Thanks for the great analysis in an easier to understand cow!
Just a couple of thought which will complicate this picture and may be already under deep discussion somewhere in the literature:
In the City B / City A framework, it is possible that historically the role of City A has been the role of the suburbs of City B. Since 1920 the automobile has reshaped and expanded cities immensely. At some point, rail and commuter lines further increased the ability for cities to expand. Initially those suburban areas had less constraints and more space while offering the same access to job centers. And, as they say, as long as it is within 30 minutes it is all good. But at some point the City B suburbs run out of roads, rails, and places that can still be reasonably called convenient. I think the phenomenon you describe started impacting suburbs much the way City A changes starting in the late 80s. So this trend is not new. The spillover to city A is a function of transportation limits first creating those effects in lower cost City B suburbs (I am not convinced of the role of underwriting on what you describe but rather those price increases inviting all of the excess capital we have).
My second point is that remote working, accelerated by the pandemic, is the next technological leap in transportation. Maybe we will get teleportation or flying cars, but not likely. Remote work allows for even more radical relocation without accepting the negative effects of lowering productivity. Agglomeration just happens in the cloud. If that’s the case, every where becomes City A (or, more specifically a suburb of City B). What we need is to take advantage of those places like Dayton Ohio that have infrastructure, lax land use, and an abundance of ready supply.
I guess my bottom line argument is that land use restrictions aren’t the problem but transportation and the limit of usable city size is the fundamental problem. The answer may not be changing zoning - that could disrupt valuable real estate, overwhelm city infrastructure, and create more congestion - but slowly encouraging more of the migration away from existing places (just like we have for the last century). At some point we will need density and will force people to give up their existing spaces, but looking at the size of our country it won’t be for a long while.
What you describe about suburbs has some truth, though I think we need to make a distinction between intra- and inter-metropolitan area migration. Cities have always changed and reorganized over time, which creates some local stresses, but doesn't tend to lead to the peculiar price trends of the past few decades. It's a lot easier, over the course of a generation, to move a few miles down the road than it is to move 1,000 miles away.
The takeaway from my work is that the expensive cities aren't expensive because of agglomeration economies. They are expensive because they have exceedingly low rates of new home building. The prices aren't high because of an unusual number of productive people moving in. They are high because of the residents resisting the need to move away. Obviously, there is more pressure on prices where demand is higher, but that isn't the primary cause of price differences between the major cities.
This point has been confirmed when new housing supply declined in all cities after 2008 and eventually rents and prices increased - from St. Louis to Seattle - as a result.
Things like remote work might reduce demand for urban living, and thus reduce the impact of the urban supply problem. But, the LA metro, for instance, has been losing more than 1/2% of its population, on net, annually, since 2017, and that hasn't been enough to quell high housing costs. It would require a deep, deep depression in order to lower urban housing demand enough to counteract how horrible urban land use rules have become.
Let me reframe it a little because I think we agree on one cause - zoning. But my point is that when historically home building has occurred to expand cities, it has not happened in the original urban core but rather in concentric circles around it. Rather than rezone downtowns we have just been expanding.
But that has limits which I think we have hit in increasingly more cities.
Take my home city of Washington DC. If we had a rapid transit line or new metro line that connected Germantown to DC in 30 minutes, it would quickly alleviate some of the price issues. At some point, germantown residents would start imposing land use restrictions. Then onto the next place. But all of that is impossible because of transportation - which, in addition to zoning - makes the problem much worse.
On your data on remote working, I agree it hasn’t had a major impact yet. But, remember, it took decades for the car to have impacts. The jury is definitely still out, but I think smart policies on remote working and revitalizing the Midwest (or, hate to say it, building Freedom cities) could really help. Just like Eisenhower did with the highway system.
I agree. In many cities, until the entry level owner occupier capital market was largely shuttered in 2008, single family greenfield construction had been ample enough to make up for sclerotic infill development and anti-apartment rules. Rules that prevent traditional dense city building do create more binding limits to city expansion.
I hadn’t connected the availability of financing with the growth of the SF suburb. That’s a very interesting point. I’ve heard defenses of SFR basically say that it fills a need for people that need the traditional trappings of schools/space/yards but cant qualify for a mortgage. Great stuff - thank you!
Great series 🚀🚀
Thanks for the great analysis in an easier to understand cow!
Just a couple of thought which will complicate this picture and may be already under deep discussion somewhere in the literature:
In the City B / City A framework, it is possible that historically the role of City A has been the role of the suburbs of City B. Since 1920 the automobile has reshaped and expanded cities immensely. At some point, rail and commuter lines further increased the ability for cities to expand. Initially those suburban areas had less constraints and more space while offering the same access to job centers. And, as they say, as long as it is within 30 minutes it is all good. But at some point the City B suburbs run out of roads, rails, and places that can still be reasonably called convenient. I think the phenomenon you describe started impacting suburbs much the way City A changes starting in the late 80s. So this trend is not new. The spillover to city A is a function of transportation limits first creating those effects in lower cost City B suburbs (I am not convinced of the role of underwriting on what you describe but rather those price increases inviting all of the excess capital we have).
My second point is that remote working, accelerated by the pandemic, is the next technological leap in transportation. Maybe we will get teleportation or flying cars, but not likely. Remote work allows for even more radical relocation without accepting the negative effects of lowering productivity. Agglomeration just happens in the cloud. If that’s the case, every where becomes City A (or, more specifically a suburb of City B). What we need is to take advantage of those places like Dayton Ohio that have infrastructure, lax land use, and an abundance of ready supply.
I guess my bottom line argument is that land use restrictions aren’t the problem but transportation and the limit of usable city size is the fundamental problem. The answer may not be changing zoning - that could disrupt valuable real estate, overwhelm city infrastructure, and create more congestion - but slowly encouraging more of the migration away from existing places (just like we have for the last century). At some point we will need density and will force people to give up their existing spaces, but looking at the size of our country it won’t be for a long while.
Thanks for engaging with the material, Lawless.
What you describe about suburbs has some truth, though I think we need to make a distinction between intra- and inter-metropolitan area migration. Cities have always changed and reorganized over time, which creates some local stresses, but doesn't tend to lead to the peculiar price trends of the past few decades. It's a lot easier, over the course of a generation, to move a few miles down the road than it is to move 1,000 miles away.
The takeaway from my work is that the expensive cities aren't expensive because of agglomeration economies. They are expensive because they have exceedingly low rates of new home building. The prices aren't high because of an unusual number of productive people moving in. They are high because of the residents resisting the need to move away. Obviously, there is more pressure on prices where demand is higher, but that isn't the primary cause of price differences between the major cities.
This point has been confirmed when new housing supply declined in all cities after 2008 and eventually rents and prices increased - from St. Louis to Seattle - as a result.
Things like remote work might reduce demand for urban living, and thus reduce the impact of the urban supply problem. But, the LA metro, for instance, has been losing more than 1/2% of its population, on net, annually, since 2017, and that hasn't been enough to quell high housing costs. It would require a deep, deep depression in order to lower urban housing demand enough to counteract how horrible urban land use rules have become.
Let me reframe it a little because I think we agree on one cause - zoning. But my point is that when historically home building has occurred to expand cities, it has not happened in the original urban core but rather in concentric circles around it. Rather than rezone downtowns we have just been expanding.
But that has limits which I think we have hit in increasingly more cities.
Take my home city of Washington DC. If we had a rapid transit line or new metro line that connected Germantown to DC in 30 minutes, it would quickly alleviate some of the price issues. At some point, germantown residents would start imposing land use restrictions. Then onto the next place. But all of that is impossible because of transportation - which, in addition to zoning - makes the problem much worse.
On your data on remote working, I agree it hasn’t had a major impact yet. But, remember, it took decades for the car to have impacts. The jury is definitely still out, but I think smart policies on remote working and revitalizing the Midwest (or, hate to say it, building Freedom cities) could really help. Just like Eisenhower did with the highway system.
I agree. In many cities, until the entry level owner occupier capital market was largely shuttered in 2008, single family greenfield construction had been ample enough to make up for sclerotic infill development and anti-apartment rules. Rules that prevent traditional dense city building do create more binding limits to city expansion.
I hadn’t connected the availability of financing with the growth of the SF suburb. That’s a very interesting point. I’ve heard defenses of SFR basically say that it fills a need for people that need the traditional trappings of schools/space/yards but cant qualify for a mortgage. Great stuff - thank you!