Happy New Year! If you’ve not already done so, please consider subscribing to this blog for more content directly delivered to your inbox.
Something that I have promoted in a number of videos and blog posts over the years is what I have called the “many-lines model” for building metro systems — modelled on Madrid.
The idea here is that instead of building any given line to a capacity X, it may be better to build 2 lines to capacity X/2; more and lower-capacity lines should allow for the same capacity, and it means that a disruption on any line is going to have a smaller impact on overall network capacity.
More lines also means putting rapid transit closer to more people. While buses can be great, a metro provides a level of service that is difficult (and arguably impossible) to provide in an economically viable way with buses.
There is also a construction capacity benefit to this approach. When you build a lot more, you have more practice and experience, which lets you keep building, which is important as cities continue to grow.
But when I look at many new(er) metro systems, the many lines model is rarely pursued. Sure, there are cities that are building a LOT, and perhaps you could argue that a city like say Shanghai is doing many lines at a large scale (few lines have >6 car trains), but Shanghai does not have the low costs of Madrid.
Lean Metros
Instead, the pattern I see emerging is in line with the broader ones you have seen as the world’s metro construction capital has shifted from Europe to Asia and the global south.
One thing I’ve always found interesting about the MTR in Hong Kong is that it isn’t very large: with less than 200 km of rail and around 100 stations, there are many significantly larger systems; however ,few match MTR on riders per kilometre. It should also be noted that Hong Kong’s system is highly interconnected with many interchanges. Hong Kong has a great system, but not really a big system. You could say this is because Hong Kong has very challenging geography, which is true — but the pattern of a “leaner” network with less track and stations can be seen in other places. Some examples that come to mind being Cairo (3 lines moving several million per day with less than 100 stations), and Sao Paulo, which also has less than 100 stations and moves millions every day.
And it’s not just cities in the global south: Copenhagen’s metro uses much smaller trains even when compared to most other European Metro systems and doesn’t have much mileage; it also has lots of interconnectivity. Another example on the microscale that comes to mind for me is Toulouse with Line C or Vancouver with the Canada line — both which are arguably “leaner” lines than the previous lines in their respective metro networks, with stubby wider trains meant to run frequently.
In fact, even looking at Toronto’s future subway network, it feels fairly “lean” — when you look at true rapid transit lines (that is lines 1, 2, 4, and the Ontario line) they will still have less than 100 stations from 2030 onwards despite probably moving around 2 million riders every day.
But, what is lean?
I think the way I look at it is that lean metros are ones that move more people on every bit of rail; calling them “maximum metros” or “high-density metros” might also make sense. In such systems, you have fewer lines with more frequency, and you move more people with the same given frequency because the trains tend to be bigger, with 6 large cars or more on main lines.
Lean metros are also more likely to have more connected networks, because connectivity enables higher “loading multipliers”, which let you move more people on less mileage.
And for what it’s worth, there’s probably a debate to be had about what’s more futureproofed between lean and many-line metros — because while many-line cities maintain institutional capacity to build, lean cities maintain space on parallel corridors to build. I don’t think I have a definitive answer on which is better, but getting space back is nigh impossible; learning to build again should not be.
Why are we seeing more lean metros?
I think the interesting question to ask is — why are we seeing more lean metro systems? Many “historic” metro networks like Paris, London, New York (which isn’t really lean), and Berlin seemed to build much more before growth slowed down.
I think the biggest reason is that, especially with modern standards with regard to disruption and construction standards, building really has become harder, while at the same time cities have become even denser on a micro scale. By building fewer stations with more intense density around them — Hong Kong stands out here — you can still put rail close to many people, but instead by focusing efforts on building more housing density rather than metro, which seems significantly simpler. This is something Alon Levy discusses in a blog post from back in 2019.
To circle back to my point about the trade off of space for institutional knowledge, I think it’s worth reiterating that if you build lots quickly, you are likely to build for lower-capacity per line and also expend you “easiest” corridors most quickly. The more infrastructure you already have, the more challenging building is going to be, especially in non-crazy geographical situations (i.e. Hong Kong).
I think that the idea of building more lines also probably makes less sense in a future where growth is less certain. While sure in countries like China and India there is a strong case for still building big, in many places (population growth in many countries is lower than we have projected, and while cities will continue to grow I think the days of a large number of cities around the world exploding with growth are probably almost gone) it probably makes sense to play it safe and heavily optimize a smaller amount of infrastructure and taking a wait-and-see approach. More lines is probably more resilient, until you’ve overbuilt and swamped yourself.
Of course the most obvious reason for lean is that it’s more cost effective, especially over the long run — you’re maintaining fewer, larger trains, but the economic value of each mile of track is also a lot higher, so it’s easier to justify long term maintenance of all of your infrastructure. Rebuilding the tunnels for lightly used metro lines in some cities in the distant future may well not pencil out if cost increases we see in the Anglosphere spread.
And I think the “lean” future is something we can already see emerging in places where the transit systems were not built lean from the start. For example, London and Paris upgrading signalling and trains on their oldest lines to try and boost capacity as much as conceivable instead of trying to build new city centre metro lines — it’s just easier in many cases to use operational and infrastructural slack to eke out more capacity than building something totally new under dense city centres.
To be clear, I’m not saying lean metros are better than many-line metros, but I do think they are becoming more common, and I believe as density increases in our cities and economic imperatives change, we will see far more of them.
Leave a Reply