Line Extension, which is light rail, it's in a trench. And the problems there are numerous but somewhat different from those of subways. So, a lot of it comes from people having technological ideas that are stuck a few generations ago. It's not exactly clear. I mean, we're trying to figure out staffing levels. We do know that there's a lot of overstaffing on the white collar side of construction, lots of supervisors that are just there to make sure everything's right, but they're not really needed.
Blue collar overstaffing is likely also the case. We have less of a smoking gun on that but it is something that is likely also the case. I mean, to put things in perspective, okay, the cheapest places in the world to build in nowadays build a subway for around $100 million per kilometer, Global Median is 250, basically everything that speaks English costs more than that. New York is two billion.
Mark: Before getting there, I guess more into kind of the why, maybe trying to get the pattern a little bit more clear. US has substantial overruns. Are there any like other generalizations can be made? Do Anglo countries also tend to have overruns? Is Northern Europe cheap? Is Southern Europe cheap? What's the pattern before we get more into kind of the granular specifics?
Alon: Sure. So, here's the pattern. The parts of the world that – so in the developed world, English is bad. But I want to make it clear that English has been only consistently bad for let's say 10, maybe 15 years.
Mark: And you say English, you mean like Britain English?
Alon: I mean, language.
Mark: Okay, language. So, this includes Singapore, this includes Hong Kong, places like that?
Mark: So, all places that operate in English tend to have cost overruns.
Alon: Yes. It’s cost overruns. It's high absolute cost. I mean, Barcelona Line Nine is actually below median cost, even with a factor of three, three and a half cost overrun, just because they thought they could build for Madrid costs, they couldn’t. But Madrid costs are so low that that line is three and a half is still below global median. So, everything that speaks English. But I mean, what do I mean by speak English? Everyone in the Netherlands speaks English. Netherlands is also expensive and everyone in Sweden speaks English and the costs in Sweden are very low.
Mark: Do you count Netherlands as an English country? Because the language within which they issued a contract, et cetera is all Dutch even though it has like 90% English penetration or like like fluency, whatever?
Alon: Yeah, that's correct. It's not how good you are at speaking English. I mean, for example, Italy. I mean, most people don't speak English, but all the planners do. Or France, it's the same thing. But they're not English. So, when I say English, I mostly mean flow of ideas and usually it's flow of bad ideas coming from Britain and the United States. And this is why I mean, stressing the English has been consistently bad, but only maybe in the last 15 years. So, Singapore right now has really high construction costs. But 20 years ago, they didn't. Canada same thing. Canada, even if you look at that – so the Canada Line, you can kind of understand why it was kind of cheaper, because they use the low cost, high impact construction method and then the merchant sued another and they had to pledge to use a more expensive construction methods for the next line in Vancouver.
But in Toronto where they didn't have that pattern, there's just a lot of learning of how to do things, and this learning of how to do things is bad. It seems like a design build contracts a lot of privatization of the state, a lot of over reliance on consultants, rather than in-house capacity. It's really easy to sell this slate of bad ideas to politicians, because what you're telling the politicians is you should have more power, your political appointees should have more power, the civil service should have less power. So, politicians love this and then things suck.
I think you see spreading from kind of its origins in the United States and the United Kingdom all over the Anglosphere in the last, let's say, 20 years.
Mark: Okay, this is really interesting. I think there's maybe several, like I guess, threads to untangle. So, one thread is just, I guess, to a certain extent, the cultural hegemony of the US where other English-speaking countries are presumably more highly influenced by US. I don't know norms. And so, the US has basically been exporting bad management practices. And this could be viewed as I don't know, maybe somewhat analogous to how kind of right lots of countries more countries are playing basketball today than they were 30 years ago, partially because of kind of US dominance in global entertainment. But this is specifically US dominance in, I don't know, understanding how to build subways, that we're now exporting a set of bad management practices.
Alon: Not just US, it’s all over the UK. So, a lot of this is dialogued between the US and the UK, in which I have a bad idea, you have a bad idea, we share them. So, now we have two bad ideas. I don't want to blame patriotism, because I do know that British construction costs rose from the ‘70s to the ‘90s. But I don't know what exactly happened in between because the Jubilee Line was ‘70s and the Jubilee Line Extension was ‘90s. I don't have anything big in between them.
In the United States, bear in mind in the United States, I mean, New York at least, the costs have been going back to the ‘30s. It's just that Americans will not learn from foreigners, maybe possibly they will acknowledge the existence of Canada from time to time, Britain. But I mean, go to an American and tell them, “Okay, you should figure out how to implement policy to become more like Germany.” This is Germany. Germany is a rich country. That will lead to like excuse mongering and they see Americans ask, “Okay, well, is there an American example that's good?” Well, 20 years ago, there was. There no longer is. There's just kind of flattening of bad practice throughout the United States.
Let’s take Seattle, for example. Seattle, even 10 years ago, was building subways not affordably, but only for a 50% premium. The next thing that they're planning to build is the Ballard, West Seattle Line. It's $650 million per kilometer and it's only 30 something percent underground. I mean, it's like a factor of four premium. LA, same thing. LA 20 years ago looked semi decent. It did no longer does. I know that for example, he brought in a consultant from New York, to help with construction in San Francisco, why would they bring a consultant from New York? New York is a failure. Well, they don't think of it that way. They think of it as, “Oh, New York is a big complex system, let's learn from them.”
So, it matters, I think that New York is the worst in the United States, because I see managers in New York, not even acknowledge the existence of things that happen every day in Boston. It's pretty awful. So, this is why English is bad. Bear in mind, France has a lot of cultural cringe toward the Anglosphere because of growth rates, and because the managerial classes are more empowered in Britain, the USA, and France, so they kind of like becoming more like America in that sense of they – because of problems that they had with Grand Paris Express, they decided to let go of France's traditional separation between the design the public planning and construction teams. And because they said, “Oh, it's a complex project, we need to maybe adopt some design build ideas.” And then they had another cost overrun as a result. When someone is big, it's really hard to say, “They suck, don't be like that.”
Mark: The US has basically begun adopting the hegemonic powers. We don't like listening to anybody, except maybe Canada and the UK occasionally. And then two, is because the biggest cities in the US are seeing as I guess, the highest status ones, people are copying from them, even though they have cost overruns. If we go back 30 years, so we're seeing a little bit at least in the kind of English-speaking world, a flattening of practices for building subways. If we go back 30 years, was there a wide variety of practices to subways?
Alon: Yeah, the design build plague is recent, but I mean, New York was not doing design build either. I mean, it's just that you need to have –
Mark: What is design build?
Alon: Design build is a system in which you award a contract to one private consultant to handle both the design and the construction of a project. This can work if you're doing something very commodity like, for example, a parking garage, a usual parking garage. It does not work for any complex project.
Mark: Why not?
Alon: Because the private consultant is not necessarily going to be under any kind of internal professional commitment to reducing costs. I mean, the headline is a bid. I mean, the bids are evaluated on cost. But if you say we can't do that, I mean, it's kind of there's more veto points to telling people now. So, if for example, some local neighborhood group says we are worried about impact during construction, it's harder for a private consultant to argue with them, and you're more at risk of not getting a contract again, because you are a political problem. It's much easier for the public sector to argue.
Mark: So, the incentive is for the private designer and builder to add on a bunch of costs, that might then minimize some neighborhood externalities. But they end up being much higher than if the public sector that could basically say, “Suck it up. You live here. So, you're going to have construction in front of your house for three months.” And that private sector is unable to do that.
Alon: I would not call it an externality. I would call it an imagined externality. A lot of it is merchant –
Alon: Yeah. So, a lot of is just surplus extraction and this is the worst when you have really high benefit projects like the Toronto electrification project. It's something that they should have done 70 years ago, but they didn't. So, at normal costs, we're talking about to figure benefit cost ratio, even with a little bit of costs problems, they thought it was going to be eight, so they added 100% contingency. And when you have 100% contingency, you are going to find grace expanded. Some people demanded noise walls for electrification. So, this is going to make the trains less noisy, but they demanded noise walls. So, they gave them a billion Canadian dollars in electric electrification and noise walls, anyway. Likewise, when they want to build some info stations, so in really cheap countries, you can build them for a single figure million, not necessarily high figure in Germany, which is a medium cross country. It's 10 million euros. Boston, which has weird things that they have to constantly outsource the design to consultants, because they think that hiring in the public sector is a sin. Then even in Boston, it's 20 to $25 million.
Toronto, because it just didn't care about cost control, because it has such a high benefit cost ratio, it's more than 100 million Canadian. So, even with the complete want and inattention, to cost control in Toronto, the benefit cost ratio is still three. So, the incentive is to say yes to everyone, rather than tell them to shove off. Once you lock in that you say yes to everyone, it's hard to say no, even for projects that have lower benefits. You see this, for example, with high speed railings in Northeast Carter that are incapable of saying no to NIMBYs, even about things that are only in their heads. They think that taking 20 houses in a suburb is going to ruin the suburb. No, it's not. It’s 40 houses. Houses are probably selected to be the closest to the freeway anyway.
So, a lot of it, again, it's an externality. It's a perceived externality to the point that it's just surplus extraction, and someone says, “I think this bad, give me money.” It's an enormously negative sum game.
Mark: You've also been touching on this point a few times, and maybe, I guess, to figure out how to articulate it a little bit more clearly, a lot of I guess the public sector has, to a certain extent been hollowed out where previously there would be a lot of engineers on staff that would be able to do some of this internally or provide much more detailed oversight of the projects and that has changed. Is that primarily in Anglo countries or English-speaking countries? Is that universal? How’s that kind of affected the cost overruns?
Alon: Okay. So, I don't know if it's universal, because I haven't studied every country. I know specifically has not happened in southern Europe. Italy has a very impact civil service, and very low costs. Scandinavia, same thing. German costs are higher. I don't know why. People here blame nimbyism. It's easier to sue in Germany. Again, over things that are mostly purely in NIMBYs heads or just purely bad faith. There's this Gigafactory that Elon Musk wants to build in the suburbs of Berlin. Every recognized environmental organization in Germany is allowed to see over basically every project that happens. So, all the environmental organizations in and around Berlin met and decided they were not going to sue because they like electric cars.
Incomes, a Bavarian based organization that is affiliated with the extreme right, they make up reasons, the real reason they don't like the Gigafactory is that it would employ Polish immigrants and that delayed the project. We delayed the project, but only about a year or year and a half. That's it. At this point, they lost a lawsuit, Musk won at this point, the Gigafactory is not opening, because Musk is not necessarily the best at the logistics, and there are things that he didn't anticipate would be linking some kind of access road or something.
So, the NIMBYs have some power. But it's often fear of NIMBYs and preemptive surrender. In Germany, as well as in the other world. I mean, in California, high speed rail, I mean, NIMBYs were incredibly powerful and had a lot of technical merit. Even in sort of technical American judge could understand like, I mean there were some smoking gun with the weighing of various factors changed to lead with a desired conclusion. And even that was only, I think, a three-year delay. It wasn't a real delay, they could keep designing throughout this. It's just three-year delay and certifying the environmental impact report.
But you see a lot of defensive design in the United States, where again, to the Northeast Carter, high speed rail, there was a lot of informal pressure on the consultants. I mean, on the junior consultants, by their senior managers to avoid touching Fairfield County, Connecticut. This is why that's kind of left as a magic asterisk. Whereas if they just said, “Screw therein”, sometimes they'll get to dictate infrastructure to the rest of the Northeast, and carve, right of way where necessary. I mean, yeah, people would sue and then they would lose. And then you could have high speed rail between New York and New Haven for a couple of billion dollars.
But instead of doing that, there were just informal pressure not to touch it. This is a lot easier to do in the private sector than in the public sector. Because in the public sector, there's still a lot of informal pressure. But it's easier to resist because you can't have informal things that you will not get this contract again.
Mark: So, how do you weigh these different influences on price differentials? For example, if I think about the environmentalism and NIMBYs that seems to be mostly kind of in the ‘70s, right, NEPA was passed in what ‘70. Now, obviously, over time, the kind of NEPA requirements have increased. If you look at the kind of hollowing out of the public sector, that's typically attributed to Thatcher and Reagan, in the ‘80s. You were mentioning that New York, the subway costs were beginning to kind of increase in a way that wasn't clear exactly what the benefit was in the ‘30s. So, how do you kind of weight these different input factors to the cost?
Alon: The issue is not that there's environmental protection laws, they have these here, too. The issue is that they're enforced by lawsuits rather than by an internal bureaucracy. So, in Italy, there's an Antiquities Protection Authority that will tell you, “No, you can't build a subway station. This might threaten Roman ruins. You have to do something more expensive.” And as a result, they're trying to build a metro station, and they're extending this line called Metro Line C into the center of Rome, a line where within the city of Rome, but in newer parts of it, let's say 18th, 19th, 20th century parts of it. The metro station might cost 14 million euros, and then closer in, we're talking maybe 100. Now, they're trying to build one at Piazza Venezia. So that's near the wedding cake, and near a bunch of ruins from the forum. And you think that one is 400, 500 million euros maybe. But that's specific to one location.
In New York, 19th century neighborhoods also build subways and kind of the same method for $700 million, instead of for the same techniques in Italy, would cost an order of magnitude less. So, a lot of it is when you have things that unfortunately lawsuit, because that's just less reliable the result and there's a lot of fear of that. The hollowing out of the public sector. I mean, yes, this is Reagan and Thatcher. But it's something that's a lot clearer in Britain in the US. In the US, it's not so much of the public sector was hollowed, it has never built up. The United States never built up the state capacity that Britain built and then dismantled, or that, for example, France built and is only starting to dismantle now or the Germany, same thing.
So, in the 1930s, it was mostly they just didn't care about costs, they didn't think it was important. The public sector was trying to drive the privately-operated subways into bankruptcy. So, they built even areas that were kind of questionable. They had a lot of overbuilding in the 30s as well, like underground flying junctions, but they were both got uncovered so they had to shave corners of buildings. So, they had to like destroy corner buildings, which were the more expensive ones, instead of trying to come up with cheaper methods afterward the way that Britain did. So, Britain, it’s the first subway line, so there’ll be the Metropolitan Line in London was very cheap. In today's money, I think $35 million per kilometer. The second one, the district line was a lot more expensive. I think maybe 90, which was a record. It was the 1860. It remained unbroken until New York in the 1930s. So, the reason was so expensive as they had to carve new right of way for cut-and-cover construction through areas that we're expensive even the like Kensington. And this is why they had to invent the boring, in the 1880s and 90s, to avoid doing that again and construction costs fell.
So, forget that deep boring, the 1950s. For example, I don't think it would save any money. But in New York, they had to carve a lot of right of way. We didn't say had, it's not because they had it, it’s because they got used to it. In the 1910s, New York carved the 1, 2, 3 trains through the village. That one was called the lower west side and was a great industrial neighborhood where you had the Triangle Shirtwaist Factory fire. And the ‘20s, early ‘30s, they carved the same thing on Sixth Avenue on the Marley's, currently the AC and E trains. But by then the village was already gentrifying, so it was just more expensive to do it. And then they did these corner shaves, and they had all these underground flying – they just didn't really care. That somehow turned into the default way to do things.
There was a lot of resentment that the system was not finished. Even today, see rail fans, post these fantasy maps that they were planning to build in 1929 or 1939 that were not completed because of either the depression or the war and say, “Oh, look back in the day, we used to dream.” And instead of saying, “No, back in the day, you used to waste money. Don't waste money now.” And because of that, you're starting to see plans that are kind of unquestioned, even though there are generations out of date, prioritization of lines that is generations out of date.
So, a lot of is just kind of decline breeds decline, and some actually decline New York City's a lot wealthier now than was in the ‘30s. But in relative terms, it absolutely has declined. New York State and forgetting in what exact year, but I think it was late 20s, or ‘30s. was by far the wealthiest in America. I think it was 65% higher per capita incomes than the rest of the country. It used to be, I think in New York. I believe it was New York, California, Massachusetts, Illinois, but New York and California were both either the two wealthiest are number one and three. So, for example, everyone thinks, “Oh, the central Valley is poor, was always poor because it was great throughout.” No, the Central Valley. I mean, yes, it was poor, but other parts of the country were poor.
Maybe California has the demographic growth sense, so California looks back to its days of greatness into let's say, ‘70s or something. But in New York, it's midcentury. Again, not because the city was better then, but because the United States was relatively speaking richer than the rest of the developed world. New York more so than the rest of the United States. The South was barely first world at the time. So, you have this kind of technological , let’s call it, we're in New York, there's widespread resistance to anything new. And this is where they kind of end up being stuck in these midcentury methods that are then exported, because everyone is saying, “New York has a subway, let's build like that.”
Mark: If I think of Italy, if I think of Southern Europe, I don't think of very, like effective governments effective bureaucracies. To me, so there’s this, interesting cultural question, they seem to have been relatively effective on this provision of large infrastructure projects, because of their internal capacity. So, is the general story of just ineffective bureaucracies in southern Italy wrong? Or What is it about the rail bureaucracies or infrastructure bureaucracies that make them comparatively effective while the rest of their bureaucracies don't function particularly well?
Alon: Okay, so I also reject the question of whether the rest of the bureaucracies don't function particularly well. I mean, yes, these countries are by European standards poor. If you look at, for example, Corona, they had higher death rates in Northern Europe. But it's entirely a matter of poverty. So, there's the stable that you can find on the internet for proportion of people aged 25 to 34, who live with their parents.
Now, this is partly cultural. So, in Scandinavia, the culture is you leave home at 18 and you'd never live with housemates. You live in a tiny apartment by yourself. So, the proportion there's I think, 5%. And then France and Germany, in the United States, in the United Kingdom, maybe that culture is not a thing. So, it's, let's say, in the teens, sometimes low teen, sometimes high teens. In southern Europe, we're talking 20%, 30%, 40%, just because they're poor, so they can't afford as much stuff. So, there's more .
But if you compare this with, for example, their ability to vaccinate, there's no obvious Southern European weakness compared with Northern Europe. Italy has this reputation for corruption, which is absolutely a thing. But in the same way that the United States has a reputation for racism, which is absolutely a thing and you can even see it still today with police brutality rates. The United States does not, however, have more job discrimination, for example against minorities than – let’s go with the European average. The European countries with less discrimination in the United States. I think Germany is one of them and there are ones with more like France. But overall, it's not like the US is materially worse on discrimination than Europe. Because the same history of racism is also why you do have strong civil rights laws, and it's the same with Italy and anti-corruption laws.
Yes, in the 1970s and ‘80s, it was awful and construction costs were high, precisely because of bribes. And then in the ‘90s, they had this process called money pulita. It actually has no connection to the word manipulation. I mean, the money is the same, it's hand but the pulita is a different word. It means clean hands were due to bribery and actually, it was an infrastructure project, it may have been the Milan Metro Line Three, maybe where they started investigating politicians and one politician was so obviously corrupt, the party kind of got them off, blabbered and gave so many names that eventually half of the Italian parliament was under indictment. All of the major political parties of Italy crashed and burned as a result. The current centralized party, the Democratic Party was formed from the carcasses of the few communists, who were moderate enough to actually be in politics, and more importantly, were not implicated in the corruption scandal.
And likewise, one of the carcasses of the center right, came Berlusconi, who was incredibly corrupt. So, as soon as he came to power, he shut these investigations down. But these investigations did succeed in forcing out a lot of the corruption and passing very strong anti-corruption laws and construction costs in Italy in the ‘80s, I think they were 300 something million dollars per kilometer, adjusted for inflation. So, higher than global. It generally costs half a secular tendency to rise. At this point, they're not 300, 350. They’re 150. Again, unless you're building underneath Roman ruins, and then it's more, but in Milan, it's 150, give or take. In parts of Rome that are not Roman, again, but maybe that.
So, it's absolutely possible to reduce costs. But the process, it'll take a while before people notice. So, in the same way that people still think that it really is as corrupt as it was 40 years ago, which is absolutely isn't. Kind of the only way people actually accept something is if Italy stops being poor. But Italy has different problems like tax evasion that are completely separate from the actual bribery. And so, people say, “Oh, well, if you think at Italy is so smart way, and get rich, I'm so smart. I do not need to listen to everyone, to anyone.” This is kind of the mentality and you see this also in Germany.
So, why does Germany not have high speed rail? I mean, we do have high speed rail lines that aren't especially well designed. But we don't have a national network. Why? Because high speed rail was invented in Japan and then was brought to Europe by France. Italy and Spain look up to France. Germany looks down on France. So, Italy and Spain built high speed rail networks connecting all their major cities. Spain, especially refined learning from France, from Germany, and picking the best age. Unfortunately, non-operating or Spanish operations still suck. But the construction is incredibly cheap. Germany thinks is better than France. So, it will never learn the French idea. Even when France doesn't think better, like in having actually fast trains connecting the major cities.
Mark: How should we think about the evolution of mega infrastructure projects, partially rail, partially other over the next kind of 20, 30 years? Because your model is basically one of what might be described as, I don't know, kind of cultural hegemony or cultural elitism, where there is this mechanism by which whatever sudden the high-status countries, I guess, because the feedback loop for these projects is so long, that whatever is done in the high-status countries, even if it's not effective, is broadly adopted. And so, if I think about the future, and the US is probably has less relative power in the world, but our cultural hegemony is arguably kind of peeking over last summer. We saw Black Lives Matter protests in places, some in Japan and New Zealand, et cetera, are we going to see the continued export of American ineffective cost mechanisms, particularly to countries that might not have English as their first language?
And then the other kind of, I guess, main consideration is, China is obviously becoming a global player. I'm not sure how their status is perceived. So, whether people are beginning to adopt their construction mechanisms is happening a little bit in low-income countries, but I don't know to what extent it might start happening in high-income countries. So, is that kind of the right framing? Or how should we expect the evolution of cost of mega infrastructure projects, particularly rails to evolve?
Alon: It's absolutely the right frame. So rich countries are not going to learn from China. It's still too poor. I mean, yes, you have people who wish that they had the tight control of the workers that they have in China. I mean, you do sometimes see really rich tech managers,
kind of wish they were Chinese tech managers and you never see this kind of, I wish we were like that toward countries that are actually richer or even stronger intact. I don't think I've ever seen, I wish we were Taiwan text about this. People discovered this with Corona and then never learned a single thing Taiwan did. In January, Taiwan had an outbreak of Corona at a hospital, 15 people got Corona. In order to prevent this from spreading, they put 4,000 people in quarantine, 6 tested positive. As a result, the corona outbreak did not spread beyond these and Taiwan is fully reopened because there's no Corona.
In all Western conversations, when I say Western, I mean the US and Europe, Australia, New Zealand are close enough that they understand. The US and Europe still do not talk about centralized quarantine. So, nobody learns from smaller countries. Now, again, China is big, why China is also not rich, but it absolutely exercises this kind of soft power and developing countries, especially ones that are not very democratic, because they also wish that they could control the opposition the way China does. Absolutely, you see people outsourcing their construction to China and it's leading to all of the same problems that you saw with Western aid or Soviet aid in the Cold War, for example. So, I imagine that you know what isomorphic mimicry is?
Alon: It's awful. Again, it’s in poor countries. One of the patterns that we found actually, is that in countries that are not first world, ex colonies tend to have higher costs. So, China has medium costs. China is not a low-cost country, it's low cost compared to the United States, because compared with United States, everything is low cost. But China's the same as the non-Chinese global median, as a non-Chinese, because about half of global subways in the last 20 years has been built there. But China is basically the same as the rest of the world median. So, same as France, for example, maybe a bit better than Germany, a lot worse than Korea, or southern Europe, or Scandinavia, or Turkey. I mean, Turkey is actually a good example. It was never colonized. It learns, but from what people talk about, “Oh, it's a cultural abnegation with capitalism.” No, Turkey is a lot less culturally self-obligating than the sort of countries that try to send all gay people to jail because they are proud of their values or something and say we're against Western values. I mean, these countries are desperate for Western approval. They're trying to imitate Turkey values and not external values, but it's the same thing and it's imitation of other values.
In Iran. Likewise, construction costs are about medium. Latin America has a range, Brazil is more expensive. I think Ecuador and Chile are the cheap ones. In Ecuador, I think, also use Spanish consultants for the metro. So, they happen to learn from the correct people. But Latin America is not very poor. I mean, some countries are. I don't think they build any subways. But let's say Chile is an upper middle-income country. And then you look at ex colonies, and it's awful. India pretty bad. Bangladesh is awful. Indonesia is awful. Vietnam is awful. The Philippines are awful. Egypt is awful. Nigeria, the construction costs of the metro, of the elevated metro they’re trying to build are bad and also the way they're building it as bad. It's basically coming from outsourcing design, often to either China or Japan, depending on geopolitics. So, they use techniques that maybe are appropriate for countries with seven times their income or eight times with predictable results.
I mean, so Addis Ababa, the construction cost of Addis Ababa we're not awful. They're okay. They're not awful. But for example, they cut corners by just closing cross streets instead of building through them and having railroad crossing gates, because maybe they were worried that people would just step onto the train. They didn't really build elevated grade separations or anything, but more importantly, Chinese consultants built this for them. In China, you have reliable electricity, and if you pay, you don't. So, whenever there's a power outage, the train stop working.
Now, early 20th century America did not have reliable electricity. So, you had streetcar companies, and you had railroads were electrifying and they made sure to get their own power plants. So, it was actually really common until the New Deal banned that as an antitrust measure. It was really common for the power company in an American city to also run the streetcars because they use the same infrastructure of wires more or less. So, there weren't going to be power outages. I mean, there might be, but the streetcar always gets first dibs. So, this is how they had reliable operations under electricity. And because it's no longer relevant to irate or even middle-income country, a lot of low-income countries that are doing this isomorphic mimicry, fail. Another thing, China really likes putting the train stations out on the moon. It's even worse in France. In France –
Mark: Out where?
Alon: On the moon.
Mark: So, what does that mean? Part of the suburbs or what?
Alon: Yeah, exactly. So, whenever you build a high-speed railway, you sometimes have to compromise and build a station that's not in city center because you can't reach out. Even Osaka is like that. Osaka is a few kilometers outside city center, but a few kilometers on a subway line, I mean, at this point to an important secondary center of Osaka. And different countries make different levels of effort. The Japanese, the German ways, you go to city center whenever you can. The French Way is major cities is going to say center stop, but anything in between can get a station, if you're really lucky is going to be tangent with a built-up area like or maybe Avignon. If you're less lucky, it's going to be 20 kilometers from the city or 30 or more. There's this station that was called a Beetroot Station because it was right in the middle between Amir and Saint-Quentin. Because it was served one or the other and they decided to stay right next to the fairway to get to a little faster. This are just bunch of places in France.
In China, the innovation is they do that, even in major cities. So, in Shanghai, they have a historic rail station that's just outside city center. I think two subway stops from People's Square. But the high-speed trains don’t go there. They built an new station far in the West, domestic airport called and they're kind of exporting that specific model of like big stations on the moon with the security control, access control. Here any boss that you can walk onto a platform to African countries where they build their railways. I know they did it in Ethiopia, and I think also Nigeria, and it just doesn't work. Because then I mean, you're leaving people several hundreds of kilometers outside city center and maybe in Shanghai, it can vaguely work because you have more growth, more housing growth. So, maybe you can turn that into a secondary center. I don't know. You also have subway lines that connect you onward. But this is –
Mark: Presumably they're also optimizing for security and not just optimizing for transportation effectiveness.
Alon: Okay, so the TGV is starting to get a little bit of access control for decoding mostly because SNCF is run by airline managers who don't know how trains work. But this is last three or last four years. At the time I left Paris, which was a little over two years ago, God you know, so the busiest station, since the second busiest TGV station still had no access control, except if you're going to London because of border control. So, the only time terrorists have ever bombed a high-speed train was in France. This was in the ‘80s, with the TGV was new, it was going on . So, Carlos the Jackal put a bomb on the train. I never remember how many people died. I think two or three more people died from a bomb placed in the locker room somewhere at the station. The two or three deaths that was viewed as a major embarrassment to the cause of international terrorism. So, terrorists stopped putting bombs on high speed trains afterward and there was never any kind of access control.
In Germany, same thing. Germany had Baader-Meinhof, the trains are not access control, the subways, I mean, don't even have fare gates. So, when you make a decision that you're going to be paranoid and do bag checks or something or make people walk through X ray machines to get into a subway station like they do in Bangkok. That's optimizing for security. That's optimizing for bullshit. I don't think that security controls anywhere in Japan on this, and they had the gas attack. They do have access control for the Shinkansen, because of some weird things with ticketing. They’ve had fewer conductors on the Shinkansen trains than they do on European trains. That's very different from doing security theater, which they do in China.
Again, it's not security. I mean, I'm sure that China thinks that every Uyghurs about to launch a 9/11 at them, but in objective reality, no.
Mark: Biden has released his infrastructure plan. You wrote a good piece about that. So, I guess, high level, is Biden's infrastructure plan good? What are the parts that could be improved? What are the parts that he's kind of focusing on for the right things?
Alon: Okay, so it's hard for me to say good or bad because I can tell you a lot about construction costs. I can't put it in legislative language other than things like, “Oh, the federal government should demand cost control. The federal government should start saying no to agencies with a bad track record.” So, the problem is because the United States is uniformly bad. I mean, you’re only in a position that you could have been in the Obama era where the FDA could have said we're not giving a cent to New York, but we are going to fund things in LA because at the time LA was cheaper or Seattle, same thing. So, first of all, they need to care more about comfortable. But again, I don't know how to put it in a legislative language. I will say that, I mean, I can find bad smoking guns in the Biden infrastructure plan. The biggest is that of the $85 billion given to modernizing transit, 55 are going to be wasted on something called State of Good Repair.
State of Good Repair is a fascinating example of how you need to constantly oversee the government using media and popular pressure. But at macro level, not micro level of people saying I don't want the tree in front of my house to go out. You want something more serious of that. The most serious oversight is, when you do an expansion, people can tell that the expansion is opening or not, right? People can tell the Second Avenue subway opened, people can tell the the Seventh Extension opened. And these actually exercise is an important disciplining function. Whereas when you do State of Good Repair, it's like saying you will create or preserve affordable housing. This is a common formulation in New York. Nobody says I will build x housing units, because that is a concrete promise that people can see your break when you say create your – you can say, “Well, the experts fade who know that if they don't reach the desired conclusion, they will never work for me again, said that if we hadn't been in power, then there would have been this many more units that would have been vacancy be controlled or something.”
It's the same with Obama in 2010, not saying a reduced unemployment, which he didn't, but saying that if he hadn't been around, unemployment would have been even higher. So that's a problem with State of Good Repair. It's not a concrete promise that people can verify. So, people say, “Oh, well, our maintenance will improve.” No, It won't. They're not going to say, “We're going to improve the state of the tracks.” And as a result, you will see the trains run 20% faster. I mean, this is what Andy Byford tried to do and you saw how Cuomo treated him.
So, State of Good Repair. It boils down to something that actually was done well in the ‘80s and ‘90s in New York. But essentially, it came out of actionable items like reducing breakdowns, reducing 10 miles an hour speed limits, not like reducing the speed limits, reducing the proportion of the system had these limits. So, people visibly saw the system get cleaner, more reliable, faster, and this is why it worked. And subsequently, it’s just been a black hole of money that goes nowhere. So, the most important actionable thing that I can say, is take the 55 billion for State of Good Repair, turn that into zero and spend all of the 55 on the expansion bucket, which is unfortunately only 30. Five will go to ADA, that’s good, 25 to everything else.
So, instead of making that 30 expansion, 55 State of Good Repair, make it 85 expansion, make it very clear. Agencies that has a track record of under maintenance shall not receive federal funds. They're also the Amtrak record, which is kind of terrible. But that's because Amtrak is kind of terrible. But I don't think the American government is ready for the conversation about wiping out Amtrak and replacing it with an agency run by people who know what electronics before concrete means.
Mark: I think that’s most government agencies.
Alon: Right, but it's actually important in rail. And because the United States is generations behind the global frontier, the usual American way of just looking to Americans to innovate, will suck, because it's not like anyone else in America knows how Swiss Railways work. I mean, electronics before concrete is a first world observation, it does not apply in really poor countries. It's the observation that it's much cheaper to invest in more reliable equipment and systems, than to build things that they call concrete. So, that would be more tracks, for example, or grade separations. Switzerland, also, Japan has ridiculously under built rail networks. You have these lines, where in America, they say, “Oh, we need four tracks there, or we need three tracks there.” In Switzerland, they could have two tracks, maybe even sections of single track. They just schedule the trains so that they don't conflict. And they build the infrastructure that they need based on the timetable.
Amtrak has no idea how to do that. I mean, you tell them about the periodic timetables and they look at you like you're an alien, and then they rattle off excuses, many of which are just fraudulent. I've heard many years ago an excuse from the FRA, which subsequently thankfully did realign its regulations with the European norms. But before they did, Steven Smith, market urbanism was asking them and they said, “Well, America has a lot of grade crossings, and American trucks are heavier than European trucks.” So, the gross load of an American truck is the gross load limit of the US Highway network is 40 short tons. This is 36 metric tons. In Europe, it's mostly 14 metric tons and there are countries that where it's higher, but in Sweden, 60 metric tons, and they're very happy with their trains. The trains don't actually crash and kill many people. People who actually compare those things will notice that in the US, they tried to reduce the weight of a track, not a track, a truck.
In Europe, they instead tried to reduce its footprint to make it narrower to fit and to narrow historic streets. But the trucks are a little bit heavier. But the funky from the ferry was tasked with making up excuses, I just thought it would be heavier in America and that nobody would check them. And again, this is an agency that actually really –
Mark: Or they didn't know the difference between metric and English tons.
Alon: No, even if they didn’t, it would be 40 versus 40. I mean, 40 versus 40 is the same number. So, no. It's just they didn't know, they don't care to know. I mean, everyone else around them doesn't know either. So, they get away with saying these things. And again, the FRA, did realign regulations with global norms. But the actual operating agencies still don't touch that. So, New Jersey Transit is still buying trains compliant with obsolete regulations at factor of two premium cost. In fact, it literally is just a scan in a Buy America article. So, I had to look into the premiums of a lot of American orders. Buy America does have a premium, it's not a factor of two premium, it's a factor of 1.5, for example, premium, maybe 1.4. And that order was I think, 1.7, maybe even more.
So, we got this obsolete, low performance, premium cost range. Amtrak, same thing with the new trains that they're getting for the Northeast Carter. I don't even know what they're compliant with, but partly because it's Buy America, partly because they decided that they don't really need. Partly because Senator Schumer sabotaged the bid to be able to brag about job creation in upstate New York. It's a factor of two premium. But the point is, nobody in America respects European expertise on this even when Europe isn't the global frontier of this and the United States isn't. It's something you need to persuade every person separately and say, “Well, the other people in my agency won't believe me.” So, there's kind of a collective hallucination and the Anglosphere in general thinks it's doing well, to the point that there are people who end up being experts in – do you know that the tendency of people to be experts in other people's social problems. Like how Israelis when you criticize their human rights record, suddenly turning to experts into every developed country developed democracy’s, human rights problems?
Alon: It's the same with Americans who suddenly turn into experts in their own eyes, in European political problems, just so that Europeans stop telling them that European trains work better, or European healthcare or whatever.
Mark: Yeah, we've talked a little bit about low-income countries and cities in low-income countries and how they can provide public transit. So, how does that differ? The point you made was that oftentimes they're kind of benchmarking on using contractors coming from countries that have incomes, or magnitude or 5x, higher. So, how do you actually adjust the kind of provision of these subways or other large transportation networks in countries that have incomes at $2,000 per capita, or $3,000 per capita?
Alon: PPP. It's all PPP adjusted. It's not income adjusted. If it were income adjusted before the cost of this country in the world Bangladesh. Bangladesh, because even with just a PPP adjustment, I think might be almost as bad as American. I know that people think that it's about the soil there, because alluvial soil does increase costs. You see this in the Netherlands, in Shanghai. These Bangladeshi projects are elevated, or mostly elevated. One of them is in fact entirely elevated and they're still really expensive.
I mean, I guess, if you let Japan build your system, then the poorer you are, the less appropriate it's going to be. Overall, by the way, the GDP per capita to cost correlation, I believe it's positive, but it's something like 0.04, which is not statistically significant.
Mark: Well, let's say you're in a country that has a purchasing power just at income of $5,000. And you need to build –
Alon: Let's say, Nigeria.
Mark: Okay, sure. You want to build a metro for the city, what what should you do?
Alon: Concretely, let's talk about Lagos, and the reason I'm going to talk about Lagos is because actually poked around there more than elsewhere. In Lagos, first of all, who do you learn from? The answer is, you don't invite foreign consultants. That's the first thing. The second thing, what do you learn from is you read whatever you can about the history of early subways in the United States, France and Germany. The reason I'm not saying Britain, is that Britain built using deep boring, which is not what Nigeria should be doing. In all other ways, Britain is also really useful to learn from, which means you learn how to use the labor-intensive methods.
Nigeria does not have expensive labor. But whenever you import capital that is expensive. So, this means you build, so you use for example the fact that the Nigerian roads and the main urban roads are very wide. You build either elevated or cut-and-cover. I'm guessing that in Lagos, they wanted to do mostly elevated because I have no idea what the geology is. I mean, the people in charge do know. But generally, when you're in an alluvial floodplain, you want to be elevated because you're going to have more cable problems. In, for example, Nairobi, I'm guessing that it's probably better to do cut-and-cover subways because the trains are different. So, you do that. You build four track lines, not two track lines, like in New York. Remember, Lagos today, is almost the same size of New York, metropolitan area to metropolitan area, and let's discuss urbanization rates in Nigeria and natural growth rates. I mean, Lagos is obviously going to become a lot larger very soon.
So, you build four track lines, local and express, like in New York. You absolutely can do some leapfrogging so you maybe want to build the lines around higher speed standards than New York did. But you want to use labor intensive methods for cut-and-cover or for elevated. You absolutely can just build concrete viaduct. Remember that most of the people in the country do not own cars, even in the city. Forgetting what Lagos is car ownership is. I want to say it's 250 per 1000 people, so same as New York City, but most of these are motorcycles, but don't quote me on that.
So, most people don't own cars. Yes, the richer people own cars. So, what? This means we for example, should be liberally taking lanes from cars, if the street is very wide, which it is. Sometimes concrete viaducts just have piles in the middle of the road. So, you can absolutely take two lanes from that. Yes, people will complain again. So, what? Build a four-track system that connects all of the major parts of Lagos to Lagos Island. In Lagos Island, you might need to bite the bullet and build things underground. But Lagos Island is small, so even with a cost premium, like in the Netherlands, it's going to be okay.
Know what system you're going to build an advance. Don't build one – I mean, it's okay to build one line at a time but know what line 12 is going to be. Because you want to make sure that all the lines intersect correctly in Lagos Island. Nairobi, same thing, except it's not going to be Lagos. It's not called Lagos Island in Nairobi; the central business district is just the Nairobi central business district like Hilton Park and such. That's what you want to do. You want to learn from the history of rich countries back when they had roughly your income level, because that is roughly the labor to capital cost ratio that you're looking for. Don't bring China and even China’s still rich. Definitely don't don't bring Japan in. Do things yourself. Yes, countries like this are incredibly self-rating incredibly self-advocating. Even India, which is not 5,000. India is, I think, pre-Corona, I think it was scratching 10,000. But crisis hitting really hard.
Even in India, I mean, basically every article you will read in Indian media about Indian infrastructure is going to talk about problems, even in cases where India is doing really well, like the electrification project. They keep talking about problems, because India keeps thinking of itself as inferior. And so, this is why you see a lot of isomorphic mimicry. Hell, you see this in southern Europe. The differences Southern Europe is similar enough to Northern Europe that isomorphic mimicry in Italy and Spain of let's say, Germany, and France actually works really well, because the labor to capital cost ratios are very similar. So, you just learn the things that France does. You will learn things that Germany does well. You synthesize, and you get Spain. Or I won’t say you get Italy, but in Italy, it's only starting to be that way. In Italy, they're starting to adopt German ideas of operations, which work really well.
But this is what you could do. If you're poor, rich country like Italy, you can't do it if you're poor, if you're actually poor country like India. So, a lot of is just understanding that you cannot privatize the public sector into prosperity. You just can't.
Mark: Yeah, so we've got a few minutes left to end. You kind of have a somewhat unusual career path and that you've sort of become a I don't know mini internet intellectual that is now influencing –
Alon: I hate that term.
Mark: Okay, well, you can give me a better term. But I find that –
Alon: I'm a researcher. I have an appointment with Marron. I don't exist just on the internet. Most of my income doesn't come from the internet.
Mark: Sure. That's true. But it feels as though you got your start on the internet.
Mark: And that you would not have that appointment with Marron, you would not have all of this. The phenomenon that I consider myself somewhat similarly in that, have started with engaging these internet communities with a different perspective, and now trying to figure out how to take this perspective and influence some of the kind of real world, on the ground challenges. I see this as kind of a broader, I don't know, class of phenomenon where there are these things that haven't really had this outside perspective before. So, I just wanted to, I guess for the last question, ask your thoughts on how you see the evolution of this with obviously yourself, but then with this kind of general phenomenon playing out as well?
Alon: So, I don't know if you've heard me talk about this, but a lot of this comes from Ezra Klein and The Health of Nations. The 2005 ratio was, I think, maybe two pages on the healthcare system of each of the seven countries. I believe, it as the US, Germany, France, Britain, Canada, Japan, which I guess is six countries. I’m forgetting if there was a seventh. We’ve kind of brought in to the forefront, in the United States healthcare isn’t just inegalitarian it’s also less efficient than in universal health care developed countries. So, they succeed, and as a result, this is kind of what led me. I mean, it was thinking in comparative terms, the fact that I mean, when you're Israeli. Again, Israel as a poor, rich country, constantly compares itself with other countries.
But I saw that the Americans actually cared to some extent. So, this way started like pushing this. But I mean, again, there’s something for everything. I mean, the costs of K 12, for example, there is an American premium, it's not a very large one. For example, teachers in Germany actually got paid better than in America. So, the US public spending is always a waste point is not always there. It just happens to be true in public transit, construction and in the operations of mainline rail. Subway operations in New York is very wasteful, but for example, Chicago isn't. So, where's this going? I don't know. I mean, in the Obama era, was yelling into the ether. And I mean, things that they were doing in the stimulus are pretty horrible. Now, I don't know if there's an improvement because, again, I don't think of it in terms of legislative language. I think of it in terms of agency competence, but again, I don't know.
Mark: Great. Well, that’s the end. Thanks for coming out.
Alon: Yeah, thanks for inviting me.
Mark: Thank you for listening to The Charter Cities Podcast. For more information about this episode and our guest, to subscribe to the show, or to connect with the Charter Cities Institute, please visit chartercitiesinstitute.org. Follow us on social media, @cci.city on Twitter and Charter Cities Institute on Facebook. I’m your host, Mark Lutter and thank you for listening to The Charter Cities Podcast.