Higher density doesn't result in more congestion. Holding population and lane-miles the same, higher density reduces congestion. The average miles traveled reduces which reduces total trip-miles.
Land use restrictions can defeat this by requiring businesses, retail, and housing far away. Also, it makes everyone commute in the same direction making poor use of half the lanes.
I know this is a bit late, but I'm curious how you mesh your anarcho-capitalism with peak load pricing roads in cities? From what I have read thus far on private road provisions and understand about the incentive structures; it seems very unlikely anything but highways would employ a peak load toll strategy, as neighborhoods do not see high loads (so transaction costs are just not worth it, nor is there really a reason) and business' abutting a road would just want as much traffic as possible so there is a strong disincentive for this. I guess maybe to avoid consumer annoyance, but I'm very annoyed at Disney and they are perfectly happy keeping me in line for rides for hours; even with 'fast passes'. It seems most consumers prefer the annoyance to non-gratis provision of those types of goods...
"Location, location, location" is out? Why no mention of developers and such promoting limited access roads that connect to existing state and county roads in sparsely populated parts of a county on the fringe of the metro area? Think Tyson's Corner. Or is this considered part of "lumpiness"?
Murray in his point #1 says that high land values might reflect good planning, but that G+G have assumed high land values are a symptom of bad planning.
You say in response and without further explanation that "G+G’s method unambiguously detects the effect of regulation on the cost of individual homes".
Could you elaborate?
I think Murray is right on this point.
Here's a thought experiment.
Suppose some jurisdiction always got their planning rules just right. Suppose the rules changed through time but were totally optimal at each point in time. They were the best set of rules imaginable, always. The rules let lots of people live there and made living there GREAT.
Would people pay a lot to live there or not much at all?
They would pay a lot. Land prices would be high.
Suppose also that buildings lasted quite a while and technologies and preferences changed a bit over time.
Might there be a few over-sized backyards which were privately optimal at the time they were created, but wouldn't be created if the land was vacant today? Of course.
So with the best of all planning rules, land would be expensive, and marginal and average land prices would not be equal.
I can see G+G's argument that marginal and average prices WOULD be equal if houses were continually torn down and rebuilt. But in the real world that destroys valuable capital, so it isn't privately profitable to do it even if the rules allow it.
In the real world, cities have a history, and good planning makes it good to live somewhere which makes land expensive.
"The standard business model for developers, however, is to buy land, develop it, then sell out. They make their money by building new stuff, not by sitting on real estate and waiting for it to appreciate. As a result, developers’ support for deregulation is exactly what the mainstream view predicts. Indeed, given the ubiquity of NIMBYism, the lobbying of developers is practically the only reason anything gets built at all."
Murray has a response post now, but to emphasize this, he has already written a lot about how this is not the case. Developers seem to be in the market of hoarding land and delaying construction, until they can get the highest price for it. Much like how if you had a monopoly on oil, you wouldn't release it all at at once for pennies in year 1, you would leak it out over how ever maybe 100 years, with duration based on how discounted the future revenue is. Oil is consumable, while land is perpetual, so this is a straight inefficiency for land. Because undeveloped land can get very favorable property tax treatment, developers face little costs in holding. I hope you address this, as I'd like to know if it is true and how it interacts with other substitutes in the market.
Then the answer to why they are pro-deregulation is probably just that they favor a little deregulation - that they don't have to pay so much to local governments to upzone - but would be against major deregulation that would make their landbank investments worthless.
A couple notes. I think a much better formulation of the 'mainstream view' (which isn't mainstream in the places that really matter), is that housing regulation decreases the supply elasticity of housing. This is a much more tenable perspective, especially for a critic, since it can make the primary question a matter of assessing the magnitude of the effect: some policies have a greater constraining effect, and some have a lesser; some provide more amenities, some less. How do we find out which policies have the greatest net negative effect, then? There are a lot of places where mainstream researchers (Glaeser and Gyourko *cough *cough) forget to properly do the cost-benefit analysis, and a critic would be very helpful.
I think Murray has a valid point in arguing that economic theories suggesting that there is a long-term equilibrium for construction & land prices is false. You hinted at this, but there are a lot of factors that go into determining housing prices -location, for example- that have little to do with the base costs of land or construction (though it's a hard thing to study properly, since so many of the variables are dependent on one another).
I'm also confused as to where Murray gets the quote in the first point of your third numbered list, where he says that G&G claim "town planning regulations do not generate ‘any benefits that manifest themselves in a higher land price.'" I know Gyourko has worked on papers that have established the value of some of the amenities that planning regulations provide, so this seems like a false claim to me. Maybe just outdated? Maybe I'm not understanding it properly?
I just did a calculation and found that the building itself costs about double due to zoning. The land cost is much higher as well, but it's harder for me to discern how that gets calculated. Here's my work on the building cost differential.
To build 100 500sq ft average units without zoning(assuming free land) it would cost five thousand for design, 3 million for labor and materials(it has been a realistic number within the last decade), 250K for connections, 100K for interest and fees, 100K to equity partner, 50K to developer, 3.55M plus land. Imagine it's in a magical non-zoned plot in the middle of a developed area. 3.5M or 35K/unit or $70/sq ft. Architect, developer and equity partner all get paid less than usual here because there's no zoning to risk the project timeline, which can be 6 months or less. Now the remainder is the land cost. In some rural area with unstable demand it might be as little as 200K which doesn't change things very much. If condo units are selling for more than 40K or rents are more than 400/mo this would possibly get built.
To build the same 100 500sq ft average units with zoning it would cost 5% for design and project management, 5 million for labor and materials(assume many building and zoning adjustments required plus moderately more expensive labor and materials due to more expensive local housing), 7%+ for interest and fees/yr, 10% to equity partner, 5% to developer, plus land and connections of utilities and roads. That comes to 7M or 70K/unit or $140/sq ft before land cost. Timeline is unknown, in some places unknowable, but figure 18+ months. Now the remainder is the land cost.
So, if zoning does what it normally does and makes housing dramatically more expensive, it will increase land cost as well as building cost, as far as the market can bear.
So, without the land, rents need to be 800 and condos need to sell for 80K. The way it seems to work in my experience is that land will be priced to reflect real all-in building costs, since if it's profitable at prices below that, someone is probably doing it. In a 100 unit project, for every 100/mo or 10K/unit increase the land goes up about a million dollars. If rents are 1500 a month the land is 7M.
Next week: “Land-use restrictions in ancient Mesopotamia.”
Higher density doesn't result in more congestion. Holding population and lane-miles the same, higher density reduces congestion. The average miles traveled reduces which reduces total trip-miles.
Land use restrictions can defeat this by requiring businesses, retail, and housing far away. Also, it makes everyone commute in the same direction making poor use of half the lanes.
I know this is a bit late, but I'm curious how you mesh your anarcho-capitalism with peak load pricing roads in cities? From what I have read thus far on private road provisions and understand about the incentive structures; it seems very unlikely anything but highways would employ a peak load toll strategy, as neighborhoods do not see high loads (so transaction costs are just not worth it, nor is there really a reason) and business' abutting a road would just want as much traffic as possible so there is a strong disincentive for this. I guess maybe to avoid consumer annoyance, but I'm very annoyed at Disney and they are perfectly happy keeping me in line for rides for hours; even with 'fast passes'. It seems most consumers prefer the annoyance to non-gratis provision of those types of goods...
"Location, location, location" is out? Why no mention of developers and such promoting limited access roads that connect to existing state and county roads in sparsely populated parts of a county on the fringe of the metro area? Think Tyson's Corner. Or is this considered part of "lumpiness"?
I find one part of your critique unconvincing.
Murray in his point #1 says that high land values might reflect good planning, but that G+G have assumed high land values are a symptom of bad planning.
You say in response and without further explanation that "G+G’s method unambiguously detects the effect of regulation on the cost of individual homes".
Could you elaborate?
I think Murray is right on this point.
Here's a thought experiment.
Suppose some jurisdiction always got their planning rules just right. Suppose the rules changed through time but were totally optimal at each point in time. They were the best set of rules imaginable, always. The rules let lots of people live there and made living there GREAT.
Would people pay a lot to live there or not much at all?
They would pay a lot. Land prices would be high.
Suppose also that buildings lasted quite a while and technologies and preferences changed a bit over time.
Might there be a few over-sized backyards which were privately optimal at the time they were created, but wouldn't be created if the land was vacant today? Of course.
So with the best of all planning rules, land would be expensive, and marginal and average land prices would not be equal.
I can see G+G's argument that marginal and average prices WOULD be equal if houses were continually torn down and rebuilt. But in the real world that destroys valuable capital, so it isn't privately profitable to do it even if the rules allow it.
In the real world, cities have a history, and good planning makes it good to live somewhere which makes land expensive.
"The standard business model for developers, however, is to buy land, develop it, then sell out. They make their money by building new stuff, not by sitting on real estate and waiting for it to appreciate. As a result, developers’ support for deregulation is exactly what the mainstream view predicts. Indeed, given the ubiquity of NIMBYism, the lobbying of developers is practically the only reason anything gets built at all."
Murray has a response post now, but to emphasize this, he has already written a lot about how this is not the case. Developers seem to be in the market of hoarding land and delaying construction, until they can get the highest price for it. Much like how if you had a monopoly on oil, you wouldn't release it all at at once for pennies in year 1, you would leak it out over how ever maybe 100 years, with duration based on how discounted the future revenue is. Oil is consumable, while land is perpetual, so this is a straight inefficiency for land. Because undeveloped land can get very favorable property tax treatment, developers face little costs in holding. I hope you address this, as I'd like to know if it is true and how it interacts with other substitutes in the market.
Then the answer to why they are pro-deregulation is probably just that they favor a little deregulation - that they don't have to pay so much to local governments to upzone - but would be against major deregulation that would make their landbank investments worthless.
Nice piece!
A couple notes. I think a much better formulation of the 'mainstream view' (which isn't mainstream in the places that really matter), is that housing regulation decreases the supply elasticity of housing. This is a much more tenable perspective, especially for a critic, since it can make the primary question a matter of assessing the magnitude of the effect: some policies have a greater constraining effect, and some have a lesser; some provide more amenities, some less. How do we find out which policies have the greatest net negative effect, then? There are a lot of places where mainstream researchers (Glaeser and Gyourko *cough *cough) forget to properly do the cost-benefit analysis, and a critic would be very helpful.
I think Murray has a valid point in arguing that economic theories suggesting that there is a long-term equilibrium for construction & land prices is false. You hinted at this, but there are a lot of factors that go into determining housing prices -location, for example- that have little to do with the base costs of land or construction (though it's a hard thing to study properly, since so many of the variables are dependent on one another).
I'm also confused as to where Murray gets the quote in the first point of your third numbered list, where he says that G&G claim "town planning regulations do not generate ‘any benefits that manifest themselves in a higher land price.'" I know Gyourko has worked on papers that have established the value of some of the amenities that planning regulations provide, so this seems like a false claim to me. Maybe just outdated? Maybe I'm not understanding it properly?
I just did a calculation and found that the building itself costs about double due to zoning. The land cost is much higher as well, but it's harder for me to discern how that gets calculated. Here's my work on the building cost differential.
To build 100 500sq ft average units without zoning(assuming free land) it would cost five thousand for design, 3 million for labor and materials(it has been a realistic number within the last decade), 250K for connections, 100K for interest and fees, 100K to equity partner, 50K to developer, 3.55M plus land. Imagine it's in a magical non-zoned plot in the middle of a developed area. 3.5M or 35K/unit or $70/sq ft. Architect, developer and equity partner all get paid less than usual here because there's no zoning to risk the project timeline, which can be 6 months or less. Now the remainder is the land cost. In some rural area with unstable demand it might be as little as 200K which doesn't change things very much. If condo units are selling for more than 40K or rents are more than 400/mo this would possibly get built.
To build the same 100 500sq ft average units with zoning it would cost 5% for design and project management, 5 million for labor and materials(assume many building and zoning adjustments required plus moderately more expensive labor and materials due to more expensive local housing), 7%+ for interest and fees/yr, 10% to equity partner, 5% to developer, plus land and connections of utilities and roads. That comes to 7M or 70K/unit or $140/sq ft before land cost. Timeline is unknown, in some places unknowable, but figure 18+ months. Now the remainder is the land cost.
So, if zoning does what it normally does and makes housing dramatically more expensive, it will increase land cost as well as building cost, as far as the market can bear.
So, without the land, rents need to be 800 and condos need to sell for 80K. The way it seems to work in my experience is that land will be priced to reflect real all-in building costs, since if it's profitable at prices below that, someone is probably doing it. In a 100 unit project, for every 100/mo or 10K/unit increase the land goes up about a million dollars. If rents are 1500 a month the land is 7M.
https://manifold.markets/SG/will-bryan-caplan-join-manifold-and Got a feeling you will geek out on this new tool Manifold Markets. Great blog!! Looking forward to reading your book!!
The best part is, he can instantly make about a 30% profit if he bets all his starting M$ on that market before his first post.