Lu Dadao of the Chinese Academy of Sciences warns that oversized, underused HSR infrastructure has ignored basic economic logic, weakened public finances, and derailed regional planning.
1. HSR can be a lot more profitable (on an operating basis) than you think. Even the Hokuriku Shinkansen, which doesn't go to any city with more than a million people (except Tokyo), has a very good operating ratio.
2. I strongly agree that Chinese HSR needs better integration with local transit, and less grand stations. Tokyo Station's Shinkansen concourse is over an order of magnitude smaller than Chongqing East, yet serves the same amount of people as Shanghai Hongqiao.
(Tokyo station shinkansen concourse size judging from the map linked, + google maps measuring tool between Yaesu Central and Marunouchi Central exits, Tokyo station's Shinkansen concourse is some tens of thousands of square meters in size: https://www.jreast.co.jp/e/stations/e1039.html)
3. When I went to Huizhou, the HSR stations/lines there seemed to be doing fine. The trains were full and there were a nonzero amount of people at the stations. It seems like China Railways has a lot of room to raise fares, do dynamic pricing, or do price discrimination.
4. HSR New Towns betray a misunderstanding of the reason why cities and suburbs exist. People go to cities because there are jobs. Suburbs exist because people want cheaper/more living space while retaining access to jobs. Most HSR new towns don't meet the first demand, and meet the latter demand no better than normal development.
5. Japanese construction was better in the past, but nowadays, China probably does not have anything to learn from Japan; many new projects are full of tunnels and bridges. Worst of all, the new Hokuriku Shinkansen extension is *projected* to be as expensive as the HS2 in the UK and over 2x as expensive as California HSR.
That said, Japanese high-speed operations are still probably the best in the world, and their integration with local transit is also great.
6. The government can't create new jobs in the long run, it can only accelerate the recovery from depressions or move jobs between parts of the economy. It gives me some schadenfreude to see that other countries also have economically illiterate policy academics.
7. Japan's current maglev troubles are political, not technical. There's one NIMBY governor that refused to let construction go through his prefecture, delaying the thing by 7+ years
8. I have a hard time taking seriously anybody who talks about negative health effects from radio waves.
HSR covers 97% of cities with >.5m population, so with population of 6m, Huizhou is not that small among the HSR cities. The overbuilding is very real. Pricing may not be effective fix, my observation is passengers in smaller cities are more sensitive to price.
6. is an answer based on the neoliberal/quasi-libertarian economic ideology that entirely dominates the economics profession in the West. From a non-ideological, reality-based perspective, infrastructure development creates opportunities for permanent economic growth because it lowers transport costs.
Re: point 7, Japan’s maglev problems are also economic: there is little reason to believe that the new line will pay off in a future of rapidly decreasing population. Jeopardizing the viability and profitability of the broader railway system for a vanity project hatched in the high growth period is highly irrational outcome driven by political and institutional inertia. With China facing a similar demographic future to Japan, the concern about overbuilding deserves to be taken quite seriously, and it seems highly likely that in a future when financial reality reasserts itself, many of these lines will need to be written off. I touch on the maglev question in the second half of this article, but the series of articles by Philip Brasor I link to offer a more detailed critique of the project’s contradictions: https://www.samholden.jp/p/trains-from-the-past-trains-to-the
I agree with much of Mr Smith’s arguments. I sense that China builds for the future of the country, also as a hedge against coming economic uncertainty. Nothing wrong with making hay while the sun shines, especially with current surpluses. Do not discount the uplifting in spirit of citizens of China’s interiors, who must see tangible evidence that they are not left behind.
Dr. Lu Dadao has provided an extremely interesting and comprehensive view and analysis of the HSR system in his country. As a member of the initial group of scholars invited work in the PRC in 1979, I recall taking a hand shoveled coal steam train from Beijing to Hefei and across to Guangzhou during my tenure there. Dr. Lu's comments on overcapacity make sense according to conventional economics. I would differ with him on the deeper implications of high speed rail and maglev for the PRC. As a U.S. citizen I have lived with the growth of suburban automobile based transportation and the change from propeller driven aircraft to jets, each with its untoward impacts on social integration and sense of community. The PRC was founded on profound and subtle principles of social commitment, obligation to the community, and consciousness of national destiny - elements the West has long forsaken with obvious results. With time, the exercise of developing and championing the HSR and maglev systems will be vindicated and become a model for the world. The key, which Dr. Lu has overlooked is the spirit of the enterprise. Despite the clear opportunities for misusing funds or overreaching practical bounds, there is an inherent virtue to creating public spaces and transport corridors with an intent to promoting communal interchange with minimal environmental load utilizing pioneering technology which will carry forward into the next century. Dr. Lu advocates for taking a reasonable path with regard to balancing real need against provision of transportation resources and suggests restrictions and redeployment of resources accordingly. In doing so, he is inadvertently advancing the proliferation of automobile and aircraft based transportation networks, both of which erode common humanity. By creating architectural and engineering marvels, temples if you will, associated with HSR based transport, the PRC is building a vison of the future for its people, guiding them toward an integrated humane future. If one is contemplating placing restrictions on investments in public infrastructure, perhaps limiting roadways and air transport would be the most practical long term means of achieving real efficiency. Los Angeles is a case in point, with enormous amounts of human time lost in automobile traffic over the past century, and it is likely that Beijing and Shanghai are falling into the same unreasonable pit. HSR hubs, reached by ever more sophisticated local high speed transport, will be the mark of the future globally with the PRC modeling this truly remarkable system for us all.
So-called 'basic economic logic' is precisely where this assessment / commentary goes off the rails, so to speak. Public finances aren't weakened by adding assets to the balance sheet, especially long-lived assets such as this. The flow that the system has enabled is negentropy in nature, which delivers both social and economic benefits that 'basic economic logic' cannot comprehend. The network is a little over 10 years old; much of it is even younger. Talk of 'under used' is standard fare when it comes to new infrastructure, but in 20 or 30 years from now, the story is different. So often one hears folk say, "if only the infrastructure was built 20-30 years ago, we wouldn't have the problems we have today". So-called 'basic economic logic' which has no real concept of time simply does not address these multi-generational issues. If the cross-generational question is one of financing (debt to future generations etc) this reflects a failure to understand that the sovereign monetary state does not have such as constraint.
The Center for Strategic and International Studies (CSIS) is funded by NATO, the U.S. Defense Department, and major defense contractors like Lockheed Martin, Raytheon, Boeing, and General Atomics. While not officially part of the State Department, its messaging consistently aligns.
From a Chinese diplomatic perspective, CSIS is one of the most systematically anti-China institutions in the U.S. policy establishment. It portrays China as inherently aggressive, expansionist, and threatening—while ignoring Beijing’s stated intent of peaceful development and defensive posture. Its war-game simulations, political warfare accusations, and maritime “militarization” dashboards provide the intellectual scaffolding for U.S. containment strategies.
To Beijing, this is propaganda disguised as analysis—used to sell weapons, shape policy, and justify encirclement.
The lines will need maintenance and repairs. Viaducts with concrete bridges are also not that long-lasting as railways built with dams and trenches are.
Anecdotally, I lived in a tier 3 city all of 2024(Shantou) to Hong Kong and the trains were consistently 70-80 full even off peak on a weekday. This is for second class cabin. It doesn’t matter which train I take. 7am. 9am. 2pm. 4pm. I don’t travel during peak because the one time I did, it was a mad house.
Well despite the issues, I will say that I noticed a great difference between how many Chinese have travelled across China and how many Canadians have travelled across Canada. The trains have made Chinese people adventurous. In Canada, where we only have absurdly slow trains, few people have ventured north and many have never seen the opposite side of the country.
China's probably learning from the Victorians who "overbuilt" an amazing rail system in Britain, providing an arterial network for the nation. Not only did it provide transport for the mass of the population, it fed the culture and became a major part of national identity in literature and on screen.
In the 1960s the Conservatives slashed it by a third in favour of the money-men building expensive polluting motorways. Typical Tory mentality, knowing the cost of everything and the value of nothing.
What this and basically all commentary on the topic misses is that transportation systems need to be considered on the network level, not the level of the individual route. It doesn't matter if HSR in less densely populated or developed areas of China loses money or is underused, because it is likely that a lot of the demand on that route also doubled as induced demand on other routes. This whole "route first" worldview is exactly what destroyed the railroads in the US, which once counted as among the most extensive in the world. The US railroad networks as a whole never once was losing money in net, the only reason so many routes were closed was to improve operating ratios. You'll notice that people don't make this mistake with roads, because they are completely socialized to begin with.
The arguments against the station cathedral are utterly stupid. Major train station in Europe are cathedral like. However the location of stations in China are sometimes indeed far from urban cores and connections are scarce. But as European station from the 19th century, we need to wait the final system to set after couple of decades. Rome wasn't build in a day. Overall the paper is a lot a blabla with few data.
Travel over short distance happens more frequently than over larger ones. The first are performed daily or about weekly, the latter only a few times per year.
Therefore, speedy connctions between large city cores and medium sized "suburban" areas can still fetch a lot of passengers. That kind of high speed routes should, however, branch out to a couple of slower routes with more stops and thus more stations served directly from the main town, rather than force passengers to change at a HSR hub.
If the "suburb" is 50 km away, a speed of 250 km/h is not unreasonable, provided that space to construct a line at reasonable costs is available. Even the Subway in New York City has slower and express lines, and even faster trains go from the City to upstate New York, even though neither are nowhere close to being as fast as HSR in China.
I am not implying that the existing routes would not be enough.
I largely agree with the presented analysis. A four by four grid cerainly helped a lot with connectivity. The next lines to form an eight by eight grid ar labeled as a hodgepot, but probably the respective sections habe some merit in themselves.
The next doubling will probably kill the system financially.
Chinese will want to travel abroad more frequently. Flying leads to large carbon emissions. The HSR network can do more to reduce these.
For example. Urumqi airport can be transformed into a rail-air hub for travel to Central Asia and Europe. Direct fligths from Urumqi to a dozen European cities would be quite convenient for travellers from many cities, if these are connected by direct trains to Urumqi: Four connecting trains with six intermediate stops each would alredy serve 24 eastern Chinese cities and together would serve 12*24 = 288 city pair connecions. The practice today where a changeover between planes is generally required at Beijng, Guangzhou or Shanghai is no better.
Travellers from the coastal areas would reach Urumqi primarily with high speed night trains, then switch to planes where high speed rails ends (and travel time would be quite long anyhow for the full route).
Yantai and Qingdao International airports could serve the majority of flights from China to Korea and Japan, also with HSR feeders
Similarly, Kunming can serve as a hub to Thailand and India (ir is also only a HSR night train ride away from Shanghai and Beijng), while other places of changeover could be used for travel to Singapore and Viet Nam.
Shorter-distance air rides like Kunming - Bangkok will also be easier to decarbonize at some point than longer-range flights like Shanghai - Bangkok.
I agree that further building up of the HSR network would be a waste, beyond some conncetions to a couple of airports near the edge of the existing HSR network that could serve as rail - air hubs.
I have a lot of reactions.
1. HSR can be a lot more profitable (on an operating basis) than you think. Even the Hokuriku Shinkansen, which doesn't go to any city with more than a million people (except Tokyo), has a very good operating ratio.
https://toyokeizai.net/articles/-/193269
2. I strongly agree that Chinese HSR needs better integration with local transit, and less grand stations. Tokyo Station's Shinkansen concourse is over an order of magnitude smaller than Chongqing East, yet serves the same amount of people as Shanghai Hongqiao.
(Tokyo station shinkansen concourse size judging from the map linked, + google maps measuring tool between Yaesu Central and Marunouchi Central exits, Tokyo station's Shinkansen concourse is some tens of thousands of square meters in size: https://www.jreast.co.jp/e/stations/e1039.html)
3. When I went to Huizhou, the HSR stations/lines there seemed to be doing fine. The trains were full and there were a nonzero amount of people at the stations. It seems like China Railways has a lot of room to raise fares, do dynamic pricing, or do price discrimination.
4. HSR New Towns betray a misunderstanding of the reason why cities and suburbs exist. People go to cities because there are jobs. Suburbs exist because people want cheaper/more living space while retaining access to jobs. Most HSR new towns don't meet the first demand, and meet the latter demand no better than normal development.
5. Japanese construction was better in the past, but nowadays, China probably does not have anything to learn from Japan; many new projects are full of tunnels and bridges. Worst of all, the new Hokuriku Shinkansen extension is *projected* to be as expensive as the HS2 in the UK and over 2x as expensive as California HSR.
That said, Japanese high-speed operations are still probably the best in the world, and their integration with local transit is also great.
https://x.com/JRUrbaneNetwork/status/1822209790392549481
https://transitcosts.com/high-speed-rail-preliminary-data-analysis/
6. The government can't create new jobs in the long run, it can only accelerate the recovery from depressions or move jobs between parts of the economy. It gives me some schadenfreude to see that other countries also have economically illiterate policy academics.
7. Japan's current maglev troubles are political, not technical. There's one NIMBY governor that refused to let construction go through his prefecture, delaying the thing by 7+ years
8. I have a hard time taking seriously anybody who talks about negative health effects from radio waves.
point 8 is great.
HSR covers 97% of cities with >.5m population, so with population of 6m, Huizhou is not that small among the HSR cities. The overbuilding is very real. Pricing may not be effective fix, my observation is passengers in smaller cities are more sensitive to price.
Your other points are spot on.
6. is an answer based on the neoliberal/quasi-libertarian economic ideology that entirely dominates the economics profession in the West. From a non-ideological, reality-based perspective, infrastructure development creates opportunities for permanent economic growth because it lowers transport costs.
Re: point 7, Japan’s maglev problems are also economic: there is little reason to believe that the new line will pay off in a future of rapidly decreasing population. Jeopardizing the viability and profitability of the broader railway system for a vanity project hatched in the high growth period is highly irrational outcome driven by political and institutional inertia. With China facing a similar demographic future to Japan, the concern about overbuilding deserves to be taken quite seriously, and it seems highly likely that in a future when financial reality reasserts itself, many of these lines will need to be written off. I touch on the maglev question in the second half of this article, but the series of articles by Philip Brasor I link to offer a more detailed critique of the project’s contradictions: https://www.samholden.jp/p/trains-from-the-past-trains-to-the
I agree with much of Mr Smith’s arguments. I sense that China builds for the future of the country, also as a hedge against coming economic uncertainty. Nothing wrong with making hay while the sun shines, especially with current surpluses. Do not discount the uplifting in spirit of citizens of China’s interiors, who must see tangible evidence that they are not left behind.
Dr. Lu Dadao has provided an extremely interesting and comprehensive view and analysis of the HSR system in his country. As a member of the initial group of scholars invited work in the PRC in 1979, I recall taking a hand shoveled coal steam train from Beijing to Hefei and across to Guangzhou during my tenure there. Dr. Lu's comments on overcapacity make sense according to conventional economics. I would differ with him on the deeper implications of high speed rail and maglev for the PRC. As a U.S. citizen I have lived with the growth of suburban automobile based transportation and the change from propeller driven aircraft to jets, each with its untoward impacts on social integration and sense of community. The PRC was founded on profound and subtle principles of social commitment, obligation to the community, and consciousness of national destiny - elements the West has long forsaken with obvious results. With time, the exercise of developing and championing the HSR and maglev systems will be vindicated and become a model for the world. The key, which Dr. Lu has overlooked is the spirit of the enterprise. Despite the clear opportunities for misusing funds or overreaching practical bounds, there is an inherent virtue to creating public spaces and transport corridors with an intent to promoting communal interchange with minimal environmental load utilizing pioneering technology which will carry forward into the next century. Dr. Lu advocates for taking a reasonable path with regard to balancing real need against provision of transportation resources and suggests restrictions and redeployment of resources accordingly. In doing so, he is inadvertently advancing the proliferation of automobile and aircraft based transportation networks, both of which erode common humanity. By creating architectural and engineering marvels, temples if you will, associated with HSR based transport, the PRC is building a vison of the future for its people, guiding them toward an integrated humane future. If one is contemplating placing restrictions on investments in public infrastructure, perhaps limiting roadways and air transport would be the most practical long term means of achieving real efficiency. Los Angeles is a case in point, with enormous amounts of human time lost in automobile traffic over the past century, and it is likely that Beijing and Shanghai are falling into the same unreasonable pit. HSR hubs, reached by ever more sophisticated local high speed transport, will be the mark of the future globally with the PRC modeling this truly remarkable system for us all.
So-called 'basic economic logic' is precisely where this assessment / commentary goes off the rails, so to speak. Public finances aren't weakened by adding assets to the balance sheet, especially long-lived assets such as this. The flow that the system has enabled is negentropy in nature, which delivers both social and economic benefits that 'basic economic logic' cannot comprehend. The network is a little over 10 years old; much of it is even younger. Talk of 'under used' is standard fare when it comes to new infrastructure, but in 20 or 30 years from now, the story is different. So often one hears folk say, "if only the infrastructure was built 20-30 years ago, we wouldn't have the problems we have today". So-called 'basic economic logic' which has no real concept of time simply does not address these multi-generational issues. If the cross-generational question is one of financing (debt to future generations etc) this reflects a failure to understand that the sovereign monetary state does not have such as constraint.
Exactly, the author sounds as if he works for some Western think tank. An amazing amount of liberal true believer scholars in China.
You nailed it.
Pekingology is a CSIS project.
The Center for Strategic and International Studies (CSIS) is funded by NATO, the U.S. Defense Department, and major defense contractors like Lockheed Martin, Raytheon, Boeing, and General Atomics. While not officially part of the State Department, its messaging consistently aligns.
From a Chinese diplomatic perspective, CSIS is one of the most systematically anti-China institutions in the U.S. policy establishment. It portrays China as inherently aggressive, expansionist, and threatening—while ignoring Beijing’s stated intent of peaceful development and defensive posture. Its war-game simulations, political warfare accusations, and maritime “militarization” dashboards provide the intellectual scaffolding for U.S. containment strategies.
To Beijing, this is propaganda disguised as analysis—used to sell weapons, shape policy, and justify encirclement.
Pekingology and Pekingnology, distinct pubs by different people and orgs.
Good catch - I did not realize.
Makes the apparent State Department slant more intriguing.
The lines will need maintenance and repairs. Viaducts with concrete bridges are also not that long-lasting as railways built with dams and trenches are.
Anecdotally, I lived in a tier 3 city all of 2024(Shantou) to Hong Kong and the trains were consistently 70-80 full even off peak on a weekday. This is for second class cabin. It doesn’t matter which train I take. 7am. 9am. 2pm. 4pm. I don’t travel during peak because the one time I did, it was a mad house.
This is an extremely interesting read, thank you so much
My first thought was: “Sour grapes”.
China: Oops, we built too much high speed rail
Everyone else: What’s that?
as opposed to doing what? Buying USD bonds?
Build it and they will come.
Well despite the issues, I will say that I noticed a great difference between how many Chinese have travelled across China and how many Canadians have travelled across Canada. The trains have made Chinese people adventurous. In Canada, where we only have absurdly slow trains, few people have ventured north and many have never seen the opposite side of the country.
China's probably learning from the Victorians who "overbuilt" an amazing rail system in Britain, providing an arterial network for the nation. Not only did it provide transport for the mass of the population, it fed the culture and became a major part of national identity in literature and on screen.
In the 1960s the Conservatives slashed it by a third in favour of the money-men building expensive polluting motorways. Typical Tory mentality, knowing the cost of everything and the value of nothing.
https://substack.com/@annachenwrites/note/c-137095173
What this and basically all commentary on the topic misses is that transportation systems need to be considered on the network level, not the level of the individual route. It doesn't matter if HSR in less densely populated or developed areas of China loses money or is underused, because it is likely that a lot of the demand on that route also doubled as induced demand on other routes. This whole "route first" worldview is exactly what destroyed the railroads in the US, which once counted as among the most extensive in the world. The US railroad networks as a whole never once was losing money in net, the only reason so many routes were closed was to improve operating ratios. You'll notice that people don't make this mistake with roads, because they are completely socialized to begin with.
The arguments against the station cathedral are utterly stupid. Major train station in Europe are cathedral like. However the location of stations in China are sometimes indeed far from urban cores and connections are scarce. But as European station from the 19th century, we need to wait the final system to set after couple of decades. Rome wasn't build in a day. Overall the paper is a lot a blabla with few data.
Remind that a lot of freight is coal, and coal use is deemed to go down.
Travel over short distance happens more frequently than over larger ones. The first are performed daily or about weekly, the latter only a few times per year.
Therefore, speedy connctions between large city cores and medium sized "suburban" areas can still fetch a lot of passengers. That kind of high speed routes should, however, branch out to a couple of slower routes with more stops and thus more stations served directly from the main town, rather than force passengers to change at a HSR hub.
If the "suburb" is 50 km away, a speed of 250 km/h is not unreasonable, provided that space to construct a line at reasonable costs is available. Even the Subway in New York City has slower and express lines, and even faster trains go from the City to upstate New York, even though neither are nowhere close to being as fast as HSR in China.
I am not implying that the existing routes would not be enough.
I largely agree with the presented analysis. A four by four grid cerainly helped a lot with connectivity. The next lines to form an eight by eight grid ar labeled as a hodgepot, but probably the respective sections habe some merit in themselves.
The next doubling will probably kill the system financially.
Chinese will want to travel abroad more frequently. Flying leads to large carbon emissions. The HSR network can do more to reduce these.
For example. Urumqi airport can be transformed into a rail-air hub for travel to Central Asia and Europe. Direct fligths from Urumqi to a dozen European cities would be quite convenient for travellers from many cities, if these are connected by direct trains to Urumqi: Four connecting trains with six intermediate stops each would alredy serve 24 eastern Chinese cities and together would serve 12*24 = 288 city pair connecions. The practice today where a changeover between planes is generally required at Beijng, Guangzhou or Shanghai is no better.
Travellers from the coastal areas would reach Urumqi primarily with high speed night trains, then switch to planes where high speed rails ends (and travel time would be quite long anyhow for the full route).
Yantai and Qingdao International airports could serve the majority of flights from China to Korea and Japan, also with HSR feeders
Similarly, Kunming can serve as a hub to Thailand and India (ir is also only a HSR night train ride away from Shanghai and Beijng), while other places of changeover could be used for travel to Singapore and Viet Nam.
Shorter-distance air rides like Kunming - Bangkok will also be easier to decarbonize at some point than longer-range flights like Shanghai - Bangkok.
I agree that further building up of the HSR network would be a waste, beyond some conncetions to a couple of airports near the edge of the existing HSR network that could serve as rail - air hubs.