6 ms·

For China, you'd expect that to skew the data in the other direction, though. Since China has over 4x the population of the US, the number of metro lines is app

by twiss 2y ago

For China, you'd expect that to skew the data in the other direction, though. Since China has over 4x the population of the US, the number of metro lines is apparently over 4x higher (since China has more metro lines per capita than the US, according to the chart).

paxys 2y ago
China has a decreasing population and is just building a ridiculous amount of everything. It is however now running into the opposite problem – entire cities and transit systems are sitting empty. A lot of development over there is happening for development's sake without trying to reasonably match actual demand.
- twiss 2y ago
  > China has a decreasing population Not in the period shown in the chart, which seems to end in 2021.
- maxglute 2y ago
  >without trying to reasonably match actual demand It's reasonably matching short/medium term demand. Actual demand is PRC increasing urbanization from current 65% to 80/90% in next 10/20 years = need infra+housing for 200m-300m people. Utilization rate of HSR increasing every year. Current excess housing sqft is enough for ~100m (persons not household), ~5-10 years of urbanization headroom. This is without accounting for depreciated housing stock, QoL upgrades. For reference 80% urbanized US is short on ~6M homes, that's 2% relative to population. PRC gunning for 80% urbanization = PRC is net short 7-10%. Realistically much higher since a lot of people in rural stuck in very shitty housing stock. Of course there's some misallocation when building so much, but few under (yet to be) occupied developements / ghost cities or 50-60m empty units reasonbly matches demand/what can be absorbed over 5-10 year horizon. Population decrease doesn't factor much until after 2050/60s, by which time need to start factoring deprecated housing stock / end of life for concrete buildings. Other "demand" consideration is demographics aging out blue collar workers, might as well as build now while labour is plentiful and cheap, better to leave future with too much built capita than too little, as many developed economies with insane building costs are finding out.
  - seanmcdirmid 2y ago
    There is just no way farmers from the villages of China are going to be able to afford the million dollar apartments speculators are buying up anticipating this continued movement from rural to urban areas. Add to that a still inflexible hukou system where even if you buy the apartment, you still can’t put your kids in local schools, it makes me think that the cities are still in for a hard time. Many buildings are never going to be actually lived in before being torn down. The construction techniques they use to soak up unskilled labor results in concrete buildings that will only last 20-30 years.
    - maxglute 2y ago
      Farmers (really the poor/underclass) directed o T3/4+ surplus that aren't worth $$$, where RE price declining. Also see central gov recently spending 300B rmb (imo just start) to transition vacancies into social housing. T1/T2 vacancies for more affluent/educated gen. Hukou likely self correcting problem (they'll keep reforming), lots of education surplus resources =going to free up YoY relative to trend of decline in past births. It's a big challenge - but IMO it's a easier problem to fill housing when you already have it then try to build enough housing when you don't/can't. Broader urban policy - central gov trying to distribute growth, urbanize/move ppl to less developed interior 2 tier cities with higher % vacancies with new industries. Big coastal hubs likely has soft upper bound in population (look at post 2014 population stalling for big T1s), I'd expect them to slowly settle 10% higher where they are by soaking up surplus to leave vacancies at healthy range, higher than 5% in hot western markets, lower than 15-20% in PRC. It's definitely a hard problem, but it's a problem where most of the physical pieces are are there. >The construction techniques they use to soak up unskilled labor results in concrete buildings that will only last 20-30 years. Unskilled gets trained by skilled, and do not remain unskilled. Bulk of building stock was built in last 15-20 years by migrant construction lifers already with 10+ years experience (entire demo skews old) with a lot of concrete pouring and steel welding experience. You don't build as much as PRC and have unskilled construction workers, half the world is built by diasphora migrant workers drawn from even less developed countries. TLDR on China collapse meme narrative of PRC construction, for what it's worth my previous field. The chinese buildings have 30-40 years life spans (no one credible guestimates 20) narrative conflates PRC construction standards in last 20 years, where vast majority of new building stock was accumuluated, and is fine for 50-75+ years (as in the concrete structure is durable), with not affluent at the time PRC choosing to cheapen out on building systems that can be easily retrofit up to modern standards. TLDR is concrete/structure/frame is FINE, "short" lifespan is really about shittier/lower quality mechanical systems, not ecological, expensive to maintain etc, and until recently because PRC so good at building fast, the answer is just to tear down and rebuild. Now PRC richer, can still choose to rebuild or retrofit (both of which they're still good at), new HVAC, facade etc, to meet performance standards now that vastly richer. This got twisted into PRC concrete last 30 years. Quality of towers being put up in CAN are comparable (and frequently worse) as stuff I see in PRC where the entire process of erecting hires has been optimized to death. There are some legit shit tier construction pre 90s (and much older) builds, i.e. all the gross mid rise you see in left behind township/village level admin divisions, exactly the kind of bad/poverty building stock reaching end of life that urbanization drive is designed to move people out of (stuff that's still all over SKR/TW). But even then, they're not collapsing out of nowhere in any numbers that actually matter, other than in event of natural disasters (i.e. Sichuan earthquake scandal), which is a real consideration, in some geographic areas. But most of building stock, built in last 20 years after - again we're really talking about the frame/structure - are fine for 50/60/70+ years like west. Also important to note vacant towers mostly not abandoned projects, there's still basic maintenance happening because they're on someone's portfolio. They're not sitting derelict for 10+ of years getting killed prematurely by elements -> they're not going to come falling down. Yes some will, but rounding error. Regardless, IMO broader consideration is RE surplus was 1-2% of GDP for ~15 years where boom was happening. Even if they sit idle/unproductive indefinitely, you're looking at in aggregate ~1 quarter of misallocation at current GDP over ~60 quarters. That's not too bad of inefficiency relative to how much was built, and inefficiency of too much > too little, especially when any solution to urban housing / infra congestion starts with having enough.
- gruez 2y ago
  Nah, they're slowly filling up: https://www.bloomberg.com/news/features/2021-09-01/chinese-ghost-cities-2021-binhai-zhengdong-new-districts-fill-up https://www.bloomberg.com/news/features/2021-09-01/chinese-g...
darth_avocado 2y ago
China has 50% more metro length than next 10 countries combined: https://medium.com/@gravitonaiconsulting/china-has-50-more-metro-length-than-the-next-10-countries-combined-3e8ebc8e9c86 https://medium.com/@gravitonaiconsulting/china-has-50-more-m...

by twiss 2y ago

paxys 2y ago

China has a decreasing population and is just building a ridiculous amount of everything. It is however now running into the opposite problem – entire cities and transit systems are sitting empty. A lot of development over there is happening for development's sake without trying to reasonably match actual demand.

twiss 2y ago
> China has a decreasing population Not in the period shown in the chart, which seems to end in 2021.
maxglute 2y ago
>without trying to reasonably match actual demand It's reasonably matching short/medium term demand. Actual demand is PRC increasing urbanization from current 65% to 80/90% in next 10/20 years = need infra+housing for 200m-300m people. Utilization rate of HSR increasing every year. Current excess housing sqft is enough for ~100m (persons not household), ~5-10 years of urbanization headroom. This is without accounting for depreciated housing stock, QoL upgrades. For reference 80% urbanized US is short on ~6M homes, that's 2% relative to population. PRC gunning for 80% urbanization = PRC is net short 7-10%. Realistically much higher since a lot of people in rural stuck in very shitty housing stock. Of course there's some misallocation when building so much, but few under (yet to be) occupied developements / ghost cities or 50-60m empty units reasonbly matches demand/what can be absorbed over 5-10 year horizon. Population decrease doesn't factor much until after 2050/60s, by which time need to start factoring deprecated housing stock / end of life for concrete buildings. Other "demand" consideration is demographics aging out blue collar workers, might as well as build now while labour is plentiful and cheap, better to leave future with too much built capita than too little, as many developed economies with insane building costs are finding out.
- seanmcdirmid 2y ago
  There is just no way farmers from the villages of China are going to be able to afford the million dollar apartments speculators are buying up anticipating this continued movement from rural to urban areas. Add to that a still inflexible hukou system where even if you buy the apartment, you still can’t put your kids in local schools, it makes me think that the cities are still in for a hard time. Many buildings are never going to be actually lived in before being torn down. The construction techniques they use to soak up unskilled labor results in concrete buildings that will only last 20-30 years.
  - maxglute 2y ago
    Farmers (really the poor/underclass) directed o T3/4+ surplus that aren't worth $$$, where RE price declining. Also see central gov recently spending 300B rmb (imo just start) to transition vacancies into social housing. T1/T2 vacancies for more affluent/educated gen. Hukou likely self correcting problem (they'll keep reforming), lots of education surplus resources =going to free up YoY relative to trend of decline in past births. It's a big challenge - but IMO it's a easier problem to fill housing when you already have it then try to build enough housing when you don't/can't. Broader urban policy - central gov trying to distribute growth, urbanize/move ppl to less developed interior 2 tier cities with higher % vacancies with new industries. Big coastal hubs likely has soft upper bound in population (look at post 2014 population stalling for big T1s), I'd expect them to slowly settle 10% higher where they are by soaking up surplus to leave vacancies at healthy range, higher than 5% in hot western markets, lower than 15-20% in PRC. It's definitely a hard problem, but it's a problem where most of the physical pieces are are there. >The construction techniques they use to soak up unskilled labor results in concrete buildings that will only last 20-30 years. Unskilled gets trained by skilled, and do not remain unskilled. Bulk of building stock was built in last 15-20 years by migrant construction lifers already with 10+ years experience (entire demo skews old) with a lot of concrete pouring and steel welding experience. You don't build as much as PRC and have unskilled construction workers, half the world is built by diasphora migrant workers drawn from even less developed countries. TLDR on China collapse meme narrative of PRC construction, for what it's worth my previous field. The chinese buildings have 30-40 years life spans (no one credible guestimates 20) narrative conflates PRC construction standards in last 20 years, where vast majority of new building stock was accumuluated, and is fine for 50-75+ years (as in the concrete structure is durable), with not affluent at the time PRC choosing to cheapen out on building systems that can be easily retrofit up to modern standards. TLDR is concrete/structure/frame is FINE, "short" lifespan is really about shittier/lower quality mechanical systems, not ecological, expensive to maintain etc, and until recently because PRC so good at building fast, the answer is just to tear down and rebuild. Now PRC richer, can still choose to rebuild or retrofit (both of which they're still good at), new HVAC, facade etc, to meet performance standards now that vastly richer. This got twisted into PRC concrete last 30 years. Quality of towers being put up in CAN are comparable (and frequently worse) as stuff I see in PRC where the entire process of erecting hires has been optimized to death. There are some legit shit tier construction pre 90s (and much older) builds, i.e. all the gross mid rise you see in left behind township/village level admin divisions, exactly the kind of bad/poverty building stock reaching end of life that urbanization drive is designed to move people out of (stuff that's still all over SKR/TW). But even then, they're not collapsing out of nowhere in any numbers that actually matter, other than in event of natural disasters (i.e. Sichuan earthquake scandal), which is a real consideration, in some geographic areas. But most of building stock, built in last 20 years after - again we're really talking about the frame/structure - are fine for 50/60/70+ years like west. Also important to note vacant towers mostly not abandoned projects, there's still basic maintenance happening because they're on someone's portfolio. They're not sitting derelict for 10+ of years getting killed prematurely by elements -> they're not going to come falling down. Yes some will, but rounding error. Regardless, IMO broader consideration is RE surplus was 1-2% of GDP for ~15 years where boom was happening. Even if they sit idle/unproductive indefinitely, you're looking at in aggregate ~1 quarter of misallocation at current GDP over ~60 quarters. That's not too bad of inefficiency relative to how much was built, and inefficiency of too much > too little, especially when any solution to urban housing / infra congestion starts with having enough.
gruez 2y ago
Nah, they're slowly filling up: https://www.bloomberg.com/news/features/2021-09-01/chinese-ghost-cities-2021-binhai-zhengdong-new-districts-fill-up https://www.bloomberg.com/news/features/2021-09-01/chinese-g...

darth_avocado 2y ago

China has 50% more metro length than next 10 countries combined: https://medium.com/@gravitonaiconsulting/china-has-50-more-metro-length-than-the-next-10-countries-combined-3e8ebc8e9c86 https://medium.com/@gravitonaiconsulting/china-has-50-more-m...